# Full Sample (N = 292)
remir_all <- read_csv("/Users/sarahcoffin/Desktop/Blueprints/FredCode/REMIR1_REMIR2_all.csv")## New names:
## Rows: 293 Columns: 263
## ── Column specification
## ──────────────────────────────────────────────────────── Delimiter: "," chr
## (143): Timestamp.Subgroups, Reviewer.Subgroups, Program.Name, Full.Citat... dbl
## (90): Program.ID, Study.Number, Study.ID, Citation.ID, Round, Citation.... lgl
## (30): Which.group.contrasts.for.LOCATION.were.examined..Check.all.that....
## ℹ Use `spec()` to retrieve the full column specification for this data. ℹ
## Specify the column types or set `show_col_types = FALSE` to quiet this message.
## • `Provide.the.key.example.supporting.your.decision..including.page.number...1`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...20`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...2`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...25`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...3`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...29`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...4`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...33`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...5`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...38`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...6`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...43`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...7`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...48`
## • `Which.of.the.groups.were.the.models.estimated.for..Check.all.that.apply...1`
## -> `Which.of.the.groups.were.the.models.estimated.for..Check.all.that.apply`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...8`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...52`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...9`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...56`
## • `Which.of.the.groups.showed.significant.beneficial.effects..Any.significant.effect.is.sufficient.to.demonstrate.a.subgroup.effect..That.is..if.any.single.subgroup.effect.is.significant..that.group.gets.coded.affirmatively..Check.all.that.apply...1`
## ->
## `Which.of.the.groups.showed.significant.beneficial.effects..Any.significant.effect.is.sufficient.to.demonstrate.a.subgroup.effect..That.is..if.any.single.subgroup.effect.is.significant..that.group.gets.coded.affirmatively..Check.all.that.apply`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...10`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...60`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...11`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...64`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...12`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...68`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...13`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...72`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...14`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...78`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...15`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...81`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...16`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...86`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...17`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...89`
## • `Were.any.group.contrasts.for.GENDER.GENDER.IDENTITY.examined...1` ->
## `Were.any.group.contrasts.for.GENDER.GENDER.IDENTITY.examined`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...18`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...94`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...19`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...97`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...20`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...102`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...21`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...105`
## • `Were.any.group.contrasts.for.SES.examined..Proxies.for.low.income.status.and.or.families.living.in.poverty..i.e...economic.disadvantage..include..FRL..Medicaid..Pell.Eligible..CHIP..etc..Do.not.rely.on.income.distribution...1`
## ->
## `Were.any.group.contrasts.for.SES.examined..Proxies.for.low.income.status.and.or.families.living.in.poverty..i.e...economic.disadvantage..include..FRL..Medicaid..Pell.Eligible..CHIP..etc..Do.not.rely.on.income.distribution`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...22`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...110`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...23`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...113`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...24`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...118`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...25`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...121`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...26`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...126`
## • `Provide.the.key.example.supporting.your.decision..including.page.number...27`
## ->
## `Provide.the.key.example.supporting.your.decision..including.page.number...129`#Check length of dataset
dim(remir_all)## [1] 293 263#Check the unique number of citations in the dataset
length(unique(remir_all$Citation.ID))## [1] 292# Clean Data
## One citation, number 4258 is listed twice, once for each of two programs
remir_all %>% filter(Citation.ID == 4258) %>% select(Study.Number, Program.Name)## To avoid counting the same citation twice, drop one for Project Personality
remir_all$drop <- ifelse(remir_all$Citation.ID == 4258 & remir_all$Study.Number == 2, 1,0)
tabyl(remir_all$drop)remir_all <- remir_all %>% filter(drop == 0)
dim(remir_all)## [1] 292 264## Look at main effect
tabyl(remir_all$main_effect)## Look at method
tabyl(remir_all$method)################## SELECT US SAMPLE ###################
tabyl(remir_all$country)remir_us <- remir_all %>% filter(country == "USA")
dim(remir_us)## [1] 240 264## Rename Main effect
remir_us$main_effect <- remir_us$REGSSLN
tabyl(remir_us$main_effect)# Create subgroup dataset (N = 100)
remir_us_subgroup <- remir_all %>% filter(country == "USA" & ! is.na(method))
dim(remir_us_subgroup)## [1] 100 264############ RACE #############
## What Race Interaction Were Examined #################
## ## Change Interaction Examined Variable Names
names(remir_all)[13] <- "inter_examine1_race"
names(remir_all)[76] <- "inter_examine2_race"
table(remir_all$inter_examine1_race, useNA = "always")##
## Black/Not Black
## 2
## Minority/majority
## 2
## Minority/Majority
## 2
## W/B (White/African American or Black)
## 2
## W/B (White/African American or Black);W/As (White/Asian or Asian American);W/PI (White/Native Hawaiian or Pacific Islander);W/NA (White/Native American or American Indian or Native Alaskan);B/As (Black or African American/Asian or Asian American);B/PI (Black or African American/Native Hawaiian or Pacific Islander);B/NA (Black or African American/Native American or American Indian or Native Alaskan);As/NA (Asian or Asian American/Native American or American Indian or Native Alaskan);PI/NA (Native Hawaiian or Pacific Islander/Native American or American Indian or Native Alaskan)
## 1
## W/B (White/African American or Black);W/As (White/Asian or Asian American);W/PI (White/Native Hawaiian or Pacific Islander);W/NA (White/Native American or American Indian or Native Alaskan);B/As (Black or African American/Asian or Asian American);B/PI (Black or African American/Native Hawaiian or Pacific Islander);B/NA (Black or African American/Native American or American Indian or Native Alaskan);As/NA (Asian or Asian American/Native American or American Indian or Native Alaskan);PI/NA (Native Hawaiian or Pacific Islander/Native American or American Indian or Native Alaskan);White/Multiracial; Black/Multiracial; Asian or Pacific Islander/Multiracial; Native American/Multiracial
## 1
## W/B (White/African American or Black);White/Other or Multiracial
## 2
## White/Minority
## 4
## White/Non-White
## 2
## White/Not White
## 1
## <NA>
## 273table(remir_all$inter_examine2_race, useNA = "always")##
## Black/Not Black
## 5
## Minority/Majority
## 5
## None
## 1
## W/As (White/Asian or Asian American);White/Minority
## 1
## W/B (White/African American or Black)
## 6
## W/B (White/African American or Black);White/Multiracial; Black/Multiracial
## 1
## White + Another Race/Black
## 3
## White/Other
## 1
## <NA>
## 269## Combine Two Tnteractions Examined
remir_all$inter_examine12_race <- ifelse(is.na(remir_all$inter_examine1_race),
remir_all$inter_examine2_race, remir_all$inter_examine1_race)
tabyl(remir_all$inter_examine12_race)############## What Race Interactions Were Significant ################
## Change Interaction Significant Variable Names
names(remir_all)[14] <- "inter_signif1_race"
names(remir_all)[77] <- "inter_signif2_race"
table(remir_all$inter_signif1_race, useNA = "always")##
## None
## 14
## W/B (White/African American or Black)
## 1
## W/B (White/African American or Black);White/Other or Multiracial
## 1
## White/Minority
## 3
## <NA>
## 273table(remir_all$inter_signif2_race, useNA = "always")##
## Black/Not Black
## 3
## Minority/Majority
## 2
## None
## 14
## W/As (White/Asian or Asian American);White/Minority
## 1
## W/B (White/African American or Black)
## 1
## W/B (White/African American or Black);White/Multiracial; Black/Multiracial
## 1
## White + Another Race/Black
## 1
## <NA>
## 269## Combine Two Tnteractions Significant
remir_all$inter_signif12_race <- ifelse(is.na(remir_all$inter_signif1_race),
remir_all$inter_signif2_race, remir_all$inter_signif1_race)
tabyl(remir_all$inter_signif12_race)############## What Race Interaction Groups Benefitted More ##############
## Change Interaction Benefitted Variable Names
names(remir_all)[16] <- "inter_benefit1_race"
names(remir_all)[79] <- "inter_benefit2_race"
table(remir_all$inter_benefit1_race, useNA = "always")##
## Black or African American;Other or Multiracial
## 1
## Minority
## 3
## None
## 14
## White
## 1
## <NA>
## 273table(remir_all$inter_benefit2_race, useNA = "always")##
## Asian or Asian American;Minority Black or African American
## 1 2
## Black or African American;White Majority
## 1 2
## None Not Black
## 14 2
## White + Another Race <NA>
## 1 269## Combine Two Tnteractions Benefitted
remir_all$inter_benefit12_race <- ifelse(is.na(remir_all$inter_benefit1_race),
remir_all$inter_benefit2_race, remir_all$inter_benefit1_race)
tabyl(remir_all$inter_benefit12_race)############# What Race Subgroup Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[47] <- "subg_examine1_race"
names(remir_all)[80] <- "subg_examine2_race"
table(remir_all$subg_examine1_race, useNA = "always")##
## Asian or Asian American;Black or African American
## 1
## Black or African American
## 3
## Black or African American;White
## 3
## White;Non-White
## 2
## <NA>
## 283table(remir_all$subg_examine2_race, useNA = "always")##
## Asian or Asian American;Black or African American;White
## 1
## Asian or Asian American;White;Minority
## 1
## Black or African American
## 3
## Black or African American;Not Black
## 1
## Black or African American;White
## 4
## Black or African American;White + Another Race
## 3
## Black or African American;White;Multiracial
## 1
## Minority
## 1
## Minority; Majority
## 2
## None
## 5
## White
## 1
## <NA>
## 269## Combine Two Subgroups Examined
remir_all$subg_examine12_race <- ifelse(is.na(remir_all$subg_examine1_race),
remir_all$subg_examine2_race, remir_all$subg_examine1_race)
tabyl(remir_all$subg_examine12_race)############## What Race Subgroups Were Significant ######################
## Change Subgroup Significant Variable Names
names(remir_all)[49] <- "subg_signif1_race"
names(remir_all)[82] <- "subg_signif2_race"
table(remir_all$subg_signif1_race, useNA = "always")##
## Asian or Asian American;Black or African American
## 1
## Black or African American
## 5
## None
## 1
## White;Non-White
## 2
## <NA>
## 283table(remir_all$subg_signif2_race, useNA = "always")##
## Asian or Asian American;Minority
## 1
## Black or African American
## 5
## Black or African American;Not Black
## 1
## Black or African American;White
## 3
## Black or African American;White + Another Race
## 3
## Black or African American;White;Multiracial
## 1
## Minority
## 1
## Minority; Majority
## 2
## None
## 5
## White
## 1
## <NA>
## 269## Combine Two Subgroup Significant
remir_all$subg_signif12_race <- ifelse(is.na(remir_all$subg_signif1_race),
remir_all$subg_signif2_race, remir_all$subg_signif1_race)
tabyl(remir_all$subg_signif12_race)## INTERACTION RACE EXAMINED
tabyl(remir_all$inter_examine12_race)## Simplify Codes
## Define as mixed if combined Black, Asian, etc into broad category
## Define as exact if referred specifically to Black, Asian, etc
## None = NA, as it means that race subgroups not examined
## Recode to Mixed and None
remir_all$inter_examine12_race_cat <-
car::recode(remir_all$inter_examine12_race, "
'Minority/Majority' = 'Mixed';
'Minority/majority' = 'Mixed';
'White/Minority' = 'Mixed';
'White/Not White' = 'Mixed';
'White/Non-White' = 'Mixed';
'White/Other' = 'Mixed';
'None' = NA ")
## Code remaining to Exact
remir_all$inter_examine12_race_cat <-
ifelse(remir_all$inter_examine12_race_cat != "Mixed", "Exact",
remir_all$inter_examine12_race_cat)
## Check -- Looks OK
tabyl(remir_all$inter_examine12_race_cat)tabyl(remir_all, inter_examine12_race, inter_examine12_race_cat)## SUBGROUP RACE EXAMINED
tabyl(remir_all$subg_examine12_race)## Simplify codes
## Define as mixed if combined Black, Asian, etcs into broad category
## Define as exact if referred specifically to Black, Asian, etc
## None = NA, as it means that race subgroups not examined
remir_all$subg_examine12_race_cat <-
ifelse(remir_all$subg_examine12_race == "White;Non-White" |
remir_all$subg_examine12_race == "Minority" |
remir_all$subg_examine12_race == "Minority; Majority", "Mixed",
ifelse(remir_all$subg_examine12_race == "None", NA,"Exact"))
## Check -- Looks OK
tabyl(remir_all, subg_examine12_race, subg_examine12_race_cat)tabyl(remir_all$subg_examine12_race_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE -- define over .75 as Homogeneous
tabyl(remir_all$inter_examine12_race_cat)remir_all$inter_examine12_race_cat <- ifelse(
(remir_all$black_all > .75 |
remir_all$asian_all > .75 |
remir_all$native_all > .75 |
remir_all$pacif_all > .75 ) &
is.na(remir_all$inter_examine12_race_cat) &
is.na(remir_all$subg_examine12_race_cat),
"Homog", remir_all$inter_examine12_race_cat)
tabyl(remir_all$inter_examine12_race_cat)## Check -- Mixed and Exact unchanged but many NAs now homogeneous
remir_all %>% filter(remir_all$inter_examine12_race_cat == 'Homog') %>%
select(black_all, asian_all, native_all, pacif_all, white_all) %>%
print(n=26)## # A tibble: 26 × 5
## black_all asian_all native_all pacif_all white_all
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 0.8 0 0 0 0
## 2 0.8 0 0 0 0
## 3 0.87 0 0 0 0
## 4 0.84 0 0 0 0
## 5 0.9 0.01 0 0 0.07
## 6 0.83 0 0 0 0.01
## 7 0.83 0 0 0 0.01
## 8 0.85 0.05 0 0 0.02
## 9 0.95 0 0 0 0
## 10 1 0 0 0 0
## 11 1 0 0 0 0
## 12 1 0 0 0 0
## 13 1 0 0 0 0
## 14 1 0 0 0 0
## 15 0.85 0 0 0 0
## 16 0.85 0 0 0 0
## 17 0.85 0 0 0 0
## 18 0.82 0 0 0 0.15
## 19 0.82 0 0 0 0.15
## 20 0.8 0 0 0 0
## 21 0.8 0 0 0 0
## 22 1 0 0 0 0
## 23 1 0 0 0 0
## 24 1 0 0 0 0
## 25 1 0 0 0 0
## 26 0.87 0 0 0 0## All Homogeneous samples are African American
## INTERACTION RACE BENEFIT
tabyl(remir_all$inter_benefit12_race)## Simplify Codes
## Need to Use Three Variables to Simplify (benefit, examine, main effect)
tabyl(remir_all, inter_benefit12_race, inter_examine12_race_cat)tabyl(remir_all, inter_benefit12_race, main_effect)## Code as Exact if benefit any specific group of Blacks, Asians, etc
## Code as Mixed if benefit a combined group of Blacks, Asians, etc
## Code as Other if benefit whites only or no group
## If None (i.e., equal effects), count as Exact or Mixed benefit if main effect
## If None (i.e., equal effects), count as no benefit or other if No main effect
## If None (i.e., equal effects) but examined is NA, then None means NA
remir_all$inter_benefit12_race_cat <-
ifelse(remir_all$inter_benefit12_race == 'Asian or Asian American;Minority' |
remir_all$inter_benefit12_race == 'Black or African American;Other or Multiracial' |
remir_all$inter_benefit12_race == 'Black or African American' |
remir_all$inter_benefit12_race == 'Black or African American;White', 'Exact',
ifelse(remir_all$inter_benefit12_race == 'Minority' |
remir_all$inter_benefit12_race == 'White + Another Race', 'Mixed',
ifelse(remir_all$inter_benefit12_race == 'Not Black' |
remir_all$inter_benefit12_race == 'White' |
remir_all$inter_benefit12_race == 'Majority', 'Other',
ifelse(remir_all$inter_benefit12_race == 'None' &
remir_all$main_effect == 'Yes' &
remir_all$inter_examine12_race_cat == 'Exact', 'Exact',
ifelse(remir_all$inter_benefit12_race == 'None' &
remir_all$main_effect == 'Yes' &
remir_all$inter_examine12_race_cat == 'Mixed', 'Mixed',
ifelse(remir_all$inter_benefit12_race == 'None' &
remir_all$main_effect == 'No' &
!is.na(remir_all$inter_examine12_race_cat), 'Other', NA))))))
## Check -- Looks OK
tabyl(remir_all$inter_benefit12_race_cat)tabyl(remir_all, inter_benefit12_race, inter_benefit12_race_cat)## SUBGROUP RACE BENEFIT
tabyl(remir_all$subg_signif12_race)## Simplify codes
## None is NA if examined subgroups is NA
## None means no subgroup effects if examined is not NA (i.e., other or no benefit)
remir_all$subg_signif12_race_cat <-
ifelse(is.na(remir_all$subg_examine12_race_cat), NA,
ifelse(remir_all$subg_signif12_race == "White;Non-White" |
remir_all$subg_signif12_race == 'Minority' |
remir_all$subg_signif12_race == 'Minority; Majority', "Mixed",
ifelse(remir_all$subg_signif12_race == "None" |
remir_all$subg_signif12_race == "White", "Other", "Exact")))
tabyl(remir_all$subg_signif12_race_cat)## Check Looks OK
tabyl(remir_all$subg_signif12_race_cat)tabyl(remir_all, subg_signif12_race, subg_signif12_race_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_all$inter_benefit12_race_cat <- ifelse(
(remir_all$black_all > .75 |
remir_all$asian_all > .75 |
remir_all$native_all > .75 |
remir_all$pacif_all > .75 ) &
is.na(remir_all$inter_benefit12_race_cat) &
is.na(remir_all$subg_signif12_race_cat),
"Homog", remir_all$inter_benefit12_race_cat)
tabyl(remir_all$inter_benefit12_race_cat)##### ETHNICITY ########################
## Change Inteaction Examined Variable Names
names(remir_all)[18] <- "inter_examine1_ethnic"
names(remir_all)[84] <- "inter_examine2_ethnic"
table(remir_all$inter_examine1_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/majority Minority/Majority
## 5 2 1
## White/Minority White/Non-White <NA>
## 3 1 280table(remir_all$inter_examine2_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/Majority None
## 14 5 2
## White/Minority White/Other <NA>
## 1 1 269## Combine Two Tnteractions Examined
remir_all$inter_examine12_ethnic <- ifelse(is.na(remir_all$inter_examine1_ethnic),
remir_all$inter_examine2_ethnic, remir_all$inter_examine1_ethnic)
tabyl(remir_all$inter_examine12_ethnic)############## What Ethnic Interactions Were Significant ###############
## Change Interaction Significant Variable Names
names(remir_all)[19] <- "inter_signif1_ethnic"
names(remir_all)[85] <- "inter_signif2_ethnic"
table(remir_all$inter_signif1_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic None White/Minority
## 2 11 3
## <NA>
## 276table(remir_all$inter_signif2_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/Majority None
## 6 2 14
## White/Minority <NA>
## 1 269## Combine Two Tnteractions Significant
remir_all$inter_signif12_ethnic <- ifelse(is.na(remir_all$inter_signif1_ethnic),
remir_all$inter_signif2_ethnic, remir_all$inter_signif1_ethnic)
tabyl(remir_all$inter_signif12_ethnic)############## What Ethnic Interaction Groups Benefited More ############
## Change Interaction Benefitted Variable Names
names(remir_all)[21] <- "inter_benefit1_ethnic"
names(remir_all)[87] <- "inter_benefit2_ethnic"
table(remir_all$inter_benefit1_ethnic, useNA = "always")##
## Hispanic Minority None <NA>
## 2 3 11 276table(remir_all$inter_benefit2_ethnic, useNA = "always")##
## Hispanic Majority Minority Non-Hispanic None <NA>
## 3 2 1 3 14 269## Combine Two Tnteractions Benefitted
remir_all$inter_benefit12_ethnic <- ifelse(is.na(remir_all$inter_benefit1_ethnic),
remir_all$inter_benefit2_ethnic, remir_all$inter_benefit1_ethnic)
tabyl(remir_all$inter_benefit12_ethnic)############# What Ethnic Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[51] <- "subg_examine1_ethnic"
names(remir_all)[88] <- "subg_examine2_ethnic"
table(remir_all$subg_examine1_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic <NA>
## 6 1 285table(remir_all$subg_examine2_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic Minority
## 12 3 2
## Minority; Majority None <NA>
## 2 4 269## Combine Two Subgroups Examined
remir_all$subg_examine12_ethnic <- ifelse(is.na(remir_all$subg_examine1_ethnic),
remir_all$subg_examine2_ethnic, remir_all$subg_examine1_ethnic)
tabyl(remir_all$subg_examine12_ethnic)############## What Ethnic Subgroups Were Significant ##################
## Change Subgroup Significant Variable Names
names(remir_all)[53] <- "subg_signif1_ethnic"
names(remir_all)[90] <- "subg_signif2_ethnic"
table(remir_all$subg_signif1_ethnic, useNA = "always")##
## Hispanic None <NA>
## 5 2 285table(remir_all$subg_signif2_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic Minority
## 10 3 2
## Minority; Majority None <NA>
## 2 6 269## Combine Two Tnteractions Significant
remir_all$subg_signif12_ethnic <- ifelse(is.na(remir_all$subg_signif1_ethnic),
remir_all$subg_signif2_ethnic, remir_all$subg_signif1_ethnic)
tabyl(remir_all$subg_signif12_ethnic)## INTERACTION ETHNIC EXAMINED
tabyl(remir_all$inter_examine12_ethnic)## Simplify Codes
## Define as mixed if combined Hispanics into broad category
## Define as exact if referred specifically to Hispanic, Non-Hispanic
## None = NA for examined, as it means that Hispanic subgroups not examined
## First, assign exact and none
remir_all$inter_examine12_ethnic_cat <-
car::recode(remir_all$inter_examine12_ethnic, "
'Hispanic/Non-Hispanic' = 'Exact';
'None' = NA ")
tabyl(remir_all$inter_examine12_ethnic_cat)## Second, remaining categores are assigned as mixed
remir_all$inter_examine12_ethnic_cat <-
ifelse(remir_all$inter_examine12_ethnic_cat != "Exact", "Mixed",
remir_all$inter_examine12_ethnic_cat)
tabyl(remir_all$inter_examine12_ethnic_cat)## Check -- Looks OK
tabyl(remir_all, inter_examine12_ethnic, inter_examine12_ethnic_cat)## SUBGROUP ETHNIC EXAMINED
tabyl(remir_all$subg_examine12_ethnic)## Simplify Codes
## Define as mixed if combined Hispanics into broad category
## Define as exact if referred specifically to Hispanic, Non-HispaniceBlack, Asian, etc
## None = NA for examined, as it means that Hispanic subgroups not examined
remir_all$subg_examine12_ethnic_cat <-
ifelse(remir_all$subg_examine12_ethnic == 'Minority'|
remir_all$subg_examine12_ethnic == 'Minority; Majority',
'Mixed',
ifelse(remir_all$subg_examine12_ethnic == 'None', NA,
'Exact'))
## Check -- Looks OK
tabyl(remir_all$subg_examine12_ethnic_cat)tabyl(remir_all, subg_examine12_ethnic, subg_examine12_ethnic_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE OF HISPANICS
tabyl(remir_all$inter_examine12_ethnic_cat)remir_all$inter_examine12_ethnic_cat <- ifelse(
(remir_all$hisp_all > .75) &
is.na(remir_all$inter_examine12_ethnic_cat) &
is.na(remir_all$subg_examine12_ethnic_cat),
"Homog", remir_all$inter_examine12_ethnic_cat)
tabyl(remir_all$inter_examine12_ethnic_cat)tabyl(remir_all, inter_examine12_ethnic, inter_examine12_ethnic_cat)## INTERACTION ETHNIC BENEFIT
tabyl(remir_all$inter_benefit12_ethnic)## Simplify Codes
## Code as Exact Benefit if Hispanic
## Code as Other if Non-Hispanic or Majority
## Code as Mixed if Minority
## If None (i.e., equal effects) count as benefit if main effect
## If None (i.e., equal effects) count as other if if no main effect
## If None and ethnic not examined, code as NA as meaning of None for interaction is unclear
## Code Unclear as NA
remir_all$inter_benefit12_ethnic_cat <-
ifelse(remir_all$inter_benefit12_ethnic == 'Hispanic', 'Exact',
ifelse(remir_all$inter_benefit12_ethnic == 'Minority', 'Mixed',
ifelse(remir_all$inter_benefit12_ethnic == 'Non-Hispanic' |
remir_all$inter_benefit12_ethnic == 'Majority', 'Other',
ifelse(remir_all$inter_benefit12_ethnic == 'None' &
remir_all$main_effect == 'Yes' &
remir_all$inter_examine12_ethnic_cat == 'Exact', 'Exact',
ifelse(remir_all$inter_benefit12_ethnic == 'None' &
remir_all$main_effect == 'Yes' &
remir_all$inter_examine12_ethnic_cat == 'Mixed', 'Mixed',
ifelse(remir_all$inter_benefit12_ethnic == 'None' &
remir_all$main_effect == 'No' &
!is.na(remir_all$inter_examine12_ethnic_cat), 'Other', NA))))))
## Check -- Looks OK
tabyl(remir_all$inter_benefit12_ethnic_cat)tabyl(remir_all, inter_benefit12_ethnic, inter_benefit12_ethnic_cat)## SUBGROUP ETHNIC BENEFIT
tabyl(remir_all$subg_signif12_ethnic)## Simplify codes
## None here means no subgroup effects, should be other or no benefit
remir_all$subg_signif12_ethnic_cat <-
ifelse(remir_all$subg_signif12_ethnic == "Hispanic" |
remir_all$subg_signif12_ethnic == "Hispanic;Non-Hispanic" , "Exact",
ifelse(remir_all$subg_signif12_ethnic == "None", "Other",
ifelse(remir_all$subg_signif12_ethnic == "Minority" |
remir_all$subg_signif12_ethnic == "Minority; Majority", "Mixed",
ifelse(is.na(remir_all$subg_examine12_ethnic), NA, 'Check'))))
## Check
tabyl(remir_all$subg_signif12_ethnic_cat)tabyl(remir_all, subg_signif12_ethnic, subg_signif12_ethnic_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE
tabyl(remir_all$inter_benefit12_ethnic_cat)remir_all$inter_benefit12_ethnic_cat <- ifelse(
(remir_all$hisp_all > .75) &
is.na(remir_all$inter_benefit12_ethnic_cat) &
is.na(remir_all$subg_signif12_ethnic_cat),
"Homog", remir_all$inter_benefit12_ethnic_cat)
tabyl(remir_all$inter_benefit12_ethnic_cat)## Problem NA n = 200 for examined n = 198 for benefited
## How can cases that don't examine ethicity show a benefit
remir_all %>% filter(is.na(inter_examine12_ethnic_cat) &
is.na(subg_examine12_ethnic_cat) &
!is.na(inter_benefit12_ethnic_cat) &
is.na(subg_signif12_ethnic_cat)) %>%
select(Citation.ID, Program.Name, inter_examine12_ethnic_cat,
subg_examine12_ethnic_cat) remir_all %>% filter(is.na(inter_examine12_ethnic_cat) &
is.na(subg_examine12_ethnic_cat) &
!is.na(inter_benefit12_ethnic_cat) &
is.na(subg_signif12_ethnic_cat)) %>%
select(Citation.ID, Program.Name, inter_benefit12_ethnic_cat,
subg_signif12_ethnic_cat)## Change 3980 and 3981 to NA so that examined and benefit are consistent
tabyl(remir_all$inter_benefit12_ethnic_cat)remir_all$inter_benefit12_ethnic_cat <-
ifelse(remir_all$Citation.ID == 3980, NA, remir_all$inter_benefit12_ethnic_cat)
tabyl(remir_all$inter_benefit12_ethnic_cat)remir_all$inter_benefit12_ethnic_cat <-
ifelse(remir_all$Citation.ID == 3981, NA, remir_all$inter_benefit12_ethnic_cat)
tabyl(remir_all$inter_benefit12_ethnic_cat)############## GENDER CODES ########################
############## What Gender Interactions Were Examined
## Change Interaction Examined Gender Variable Names
names(remir_all)[23] <- "inter_examine1_gender"
names(remir_all)[92] <- "inter_examine2_gender"
table(remir_all$inter_examine1_gender, useNA = "always")##
## Male/Female <NA>
## 41 251table(remir_all$inter_examine2_gender, useNA = "always")##
## Male/Female <NA>
## 37 255## Combine Two Gender Tnteractions Examined
remir_all$inter_examine12_gender <- ifelse(is.na(remir_all$inter_examine1_gender),
remir_all$inter_examine2_gender, remir_all$inter_examine1_gender)
tabyl(remir_all$inter_examine12_gender)############## What Gender interactions Were Significant ###############
## Change Interaction Significant Variable Names
names(remir_all)[24] <- "inter_signif1_gender"
names(remir_all)[93] <- "inter_signif2_gender"
table(remir_all$inter_signif1_gender, useNA = "always")##
## Male/Female None <NA>
## 6 35 251table(remir_all$inter_signif2_gender, useNA = "always")##
## Male/Female None <NA>
## 18 19 255## Combine Two Tnteractions Significant
remir_all$inter_signif12_gender <- ifelse(is.na(remir_all$inter_signif1_gender),
remir_all$inter_signif2_gender, remir_all$inter_signif1_gender)
tabyl(remir_all$inter_signif12_gender)########## What Gender Interaction Groups Benefited More ##################
## Change Interaction Benefitted Variable Names
names(remir_all)[26] <- "inter_benefit1_gender"
names(remir_all)[95] <- "inter_benefit2_gender"
table(remir_all$inter_benefit1_gender, useNA = "always")##
## Female Male None Unclear <NA>
## 4 1 35 1 251table(remir_all$inter_benefit2_gender, useNA = "always")##
## Female Male None Unclear <NA>
## 7 7 19 4 255## Combine Two Tnteractions
remir_all$inter_benefit12_gender <- ifelse(is.na(remir_all$inter_benefit1_gender),
remir_all$inter_benefit2_gender, remir_all$inter_benefit1_gender)
tabyl(remir_all$inter_benefit12_gender)############# What Gender Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[55] <- "subg_examine1_gender"
names(remir_all)[96] <- "subg_examine2_gender"
table(remir_all$subg_examine1_gender, useNA = "always")##
## Male;Female <NA>
## 13 279table(remir_all$subg_examine2_gender, useNA = "always")##
## Female Male;Female None <NA>
## 1 34 2 255## Combine Two Subgroups Examined
remir_all$subg_examine12_gender <- ifelse(is.na(remir_all$subg_examine1_gender),
remir_all$subg_examine2_gender, remir_all$subg_examine1_gender)
tabyl(remir_all$subg_examine12_gender)############## What Gender Subgroups Were Significant #####################
## Change Subgroup Significant Variable Names
names(remir_all)[57] <- "subg_signif1_gender"
names(remir_all)[98] <- "subg_signif2_gender"
table(remir_all$subg_signif1_gender, useNA = "always")##
## Female Male Male;Female None <NA>
## 2 3 6 2 279table(remir_all$subg_signif2_gender, useNA = "always")##
## Female Male Male;Female None <NA>
## 3 2 25 7 255## Combine Two Tnteractions Significant
remir_all$subg_signif12_gender <- ifelse(is.na(remir_all$subg_signif1_gender),
remir_all$subg_signif2_gender, remir_all$subg_signif1_gender)
tabyl(remir_all$subg_signif12_gender)## INTERACTION GENDER EXAMINED
tabyl(remir_all$inter_examine12_gender)## SUBGROUP GENDER EXAMINED
tabyl(remir_all$subg_examine12_gender)## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_all$inter_examine12_gender <- ifelse(
(remir_all$female_all > .75 ) &
is.na(remir_all$inter_examine12_gender) &
is.na(remir_all$subg_examine12_gender),
"Homog", remir_all$inter_examine12_gender)
tabyl(remir_all$inter_examine12_gender)## INTERACTION GENDER BENEFITTTED
tabyl(remir_all$inter_benefit12_gender)## Simplify Codes
remir_all$inter_benefit12_gender_cat <-
ifelse(remir_all$inter_benefit12_gender == 'None' &
remir_all$main_effect == 'Yes', 'Both',
ifelse(remir_all$inter_benefit12_gender == 'None' &
remir_all$main_effect == 'No', 'Neither',
remir_all$inter_benefit12_gender))
## Check -- Look OK
tabyl(remir_all, inter_benefit12_gender, inter_benefit12_gender_cat)tabyl(remir_all$inter_benefit12_gender_cat)## SUBGROUP GENDER BENEFITTED
tabyl(remir_all$subg_signif12_gender)## SEPARATE OUT HOMOGENOUS SAMPLE
remir_all$inter_benefit12_gender_cat <- ifelse(
(remir_all$female_all > .75) &
is.na(remir_all$inter_benefit12_gender_cat) &
is.na(remir_all$subg_signif12_gender),
"Homog", remir_all$inter_benefit12_gender_cat)
tabyl(remir_all$inter_benefit12_gender_cat)## More Simplification
remir_all$inter_benefit12_gender_cat2 <-
ifelse(remir_all$inter_benefit12_gender_cat == 'Male' |
remir_all$inter_benefit12_gender_cat == 'Neither', 'Other',
ifelse(remir_all$inter_benefit12_gender_cat == 'Both' |
remir_all$inter_benefit12_gender_cat == 'Female', 'Female',
ifelse(remir_all$inter_benefit12_gender_cat == 'Unclear', NA,
remir_all$inter_benefit12_gender_cat)))
tabyl(remir_all, inter_benefit12_gender_cat, inter_benefit12_gender_cat2)tabyl(remir_all, inter_benefit12_gender_cat2)remir_all$subg_signif12_gender_cat <-
ifelse(remir_all$subg_signif12_gender == 'Female' |
remir_all$subg_signif12_gender == 'Male;Female', 'Female',
ifelse(remir_all$subg_signif12_gender == 'Male' |
remir_all$subg_signif12_gender == 'None', 'Other',
remir_all$subg_signif12_gender))
tabyl(remir_all, subg_signif12_gender, subg_signif12_gender_cat)tabyl(remir_all, subg_signif12_gender_cat)############## SEX ########################
############## What Sex Interaction Groups Were Examined ####################
## Change Interaction Examined Variable Names
names(remir_all)[27] <- "inter_examine1_sex"
names(remir_all)[100] <- "inter_examine2_sex"
table(remir_all$inter_examine1_sex, useNA = "always")##
## No Yes <NA>
## 51 2 239table(remir_all$inter_examine2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Examined
remir_all$inter_examine12_sex <- ifelse(is.na(remir_all$inter_examine1_sex),
remir_all$inter_examine2_sex, remir_all$inter_examine1_sex)
tabyl(remir_all$inter_examine12_sex)############## What Sex interactions Were Significant ###################
## Change Interaction Significant Variable Names
names(remir_all)[28] <- "inter_signif1_sex"
names(remir_all)[101] <- "inter_signif2_sex"
table(remir_all$inter_signif1_sex, useNA = "always")##
## None <NA>
## 2 290table(remir_all$inter_signif2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Significant
remir_all$inter_signif12_sex <- ifelse(is.na(remir_all$inter_signif1_sex),
remir_all$inter_signif2_sex, remir_all$inter_signif1_sex)
tabyl(remir_all$inter_signif12_sex)############ What Sex Interaction Groups Benefited More ####################
## Change Interactions Benefitted Variable Names
names(remir_all)[30] <- "inter_benefit1_sex"
names(remir_all)[103] <- "inter_benefit2_sex"
table(remir_all$inter_benefit1_sex, useNA = "always")##
## None <NA>
## 2 290table(remir_all$inter_benefit2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Benefitted
remir_all$inter_benefit12_sex <- ifelse(is.na(remir_all$inter_benefit1_sex),
remir_all$inter_benefit2_sex, remir_all$inter_benefit1_sex)
tabyl(remir_all$inter_benefit12_sex)############# What Sex Subgroup Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[59] <- "subg_examine1_sex"
names(remir_all)[104] <- "subg_examine2_sex"
table(remir_all$subg_examine1_sex, useNA = "always")##
## <NA>
## 292table(remir_all$subg_examine2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Subgroups Examined
remir_all$subg_examine12_sex <- ifelse(is.na(remir_all$subg_examine1_sex),
remir_all$subg_examine2_sex, remir_all$subg_examine1_sex)
tabyl(remir_all$subg_examine12_sex)############## What Sex Subgroups Were Significant #######################
## Change Subgroup Significant Variable Names
names(remir_all)[61] <- "subg_signif1_sex"
names(remir_all)[106] <- "subg_signif2_sex"
table(remir_all$subg_signif1_sex, useNA = "always")##
## <NA>
## 292table(remir_all$subg_signif2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Significant
remir_all$subg_signif12_sex <- ifelse(is.na(remir_all$subg_signif1_sex),
remir_all$subg_signif2_sex, remir_all$subg_signif1_sex)
tabyl(remir_all$subg_signif12_sex)## INTERACTION SEX EXAMINED
## Codebook didn't ask which groups examined in interaction
## Used were any groups examined
tabyl(remir_all$inter_examine12_sex)## SUBGROUP SEX EXAMINED
tabyl(remir_all$subg_examine12_sex)## SEPARATE OUT HOMOGENEOUS SAMPLE
## Not possible, no measure of lgbt sample in Remir 1
## INTERACTION SEX BENEFITTED
tabyl(remir_all$inter_benefit12_sex)tabyl(remir_all, inter_benefit12_sex, main_effect)## None with main effect means benefit
remir_all$inter_benefit12_sex_cat <-
ifelse(remir_all$inter_benefit12_sex == 'None' &
remir_all$main_effect == 'Yes', 'Yes', remir_all$inter_benefit12_sex)
tabyl(remir_all$inter_benefit12_sex_cat)## SUBGROUP SEX BENEFITTED
tabyl(remir_all$subg_signif12_sex)############## SES ########################
############## What SES Interaction Group Were Examined ########################
## Change Interaction SES Varialbe Names
names(remir_all)[31] <- "inter_examine1_ses"
names(remir_all)[108] <- "inter_examine2_ses"
table(remir_all$inter_examine1_ses, useNA = "always")##
## No Yes <NA>
## 40 13 239table(remir_all$inter_examine2_ses, useNA = "always")##
## Low/High <NA>
## 13 279## Combine Two Tnteractions Examined
remir_all$inter_examine12_ses <- ifelse(is.na(remir_all$inter_examine1_ses),
remir_all$inter_examine2_ses, remir_all$inter_examine1_ses)
tabyl(remir_all$inter_examine12_ses)############## What SES Interactions Were Significant ####################
## Change Interaction Significant Variable Names
names(remir_all)[32] <- "inter_signif1_ses"
names(remir_all)[109] <- "inter_signif2_ses"
table(remir_all$inter_signif1_ses, useNA = "always")##
## Low/High None <NA>
## 5 8 279table(remir_all$inter_signif2_ses, useNA = "always")##
## Low/High None <NA>
## 4 9 279## Combine Two Tnteractions Significant
remir_all$inter_signif12_ses <- ifelse(is.na(remir_all$inter_signif1_ses),
remir_all$inter_signif2_ses, remir_all$inter_signif1_ses)
tabyl(remir_all$inter_signif12_ses)############ What SES Interaction Groups Benefited More ##################
## Change Interaction Benefitted Variable Names
names(remir_all)[34] <- "inter_benefit1_ses"
names(remir_all)[111] <- "inter_benefit2_ses"
table(remir_all$inter_benefit1_ses, useNA = "always")##
## High Low None <NA>
## 1 4 8 279table(remir_all$inter_benefit2_ses, useNA = "always")##
## High Low None Unclear <NA>
## 1 2 9 1 279## Combine Two Tnteractions
remir_all$inter_benefit12_ses <- ifelse(is.na(remir_all$inter_benefit1_ses),
remir_all$inter_benefit2_ses, remir_all$inter_benefit1_ses)
tabyl(remir_all$inter_benefit12_ses)############# What SES Subgroups Were Examined ####################
## Change Subgroups Examined Variable Names
names(remir_all)[63] <- "subg_examine1_ses"
names(remir_all)[112] <- "subg_examine2_ses"
table(remir_all$subg_examine1_ses, useNA = "always")##
## Low Low;High <NA>
## 3 6 283table(remir_all$subg_examine2_ses, useNA = "always")##
## Low Low;High <NA>
## 4 9 279## Combine Two Subgroups Examined
remir_all$subg_examine12_ses <- ifelse(is.na(remir_all$subg_examine1_ses),
remir_all$subg_examine2_ses, remir_all$subg_examine1_ses)
tabyl(remir_all$subg_examine12_ses)############## What SES Subgroups Were Significant #########################
## Change Subgroups Significant Variable Names
names(remir_all)[65] <- "subg_signif1_ses"
names(remir_all)[114] <- "subg_signif2_ses"
table(remir_all$subg_signif1_ses, useNA = "always")##
## Low Low;High None <NA>
## 5 2 2 283table(remir_all$subg_signif2_ses, useNA = "always")##
## High Low Low;High <NA>
## 1 5 7 279## Combine Two Tnteractions Significant
remir_all$subg_signif12_ses <- ifelse(is.na(remir_all$subg_signif1_ses),
remir_all$subg_signif2_ses, remir_all$subg_signif1_ses)
tabyl(remir_all$subg_signif12_ses)## INTERACTION SES EXAMINED
tabyl(remir_all$inter_examine12_ses)## Change No to NA
remir_all$inter_examine12_ses_cat <- ifelse(
remir_all$inter_examine12_ses == 'No', NA, remir_all$inter_examine12_ses)
tabyl(remir_all$inter_examine12_ses_cat)## SUBGROUP SES EXAMINED
tabyl(remir_all$subg_examine12_ses)## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_all$inter_examine12_ses_cat <- ifelse(
(remir_all$econ_all > .75) &
is.na(remir_all$inter_examine12_ses_cat) &
is.na(remir_all$subg_examine12_ses),
"Homog", remir_all$inter_examine12_ses_cat)
tabyl(remir_all$inter_examine12_ses_cat)## INTERACTION SES BENEFITTED
tabyl(remir_all$inter_benefit12_ses)## Simplify Codes
tabyl(remir_all, inter_benefit12_ses, inter_examine12_ses)remir_all$inter_benefit12_ses_cat <-
ifelse(remir_all$inter_benefit12_ses == 'None' &
remir_all$main_effect == 'Yes', 'Both',
ifelse(remir_all$inter_benefit12_ses == 'None' &
remir_all$main_effect == 'No', 'Neither',
remir_all$inter_benefit12_ses))
tabyl(remir_all, inter_benefit12_ses, inter_benefit12_ses_cat)tabyl(remir_all, inter_benefit12_ses_cat)## Simplify further
remir_all$inter_benefit12_ses_cat2 <-
ifelse(remir_all$inter_benefit12_ses_cat == 'Unclear', NA,
ifelse(remir_all$inter_benefit12_ses_cat == 'Both' |
remir_all$inter_benefit12_ses_cat == 'Low', 'Low',
remir_all$inter_benefit12_ses_cat))
tabyl(remir_all, inter_benefit12_ses_cat, inter_benefit12_ses_cat2)tabyl(remir_all, inter_benefit12_ses_cat2)## SUBGROUP SES BENEFITTED
tabyl(remir_all$subg_signif12_ses)## Simplify
remir_all$subg_signif12_ses_cat <-
ifelse(remir_all$subg_signif12_ses == 'High' |
remir_all$subg_signif12_ses == 'None', 'Other',
ifelse(remir_all$subg_signif12_ses == 'Low' |
remir_all$subg_signif12_ses == 'Low;High', 'Low',
remir_all$subg_signif12_ses))
tabyl(remir_all, subg_signif12_ses, subg_signif12_ses_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_all$inter_benefit12_ses_cat2 <- ifelse(
(remir_all$econ_all > .75) &
is.na(remir_all$inter_benefit12_ses_cat2) &
is.na(remir_all$subg_signif12_ses_cat),
"Homog", remir_all$inter_benefit12_ses_cat2)
tabyl(remir_all$inter_benefit12_ses_cat2)############## LOCATION ########################
############## What Location Interaction Groups Were Examined ###################
## Change Interaction Examined Variable Names
names(remir_all)[36] <- "inter_examine1_loc"
names(remir_all)[116] <- "inter_examine2_loc"
table(remir_all$inter_examine1_loc, useNA = "always")##
## <NA>
## 292table(remir_all$inter_examine2_loc, useNA = "always")##
## Urban/Non-Urban <NA>
## 2 290## Combine Two Tnteractions Examined
remir_all$inter_examine12_loc <- ifelse(is.na(remir_all$inter_examine1_loc),
remir_all$inter_examine2_loc, remir_all$inter_examine1_loc)
tabyl(remir_all$inter_examine12_loc)############ What Location Interactions Were Significant ################
## Change Interaction Significant Variable Names
names(remir_all)[37] <- "inter_signif1_loc"
names(remir_all)[117] <- "inter_signif2_loc"
table(remir_all$inter_signif1_loc, useNA = "always")##
## <NA>
## 292table(remir_all$inter_signif2_loc, useNA = "always")##
## None <NA>
## 2 290## Combine Two Tnteractions Significant
remir_all$inter_signif12_loc <- ifelse(is.na(remir_all$inter_signif1_loc),
remir_all$inter_signif2_loc, remir_all$inter_signif1_loc)
tabyl(remir_all$inter_signif12_loc)############ What Location Interaction Groups Benefited More #############
## Change Interaction Benefitted Variable Names
names(remir_all)[39] <- "inter_benefit1_loc"
names(remir_all)[119] <- "inter_benefit2_loc"
table(remir_all$inter_benefit1_loc, useNA = "always")##
## <NA>
## 292table(remir_all$inter_benefit2_loc, useNA = "always")##
## None <NA>
## 2 290## Combine Two Tnteractions
remir_all$inter_benefit12_loc <- ifelse(is.na(remir_all$inter_benefit1_loc),
remir_all$inter_benefit2_loc, remir_all$inter_benefit1_loc)
tabyl(remir_all$inter_benefit12_loc)############# What Location Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[67] <- "subg_examine1_loc"
names(remir_all)[120] <- "subg_examine2_loc"
table(remir_all$subg_examine1_loc, useNA = "always")##
## Rural <NA>
## 2 290table(remir_all$subg_examine2_loc, useNA = "always")##
## Urban;Non-Urban <NA>
## 2 290## Combine Two Subgroups Examined
remir_all$subg_examine12_loc <- ifelse(is.na(remir_all$subg_examine1_loc),
remir_all$subg_examine2_loc, remir_all$subg_examine1_loc)
tabyl(remir_all$subg_examine12_loc)############## What Location Subgroups Were Significant ####################
## Change Subgroup Significant Variable Names
names(remir_all)[69] <- "subg_signif1_loc"
names(remir_all)[122] <- "subg_signif2_loc"
table(remir_all$subg_signif1_loc, useNA = "always")##
## Rural <NA>
## 2 290table(remir_all$subg_signif2_loc, useNA = "always")##
## None Urban;Non-Urban <NA>
## 1 1 290## Combine Two Tnteractions Significant
remir_all$subg_signif12_loc <- ifelse(is.na(remir_all$subg_signif1_loc),
remir_all$subg_signif2_loc, remir_all$subg_signif1_loc)
tabyl(remir_all$subg_signif12_loc)## INTERACTION LOCATION EXAMINED
tabyl(remir_all$inter_examine12_loc)## SUBGROUP LOCATION EXAMINED
tabyl(remir_all$subg_examine12_loc)## SEPARATE OUT HOMOGENEOUS SAMPLE
## No measure of sample proportion for urban or rural
## INTERACTION LOCATION BENEFITTED
tabyl(remir_all$inter_benefit12_loc)## SUBGROUP LOCATION BENEFITTED
tabyl(remir_all$subg_signif12_loc)############## NATIVITY ########################
############## What Nativity Interaction Groups Were Examined ################################
## Change Interaction Nativity Variable Names
names(remir_all)[41] <- "inter_examine1_nat"
names(remir_all)[124] <- "inter_examine2_nat"
table(remir_all$inter_examine1_nat, useNA = "always")##
## Immigrant/Nonimmigrant <NA>
## 1 291table(remir_all$inter_examine2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Examined
remir_all$inter_examine12_nat <- ifelse(is.na(remir_all$inter_examine1_nat),
remir_all$inter_examine2_nat, remir_all$inter_examine1_nat)
tabyl(remir_all$inter_examine12_nat)############## What Nativity Interactions Were Significant ##############################
## Change Interaction Significant Variable Names
names(remir_all)[42] <- "inter_signif1_nat"
names(remir_all)[125] <- "inter_signif2_nat"
table(remir_all$inter_signif1_nat, useNA = "always")##
## None <NA>
## 1 291table(remir_all$inter_signif2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Significant
remir_all$inter_signif12_nat <- ifelse(is.na(remir_all$inter_signif1_nat),
remir_all$inter_signif2_nat, remir_all$inter_signif1_nat)
tabyl(remir_all$inter_signif12_nat)############## What Nativity Interaction Groups Benefited More ##############################
## Change Interaction Benfitted Variable Names
names(remir_all)[44] <- "inter_benefit1_nat"
names(remir_all)[127] <- "inter_benefit2_nat"
table(remir_all$inter_benefit1_nat, useNA = "always")##
## None <NA>
## 1 291table(remir_all$inter_benefit2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions
remir_all$inter_benefit12_nat <- ifelse(is.na(remir_all$inter_benefit1_nat),
remir_all$inter_benefit2_nat, remir_all$inter_benefit1_nat)
tabyl(remir_all$inter_benefit12_nat)############# What Nativity Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[71] <- "subg_examine1_nat"
names(remir_all)[128] <- "subg_examine2_nat"
table(remir_all$subg_examine1_nat, useNA = "always")##
## <NA>
## 292table(remir_all$subg_examine2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Subgroups Examined
remir_all$subg_examine12_nat <- ifelse(is.na(remir_all$subg_examine1_nat),
remir_all$subg_examine2_nat, remir_all$subg_examine1_nat)
tabyl(remir_all$subg_examine12_nat)############## What Nativity Subgroups Were Significant ##############################
## Change Subgroup Significant Variable Names
names(remir_all)[73] <- "subg_signif1_nat"
names(remir_all)[130] <- "subg_signif2_nat"
table(remir_all$subg_signif1_nat, useNA = "always")##
## <NA>
## 292table(remir_all$subg_signif2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Significant
remir_all$subg_signif12_nat <- ifelse(is.na(remir_all$subg_signif1_nat),
remir_all$subg_signif2_nat, remir_all$subg_signif1_nat)
tabyl(remir_all$subg_signif12_nat)## INTERACTION NATIVITY EXAMINED
tabyl(remir_all$inter_examine12_nat)## SUBGROUP NATIVITY EXAMINED
tabyl(remir_all$subg_examine12_nat)## INTERACTION NATIVITY BENEFITTED
tabyl(remir_all$inter_benefit12_nat)## SUBGROUP NATIVITY BENEFITTED
tabyl(remir_all$subg_signif12_nat)#########################################################################
######## SECOND. CREATE SHORT VARIABLES WITH THREE CATEGORIES ############
############## TO BE USED FOR ONLINE SUPPLEMENT ########################
##################### USE FULL SAMPLE ##################################
#########################################################################
## Race
tabyl(remir_all, inter_examine12_race_cat, subg_examine12_race_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_all$examine12_race_short <-
ifelse(is.na(remir_all$inter_examine12_race_cat) &
is.na(remir_all$subg_examine12_race_cat), 'None',
ifelse(remir_all$inter_examine12_race_cat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_all$examine12_race_short)## Fix remaining NAs -- Should be yes since they are missing for one test only
remir_all$examine12_race_short <- ifelse(
is.na(remir_all$examine12_race_short), 'Yes', remir_all$examine12_race_short)
tabyl(remir_all$examine12_race_short)## Ethnicity
tabyl(remir_all, inter_examine12_ethnic_cat, subg_examine12_ethnic_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_all$examine12_ethnic_short <-
ifelse(is.na(remir_all$inter_examine12_ethnic_cat) &
is.na(remir_all$subg_examine12_ethnic_cat), 'None',
ifelse(remir_all$inter_examine12_ethnic_cat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_all$examine12_ethnic_short)## Fix remaining NAs -- Should be yes since they are missing for one test only
remir_all$examine12_ethnic_short <- ifelse(
is.na(remir_all$examine12_ethnic_short), 'Yes', remir_all$examine12_ethnic_short)
tabyl(remir_all$examine12_ethnic_short)## Gender
tabyl(remir_all, inter_examine12_gender, subg_examine12_gender) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_all$examine12_gender_short <-
ifelse(is.na(remir_all$inter_examine12_gender) &
is.na(remir_all$subg_examine12_gender), 'None',
ifelse(remir_all$inter_examine12_gender == 'Homog', 'Homog', 'Yes'))
tabyl(remir_all$examine12_gender_short)## Fix remaining NAs -- Should be yes since they are missing for one test only
remir_all$examine12_gender_short <- ifelse(
is.na(remir_all$examine12_gender_short), 'Yes', remir_all$examine12_gender_short)
tabyl(remir_all$examine12_gender_short)## Sex
tabyl(remir_all, inter_examine12_sex, subg_examine12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_all$examine12_sex_short <-
ifelse(is.na(remir_all$inter_examine12_sex), 'No',
remir_all$inter_examine12_sex)
tabyl(remir_all$examine12_sex_short)## SES
tabyl(remir_all, inter_examine12_ses_cat, subg_examine12_ses) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_all$examine12_ses_short <-
ifelse(is.na(remir_all$inter_examine12_ses_cat) &
is.na(remir_all$subg_examine12_ses), 'None',
ifelse(remir_all$inter_examine12_ses_cat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_all$examine12_ses_short)## Fix remaining NAs -- Should be yes since they are missing for one test only
remir_all$examine12_ses_short <- ifelse(
is.na(remir_all$examine12_ses_short), 'Yes', remir_all$examine12_ses_short)
tabyl(remir_all$examine12_ses_short)## Location
tabyl(remir_all, inter_examine12_loc, subg_examine12_loc) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_all$examine12_loc_short <-
ifelse(is.na(remir_all$inter_examine12_loc) &
is.na(remir_all$subg_examine12_loc), 'None',
ifelse(remir_all$inter_examine12_loc == 'Homog', 'Homog', 'Yes'))
tabyl(remir_all$examine12_loc_short)## Fix remaining NAs -- Should be yes since they are missing for one test only
remir_all$examine12_loc_short <- ifelse(
is.na(remir_all$examine12_loc_short), 'Yes', remir_all$examine12_loc_short)
tabyl(remir_all$examine12_loc_short)## Nativity
tabyl(remir_all, inter_examine12_nat, subg_examine12_nat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_all$examine12_nat_short <-
ifelse(is.na(remir_all$inter_examine12_nat) &
is.na(remir_all$subg_examine12_nat), 'None',
ifelse(remir_all$inter_examine12_nat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_all$examine12_nat_short)# Fix remaining NAs for all
remir_all$examine12_nat_short <- ifelse(
is.na(remir_all$examine12_nat_short), 'Yes', remir_all$examine12_nat_short)## i. Define Variables for Table 2
### RACE (Table 2: Column 1)###
# Interaction Race Examined
names(remir_us)[13] <- "inter_examine1_race"
names(remir_us)[76] <- "inter_examine2_race"
remir_us$inter_examine12_race <- ifelse(is.na(remir_us$inter_examine1_race), remir_us$inter_examine2_race, remir_us$inter_examine1_race)
# Simplify Codes
remir_us$inter_examine12_race_cat <- remir_us$inter_examine12_race
remir_us$inter_examine12_race_cat <- gsub('Minority/Majority|Minority/majority|White/Minority|White/Not White|White/Non-White|White/Other', 'Mixed', remir_us$inter_examine12_race_cat)
remir_us$inter_examine12_race_cat <- ifelse(remir_us$inter_examine12_race_cat == 'None', NA, remir_us$inter_examine12_race_cat)
remir_us$inter_examine12_race_cat <- ifelse(remir_us$inter_examine12_race_cat != "Mixed", "Exact", remir_us$inter_examine12_race_cat)
# Interaction Race Significant
names(remir_us)[14] <- "inter_signif1_race"
names(remir_us)[77] <- "inter_signif2_race"
remir_us$inter_signif12_race <- ifelse(is.na(remir_us$inter_signif1_race), remir_us$inter_signif2_race, remir_us$inter_signif1_race)
# Interaction Race Benefitted
names(remir_us)[16] <- "inter_benefit1_race"
names(remir_us)[79] <- "inter_benefit2_race"
remir_us$inter_benefit12_race <- ifelse(is.na(remir_us$inter_benefit1_race), remir_us$inter_benefit2_race, remir_us$inter_benefit1_race)
# Subgroup Race Examined
names(remir_us)[47] <- "subg_examine1_race"
names(remir_us)[80] <- "subg_examine2_race"
remir_us$subg_examine12_race <- ifelse(is.na(remir_us$subg_examine1_race), remir_us$subg_examine2_race, remir_us$subg_examine1_race)
remir_us$subg_examine12_race_cat <- ifelse(remir_us$subg_examine12_race %in% c("White;Non-White", "Minority", "Minority; Majority"), "Mixed", ifelse(remir_us$subg_examine12_race == "None", NA, "Exact"))
# Subgroup Race Significant
names(remir_us)[49] <- "subg_signif1_race"
names(remir_us)[82] <- "subg_signif2_race"
remir_us$subg_signif12_race <- ifelse(is.na(remir_us$subg_signif1_race), remir_us$subg_signif2_race, remir_us$subg_signif1_race)
remir_us$subg_signif12_race_cat <- ifelse(is.na(remir_us$subg_examine12_race_cat), NA, ifelse(remir_us$subg_signif12_race %in% c("White;Non-White", "Minority", "Minority; Majority"), "Mixed", ifelse(remir_us$subg_signif12_race %in% c("None", "White"), "Other", "Exact")))
# Homogeneous Sample
remir_us$inter_examine12_race_cat <- ifelse((remir_us$black_all > .75 | remir_us$asian_all > .75 | remir_us$native_all > .75 | remir_us$pacif_all > .75) & is.na(remir_us$inter_examine12_race_cat) & is.na(remir_us$subg_examine12_race_cat), "Homog", remir_us$inter_examine12_race_cat)
# Interaction Race Benefit Category
remir_us$inter_benefit12_race_cat <- ifelse(remir_us$inter_benefit12_race %in% c('Asian or Asian American;Minority', 'Black or African American;Other or Multiracial', 'Black or African American', 'Black or African American;White'), 'Exact', ifelse(remir_us$inter_benefit12_race %in% c('Minority', 'White + Another Race'), 'Mixed', ifelse(remir_us$inter_benefit12_race %in% c('Not Black', 'White', 'Majority'), 'Other', ifelse(remir_us$inter_benefit12_race == 'None' & remir_us$main_effect == 'Yes' & remir_us$inter_examine12_race_cat == 'Exact', 'Exact', ifelse(remir_us$inter_benefit12_race == 'None' & remir_us$main_effect == 'Yes' & remir_us$inter_examine12_race_cat == 'Mixed', 'Mixed', ifelse(remir_us$inter_benefit12_race == 'None' & remir_us$main_effect == 'No' & !is.na(remir_us$inter_examine12_race_cat), 'Other', NA))))))
# Homogeneous Sample for Benefit
remir_us$inter_benefit12_race_cat <- ifelse((remir_us$black_all > .75 | remir_us$asian_all > .75 | remir_us$native_all > .75 | remir_us$pacif_all > .75) & is.na(remir_us$inter_benefit12_race_cat) & is.na(remir_us$subg_signif12_race_cat), "Homog", remir_us$inter_benefit12_race_cat)
# Simplified Short Race Examined Category
remir_us$examine12_race_short <- ifelse(is.na(remir_us$inter_examine12_race_cat) & is.na(remir_us$subg_examine12_race_cat), 'None', ifelse(remir_us$inter_examine12_race_cat == 'Homog', 'Homog', 'Yes'))
remir_us$examine12_race_short <- ifelse(is.na(remir_us$examine12_race_short), 'Yes', remir_us$examine12_race_short)
### ETHNICITY (Table 2, Column 3&4) ###
# PREP: Creation of New Variables
############## What Ethnic Interaction Groups Examined ################
## Change Interaction Examined Variable Names
names(remir_us)[18] <- "inter_examine1_ethnic"
names(remir_us)[84] <- "inter_examine2_ethnic"
## Combine Two Interactions Examined
remir_us$inter_examine12_ethnic <- ifelse(is.na(remir_us$inter_examine1_ethnic),
remir_us$inter_examine2_ethnic, remir_us$inter_examine1_ethnic)
############## What Ethnic Interactions Were Significant ###############
## Change Interaction Significant Variable Names
names(remir_us)[19] <- "inter_signif1_ethnic"
names(remir_us)[85] <- "inter_signif2_ethnic"
## Combine Two Interactions Significant
remir_us$inter_signif12_ethnic <- ifelse(is.na(remir_us$inter_signif1_ethnic),
remir_us$inter_signif2_ethnic, remir_us$inter_signif1_ethnic)
############## What Ethnic Interaction Groups Benefited More ############
## Change Interaction Benefitted Variable Names
names(remir_us)[21] <- "inter_benefit1_ethnic"
names(remir_us)[87] <- "inter_benefit2_ethnic"
## Combine Two Interactions Benefitted
remir_us$inter_benefit12_ethnic <- ifelse(is.na(remir_us$inter_benefit1_ethnic),
remir_us$inter_benefit2_ethnic, remir_us$inter_benefit1_ethnic)
############# What Ethnic Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_us)[51] <- "subg_examine1_ethnic"
names(remir_us)[88] <- "subg_examine2_ethnic"
## Combine Two Subgroups Examined
remir_us$subg_examine12_ethnic <- ifelse(is.na(remir_us$subg_examine1_ethnic),
remir_us$subg_examine2_ethnic, remir_us$subg_examine1_ethnic)
############## What Ethnic Subgroups Were Significant ##################
## Change Subgroup Significant Variable Names
names(remir_us)[53] <- "subg_signif1_ethnic"
names(remir_us)[90] <- "subg_signif2_ethnic"
## Combine Two Interactions Significant
remir_us$subg_signif12_ethnic <- ifelse(is.na(remir_us$subg_signif1_ethnic),
remir_us$subg_signif2_ethnic, remir_us$subg_signif1_ethnic)
## INTERACTION ETHNIC EXAMINED
remir_us$inter_examine12_ethnic_cat <-
car::recode(remir_us$inter_examine12_ethnic, "
'Hispanic/Non-Hispanic' = 'Exact';
'None' = NA ")
remir_us$inter_examine12_ethnic_cat <-
ifelse(remir_us$inter_examine12_ethnic_cat != "Exact", "Mixed",
remir_us$inter_examine12_ethnic_cat)
## SUBGROUP ETHNIC EXAMINED
remir_us$subg_examine12_ethnic_cat <-
ifelse(remir_us$subg_examine12_ethnic == 'Minority'|
remir_us$subg_examine12_ethnic == 'Minority; Majority',
'Mixed',
ifelse(remir_us$subg_examine12_ethnic == 'None', NA,
'Exact'))
## SEPARATE OUT HOMOGENEOUS SAMPLE OF HISPANICS
remir_us$inter_examine12_ethnic_cat <- ifelse(
(remir_us$hisp_all > .75) &
is.na(remir_us$inter_examine12_ethnic_cat) &
is.na(remir_us$subg_examine12_ethnic_cat),
"Homog", remir_us$inter_examine12_ethnic_cat)
# Check Variables
table(remir_us$inter_examine1_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/majority Minority/Majority
## 5 2 1
## White/Minority White/Non-White <NA>
## 3 1 228table(remir_us$inter_examine2_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/Majority None
## 14 5 2
## White/Minority White/Other <NA>
## 1 1 217tabyl(remir_us$inter_examine12_ethnic)table(remir_us$inter_signif1_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic None White/Minority
## 2 11 3
## <NA>
## 224table(remir_us$inter_signif2_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/Majority None
## 6 2 14
## White/Minority <NA>
## 1 217tabyl(remir_us$inter_signif12_ethnic)table(remir_us$inter_benefit1_ethnic, useNA = "always")##
## Hispanic Minority None <NA>
## 2 3 11 224table(remir_us$inter_benefit2_ethnic, useNA = "always")##
## Hispanic Majority Minority Non-Hispanic None <NA>
## 3 2 1 3 14 217tabyl(remir_us$inter_benefit12_ethnic)table(remir_us$subg_examine1_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic <NA>
## 6 1 233table(remir_us$subg_examine2_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic Minority
## 12 3 2
## Minority; Majority None <NA>
## 2 4 217tabyl(remir_us$subg_examine12_ethnic)table(remir_us$subg_signif1_ethnic, useNA = "always")##
## Hispanic None <NA>
## 5 2 233table(remir_us$subg_signif2_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic Minority
## 10 3 2
## Minority; Majority None <NA>
## 2 6 217tabyl(remir_us$subg_signif12_ethnic)tabyl(remir_us$inter_examine12_ethnic_cat)tabyl(remir_us, inter_examine12_ethnic, inter_examine12_ethnic_cat)tabyl(remir_us$subg_examine12_ethnic_cat)tabyl(remir_us, subg_examine12_ethnic, subg_examine12_ethnic_cat)tabyl(remir_us$inter_examine12_ethnic_cat)tabyl(remir_us, inter_examine12_ethnic, inter_examine12_ethnic_cat)## INTERACTION ETHNIC BENEFIT
remir_us$inter_benefit12_ethnic_cat <-
ifelse(remir_us$inter_benefit12_ethnic == 'Hispanic', 'Exact',
ifelse(remir_us$inter_benefit12_ethnic == 'Minority', 'Mixed',
ifelse(remir_us$inter_benefit12_ethnic == 'Non-Hispanic' |
remir_us$inter_benefit12_ethnic == 'Majority', 'Other',
ifelse(remir_us$inter_benefit12_ethnic == 'None' &
remir_us$main_effect == 'Yes' &
remir_us$inter_examine12_ethnic_cat == 'Exact', 'Exact',
ifelse(remir_us$inter_benefit12_ethnic == 'None' &
remir_us$main_effect == 'Yes' &
remir_us$inter_examine12_ethnic_cat == 'Mixed', 'Mixed',
ifelse(remir_us$inter_benefit12_ethnic == 'None' &
remir_us$main_effect == 'No' &
!is.na(remir_us$inter_examine12_ethnic_cat), 'Other', NA))))))
## SUBGROUP ETHNIC BENEFIT
remir_us$subg_signif12_ethnic_cat <-
ifelse(remir_us$subg_signif12_ethnic == "Hispanic" |
remir_us$subg_signif12_ethnic == "Hispanic;Non-Hispanic" , "Exact",
ifelse(remir_us$subg_signif12_ethnic == "None", "Other",
ifelse(remir_us$subg_signif12_ethnic == "Minority" |
remir_us$subg_signif12_ethnic == "Minority; Majority", "Mixed",
ifelse(is.na(remir_us$subg_examine12_ethnic), NA, 'Check'))))
## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_us$inter_benefit12_ethnic_cat <- ifelse(
(remir_us$hisp_all > .75) &
is.na(remir_us$inter_benefit12_ethnic_cat) &
is.na(remir_us$subg_signif12_ethnic_cat),
"Homog", remir_us$inter_benefit12_ethnic_cat)
## Change 3980 and 3981 to NA so that examined and benefit are consistent
remir_us$inter_benefit12_ethnic_cat <-
ifelse(remir_us$Citation.ID == 3980, NA, remir_us$inter_benefit12_ethnic_cat)
remir_us$inter_benefit12_ethnic_cat <-
ifelse(remir_us$Citation.ID == 3981, NA, remir_us$inter_benefit12_ethnic_cat)
remir_us$examine12_ethnic_short <-
ifelse(is.na(remir_us$inter_examine12_ethnic_cat) &
is.na(remir_us$subg_examine12_ethnic_cat), 'None',
ifelse(remir_us$inter_examine12_ethnic_cat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_us$examine12_ethnic_short)## Fix remaining NAs -- Should be yes since they are missing for one test only
remir_us$examine12_ethnic_short <- ifelse(
is.na(remir_us$examine12_ethnic_short), 'Yes', remir_us$examine12_ethnic_short)
### GENDER ###
## Column 5: Measured gender according to a binary category of female
# Combine Interaction Examined Gender Variables for all data
names(remir_us)[23] <- "inter_examine1_gender"
names(remir_us)[92] <- "inter_examine2_gender"
remir_us$inter_examine12_gender <- ifelse(is.na(remir_us$inter_examine1_gender),
remir_us$inter_examine2_gender, remir_us$inter_examine1_gender)
# Combine Interaction Significant Gender Variables for all data
names(remir_us)[24] <- "inter_signif1_gender"
names(remir_us)[93] <- "inter_signif2_gender"
remir_us$inter_signif12_gender <- ifelse(is.na(remir_us$inter_signif1_gender),
remir_us$inter_signif2_gender, remir_us$inter_signif1_gender)
# Combine Interaction Benefitted Gender Variables for all data
names(remir_us)[26] <- "inter_benefit1_gender"
names(remir_us)[95] <- "inter_benefit2_gender"
remir_us$inter_benefit12_gender <- ifelse(is.na(remir_us$inter_benefit1_gender),
remir_us$inter_benefit2_gender, remir_us$inter_benefit1_gender)
# Combine Subgroup Examined Gender Variables for all data
names(remir_us)[55] <- "subg_examine1_gender"
names(remir_us)[96] <- "subg_examine2_gender"
remir_us$subg_examine12_gender <- ifelse(is.na(remir_us$subg_examine1_gender),
remir_us$subg_examine2_gender, remir_us$subg_examine1_gender)
# Combine Subgroup Significant Gender Variables for all data
names(remir_us)[57] <- "subg_signif1_gender"
names(remir_us)[98] <- "subg_signif2_gender"
remir_us$subg_signif12_gender <- ifelse(is.na(remir_us$subg_signif1_gender),
remir_us$subg_signif2_gender, remir_us$subg_signif1_gender)
# Create a homogeneous sample variable for interaction examined
remir_us$inter_examine12_gender <- ifelse(
(remir_us$female_all > .75) &
is.na(remir_us$inter_examine12_gender) &
is.na(remir_us$subg_examine12_gender),
"Homog", remir_us$inter_examine12_gender)
# Simplify Interaction Gender Benefitted codes
remir_us$inter_benefit12_gender_cat <- ifelse(
remir_us$inter_benefit12_gender == 'None' & remir_us$main_effect == 'Yes', 'Both',
ifelse(remir_us$inter_benefit12_gender == 'None' & remir_us$main_effect == 'No', 'Neither',
remir_us$inter_benefit12_gender))
# Create a homogeneous sample variable for interaction benefitted
remir_us$inter_benefit12_gender_cat <- ifelse(
(remir_us$female_all > .75) &
is.na(remir_us$inter_benefit12_gender_cat) &
is.na(remir_us$subg_signif12_gender),
"Homog", remir_us$inter_benefit12_gender_cat)
# More simplification for interaction benefitted
remir_us$inter_benefit12_gender_cat2 <- ifelse(
remir_us$inter_benefit12_gender_cat %in% c('Male', 'Neither'), 'Other',
ifelse(remir_us$inter_benefit12_gender_cat %in% c('Both', 'Female'), 'Female',
ifelse(remir_us$inter_benefit12_gender_cat == 'Unclear', NA, remir_us$inter_benefit12_gender_cat)))
# Simplify subgroup significant gender codes
remir_us$subg_signif12_gender_cat <- ifelse(
remir_us$subg_signif12_gender %in% c('Female', 'Male;Female'), 'Female',
ifelse(remir_us$subg_signif12_gender %in% c('Male', 'None'), 'Other',
remir_us$subg_signif12_gender))
# Create a combined variable for examined gender subgroups
remir_us$examine12_gender_short <-
ifelse(is.na(remir_us$inter_examine12_gender) &
is.na(remir_us$subg_examine12_gender), 'None',
ifelse(remir_us$inter_examine12_gender == 'Homog', 'Homog', 'Yes'))
# Fix remaining NAs for examined gender subgroups
remir_us$examine12_gender_short <- ifelse(
is.na(remir_us$examine12_gender_short), 'Yes', remir_us$examine12_gender_short)
# Combine Interaction Examined Gender Variables for all data
names(remir_all)[23] <- "inter_examine1_gender"
names(remir_all)[92] <- "inter_examine2_gender"
remir_all$inter_examine12_gender <- ifelse(is.na(remir_all$inter_examine1_gender),
remir_all$inter_examine2_gender, remir_all$inter_examine1_gender)
# Combine Interaction Significant Gender Variables for all data
names(remir_all)[24] <- "inter_signif1_gender"
names(remir_all)[93] <- "inter_signif2_gender"
remir_all$inter_signif12_gender <- ifelse(is.na(remir_all$inter_signif1_gender),
remir_all$inter_signif2_gender, remir_all$inter_signif1_gender)
# Combine Interaction Benefitted Gender Variables for all data
names(remir_all)[26] <- "inter_benefit1_gender"
names(remir_all)[95] <- "inter_benefit2_gender"
remir_all$inter_benefit12_gender <- ifelse(is.na(remir_all$inter_benefit1_gender),
remir_all$inter_benefit2_gender, remir_all$inter_benefit1_gender)
# Combine Subgroup Examined Gender Variables for all data
names(remir_all)[55] <- "subg_examine1_gender"
names(remir_all)[96] <- "subg_examine2_gender"
remir_all$subg_examine12_gender <- ifelse(is.na(remir_all$subg_examine1_gender),
remir_all$subg_examine2_gender, remir_all$subg_examine1_gender)
# Combine Subgroup Significant Gender Variables for all data
names(remir_all)[57] <- "subg_signif1_gender"
names(remir_all)[98] <- "subg_signif2_gender"
remir_all$subg_signif12_gender <- ifelse(is.na(remir_all$subg_signif1_gender),
remir_all$subg_signif2_gender, remir_all$subg_signif1_gender)
# Create a homogeneous sample variable for interaction examined
remir_all$inter_examine12_gender <- ifelse(
(remir_all$female_all > .75) &
is.na(remir_all$inter_examine12_gender) &
is.na(remir_all$subg_examine12_gender),
"Homog", remir_all$inter_examine12_gender)
# Simplify Interaction Gender Benefitted codes
remir_all$inter_benefit12_gender_cat <- ifelse(
remir_all$inter_benefit12_gender == 'None' & remir_all$main_effect == 'Yes', 'Both',
ifelse(remir_all$inter_benefit12_gender == 'None' & remir_all$main_effect == 'No', 'Neither',
remir_all$inter_benefit12_gender))
# Create a homogeneous sample variable for interaction benefitted
remir_all$inter_benefit12_gender_cat <- ifelse(
(remir_all$female_all > .75) &
is.na(remir_all$inter_benefit12_gender_cat) &
is.na(remir_all$subg_signif12_gender),
"Homog", remir_all$inter_benefit12_gender_cat)
# More simplification for interaction benefitted
remir_all$inter_benefit12_gender_cat2 <- ifelse(
remir_all$inter_benefit12_gender_cat %in% c('Male', 'Neither'), 'Other',
ifelse(remir_all$inter_benefit12_gender_cat %in% c('Both', 'Female'), 'Female',
ifelse(remir_all$inter_benefit12_gender_cat == 'Unclear', NA, remir_all$inter_benefit12_gender_cat)))
# Simplify subgroup significant gender codes
remir_all$subg_signif12_gender_cat <- ifelse(
remir_all$subg_signif12_gender %in% c('Female', 'Male;Female'), 'Female',
ifelse(remir_all$subg_signif12_gender %in% c('Male', 'None'), 'Other',
remir_all$subg_signif12_gender))
# Generate summary tables
tabyl(remir_us, inter_examine12_gender, subg_examine12_gender) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)tabyl(remir_us$examine12_gender_short)tabyl(remir_us$inter_examine12_gender)tabyl(remir_us$inter_signif12_gender)tabyl(remir_us$inter_benefit12_gender)tabyl(remir_us$subg_examine12_gender)tabyl(remir_us$subg_signif12_gender)# Tables for all data
tabyl(remir_all$inter_examine12_gender)tabyl(remir_all$inter_signif12_gender)tabyl(remir_all$inter_benefit12_gender)tabyl(remir_all$subg_examine12_gender)tabyl(remir_all$subg_signif12_gender)# Check simplified interaction benefitted codes
tabyl(remir_all, inter_benefit12_gender, inter_benefit12_gender_cat)tabyl(remir_all$inter_benefit12_gender_cat)tabyl(remir_all, inter_benefit12_gender_cat, inter_benefit12_gender_cat2)# Check simplified subgroup significant gender codes
tabyl(remir_all, subg_signif12_gender, subg_signif12_gender_cat)tabyl(remir_all$subg_signif12_gender_cat)### SEXUAL IDENTITY ###
## Change Interaction Examined Variable Names
names(remir_us)[27] <- "inter_examine1_sex"
names(remir_us)[100] <- "inter_examine2_sex"
remir_us$inter_examine12_sex <- ifelse(is.na(remir_us$inter_examine1_sex),
remir_us$inter_examine2_sex, remir_us$inter_examine1_sex)
## Change Interaction Significant Variable Names
names(remir_us)[28] <- "inter_signif1_sex"
names(remir_us)[101] <- "inter_signif2_sex"
remir_us$inter_signif12_sex <- ifelse(is.na(remir_us$inter_signif1_sex),
remir_us$inter_signif2_sex, remir_us$inter_signif1_sex)
## Change Interactions Benefitted Variable Names
names(remir_us)[30] <- "inter_benefit1_sex"
names(remir_us)[103] <- "inter_benefit2_sex"
remir_us$inter_benefit12_sex <- ifelse(is.na(remir_us$inter_benefit1_sex),
remir_us$inter_benefit2_sex, remir_us$inter_benefit1_sex)
## Change Subgroup Examined Variable Names
names(remir_us)[59] <- "subg_examine1_sex"
names(remir_us)[104] <- "subg_examine2_sex"
remir_us$subg_examine12_sex <- ifelse(is.na(remir_us$subg_examine1_sex),
remir_us$subg_examine2_sex, remir_us$subg_examine1_sex)
## Change Subgroup Significant Variable Names
names(remir_us)[61] <- "subg_signif1_sex"
names(remir_us)[106] <- "subg_signif2_sex"
remir_us$subg_signif12_sex <- ifelse(is.na(remir_us$subg_signif1_sex),
remir_us$subg_signif2_sex, remir_us$subg_signif1_sex)
## Create a short variable for examined sex subgroups
remir_us$examine12_sex_short <-
ifelse(is.na(remir_us$inter_examine12_sex), 'No', remir_us$inter_examine12_sex)
## Combine Interaction Benefitted Variables for Category
remir_us$inter_benefit12_sex <- ifelse(is.na(remir_us$inter_benefit1_sex),
remir_us$inter_benefit2_sex, remir_us$inter_benefit1_sex)
remir_us$inter_benefit12_sex_cat <- ifelse(remir_us$inter_benefit12_sex == 'None' &
remir_us$main_effect == 'Yes', 'Yes', remir_us$inter_benefit12_sex)
# Generate summary tables
table(remir_us$inter_examine1_sex, useNA = "always")##
## No Yes <NA>
## 40 2 198table(remir_us$inter_examine2_sex, useNA = "always")##
## <NA>
## 240tabyl(remir_us$inter_examine12_sex)table(remir_us$inter_signif1_sex, useNA = "always")##
## None <NA>
## 2 238table(remir_us$inter_signif2_sex, useNA = "always")##
## <NA>
## 240tabyl(remir_us$inter_signif12_sex)table(remir_us$inter_benefit1_sex, useNA = "always")##
## None <NA>
## 2 238table(remir_us$inter_benefit2_sex, useNA = "always")##
## <NA>
## 240tabyl(remir_us$inter_benefit12_sex)table(remir_us$subg_examine1_sex, useNA = "always")##
## <NA>
## 240table(remir_us$subg_examine2_sex, useNA = "always")##
## <NA>
## 240tabyl(remir_us$subg_examine12_sex)table(remir_us$subg_signif1_sex, useNA = "always")##
## <NA>
## 240table(remir_us$subg_signif2_sex, useNA = "always")##
## <NA>
## 240tabyl(remir_us$subg_signif12_sex)tabyl(remir_us, inter_examine12_sex, subg_examine12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)tabyl(remir_us$examine12_sex_short)tabyl(remir_us$inter_benefit12_sex)tabyl(remir_us$inter_benefit12_sex_cat)### Economic Disadvantage ###
# Change Interaction SES Variable Names
names(remir_us)[31] <- "inter_examine1_ses"
names(remir_us)[108] <- "inter_examine2_ses"
# Combine Two Interactions Examined
remir_us$inter_examine12_ses <- ifelse(is.na(remir_us$inter_examine1_ses),
remir_us$inter_examine2_ses, remir_us$inter_examine1_ses)
remir_us$inter_examine12_ses_cat <- ifelse(
remir_us$inter_examine12_ses == 'No', NA, remir_us$inter_examine12_ses)
# Change Interaction Significant Variable Names
names(remir_us)[32] <- "inter_signif1_ses"
names(remir_us)[109] <- "inter_signif2_ses"
# Combine Two Interactions Significant
remir_us$inter_signif12_ses <- ifelse(is.na(remir_us$inter_signif1_ses),
remir_us$inter_signif2_ses, remir_us$inter_signif1_ses)
# Change Subgroups Significant Variable Names
names(remir_us)[65] <- "subg_signif1_ses"
names(remir_us)[114] <- "subg_signif2_ses"
# Combine Two Interactions Significant
remir_us$subg_signif12_ses <- ifelse(is.na(remir_us$subg_signif1_ses),
remir_us$subg_signif2_ses, remir_us$subg_signif1_ses)
remir_us$subg_signif12_ses_cat <- ifelse(remir_us$subg_signif12_ses == 'High' |
remir_us$subg_signif12_ses == 'None', 'Other',
ifelse(remir_us$subg_signif12_ses == 'Low' |
remir_us$subg_signif12_ses == 'Low;High', 'Low',
remir_us$subg_signif12_ses))
# Change Interaction Benefitted Variable Names
names(remir_us)[34] <- "inter_benefit1_ses"
names(remir_us)[111] <- "inter_benefit2_ses"
# Combine Two Interactions Benefitted
remir_us$inter_benefit12_ses <- ifelse(is.na(remir_us$inter_benefit1_ses),
remir_us$inter_benefit2_ses, remir_us$inter_benefit1_ses)
remir_us$inter_benefit12_ses_cat <- ifelse(remir_us$inter_benefit12_ses == 'None' &
remir_us$main_effect == 'Yes', 'Both',
ifelse(remir_us$inter_benefit12_ses == 'None' &
remir_us$main_effect == 'No', 'Neither',
remir_us$inter_benefit12_ses))
# Change Subgroups Examined Variable Names
names(remir_us)[63] <- "subg_examine1_ses"
names(remir_us)[112] <- "subg_examine2_ses"
# Combine Two Subgroups Examined
remir_us$subg_examine12_ses <- ifelse(is.na(remir_us$subg_examine1_ses),
remir_us$subg_examine2_ses, remir_us$subg_examine1_ses)
# Separate Out Homogeneous Sample
remir_us$inter_benefit12_ses_cat <- ifelse(
(remir_us$econ_all > .75) & is.na(remir_us$inter_benefit12_ses_cat) &
is.na(remir_us$subg_signif12_ses), "Homog", remir_us$inter_benefit12_ses_cat)
# Simplify inter_benefit12_ses_cat
remir_us$inter_benefit12_ses_cat2 <- ifelse(remir_us$inter_benefit12_ses_cat == 'Unclear', NA,
ifelse(remir_us$inter_benefit12_ses_cat == 'Both' |
remir_us$inter_benefit12_ses_cat == 'Low', 'Low',
remir_us$inter_benefit12_ses_cat))
# Simplify subg_signif12_ses_cat
remir_us$subg_signif12_ses_cat <- ifelse(remir_us$subg_signif12_ses == 'High' |
remir_us$subg_signif12_ses == 'None', 'Other',
ifelse(remir_us$subg_signif12_ses == 'Low' |
remir_us$subg_signif12_ses == 'Low;High', 'Low',
remir_us$subg_signif12_ses))
# Simplify examine12_ses
remir_us$examine12_ses_short <- ifelse(
is.na(remir_us$inter_examine12_ses_cat) & is.na(remir_us$subg_examine12_ses), 'None',
ifelse(remir_us$inter_examine12_ses_cat == 'Homog', 'Homog', 'Yes'))
# Fix remaining NAs
remir_us$examine12_ses_short <- ifelse(
is.na(remir_us$examine12_ses_short), 'Yes', remir_us$examine12_ses_short)
# Generate summary tables for interactions and subgroups examined SES
tabyl(remir_us$inter_examine12_ses)tabyl(remir_us$subg_examine12_ses)# Compute examined proportions for low SES
table(remir_us$inter_examine12_ses_cat, remir_us$subg_examine12_ses, useNA = "always")##
## Low Low;High <NA>
## Low/High 2 8 0
## Yes 0 0 10
## <NA> 2 6 212tabyl(remir_us, inter_examine12_ses_cat, subg_examine12_ses) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)# Generate summary tables for examine12_ses_short
tabyl(remir_us$examine12_ses_short)# Generate summary tables for interactions benefitted SES
tabyl(remir_us, inter_benefit12_ses, inter_examine12_ses)tabyl(remir_us, inter_benefit12_ses, inter_benefit12_ses_cat)# Subgroup SES Benefitted
tabyl(remir_us$subg_signif12_ses)# Separate Out Homogeneous Sample
tabyl(remir_us$inter_benefit12_ses_cat)# Print homogeneous sample summary for remir_us
tabyl(remir_us$inter_benefit12_ses_cat2)### LOCATION ###
## Change Interaction Examined Variable Names
names(remir_all)[36] <- "inter_examine1_loc"
names(remir_all)[116] <- "inter_examine2_loc"
# Combine Two Interactions Examined
remir_all$inter_examine12_loc <- ifelse(is.na(remir_all$inter_examine1_loc),
remir_all$inter_examine2_loc, remir_all$inter_examine1_loc)
## Change Interaction Examined Variable Names
names(remir_us)[36] <- "inter_examine1_loc"
names(remir_us)[116] <- "inter_examine2_loc"
# Combine Two Interactions Examined
remir_us$inter_examine12_loc <- ifelse(is.na(remir_us$inter_examine1_loc),
remir_us$inter_examine2_loc, remir_us$inter_examine1_loc)
############ What Location Interactions Were Significant ################
## Change Interaction Significant Variable Names
names(remir_us)[37] <- "inter_signif1_loc"
names(remir_us)[117] <- "inter_signif2_loc"
# Combine Two Interactions Significant
remir_us$inter_signif12_loc <- ifelse(is.na(remir_us$inter_signif1_loc),
remir_us$inter_signif2_loc, remir_us$inter_signif1_loc)
############ What Location Interaction Groups Benefited More #############
## Change Interaction Benefitted Variable Names
names(remir_us)[39] <- "inter_benefit1_loc"
names(remir_us)[119] <- "inter_benefit2_loc"
# Combine Two Interactions Benefitted
remir_us$inter_benefit12_loc <- ifelse(is.na(remir_us$inter_benefit1_loc),
remir_us$inter_benefit2_loc, remir_us$inter_benefit1_loc)
############# What Location Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[67] <- "subg_examine1_loc"
names(remir_all)[120] <- "subg_examine2_loc"
# Combine Two Subgroups Examined
remir_all$subg_examine12_loc <- ifelse(is.na(remir_all$subg_examine1_loc),
remir_all$subg_examine2_loc, remir_all$subg_examine1_loc)
## Change Subgroup Examined Variable Names
names(remir_us)[67] <- "subg_examine1_loc"
names(remir_us)[120] <- "subg_examine2_loc"
# Combine Two Subgroups Examined
remir_us$subg_examine12_loc <- ifelse(is.na(remir_us$subg_examine1_loc),
remir_us$subg_examine2_loc, remir_us$subg_examine1_loc)
############## What Location Subgroups Were Significant ####################
## Change Subgroup Significant Variable Names
names(remir_us)[69] <- "subg_signif1_loc"
names(remir_us)[122] <- "subg_signif2_loc"
# Combine Two Interactions Significant
remir_us$subg_signif12_loc <- ifelse(is.na(remir_us$subg_signif1_loc),
remir_us$subg_signif2_loc, remir_us$subg_signif1_loc)
# Simplify examine12_loc for all
remir_all$examine12_loc_short <-
ifelse(is.na(remir_all$inter_examine12_loc) &
is.na(remir_all$subg_examine12_loc), 'None',
ifelse(remir_all$inter_examine12_loc == 'Homog', 'Homog', 'Yes'))
# Fix remaining NAs for all
remir_all$examine12_loc_short <- ifelse(
is.na(remir_all$examine12_loc_short), 'Yes', remir_all$examine12_loc_short)
# Simplify examine12_loc for US
remir_us$examine12_loc_short <-
ifelse(is.na(remir_us$inter_examine12_loc) &
is.na(remir_us$subg_examine12_loc), 'None',
ifelse(remir_us$inter_examine12_loc == 'Homog', 'Homog', 'Yes'))
# Fix remaining NAs for US
remir_us$examine12_loc_short <- ifelse(
is.na(remir_us$examine12_loc_short), 'Yes', remir_us$examine12_loc_short)
# Generate summary tables for interactions and subgroups examined location for all
tabyl(remir_all$inter_examine12_loc)tabyl(remir_all$subg_examine12_loc)# Compute examined proportions for location for all
table(remir_all$inter_examine12_loc, remir_all$subg_examine12_loc, useNA = "always")##
## Rural Urban;Non-Urban <NA>
## Urban/Non-Urban 0 2 0
## <NA> 2 0 288tabyl(remir_all, inter_examine12_loc, subg_examine12_loc) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)# Generate summary tables for examine12_loc_short for all
tabyl(remir_all$examine12_loc_short)# Generate summary tables for interactions and subgroups examined location for US
tabyl(remir_us$inter_examine12_loc)tabyl(remir_us$subg_examine12_loc)### Nativity ###
# What Nativity Interaction Groups Were Examined
## Change Interaction Nativity Variable Names
names(remir_us)[41] <- "inter_examine1_nat"
names(remir_us)[124] <- "inter_examine2_nat"
## Change Interaction Nativity Variable Names
names(remir_all)[41] <- "inter_examine1_nat"
names(remir_all)[124] <- "inter_examine2_nat"
# Combine Two Interactions Examined
remir_all$inter_examine12_nat <- ifelse(is.na(remir_all$inter_examine1_nat),
remir_all$inter_examine2_nat, remir_all$inter_examine1_nat)
remir_us$inter_examine12_nat <- ifelse(is.na(remir_us$inter_examine1_nat),
remir_us$inter_examine2_nat, remir_us$inter_examine1_nat)
############## What Nativity Interactions Were Significant
## Change Interaction Significant Variable Names
names(remir_us)[42] <- "inter_signif1_nat"
names(remir_us)[125] <- "inter_signif2_nat"
# Combine Two Interactions Significant
remir_us$inter_signif12_nat <- ifelse(is.na(remir_us$inter_signif1_nat),
remir_us$inter_signif2_nat, remir_us$inter_signif1_nat)
############## What Nativity Interaction Groups Benefited More
## Change Interaction Benefitted Variable Names
names(remir_us)[44] <- "inter_benefit1_nat"
names(remir_us)[127] <- "inter_benefit2_nat"
# Combine Two Interactions Benefitted
remir_us$inter_benefit12_nat <- ifelse(is.na(remir_us$inter_benefit1_nat),
remir_us$inter_benefit2_nat, remir_us$inter_benefit1_nat)
############# What Nativity Subgroups Were Examined
## Change Subgroup Examined Variable Names
names(remir_us)[71] <- "subg_examine1_nat"
names(remir_us)[128] <- "subg_examine2_nat"
names(remir_all)[71] <- "subg_examine1_nat"
names(remir_all)[128] <- "subg_examine2_nat"
# Combine Two Subgroups Examined
remir_us$subg_examine12_nat <- ifelse(is.na(remir_us$subg_examine1_nat),
remir_us$subg_examine2_nat, remir_us$subg_examine1_nat)
remir_all$subg_examine12_nat <- ifelse(is.na(remir_all$subg_examine1_nat),
remir_all$subg_examine2_nat, remir_all$subg_examine1_nat)
############## What Nativity Subgroups Were Significant
## Change Subgroup Significant Variable Names
names(remir_us)[73] <- "subg_signif1_nat"
names(remir_us)[130] <- "subg_signif2_nat"
# Combine Two Interactions Significant
remir_us$subg_signif12_nat <- ifelse(is.na(remir_us$subg_signif1_nat),
remir_us$subg_signif2_nat, remir_us$subg_signif1_nat)
# Simplify examine12_nat for US
remir_us$examine12_nat_short <-
ifelse(is.na(remir_us$inter_examine12_nat) &
is.na(remir_us$subg_examine12_nat), 'None',
ifelse(remir_us$inter_examine12_nat == 'Homog', 'Homog', 'Yes'))
# Fix remaining NAs for US
remir_us$examine12_nat_short <- ifelse(
is.na(remir_us$examine12_nat_short), 'Yes', remir_us$examine12_nat_short)# Interaction Race Examined
names(remir_us_subgroup)[13] <- "inter_examine1_race"
names(remir_us_subgroup)[76] <- "inter_examine2_race"
remir_us_subgroup$inter_examine12_race <- ifelse(is.na(remir_us_subgroup$inter_examine1_race), remir_us_subgroup$inter_examine2_race, remir_us_subgroup$inter_examine1_race)
# Simplify Codes
remir_us_subgroup$inter_examine12_race_cat <- remir_us_subgroup$inter_examine12_race
remir_us_subgroup$inter_examine12_race_cat <- gsub('Minority/Majority|Minority/majority|White/Minority|White/Not White|White/Non-White|White/Other', 'Mixed', remir_us_subgroup$inter_examine12_race_cat)
remir_us_subgroup$inter_examine12_race_cat <- ifelse(remir_us_subgroup$inter_examine12_race_cat == 'None', NA, remir_us_subgroup$inter_examine12_race_cat)
remir_us_subgroup$inter_examine12_race_cat <- ifelse(remir_us_subgroup$inter_examine12_race_cat != "Mixed", "Exact", remir_us_subgroup$inter_examine12_race_cat)
# Interaction Race Significant
names(remir_us_subgroup)[14] <- "inter_signif1_race"
names(remir_us_subgroup)[77] <- "inter_signif2_race"
remir_us_subgroup$inter_signif12_race <- ifelse(is.na(remir_us_subgroup$inter_signif1_race), remir_us_subgroup$inter_signif2_race, remir_us_subgroup$inter_signif1_race)
# Interaction Race Benefitted
names(remir_us_subgroup)[16] <- "inter_benefit1_race"
names(remir_us_subgroup)[79] <- "inter_benefit2_race"
remir_us_subgroup$inter_benefit12_race <- ifelse(is.na(remir_us_subgroup$inter_benefit1_race), remir_us_subgroup$inter_benefit2_race, remir_us_subgroup$inter_benefit1_race)
# Subgroup Race Examined
names(remir_us_subgroup)[47] <- "subg_examine1_race"
names(remir_us_subgroup)[80] <- "subg_examine2_race"
remir_us_subgroup$subg_examine12_race <- ifelse(is.na(remir_us_subgroup$subg_examine1_race), remir_us_subgroup$subg_examine2_race, remir_us_subgroup$subg_examine1_race)
remir_us_subgroup$subg_examine12_race_cat <- ifelse(remir_us_subgroup$subg_examine12_race %in% c("White;Non-White", "Minority", "Minority; Majority"), "Mixed", ifelse(remir_us_subgroup$subg_examine12_race == "None", NA, "Exact"))
# Subgroup Race Significant
names(remir_us_subgroup)[49] <- "subg_signif1_race"
names(remir_us_subgroup)[82] <- "subg_signif2_race"
remir_us_subgroup$subg_signif12_race <- ifelse(is.na(remir_us_subgroup$subg_signif1_race), remir_us_subgroup$subg_signif2_race, remir_us_subgroup$subg_signif1_race)
remir_us_subgroup$subg_signif12_race_cat <- ifelse(is.na(remir_us_subgroup$subg_examine12_race_cat), NA, ifelse(remir_us_subgroup$subg_signif12_race %in% c("White;Non-White", "Minority", "Minority; Majority"), "Mixed", ifelse(remir_us_subgroup$subg_signif12_race %in% c("None", "White"), "Other", "Exact")))
# Homogeneous Sample
remir_us_subgroup$inter_examine12_race_cat <- ifelse((remir_us_subgroup$black_all > .75 | remir_us_subgroup$asian_all > .75 | remir_us_subgroup$native_all > .75 | remir_us_subgroup$pacif_all > .75) & is.na(remir_us_subgroup$inter_examine12_race_cat) & is.na(remir_us_subgroup$subg_examine12_race_cat), "Homog", remir_us_subgroup$inter_examine12_race_cat)
# Interaction Race Benefit Category
remir_us_subgroup$inter_benefit12_race_cat <- ifelse(remir_us_subgroup$inter_benefit12_race %in% c('Asian or Asian American;Minority', 'Black or African American;Other or Multiracial', 'Black or African American', 'Black or African American;White'), 'Exact', ifelse(remir_us_subgroup$inter_benefit12_race %in% c('Minority', 'White + Another Race'), 'Mixed', ifelse(remir_us_subgroup$inter_benefit12_race %in% c('Not Black', 'White', 'Majority'), 'Other', ifelse(remir_us_subgroup$inter_benefit12_race == 'None' & remir_us_subgroup$main_effect == 'Yes' & remir_us_subgroup$inter_examine12_race_cat == 'Exact', 'Exact', ifelse(remir_us_subgroup$inter_benefit12_race == 'None' & remir_us_subgroup$main_effect == 'Yes' & remir_us_subgroup$inter_examine12_race_cat == 'Mixed', 'Mixed', ifelse(remir_us_subgroup$inter_benefit12_race == 'None' & remir_us_subgroup$main_effect == 'No' & !is.na(remir_us_subgroup$inter_examine12_race_cat), 'Other', NA))))))
# Homogeneous Sample for Benefit
remir_us_subgroup$inter_benefit12_race_cat <- ifelse((remir_us_subgroup$black_all > .75 | remir_us_subgroup$asian_all > .75 | remir_us_subgroup$native_all > .75 | remir_us_subgroup$pacif_all > .75) & is.na(remir_us_subgroup$inter_benefit12_race_cat) & is.na(remir_us_subgroup$subg_signif12_race_cat), "Homog", remir_us_subgroup$inter_benefit12_race_cat)
# Simplified Short Race Examined Category
remir_us_subgroup$examine12_race_short <- ifelse(is.na(remir_us_subgroup$inter_examine12_race_cat) & is.na(remir_us_subgroup$subg_examine12_race_cat), 'None', ifelse(remir_us_subgroup$inter_examine12_race_cat == 'Homog', 'Homog', 'Yes'))
remir_us_subgroup$examine12_race_short <- ifelse(is.na(remir_us_subgroup$examine12_race_short), 'Yes', remir_us_subgroup$examine12_race_short)
### ETHNICITY (Table 2, Column 3&4) ###
# PREP: Creation of New Variables
############## What Ethnic Interaction Groups Examined ################
## Change Interaction Examined Variable Names
names(remir_us_subgroup)[18] <- "inter_examine1_ethnic"
names(remir_us_subgroup)[84] <- "inter_examine2_ethnic"
## Combine Two Interactions Examined
remir_us_subgroup$inter_examine12_ethnic <- ifelse(is.na(remir_us_subgroup$inter_examine1_ethnic),
remir_us_subgroup$inter_examine2_ethnic, remir_us_subgroup$inter_examine1_ethnic)
############## What Ethnic Interactions Were Significant ###############
## Change Interaction Significant Variable Names
names(remir_us_subgroup)[19] <- "inter_signif1_ethnic"
names(remir_us_subgroup)[85] <- "inter_signif2_ethnic"
## Combine Two Interactions Significant
remir_us_subgroup$inter_signif12_ethnic <- ifelse(is.na(remir_us_subgroup$inter_signif1_ethnic),
remir_us_subgroup$inter_signif2_ethnic, remir_us_subgroup$inter_signif1_ethnic)
############## What Ethnic Interaction Groups Benefited More ############
## Change Interaction Benefitted Variable Names
names(remir_us_subgroup)[21] <- "inter_benefit1_ethnic"
names(remir_us_subgroup)[87] <- "inter_benefit2_ethnic"
## Combine Two Interactions Benefitted
remir_us_subgroup$inter_benefit12_ethnic <- ifelse(is.na(remir_us_subgroup$inter_benefit1_ethnic),
remir_us_subgroup$inter_benefit2_ethnic, remir_us_subgroup$inter_benefit1_ethnic)
############# What Ethnic Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_us_subgroup)[51] <- "subg_examine1_ethnic"
names(remir_us_subgroup)[88] <- "subg_examine2_ethnic"
## Combine Two Subgroups Examined
remir_us_subgroup$subg_examine12_ethnic <- ifelse(is.na(remir_us_subgroup$subg_examine1_ethnic),
remir_us_subgroup$subg_examine2_ethnic, remir_us_subgroup$subg_examine1_ethnic)
############## What Ethnic Subgroups Were Significant ##################
## Change Subgroup Significant Variable Names
names(remir_us_subgroup)[53] <- "subg_signif1_ethnic"
names(remir_us_subgroup)[90] <- "subg_signif2_ethnic"
## Combine Two Interactions Significant
remir_us_subgroup$subg_signif12_ethnic <- ifelse(is.na(remir_us_subgroup$subg_signif1_ethnic),
remir_us_subgroup$subg_signif2_ethnic, remir_us_subgroup$subg_signif1_ethnic)
## INTERACTION ETHNIC EXAMINED
remir_us_subgroup$inter_examine12_ethnic_cat <-
car::recode(remir_us_subgroup$inter_examine12_ethnic, "
'Hispanic/Non-Hispanic' = 'Exact';
'None' = NA ")
remir_us_subgroup$inter_examine12_ethnic_cat <-
ifelse(remir_us_subgroup$inter_examine12_ethnic_cat != "Exact", "Mixed",
remir_us_subgroup$inter_examine12_ethnic_cat)
## SUBGROUP ETHNIC EXAMINED
remir_us_subgroup$subg_examine12_ethnic_cat <-
ifelse(remir_us_subgroup$subg_examine12_ethnic == 'Minority'|
remir_us_subgroup$subg_examine12_ethnic == 'Minority; Majority',
'Mixed',
ifelse(remir_us_subgroup$subg_examine12_ethnic == 'None', NA,
'Exact'))
## SEPARATE OUT HOMOGENEOUS SAMPLE OF HISPANICS
remir_us_subgroup$inter_examine12_ethnic_cat <- ifelse(
(remir_us_subgroup$hisp_all > .75) &
is.na(remir_us_subgroup$inter_examine12_ethnic_cat) &
is.na(remir_us_subgroup$subg_examine12_ethnic_cat),
"Homog", remir_us_subgroup$inter_examine12_ethnic_cat)
# Check Variables
table(remir_us_subgroup$inter_examine1_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/majority Minority/Majority
## 5 2 1
## White/Minority White/Non-White <NA>
## 3 1 88table(remir_us_subgroup$inter_examine2_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/Majority None
## 14 5 2
## White/Minority White/Other <NA>
## 1 1 77tabyl(remir_us_subgroup$inter_examine12_ethnic)table(remir_us_subgroup$inter_signif1_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic None White/Minority
## 2 11 3
## <NA>
## 84table(remir_us_subgroup$inter_signif2_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/Majority None
## 6 2 14
## White/Minority <NA>
## 1 77tabyl(remir_us_subgroup$inter_signif12_ethnic)table(remir_us_subgroup$inter_benefit1_ethnic, useNA = "always")##
## Hispanic Minority None <NA>
## 2 3 11 84table(remir_us_subgroup$inter_benefit2_ethnic, useNA = "always")##
## Hispanic Majority Minority Non-Hispanic None <NA>
## 3 2 1 3 14 77tabyl(remir_us_subgroup$inter_benefit12_ethnic)table(remir_us_subgroup$subg_examine1_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic <NA>
## 6 1 93table(remir_us_subgroup$subg_examine2_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic Minority
## 12 3 2
## Minority; Majority None <NA>
## 2 4 77tabyl(remir_us_subgroup$subg_examine12_ethnic)table(remir_us_subgroup$subg_signif1_ethnic, useNA = "always")##
## Hispanic None <NA>
## 5 2 93table(remir_us_subgroup$subg_signif2_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic Minority
## 10 3 2
## Minority; Majority None <NA>
## 2 6 77tabyl(remir_us_subgroup$subg_signif12_ethnic)tabyl(remir_us_subgroup$inter_examine12_ethnic_cat)tabyl(remir_us_subgroup, inter_examine12_ethnic, inter_examine12_ethnic_cat)tabyl(remir_us_subgroup$subg_examine12_ethnic_cat)tabyl(remir_us_subgroup, subg_examine12_ethnic, subg_examine12_ethnic_cat)tabyl(remir_us_subgroup$inter_examine12_ethnic_cat)tabyl(remir_us_subgroup, inter_examine12_ethnic, inter_examine12_ethnic_cat)## INTERACTION ETHNIC BENEFIT
remir_us_subgroup$inter_benefit12_ethnic_cat <-
ifelse(remir_us_subgroup$inter_benefit12_ethnic == 'Hispanic', 'Exact',
ifelse(remir_us_subgroup$inter_benefit12_ethnic == 'Minority', 'Mixed',
ifelse(remir_us_subgroup$inter_benefit12_ethnic == 'Non-Hispanic' |
remir_us_subgroup$inter_benefit12_ethnic == 'Majority', 'Other',
ifelse(remir_us_subgroup$inter_benefit12_ethnic == 'None' &
remir_us_subgroup$main_effect == 'Yes' &
remir_us_subgroup$inter_examine12_ethnic_cat == 'Exact', 'Exact',
ifelse(remir_us_subgroup$inter_benefit12_ethnic == 'None' &
remir_us_subgroup$main_effect == 'Yes' &
remir_us_subgroup$inter_examine12_ethnic_cat == 'Mixed', 'Mixed',
ifelse(remir_us_subgroup$inter_benefit12_ethnic == 'None' &
remir_us_subgroup$main_effect == 'No' &
!is.na(remir_us_subgroup$inter_examine12_ethnic_cat), 'Other', NA))))))
## SUBGROUP ETHNIC BENEFIT
remir_us_subgroup$subg_signif12_ethnic_cat <-
ifelse(remir_us_subgroup$subg_signif12_ethnic == "Hispanic" |
remir_us_subgroup$subg_signif12_ethnic == "Hispanic;Non-Hispanic" , "Exact",
ifelse(remir_us_subgroup$subg_signif12_ethnic == "None", "Other",
ifelse(remir_us_subgroup$subg_signif12_ethnic == "Minority" |
remir_us_subgroup$subg_signif12_ethnic == "Minority; Majority", "Mixed",
ifelse(is.na(remir_us_subgroup$subg_examine12_ethnic), NA, 'Check'))))
## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_us_subgroup$inter_benefit12_ethnic_cat <- ifelse(
(remir_us_subgroup$hisp_all > .75) &
is.na(remir_us_subgroup$inter_benefit12_ethnic_cat) &
is.na(remir_us_subgroup$subg_signif12_ethnic_cat),
"Homog", remir_us_subgroup$inter_benefit12_ethnic_cat)
## Change 3980 and 3981 to NA so that examined and benefit are consistent
remir_us_subgroup$inter_benefit12_ethnic_cat <-
ifelse(remir_us_subgroup$Citation.ID == 3980, NA, remir_us_subgroup$inter_benefit12_ethnic_cat)
remir_us_subgroup$inter_benefit12_ethnic_cat <-
ifelse(remir_us_subgroup$Citation.ID == 3981, NA, remir_us_subgroup$inter_benefit12_ethnic_cat)
remir_us_subgroup$examine12_ethnic_short <-
ifelse(is.na(remir_us_subgroup$inter_examine12_ethnic_cat) &
is.na(remir_us_subgroup$subg_examine12_ethnic_cat), 'None',
ifelse(remir_us_subgroup$inter_examine12_ethnic_cat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_us_subgroup$examine12_ethnic_short)## Fix remaining NAs -- Should be yes since they are missing for one test only
remir_us_subgroup$examine12_ethnic_short <- ifelse(
is.na(remir_us_subgroup$examine12_ethnic_short), 'Yes', remir_us_subgroup$examine12_ethnic_short)
### GENDER ###
## Column 5: Measured gender according to a binary category of female
# Combine Interaction Examined Gender Variables for all data
names(remir_us_subgroup)[23] <- "inter_examine1_gender"
names(remir_us_subgroup)[92] <- "inter_examine2_gender"
remir_us_subgroup$inter_examine12_gender <- ifelse(is.na(remir_us_subgroup$inter_examine1_gender),
remir_us_subgroup$inter_examine2_gender, remir_us_subgroup$inter_examine1_gender)
# Combine Interaction Significant Gender Variables for all data
names(remir_us_subgroup)[24] <- "inter_signif1_gender"
names(remir_us_subgroup)[93] <- "inter_signif2_gender"
remir_us_subgroup$inter_signif12_gender <- ifelse(is.na(remir_us_subgroup$inter_signif1_gender),
remir_us_subgroup$inter_signif2_gender, remir_us_subgroup$inter_signif1_gender)
# Combine Interaction Benefitted Gender Variables for all data
names(remir_us_subgroup)[26] <- "inter_benefit1_gender"
names(remir_us_subgroup)[95] <- "inter_benefit2_gender"
remir_us_subgroup$inter_benefit12_gender <- ifelse(is.na(remir_us_subgroup$inter_benefit1_gender),
remir_us_subgroup$inter_benefit2_gender, remir_us_subgroup$inter_benefit1_gender)
# Combine Subgroup Examined Gender Variables for all data
names(remir_us_subgroup)[55] <- "subg_examine1_gender"
names(remir_us_subgroup)[96] <- "subg_examine2_gender"
remir_us_subgroup$subg_examine12_gender <- ifelse(is.na(remir_us_subgroup$subg_examine1_gender),
remir_us_subgroup$subg_examine2_gender, remir_us_subgroup$subg_examine1_gender)
# Combine Subgroup Significant Gender Variables for all data
names(remir_us_subgroup)[57] <- "subg_signif1_gender"
names(remir_us_subgroup)[98] <- "subg_signif2_gender"
remir_us_subgroup$subg_signif12_gender <- ifelse(is.na(remir_us_subgroup$subg_signif1_gender),
remir_us_subgroup$subg_signif2_gender, remir_us_subgroup$subg_signif1_gender)
# Create a homogeneous sample variable for interaction examined
remir_us_subgroup$inter_examine12_gender <- ifelse(
(remir_us_subgroup$female_all > .75) &
is.na(remir_us_subgroup$inter_examine12_gender) &
is.na(remir_us_subgroup$subg_examine12_gender),
"Homog", remir_us_subgroup$inter_examine12_gender)
# Simplify Interaction Gender Benefitted codes
remir_us_subgroup$inter_benefit12_gender_cat <- ifelse(
remir_us_subgroup$inter_benefit12_gender == 'None' & remir_us_subgroup$main_effect == 'Yes', 'Both',
ifelse(remir_us_subgroup$inter_benefit12_gender == 'None' & remir_us_subgroup$main_effect == 'No', 'Neither',
remir_us_subgroup$inter_benefit12_gender))
# Create a homogeneous sample variable for interaction benefitted
remir_us_subgroup$inter_benefit12_gender_cat <- ifelse(
(remir_us_subgroup$female_all > .75) &
is.na(remir_us_subgroup$inter_benefit12_gender_cat) &
is.na(remir_us_subgroup$subg_signif12_gender),
"Homog", remir_us_subgroup$inter_benefit12_gender_cat)
# More simplification for interaction benefitted
remir_us_subgroup$inter_benefit12_gender_cat2 <- ifelse(
remir_us_subgroup$inter_benefit12_gender_cat %in% c('Male', 'Neither'), 'Other',
ifelse(remir_us_subgroup$inter_benefit12_gender_cat %in% c('Both', 'Female'), 'Female',
ifelse(remir_us_subgroup$inter_benefit12_gender_cat == 'Unclear', NA, remir_us_subgroup$inter_benefit12_gender_cat)))
# Simplify subgroup significant gender codes
remir_us_subgroup$subg_signif12_gender_cat <- ifelse(
remir_us_subgroup$subg_signif12_gender %in% c('Female', 'Male;Female'), 'Female',
ifelse(remir_us_subgroup$subg_signif12_gender %in% c('Male', 'None'), 'Other',
remir_us_subgroup$subg_signif12_gender))
# Create a combined variable for examined gender subgroups
remir_us_subgroup$examine12_gender_short <-
ifelse(is.na(remir_us_subgroup$inter_examine12_gender) &
is.na(remir_us_subgroup$subg_examine12_gender), 'None',
ifelse(remir_us_subgroup$inter_examine12_gender == 'Homog', 'Homog', 'Yes'))
# Fix remaining NAs for examined gender subgroups
remir_us_subgroup$examine12_gender_short <- ifelse(
is.na(remir_us_subgroup$examine12_gender_short), 'Yes', remir_us_subgroup$examine12_gender_short)
# Combine Interaction Examined Gender Variables for all data
names(remir_us_subgroup)[23] <- "inter_examine1_gender"
names(remir_us_subgroup)[92] <- "inter_examine2_gender"
remir_us_subgroup$inter_examine12_gender <- ifelse(is.na(remir_us_subgroup$inter_examine1_gender),
remir_us_subgroup$inter_examine2_gender, remir_us_subgroup$inter_examine1_gender)
# Combine Interaction Significant Gender Variables for all data
names(remir_us_subgroup)[24] <- "inter_signif1_gender"
names(remir_us_subgroup)[93] <- "inter_signif2_gender"
remir_us_subgroup$inter_signif12_gender <- ifelse(is.na(remir_us_subgroup$inter_signif1_gender),
remir_us_subgroup$inter_signif2_gender, remir_us_subgroup$inter_signif1_gender)
# Combine Interaction Benefitted Gender Variables for all data
names(remir_us_subgroup)[26] <- "inter_benefit1_gender"
names(remir_us_subgroup)[95] <- "inter_benefit2_gender"
remir_us_subgroup$inter_benefit12_gender <- ifelse(is.na(remir_us_subgroup$inter_benefit1_gender),
remir_us_subgroup$inter_benefit2_gender, remir_us_subgroup$inter_benefit1_gender)
# Combine Subgroup Examined Gender Variables for all data
names(remir_us_subgroup)[55] <- "subg_examine1_gender"
names(remir_us_subgroup)[96] <- "subg_examine2_gender"
remir_us_subgroup$subg_examine12_gender <- ifelse(is.na(remir_us_subgroup$subg_examine1_gender),
remir_us_subgroup$subg_examine2_gender, remir_us_subgroup$subg_examine1_gender)
# Combine Subgroup Significant Gender Variables for all data
names(remir_us_subgroup)[57] <- "subg_signif1_gender"
names(remir_us_subgroup)[98] <- "subg_signif2_gender"
remir_us_subgroup$subg_signif12_gender <- ifelse(is.na(remir_us_subgroup$subg_signif1_gender),
remir_us_subgroup$subg_signif2_gender, remir_us_subgroup$subg_signif1_gender)
# Create a homogeneous sample variable for interaction examined
remir_us_subgroup$inter_examine12_gender <- ifelse(
(remir_us_subgroup$female_all > .75) &
is.na(remir_us_subgroup$inter_examine12_gender) &
is.na(remir_us_subgroup$subg_examine12_gender),
"Homog", remir_us_subgroup$inter_examine12_gender)
# Simplify Interaction Gender Benefitted codes
remir_us_subgroup$inter_benefit12_gender_cat <- ifelse(
remir_us_subgroup$inter_benefit12_gender == 'None' & remir_us_subgroup$main_effect == 'Yes', 'Both',
ifelse(remir_us_subgroup$inter_benefit12_gender == 'None' & remir_us_subgroup$main_effect == 'No', 'Neither',
remir_us_subgroup$inter_benefit12_gender))
# Create a homogeneous sample variable for interaction benefitted
remir_us_subgroup$inter_benefit12_gender_cat <- ifelse(
(remir_us_subgroup$female_all > .75) &
is.na(remir_us_subgroup$inter_benefit12_gender_cat) &
is.na(remir_us_subgroup$subg_signif12_gender),
"Homog", remir_us_subgroup$inter_benefit12_gender_cat)
# More simplification for interaction benefitted
remir_us_subgroup$inter_benefit12_gender_cat2 <- ifelse(
remir_us_subgroup$inter_benefit12_gender_cat %in% c('Male', 'Neither'), 'Other',
ifelse(remir_us_subgroup$inter_benefit12_gender_cat %in% c('Both', 'Female'), 'Female',
ifelse(remir_us_subgroup$inter_benefit12_gender_cat == 'Unclear', NA, remir_us_subgroup$inter_benefit12_gender_cat)))
# Simplify subgroup significant gender codes
remir_us_subgroup$subg_signif12_gender_cat <- ifelse(
remir_us_subgroup$subg_signif12_gender %in% c('Female', 'Male;Female'), 'Female',
ifelse(remir_us_subgroup$subg_signif12_gender %in% c('Male', 'None'), 'Other',
remir_us_subgroup$subg_signif12_gender))
# Generate summary tables
tabyl(remir_us_subgroup, inter_examine12_gender, subg_examine12_gender) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)tabyl(remir_us_subgroup$examine12_gender_short)tabyl(remir_us_subgroup$inter_examine12_gender)tabyl(remir_us_subgroup$inter_signif12_gender)tabyl(remir_us_subgroup$inter_benefit12_gender)tabyl(remir_us_subgroup$subg_examine12_gender)tabyl(remir_us_subgroup$subg_signif12_gender)# Tables for all data
tabyl(remir_us_subgroup$inter_examine12_gender)tabyl(remir_us_subgroup$inter_signif12_gender)tabyl(remir_us_subgroup$inter_benefit12_gender)tabyl(remir_us_subgroup$subg_examine12_gender)tabyl(remir_us_subgroup$subg_signif12_gender)# Check simplified interaction benefitted codes
tabyl(remir_us_subgroup, inter_benefit12_gender, inter_benefit12_gender_cat)tabyl(remir_us_subgroup$inter_benefit12_gender_cat)tabyl(remir_us_subgroup, inter_benefit12_gender_cat, inter_benefit12_gender_cat2)# Check simplified subgroup significant gender codes
tabyl(remir_us_subgroup, subg_signif12_gender, subg_signif12_gender_cat)tabyl(remir_us_subgroup$subg_signif12_gender_cat)### SEXUAL IDENTITY ###
## Change Interaction Examined Variable Names
names(remir_us_subgroup)[27] <- "inter_examine1_sex"
names(remir_us_subgroup)[100] <- "inter_examine2_sex"
remir_us_subgroup$inter_examine12_sex <- ifelse(is.na(remir_us_subgroup$inter_examine1_sex),
remir_us_subgroup$inter_examine2_sex, remir_us_subgroup$inter_examine1_sex)
## Change Interaction Significant Variable Names
names(remir_us_subgroup)[28] <- "inter_signif1_sex"
names(remir_us_subgroup)[101] <- "inter_signif2_sex"
remir_us_subgroup$inter_signif12_sex <- ifelse(is.na(remir_us_subgroup$inter_signif1_sex),
remir_us_subgroup$inter_signif2_sex, remir_us_subgroup$inter_signif1_sex)
## Change Interactions Benefitted Variable Names
names(remir_us_subgroup)[30] <- "inter_benefit1_sex"
names(remir_us_subgroup)[103] <- "inter_benefit2_sex"
remir_us_subgroup$inter_benefit12_sex <- ifelse(is.na(remir_us_subgroup$inter_benefit1_sex),
remir_us_subgroup$inter_benefit2_sex, remir_us_subgroup$inter_benefit1_sex)
## Change Subgroup Examined Variable Names
names(remir_us_subgroup)[59] <- "subg_examine1_sex"
names(remir_us_subgroup)[104] <- "subg_examine2_sex"
remir_us_subgroup$subg_examine12_sex <- ifelse(is.na(remir_us_subgroup$subg_examine1_sex),
remir_us_subgroup$subg_examine2_sex, remir_us_subgroup$subg_examine1_sex)
## Change Subgroup Significant Variable Names
names(remir_us_subgroup)[61] <- "subg_signif1_sex"
names(remir_us_subgroup)[106] <- "subg_signif2_sex"
remir_us_subgroup$subg_signif12_sex <- ifelse(is.na(remir_us_subgroup$subg_signif1_sex),
remir_us_subgroup$subg_signif2_sex, remir_us_subgroup$subg_signif1_sex)
## Create a short variable for examined sex subgroups
remir_us_subgroup$examine12_sex_short <-
ifelse(is.na(remir_us_subgroup$inter_examine12_sex), 'No', remir_us_subgroup$inter_examine12_sex)
## Combine Interaction Benefitted Variables for Category
remir_us_subgroup$inter_benefit12_sex <- ifelse(is.na(remir_us_subgroup$inter_benefit1_sex),
remir_us_subgroup$inter_benefit2_sex, remir_us_subgroup$inter_benefit1_sex)
remir_us_subgroup$inter_benefit12_sex_cat <- ifelse(remir_us_subgroup$inter_benefit12_sex == 'None' &
remir_us_subgroup$main_effect == 'Yes', 'Yes', remir_us_subgroup$inter_benefit12_sex)
# Generate summary tables
table(remir_us_subgroup$inter_examine1_sex, useNA = "always")##
## No Yes <NA>
## 40 2 58table(remir_us_subgroup$inter_examine2_sex, useNA = "always")##
## <NA>
## 100tabyl(remir_us_subgroup$inter_examine12_sex)table(remir_us_subgroup$inter_signif1_sex, useNA = "always")##
## None <NA>
## 2 98table(remir_us_subgroup$inter_signif2_sex, useNA = "always")##
## <NA>
## 100tabyl(remir_us_subgroup$inter_signif12_sex)table(remir_us_subgroup$inter_benefit1_sex, useNA = "always")##
## None <NA>
## 2 98table(remir_us_subgroup$inter_benefit2_sex, useNA = "always")##
## <NA>
## 100tabyl(remir_us_subgroup$inter_benefit12_sex)table(remir_us_subgroup$subg_examine1_sex, useNA = "always")##
## <NA>
## 100table(remir_us_subgroup$subg_examine2_sex, useNA = "always")##
## <NA>
## 100tabyl(remir_us_subgroup$subg_examine12_sex)table(remir_us_subgroup$subg_signif1_sex, useNA = "always")##
## <NA>
## 100table(remir_us_subgroup$subg_signif2_sex, useNA = "always")##
## <NA>
## 100tabyl(remir_us_subgroup$subg_signif12_sex)tabyl(remir_us_subgroup, inter_examine12_sex, subg_examine12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)tabyl(remir_us_subgroup$examine12_sex_short)tabyl(remir_us_subgroup$inter_benefit12_sex)tabyl(remir_us_subgroup$inter_benefit12_sex_cat)### Economic Disadvantage ###
# Change Interaction SES Variable Names
names(remir_us_subgroup)[31] <- "inter_examine1_ses"
names(remir_us_subgroup)[108] <- "inter_examine2_ses"
# Combine Two Interactions Examined
remir_us_subgroup$inter_examine12_ses <- ifelse(is.na(remir_us_subgroup$inter_examine1_ses),
remir_us_subgroup$inter_examine2_ses, remir_us_subgroup$inter_examine1_ses)
remir_us_subgroup$inter_examine12_ses_cat <- ifelse(
remir_us_subgroup$inter_examine12_ses == 'No', NA, remir_us_subgroup$inter_examine12_ses)
# Change Interaction Significant Variable Names
names(remir_us_subgroup)[32] <- "inter_signif1_ses"
names(remir_us_subgroup)[109] <- "inter_signif2_ses"
# Combine Two Interactions Significant
remir_us_subgroup$inter_signif12_ses <- ifelse(is.na(remir_us_subgroup$inter_signif1_ses),
remir_us_subgroup$inter_signif2_ses, remir_us_subgroup$inter_signif1_ses)
# Change Subgroups Significant Variable Names
names(remir_us_subgroup)[65] <- "subg_signif1_ses"
names(remir_us_subgroup)[114] <- "subg_signif2_ses"
# Combine Two Interactions Significant
remir_us_subgroup$subg_signif12_ses <- ifelse(is.na(remir_us_subgroup$subg_signif1_ses),
remir_us_subgroup$subg_signif2_ses, remir_us_subgroup$subg_signif1_ses)
remir_us_subgroup$subg_signif12_ses_cat <- ifelse(remir_us_subgroup$subg_signif12_ses == 'High' |
remir_us_subgroup$subg_signif12_ses == 'None', 'Other',
ifelse(remir_us_subgroup$subg_signif12_ses == 'Low' |
remir_us_subgroup$subg_signif12_ses == 'Low;High', 'Low',
remir_us_subgroup$subg_signif12_ses))
# Change Interaction Benefitted Variable Names
names(remir_us_subgroup)[34] <- "inter_benefit1_ses"
names(remir_us_subgroup)[111] <- "inter_benefit2_ses"
# Combine Two Interactions Benefitted
remir_us_subgroup$inter_benefit12_ses <- ifelse(is.na(remir_us_subgroup$inter_benefit1_ses),
remir_us_subgroup$inter_benefit2_ses, remir_us_subgroup$inter_benefit1_ses)
remir_us_subgroup$inter_benefit12_ses_cat <- ifelse(remir_us_subgroup$inter_benefit12_ses == 'None' &
remir_us_subgroup$main_effect == 'Yes', 'Both',
ifelse(remir_us_subgroup$inter_benefit12_ses == 'None' &
remir_us_subgroup$main_effect == 'No', 'Neither',
remir_us_subgroup$inter_benefit12_ses))
# Change Subgroups Examined Variable Names
names(remir_us_subgroup)[63] <- "subg_examine1_ses"
names(remir_us_subgroup)[112] <- "subg_examine2_ses"
# Combine Two Subgroups Examined
remir_us_subgroup$subg_examine12_ses <- ifelse(is.na(remir_us_subgroup$subg_examine1_ses),
remir_us_subgroup$subg_examine2_ses, remir_us_subgroup$subg_examine1_ses)
# Separate Out Homogeneous Sample
remir_us_subgroup$inter_benefit12_ses_cat <- ifelse(
(remir_us_subgroup$econ_all > .75) & is.na(remir_us_subgroup$inter_benefit12_ses_cat) &
is.na(remir_us_subgroup$subg_signif12_ses), "Homog", remir_us_subgroup$inter_benefit12_ses_cat)
# Simplify inter_benefit12_ses_cat
remir_us_subgroup$inter_benefit12_ses_cat2 <- ifelse(remir_us_subgroup$inter_benefit12_ses_cat == 'Unclear', NA,
ifelse(remir_us_subgroup$inter_benefit12_ses_cat == 'Both' |
remir_us_subgroup$inter_benefit12_ses_cat == 'Low', 'Low',
remir_us_subgroup$inter_benefit12_ses_cat))
# Simplify subg_signif12_ses_cat
remir_us_subgroup$subg_signif12_ses_cat <- ifelse(remir_us_subgroup$subg_signif12_ses == 'High' |
remir_us_subgroup$subg_signif12_ses == 'None', 'Other',
ifelse(remir_us_subgroup$subg_signif12_ses == 'Low' |
remir_us_subgroup$subg_signif12_ses == 'Low;High', 'Low',
remir_us_subgroup$subg_signif12_ses))
# Simplify examine12_ses
remir_us_subgroup$examine12_ses_short <- ifelse(
is.na(remir_us_subgroup$inter_examine12_ses_cat) & is.na(remir_us_subgroup$subg_examine12_ses), 'None',
ifelse(remir_us_subgroup$inter_examine12_ses_cat == 'Homog', 'Homog', 'Yes'))
# Fix remaining NAs
remir_us_subgroup$examine12_ses_short <- ifelse(
is.na(remir_us_subgroup$examine12_ses_short), 'Yes', remir_us_subgroup$examine12_ses_short)
# Generate summary tables for interactions and subgroups examined SES
tabyl(remir_us_subgroup$inter_examine12_ses)tabyl(remir_us_subgroup$subg_examine12_ses)# Compute examined proportions for low SES
table(remir_us_subgroup$inter_examine12_ses_cat, remir_us_subgroup$subg_examine12_ses, useNA = "always")##
## Low Low;High <NA>
## Low/High 2 8 0
## Yes 0 0 10
## <NA> 2 6 72tabyl(remir_us_subgroup, inter_examine12_ses_cat, subg_examine12_ses) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)# Generate summary tables for examine12_ses_short
tabyl(remir_us_subgroup$examine12_ses_short)# Generate summary tables for interactions benefitted SES
tabyl(remir_us_subgroup, inter_benefit12_ses, inter_examine12_ses)tabyl(remir_us_subgroup, inter_benefit12_ses, inter_benefit12_ses_cat)# Subgroup SES Benefitted
tabyl(remir_us_subgroup$subg_signif12_ses)# Separate Out Homogeneous Sample
tabyl(remir_us_subgroup$inter_benefit12_ses_cat)# Print homogeneous sample summary for remir_us_subgroup
tabyl(remir_us_subgroup$inter_benefit12_ses_cat2)### LOCATION ###
## Change Interaction Examined Variable Names
names(remir_us_subgroup)[36] <- "inter_examine1_loc"
names(remir_us_subgroup)[116] <- "inter_examine2_loc"
# Combine Two Interactions Examined
remir_us_subgroup$inter_examine12_loc <- ifelse(is.na(remir_us_subgroup$inter_examine1_loc),
remir_us_subgroup$inter_examine2_loc, remir_us_subgroup$inter_examine1_loc)
############ What Location Interactions Were Significant ################
## Change Interaction Significant Variable Names
names(remir_us_subgroup)[37] <- "inter_signif1_loc"
names(remir_us_subgroup)[117] <- "inter_signif2_loc"
# Combine Two Interactions Significant
remir_us_subgroup$inter_signif12_loc <- ifelse(is.na(remir_us_subgroup$inter_signif1_loc),
remir_us_subgroup$inter_signif2_loc, remir_us_subgroup$inter_signif1_loc)
############ What Location Interaction Groups Benefited More #############
## Change Interaction Benefitted Variable Names
names(remir_us_subgroup)[39] <- "inter_benefit1_loc"
names(remir_us_subgroup)[119] <- "inter_benefit2_loc"
# Combine Two Interactions Benefitted
remir_us_subgroup$inter_benefit12_loc <- ifelse(is.na(remir_us_subgroup$inter_benefit1_loc),
remir_us_subgroup$inter_benefit2_loc, remir_us_subgroup$inter_benefit1_loc)
############# What Location Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_us_subgroup)[67] <- "subg_examine1_loc"
names(remir_us_subgroup)[120] <- "subg_examine2_loc"
# Combine Two Subgroups Examined
remir_us_subgroup$subg_examine12_loc <- ifelse(is.na(remir_us_subgroup$subg_examine1_loc),
remir_us_subgroup$subg_examine2_loc, remir_us_subgroup$subg_examine1_loc)
############## What Location Subgroups Were Significant ####################
## Change Subgroup Significant Variable Names
names(remir_us_subgroup)[69] <- "subg_signif1_loc"
names(remir_us_subgroup)[122] <- "subg_signif2_loc"
# Combine Two Interactions Significant
remir_us_subgroup$subg_signif12_loc <- ifelse(is.na(remir_us_subgroup$subg_signif1_loc),
remir_us_subgroup$subg_signif2_loc, remir_us_subgroup$subg_signif1_loc)
# Simplify examine12_loc for US
remir_us_subgroup$examine12_loc_short <-
ifelse(is.na(remir_us_subgroup$inter_examine12_loc) &
is.na(remir_us_subgroup$subg_examine12_loc), 'None',
ifelse(remir_us_subgroup$inter_examine12_loc == 'Homog', 'Homog', 'Yes'))
# Fix remaining NAs for US
remir_us_subgroup$examine12_loc_short <- ifelse(
is.na(remir_us_subgroup$examine12_loc_short), 'Yes', remir_us_subgroup$examine12_loc_short)
# Generate summary tables for interactions and subgroups examined location for US
tabyl(remir_us_subgroup$inter_examine12_loc)tabyl(remir_us_subgroup$subg_examine12_loc)# Compute examined proportions for location for US
table(remir_us_subgroup$inter_examine12_loc, remir_us_subgroup$subg_examine12_loc, useNA = "always")##
## Rural Urban;Non-Urban <NA>
## Urban/Non-Urban 0 2 0
## <NA> 2 0 96tabyl(remir_us_subgroup, inter_examine12_loc, subg_examine12_loc) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)# Generate summary tables for examine12_loc_short for US
tabyl(remir_us_subgroup$examine12_loc_short)### Nativity ###
# What Nativity Interaction Groups Were Examined
## Change Interaction Nativity Variable Names
names(remir_us_subgroup)[41] <- "inter_examine1_nat"
names(remir_us_subgroup)[124] <- "inter_examine2_nat"
# Combine Two Interactions Examined
remir_us_subgroup$inter_examine12_nat <- ifelse(is.na(remir_us_subgroup$inter_examine1_nat),
remir_us_subgroup$inter_examine2_nat, remir_us_subgroup$inter_examine1_nat)
############## What Nativity Interactions Were Significant
## Change Interaction Significant Variable Names
names(remir_us_subgroup)[42] <- "inter_signif1_nat"
names(remir_us_subgroup)[125] <- "inter_signif2_nat"
# Combine Two Interactions Significant
remir_us_subgroup$inter_signif12_nat <- ifelse(is.na(remir_us_subgroup$inter_signif1_nat),
remir_us_subgroup$inter_signif2_nat, remir_us_subgroup$inter_signif1_nat)
############## What Nativity Interaction Groups Benefited More
## Change Interaction Benefitted Variable Names
names(remir_us_subgroup)[44] <- "inter_benefit1_nat"
names(remir_us_subgroup)[127] <- "inter_benefit2_nat"
# Combine Two Interactions Benefitted
remir_us_subgroup$inter_benefit12_nat <- ifelse(is.na(remir_us_subgroup$inter_benefit1_nat),
remir_us_subgroup$inter_benefit2_nat, remir_us_subgroup$inter_benefit1_nat)
############# What Nativity Subgroups Were Examined
## Change Subgroup Examined Variable Names
names(remir_us_subgroup)[71] <- "subg_examine1_nat"
names(remir_us_subgroup)[128] <- "subg_examine2_nat"
# Combine Two Subgroups Examined
remir_us_subgroup$subg_examine12_nat <- ifelse(is.na(remir_us_subgroup$subg_examine1_nat),
remir_us_subgroup$subg_examine2_nat, remir_us_subgroup$subg_examine1_nat)
############## What Nativity Subgroups Were Significant
## Change Subgroup Significant Variable Names
names(remir_us_subgroup)[73] <- "subg_signif1_nat"
names(remir_us_subgroup)[130] <- "subg_signif2_nat"
# Combine Two Interactions Significant
remir_us_subgroup$subg_signif12_nat <- ifelse(is.na(remir_us_subgroup$subg_signif1_nat),
remir_us_subgroup$subg_signif2_nat, remir_us_subgroup$subg_signif1_nat)
# Simplify examine12_nat for US
remir_us_subgroup$examine12_nat_short <-
ifelse(is.na(remir_us_subgroup$inter_examine12_nat) &
is.na(remir_us_subgroup$subg_examine12_nat), 'None',
ifelse(remir_us_subgroup$inter_examine12_nat == 'Homog', 'Homog', 'Yes'))
# Fix remaining NAs for US
remir_us_subgroup$examine12_nat_short <- ifelse(
is.na(remir_us_subgroup$examine12_nat_short), 'Yes', remir_us_subgroup$examine12_nat_short)
# Generate summary tables for interactions and subgroups examined nativity for US
tabyl(remir_us_subgroup$inter_examine12_nat)tabyl(remir_us_subgroup$subg_examine12_nat)# Compute examined proportions for nativity for US
table(remir_us_subgroup$inter_examine12_nat, remir_us_subgroup$subg_examine12_nat, useNA = "always")##
## <NA>
## Immigrant/Nonimmigrant 1
## <NA> 99tabyl(remir_us_subgroup, inter_examine12_nat, subg_examine12_nat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)# Generate summary tables for examine12_nat_short for US
tabyl(remir_us_subgroup$examine12_nat_short)###############################################################About Dummy Codes: Dummy code variables were created for each subgroup, per citation. If a citation included a culturally tailored program, it received a "1". Otherwise, it received a "0".
#########################################################
### ANY TAILORING ###
## Description: Was the program culturally tailored to any subgroups?
### FULL SAMPLE (N = 292)
## Dummy Variable for Any Culturally Tailored Program
remir_all$target_any <- ifelse(remir_all$target ==
'No group explicitly targeted', 1,
ifelse(remir_all$target == 'Families/individuals in foster care or the child welfare system', 1, 0))
tabyl(remir_all, target, target_any)### RACE ###
## Description: Was the program culturally tailored for one or more racial identity groups?
## Create Dummy Variable for Targeted Race Group
remir_all$target_race <- ifelse(
remir_all$target == 'Black or African American' |
remir_all$target == 'Black or African American, Hispanic or Latino' |
remir_all$target == 'Black or African American, Youth in rural communities',
1,0)
tabyl(remir_all, target, target_race)### ETHNICITY ###
## Create Dummy Variable for Targeted Ethnicity Group
remir_all$target_ethnic <- ifelse(
remir_all$target == 'Hispanic or Latino' |
remir_all$target == 'Black or African American, Hispanic or Latino',
1,0)
tabyl(remir_all, target, target_ethnic)## Create Dummy Variable for Targeted Gender Group
remir_all$target_gender <- ifelse(
remir_all$target == 'Gender' |
remir_all$target == 'Gender, Families/individuals in foster care or the child welfare system' |
remir_all$target == 'Gender, Low-income youth/families',
1,0)
tabyl(remir_all, target, target_gender)### SEXUAL IDENTITY ###
## No Culturally Tailored Program for Sex
## Create Dummy Variable for Targeted SES Group
remir_all$target_ses <- ifelse(
remir_all$target == 'Gender, Low-income youth/families' |
remir_all$target == 'Low-income youth/families' ,
1,0)
tabyl(remir_all, target, target_ses)## Create Dummy Variable for Targeted Location Group
remir_all$target_loc <- ifelse(
remir_all$target == 'Black or African American, Youth in rural communities' |
remir_all$target == 'Youth in rural communities' |
remir_all$target == 'Youth in urban communities',
1,0)
tabyl(remir_all, target, target_loc)## No culturally tailored programs for nativity
## Analysis Sample (N = 240)
## Dummy Variable for Any Culturally Tailored Program
remir_us$target_any <- ifelse(remir_us$target ==
'No group explicitly targeted', 1,
ifelse(remir_us$target == 'Families/individuals in foster care or the child welfare system', 1, 0))
tabyl(remir_us, target, target_any)### RACE ###
## Description: Was the program culturally tailored for one or more racial identity groups?
## Create Dummy Variable for Targeted Race Group
remir_us$target_race <- ifelse(
remir_us$target == 'Black or African American' |
remir_us$target == 'Black or African American, Hispanic or Latino' |
remir_us$target == 'Black or African American, Youth in rural communities',
1,0)
tabyl(remir_us, target, target_race)### ETHNICITY ###
## Create Dummy Variable for Targeted Ethnicity Group
remir_us$target_ethnic <- ifelse(
remir_us$target == 'Hispanic or Latino' |
remir_us$target == 'Black or African American, Hispanic or Latino',
1,0)
tabyl(remir_us, target, target_ethnic)## Create Dummy Variable for Targeted Gender Group
remir_us$target_gender <- ifelse(
remir_us$target == 'Gender' |
remir_us$target == 'Gender, Families/individuals in foster care or the child welfare system' |
remir_us$target == 'Gender, Low-income youth/families',
1,0)
tabyl(remir_us, target, target_gender)### SEXUAL IDENTITY ###
## No Culturally Tailored Program for Sex
## Create Dummy Variable for Targeted SES Group
remir_us$target_ses <- ifelse(
remir_us$target == 'Gender, Low-income youth/families' |
remir_us$target == 'Low-income youth/families' ,
1,0)
tabyl(remir_us, target, target_ses)## Create Dummy Variable for Targeted Location Group
remir_us$target_loc <- ifelse(
remir_us$target == 'Black or African American, Youth in rural communities' |
remir_us$target == 'Youth in rural communities' |
remir_us$target == 'Youth in urban communities',
1,0)
tabyl(remir_us, target, target_loc)## No culturally tailored programs for nativity
### SUB-ANALYSIS SAMPLE (N = 100)
## Dummy Variable for Any Culturally Tailored Program
remir_us_subgroup$target_any <- ifelse(remir_us_subgroup$target ==
'No group explicitly targeted', 1,
ifelse(remir_us_subgroup$target == 'Families/individuals in foster care or the child welfare system', 1, 0))
tabyl(remir_us_subgroup, target, target_any)### RACE ###
## Description: Was the program culturally tailored for one or more racial identity groups?
## Create Dummy Variable for Targeted Race Group
remir_us_subgroup$target_race <- ifelse(
remir_us_subgroup$target == 'Black or African American' |
remir_us_subgroup$target == 'Black or African American, Hispanic or Latino' |
remir_us_subgroup$target == 'Black or African American, Youth in rural communities',
1,0)
tabyl(remir_us_subgroup, target, target_race)### ETHNICITY ###
## Create Dummy Variable for Targeted Ethnicity Group
remir_us_subgroup$target_ethnic <- ifelse(
remir_us_subgroup$target == 'Hispanic or Latino' |
remir_us_subgroup$target == 'Black or African American, Hispanic or Latino',
1,0)
tabyl(remir_us_subgroup, target, target_ethnic)## Create Dummy Variable for Targeted Gender Group
remir_us_subgroup$target_gender <- ifelse(
remir_us_subgroup$target == 'Gender' |
remir_us_subgroup$target == 'Gender, Families/individuals in foster care or the child welfare system' |
remir_us_subgroup$target == 'Gender, Low-income youth/families',
1,0)
tabyl(remir_us_subgroup, target, target_gender)### SEXUAL IDENTITY ###
## No Culturally Tailored Program for Sex
## Create Dummy Variable for Targeted SES Group
remir_us_subgroup$target_ses <- ifelse(
remir_us_subgroup$target == 'Gender, Low-income youth/families' |
remir_us_subgroup$target == 'Low-income youth/families' ,
1,0)
tabyl(remir_us_subgroup, target, target_ses)## Create Dummy Variable for Targeted Location Group
remir_us_subgroup$target_loc <- ifelse(
remir_us_subgroup$target == 'Black or African American, Youth in rural communities' |
remir_us_subgroup$target == 'Youth in rural communities' |
remir_us_subgroup$target == 'Youth in urban communities',
1,0)
tabyl(remir_us_subgroup, target, target_loc)## No culturally tailored programs for nativity## About Table 1
# Table 1. Ns and Proportions for Culturally Tailored EBPIs by Subgroup
# Subgroups: Culturally Tailored EBPI; Target Population of Culturally Tailored EBPI; Race (Asian or Asian American, Black or African American, Native American, American Indian, Alaska Native, Native Hawaiian or Pacific Islander, White), Ethnicity – Hispanic or Latino; Gender (dichotomy of male or female); Economic disadvantage; Location – Urban; Location – Rural
# Notes: EBPIs – Evidence-Based Preventive Interventions
# a EBPIs may target multiple groups so that the proportions for the targeted groups sum to more than the total of culturally tailored EBPIs in the first row.
# b Reports conducted in both within and outside of the United States.
# c Reports conducted in the United States.
# d Reports conducted in the United States that tested for one or more of the following subgroups: race, ethnicity, gender, sexual identify, economic disadvantage, location (rural, urban, suburban), nativity status (foreign-born – yes/no). # Culturally Tailored EBPI
tabyl(remir_all, target_any)# Target Population of Culturally Tailored EBPI: Race
## Note: Culturally-tailored programs were designed for Black or African American subgroups only, so the output of this variable represents column I, row III for Table 1.
tabyl(remir_all, target_race)# Ethnicity – Hispanic or Latino
tabyl(remir_all, target_ethnic)# Gender (dichotomy of male or female)
tabyl(remir_all, target_gender)# Economic disadvantage
tabyl(remir_all, target_ses)#Location – Urban and Rural
tabyl(remir_all, target_loc)# Culturally Tailored EBPI
remir_all %>% filter(country == "USA") %>% tabyl(target_any)# Target Population of Culturally Tailored EBPI: Race
## Note: Culturally-tailored programs were designed for Black or African American subgroups only, so the output of this variable represents column II, row III for Table 1.
remir_all %>% filter(country == "USA") %>% tabyl(target_race)# Ethnicity – Hispanic or Latino
remir_all %>% filter(country == "USA") %>% tabyl(target_ethnic)# Gender (dichotomy of male or female)
remir_all %>% filter(country == "USA") %>% tabyl(target_gender)# Economic disadvantage
remir_all %>% filter(country == "USA") %>% tabyl(target_ses)#Location – Urban and Rural
remir_all %>% filter(country == "USA") %>% tabyl(target_loc)# Culturally Tailored EBPI
remir_all %>% filter(country == "USA" & !is.na(method)) %>% tabyl(target_any)# Target Population of Culturally Tailored EBPI: Race
## Note: Culturally-tailored programs were designed for Black or African American subgroups only, so the output of this variable represents column II, row III for Table 1.
remir_all %>% filter(country == "USA" & !is.na(method)) %>% tabyl(target_race)# Ethnicity – Hispanic or Latino
remir_all %>% filter(country == "USA" & !is.na(method)) %>% tabyl(target_ethnic)# Gender (dichotomy of male or female)
remir_all %>% filter(country == "USA" & !is.na(method)) %>% tabyl(target_gender)# Economic disadvantage
remir_all %>% filter(country == "USA" & !is.na(method)) %>% tabyl(target_ses)#Location – Urban and Rural
remir_all %>% filter(country == "USA" & !is.na(method)) %>% tabyl(target_loc)table(remir_us$inter_examine12_race_cat, remir_us$subg_examine12_race_cat,
useNA = "always")##
## Exact Mixed <NA>
## Exact 13 0 11
## Homog 0 0 26
## Mixed 1 3 13
## <NA> 8 2 163tabyl(remir_us, inter_examine12_race_cat, subg_examine12_race_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)# No - did not test subgroups
## All of these outputs were added to sum to .754, the proportion of programs that did NOT test for subgroup effects.
1.3+5.4+0.8+67.9 ## Did Not Examine Exact, including NA ## [1] 75.4## Did Not Examine Benefit, including NA and Mixed
remir_us %>% filter(
(is.na(subg_signif12_race_cat) &
is.na(inter_benefit12_race_cat)) |
(subg_signif12_race_cat == 'Mixed' &
inter_benefit12_race_cat == 'Mixed') |
(is.na(subg_signif12_race_cat) &
inter_benefit12_race_cat == 'Mixed') |
(subg_signif12_race_cat == 'Mixed' &
is.na(inter_benefit12_race_cat) |
(subg_signif12_race_cat == 'Other' &
inter_benefit12_race_cat == 'Mixed') |
(subg_signif12_race_cat == 'Mixed' &
inter_benefit12_race_cat == 'Other') )) %>% count()/240# Yes - tested subgroups
## All of these outputs were added to sum to .246, the proportion of programs that tested for subgroup effects.
4.6 ## Examined Exact Interaction only (NA for Subgroup)## [1] 4.63.3 ## Examined Exact Subgroup only (NA for Interaction)## [1] 3.35.4+0.4 ## Examined Exact Interaction and Subgroup (Include Mixed&Exact)## [1] 5.810.8 ## No Tests but Homogeneous Sample## [1] 10.8#### a. Examined Interaction
remir_us %>% filter(inter_examine12_race_cat == 'Exact' &
is.na(subg_examine12_race_cat)) %>% count()/240#### b. Examined Subgroup
remir_us %>% filter(subg_examine12_race_cat == 'Exact' &
is.na(inter_examine12_race_cat)) %>% count()/240## c. Examined Both
remir_us %>% filter((subg_examine12_race_cat == 'Exact' &
inter_examine12_race_cat == 'Exact') |
(subg_examine12_race_cat == 'Exact' &
inter_examine12_race_cat == 'Mixed') |
(subg_examine12_race_cat == 'Mixed' &
inter_examine12_race_cat == 'Exact')) %>% count()/240#### d. Homogeneous Sample
remir_us %>% filter(inter_examine12_race_cat == 'Homog') %>% count()/240##### e. Relative Benefit in Interaction
table(remir_us$inter_benefit12_race_cat, remir_us$subg_signif12_race_cat,
useNA = 'always') ##
## Exact Mixed Other <NA>
## Exact 12 0 0 8
## Homog 0 0 0 26
## Mixed 1 1 1 13
## Other 0 2 0 3
## <NA> 7 2 1 163tabyl(remir_us, inter_benefit12_race_cat, subg_signif12_race_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## Benefitted Exact Interaction only (NA for Subgroup)
remir_us %>% filter(inter_benefit12_race_cat == 'Exact' &
is.na(subg_signif12_race_cat)) %>% count()/240##### f. Absolute Benefit in Subgroup
## Benefitted Exact Subgroup only (NA for Interaction)
remir_us %>% filter(subg_signif12_race_cat == 'Exact' &
is.na(inter_benefit12_race_cat)) %>% count()/240##### g. Benefit in Interaction and Subgroup
## Benefitted Exact Interaction and Subgroup (Include if one Exact)
remir_us %>% filter((subg_signif12_race_cat == 'Exact' &
inter_benefit12_race_cat == 'Exact') |
(subg_signif12_race_cat == 'Exact' &
inter_benefit12_race_cat == 'Mixed') |
(subg_signif12_race_cat == 'Mixed' &
inter_examine12_race_cat == 'Exact')) %>% count()/240##### h. Homogenous Sample
## No Tests but Homogeneous Sample
remir_us %>% filter(inter_benefit12_race_cat == 'Homog') %>% count()/240##### i. No Benefit Interaction or Subgroup
## Other: Benefit for Whites or No Benefit Overall
remir_us %>% filter((subg_signif12_race_cat == 'Other' &
inter_benefit12_race_cat == 'Other') |
(subg_signif12_race_cat == 'Other' &
is.na(inter_benefit12_race_cat)) |
(is.na(subg_signif12_race_cat) &
inter_benefit12_race_cat == 'Other')) %>% count()/240## Check Exact Benefit Results Manually
2.9 ## Benefit Exact Interaction only## [1] 2.92.9 ## Benefit Exact Subgroup only## [1] 2.95.0+0.4 ## Benefit Exact Interaction and Subgroup## [1] 5.410.8 ## No Tests but Homogeneous Sample## [1] 10.82.1+0.4 ## Did Not Benefit Exact Group -- Benefit for Other ## [1] 2.50.4+0.4+5.0+0.8+0.8+67.9 ## Not Examined -- Include Mixed Only, Mixed Other## [1] 75.32.9+2.9+5.4+10.8+2.5+75.3 ## Sums of 99.8## [1] 99.8table(remir_us$inter_examine12_race_cat, remir_us$subg_examine12_race_cat,
useNA = "always")##
## Exact Mixed <NA>
## Exact 13 0 11
## Homog 0 0 26
## Mixed 1 3 13
## <NA> 8 2 163tabyl(remir_us, inter_examine12_race_cat, subg_examine12_race_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)# No - did not test subgroups
## All of these outputs were added to sum to .679, the proportion of programs that did NOT test for subgroup effects.
67.9 ## Examined Neither ## [1] 67.9remir_us %>% filter(
(is.na(subg_examine12_race_cat) &
is.na(inter_examine12_race_cat))) %>% count()/240# Yes - tested subgroups
## All of these outputs were added to sum to .321, the proportion of programs that tested for subgroup effects.
4.6+5.4 ## Examined Exact or Mixed Interaction only (NA for Subgroup)## [1] 103.3+0.8 ## Examined Mixed Subgroup only (NA for Interaction)## [1] 4.15.4+0.4+1.3 ## Examined Exact Interaction and Subgroup## [1] 7.110.8 ## No Tests but Homogeneous Sample## [1] 10.8#### a. Examined Interaction
remir_us %>% filter(
(inter_examine12_race_cat == 'Exact' |
inter_examine12_race_cat == 'Mixed') &
is.na(subg_examine12_race_cat)) %>% count()/240#### b. Examined Subgroup
remir_us %>% filter(
(subg_examine12_race_cat == 'Exact' |
subg_examine12_race_cat == 'Mixed') &
is.na(inter_examine12_race_cat)) %>% count()/240#### c. Examined Both
remir_us %>% filter((subg_examine12_race_cat == 'Exact' &
inter_examine12_race_cat == 'Exact') |
(subg_examine12_race_cat == 'Exact' &
inter_examine12_race_cat == 'Mixed') |
(subg_examine12_race_cat == 'Mixed' &
inter_examine12_race_cat == 'Exact') |
(subg_examine12_race_cat == 'Mixed' &
inter_examine12_race_cat == 'Mixed') ) %>% count()/240#### d. Homogeneous Sample
remir_us %>% filter(inter_examine12_race_cat == 'Homog') %>% count()/240#### e. Relative Benefit in Interaction
table(remir_us$inter_benefit12_race_cat, remir_us$subg_signif12_race_cat,
useNA = 'always') ##
## Exact Mixed Other <NA>
## Exact 12 0 0 8
## Homog 0 0 0 26
## Mixed 1 1 1 13
## Other 0 2 0 3
## <NA> 7 2 1 163tabyl(remir_us, inter_benefit12_race_cat, subg_signif12_race_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## Benefitted Exact or Mixed Interaction only (NA for Subgroup)
remir_us %>% filter(
(inter_benefit12_race_cat == 'Exact' |
inter_benefit12_race_cat == 'Mixed') &
is.na(subg_signif12_race_cat)) %>% count()/240#### f. Absolute Benefit in Subgroup
## Benefitted Exact or Mixed Subgroup only (NA for Interaction)
remir_us %>% filter(
(subg_signif12_race_cat == 'Exact' |
subg_signif12_race_cat == 'Mixed') &
is.na(inter_benefit12_race_cat)) %>% count()/240#### g. Benefit in Interaction & Subgroup
## Benefitted Exact or Mixed Interaction and Subgroup (Include Mixed&Exact)
remir_us %>% filter((subg_signif12_race_cat == 'Exact' &
inter_benefit12_race_cat == 'Exact') |
(subg_signif12_race_cat == 'Exact' &
inter_benefit12_race_cat == 'Mixed') |
(subg_signif12_race_cat == 'Mixed' &
inter_benefit12_race_cat == 'Exact') |
(subg_signif12_race_cat == 'Mixed' &
inter_benefit12_race_cat == 'Mixed') ) %>% count()/240#### h. Homogeneous Sample
remir_us %>% filter(inter_benefit12_race_cat == 'Homog') %>% count()/240#### i. No Benefit Interaction or Subgroup
## Other: Benefit for Whites or No Benefit Overall
remir_us %>% filter((subg_signif12_race_cat == 'Other' &
inter_benefit12_race_cat == 'Other') |
(subg_signif12_race_cat == 'Other' &
is.na(inter_benefit12_race_cat)) |
(is.na(subg_signif12_race_cat) &
inter_benefit12_race_cat == 'Other') |
(subg_signif12_race_cat == 'Other' &
inter_benefit12_race_cat == 'Mixed') |
(subg_signif12_race_cat == 'Mixed' &
inter_benefit12_race_cat == 'Other') ) %>% count()/240## Manually Check Benefit Percentages for Exact or Mixed
2.9+5.0 ## Benefit Exact and Mixed Interaction only## [1] 7.92.9+0.8 ## Benefit exact and Mixed Subgroup only## [1] 3.75.0+0.4+0.4 ## Benefit Exact and Mixed Both Tests## [1] 5.810.8 ## No tests but Homogeneous## [1] 10.80.8+2.1+0.4+0.4 ## Benefit Other, including Mixed/Other## [1] 3.767.9 ## Did not Benefit Group## [1] 67.97.9+3.7+5.8+3.7+10.8+67.9 ## Sums to 99.8## [1] 99.8# About Columns 3 & 4, Table 2
# Column 3: Measured ethnicity according to one or more of the U.S. Census categories of Hispanic or Latino.
#Column 4: Measured ethnicity according to the U.S. Census categories (see #3) or mixed categories with racial groups.
# Interpretation Key
## Exact: Ethnicity was Hispanic or Latino
## Mixed: Did not use distinct ethnic categories. Instead, used mixed categories, such as "Minority vs. Non-minority", where ethnic groups were categorized together.
## Homog: Homogenous sample with 75% or greater that identified with only one subgroup identity.
## NA: Ethnicity information was not reported
# To represent the output in Table 2, values have been calculated based on the table outputs. tabyl(remir_us, inter_examine12_ethnic_cat, subg_examine12_ethnic_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)1.7+4.2+79.6 ## Did NOT test subgroups (Column 3)## [1] 85.579.6 ## Did NOT test subgroups (Column 4)## [1] 79.6## COMBINE ETHNIC INTERACTION AND SUBGROUP EXAMINED
table(remir_us$inter_examine12_ethnic_cat, remir_us$subg_examine12_ethnic_cat,
useNA = "always")##
## Exact Mixed <NA>
## Exact 13 0 6
## Homog 0 0 7
## Mixed 0 4 10
## <NA> 9 0 191tabyl(remir_us, inter_examine12_ethnic_cat, subg_examine12_ethnic_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## Compute Percentages for Exact Four Categories
2.5 ## Examined Exact Hispanic Interaction only## [1] 2.53.8 ## Examined Exact Hispanic Subgroup only## [1] 3.85.4 ## Examined Hispanic Interaction and Subgroup## [1] 5.42.9 ## No Tests but Homogeneous Subgroup## [1] 2.91.7+4.2+79.6 ## Examined Neither Test or Mixed## [1] 85.52.5+3.8+5.4+2.9+85.5 ## Sums to 100.1## [1] 100.1## Compute Percentages for Exact and Mixed Categories
2.5+4.2 ## Examined Interaction only## [1] 6.73.8+0.0 ## Examined Subgroup only## [1] 3.85.4+1.7 ## Examined Interaction and Subgroup## [1] 7.12.9 ## No Test but Homogeneous Sample## [1] 2.979.6 ## Examined Neither ## [1] 79.66.7+3.8+7.1+2.9+79.6 ## Sum to 100.1## [1] 100.1## Other: All other commands
## INTERACTION ETHNIC BENEFIT
tabyl(remir_us$inter_benefit12_ethnic)## Check -- Looks OK
tabyl(remir_us$inter_benefit12_ethnic_cat)tabyl(remir_us, inter_benefit12_ethnic, inter_benefit12_ethnic_cat)## SUBGROUP ETHNIC BENEFIT
tabyl(remir_us$subg_signif12_ethnic)## Check
tabyl(remir_us$subg_signif12_ethnic_cat)tabyl(remir_us, subg_signif12_ethnic, subg_signif12_ethnic_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE
tabyl(remir_us$inter_benefit12_ethnic_cat)tabyl(remir_us$inter_benefit12_ethnic_cat)## Problem NA n = 200 for examined n = 198 for benefited
## How can cases that don't examine ethicity show a benefit
remir_us %>% filter(is.na(inter_examine12_ethnic_cat) &
is.na(subg_examine12_ethnic_cat) &
!is.na(inter_benefit12_ethnic_cat) &
is.na(subg_signif12_ethnic_cat)) %>%
select(Citation.ID, Program.Name, inter_examine12_ethnic_cat,
subg_examine12_ethnic_cat) remir_us %>% filter(is.na(inter_examine12_ethnic_cat) &
is.na(subg_examine12_ethnic_cat) &
!is.na(inter_benefit12_ethnic_cat) &
is.na(subg_signif12_ethnic_cat)) %>%
select(Citation.ID, Program.Name, inter_benefit12_ethnic_cat,
subg_signif12_ethnic_cat)## COMBINE ETHNIC INTERACTION AND SUBGROUP BENEFIT
## Other category is problematic -- compute as residual
table(remir_us$inter_benefit12_ethnic_cat, remir_us$subg_signif12_ethnic_cat,
useNA = 'always') ##
## Exact Mixed Other <NA>
## Exact 8 0 3 5
## Homog 0 0 0 7
## Mixed 0 2 3 7
## Other 3 2 0 0
## <NA> 7 0 2 191tabyl(remir_us, inter_benefit12_ethnic_cat, subg_signif12_ethnic_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## Compute Percentages for Exact Categories (Column 3)
2.1 ## Benefitted Exact Hispanic Interaction only## [1] 2.12.9 ## Benefitted Exact Hispanic Subgroup only## [1] 2.93.3 ## Benefitted Hispanic Interaction and Subgroup## [1] 3.31.3+0.4+1.3+0.8 ## Benefitted Other## [1] 3.82.9 ## No tests but Homogeneous Sample## [1] 2.90.8+0.8+2.9+0.8+79.6 ## No tests -- use same percentage of no tests for examined## [1] 84.9100-2.1-2.9-3.3-2.9-79.6 ## Residual = Did Not Benefit Hispanic Exact ## [1] 9.22.1+2.9+3.3+44.2+44.2+3.3 ## Sums to 100## [1] 100## Compute Percentages for Mixed Categories (Column 4)
2.1+2.9 ## Benefitted Mixed Hispanic Interaction only## [1] 52.9+0.0 ## Benefitted Mixed Hispanic Subgroup only## [1] 2.93.3+0.8 ## Benefitted Hispanic Interaction and Subgroup## [1] 4.12.9 ## No Tests but Homogeneous Sample## [1] 2.979.6 ## Did Not Examine Ethnic Mixed## [1] 79.61.3+0.8+0.4+1.3+0.8+0.8 ## Residual = Did Not Benefit Hispanic Exact ## [1] 5.45.0+2.9+4.1+2.9+5.4+79.6 ## Sums to 99.9## [1] 99.9## Did Not Examine Gender
remir_us %>% filter(
(is.na(subg_examine12_gender) &
is.na(inter_examine12_gender))) %>% count()/24053.3 ## Examined Neither## [1] 53.3##### a. Examined Interaction
remir_us %>% filter(inter_examine12_gender == 'Male/Female' &
(is.na(subg_examine12_gender) | subg_examine12_gender == 'None')) %>% count()/240##### b. Examined Subgroup
remir_us %>% filter((subg_examine12_gender == 'Male;Female' |
subg_examine12_gender == 'Female') &
is.na(inter_examine12_gender)) %>% count()/240##### c. Examined Both
remir_us %>% filter((subg_examine12_gender == 'Male;Female' &
inter_examine12_gender == 'Male/Female') |
(subg_examine12_gender == 'Female' &
inter_examine12_gender == 'Male/Female')) %>% count()/240##### d. Homogenous Sample
remir_us %>% filter(inter_examine12_gender == 'Homog') %>% count()/240## Manually Check Percentages
12.9+0.8 ## a. Examined Interaction## [1] 13.70.0+4.2 ## b. Examined Subgroup only## [1] 4.20.4+13.3 ## c. Examined Interaction and Subgroup## [1] 13.715.0 ## d. Homogeneous Sample## [1] 1553.3 ## No: I. Tested Subgroups## [1] 53.313.7+4.2+13.7+15.0+53.3 ## Sums to 99.9## [1] 99.9##### e. Relative Benefit in Interaction
## COMBINED INTERACTION AND SUBGROUP GENDER BENEFITTED
table(remir_us$inter_benefit12_gender_cat2, remir_us$subg_signif12_gender_cat,
useNA = "always")##
## Female Other <NA>
## Female 23 3 30
## Homog 0 0 36
## Other 4 1 0
## <NA> 6 8 129tabyl(remir_us, inter_benefit12_gender_cat2, subg_signif12_gender_cat) %>%
adorn_percentages("all") %>%
adorn_pct_formatting(rounding = "half up", digits = 1) ## Benefit Interaction only (NA for Subgroup)
remir_us %>% filter(inter_benefit12_gender_cat2 == 'Female' &
is.na(subg_signif12_gender_cat)) %>% count()/240##### f. Absolute Benefit in Subgroup
remir_us %>% filter(subg_signif12_gender_cat == 'Female' &
is.na(inter_benefit12_gender_cat2)) %>% count()/240##### g. Benefit in Interaction & Subgroup
remir_us %>% filter( (subg_signif12_gender_cat == 'Female' &
inter_benefit12_gender_cat2 == 'Female') |
(subg_signif12_gender_cat == 'Female' &
inter_benefit12_gender_cat2 == 'Other') |
(inter_benefit12_gender_cat2 == 'Female' &
subg_signif12_gender_cat == 'Other') ) %>% count()/240##### h. Homogenous Sample
remir_us %>% filter(inter_benefit12_gender_cat2 == 'Homog') %>% count()/240##### i. No Benefit Interaction or Subgroup
remir_us %>% filter(
(subg_signif12_gender_cat == 'Other' &
inter_benefit12_gender_cat2 == 'Other') |
(subg_signif12_gender_cat == 'Other' &
is.na(inter_benefit12_gender_cat2)) |
(inter_benefit12_gender_cat2 == 'Other' &
is.na(subg_signif12_gender_cat))) %>% count()/240## Manually Check Percentages
12.5 ## Benefitted Females Interaction only## [1] 12.52.5 ## Benefitted Females Subgroup only## [1] 2.59.6+0.8+1.7 ## Benefitted Females in Interaction or Subgroup## [1] 12.115.0 ## No Tests But Homogeneous Subgroup## [1] 150.8+3.3 ## Benefitted Other## [1] 4.153.8 ## Unclear and Did Not Examine ## [1] 53.812.5+2.5+12.1+15.0+4.1+53.8 ## Sums to 100## [1] 100## COMBINED INTERACTION SUBGROUP SEX EXAMINED
table(remir_us$inter_examine12_sex, remir_us$subg_examine12_sex,
useNA = "always")##
## <NA>
## No 40
## Yes 2
## <NA> 198tabyl(remir_us, inter_examine12_sex, subg_examine12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)82.5 + 16.7 # NA and "No" added together to produce the total number of citations that did NOT test subgroups (99.2% or .992)## [1] 99.2##### a. Examined Interaction
tabyl(remir_us, inter_examine12_sex, subg_examine12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## Examined Interaction only
0.8 ## [1] 0.8##### b. Examined Subgroup
tabyl(remir_us, inter_examine12_sex, subg_examine12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1).00 ## [1] 0##### c. Examined Both
tabyl(remir_us, inter_examine12_sex, subg_examine12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1).00 ## [1] 0##### d. Homogenous Sample
tabyl(remir_us, inter_examine12_sex, subg_examine12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)##### e. Relative Benefit in Interaction
tabyl(remir_us, inter_benefit12_sex_cat, subg_signif12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)0.8 ## Benefitted LGPT Interaction only## [1] 0.8##### f. Absolute Benefit in Subgroup
tabyl(remir_us, inter_benefit12_sex_cat, subg_signif12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)0 ## Benefitted LGPT Subgroup only## [1] 0##### g. Benefit in Interaction & Subgroup
tabyl(remir_us, inter_benefit12_sex_cat, subg_signif12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)0 ## Benefitted LGPT Interaction and Subgroup## [1] 0##### h. Homogenous Sample
tabyl(remir_us, inter_benefit12_sex_cat, subg_signif12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)0 ## Benefitted non-LGPT or Neither## [1] 0##### i. No Benefit Interaction or Subgroup
tabyl(remir_us, inter_benefit12_sex_cat, subg_signif12_sex) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)0 ## No Homogeneous sample measure## [1] 0## COMBINED INTERACTION AND SUBGROUP SES EXAMINED
table(remir_us$inter_examine12_ses_cat, remir_us$subg_examine12_ses,
useNA = "always")##
## Low Low;High <NA>
## Low/High 2 8 0
## Yes 0 0 10
## <NA> 2 6 212tabyl(remir_us, inter_examine12_ses_cat, subg_examine12_ses) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## Did Not Examine SES
remir_us %>% filter(is.na(subg_signif12_ses_cat) &
is.na(inter_benefit12_ses_cat2)) %>% count()/240## a. Examined Interaction
remir_us %>% filter( (inter_examine12_ses_cat == 'Low/High' |
inter_examine12_ses_cat == 'Yes') &
(is.na(subg_examine12_ses))) %>% count()/240## b. Examined Subgroup
remir_us %>% filter( (subg_examine12_ses == 'Low;High' |
subg_examine12_ses == 'Low') &
(is.na(inter_examine12_ses_cat))) %>% count()/240## c. Examined Both
remir_us %>% filter( (subg_examine12_ses == 'Low;High' |
subg_examine12_ses == 'Low') &
(inter_examine12_ses_cat == 'Low/High' |
inter_examine12_ses_cat == 'Yes')) %>% count()/240## d. Homogenous Sample
remir_us$inter_examine12_ses_cat <- ifelse(
(remir_us$econ_all > .75) &
is.na(remir_us$inter_examine12_ses_cat) &
is.na(remir_us$subg_examine12_ses),
"Homog", remir_us$inter_examine12_ses_cat)
tabyl(remir_us$inter_examine12_ses_cat)remir_us %>% filter(inter_examine12_ses_cat == 'Homog') %>% count()/240## Manually Check SES Percentages
0.0+4.2 ## Examined Interaction only## [1] 4.20.8+2.5 ## Examined Subgroup only## [1] 3.30.8+3.3 ## Examined Interaction and Subgroup## [1] 4.116.3 ## No Tests but Homogeneous Sample## [1] 16.372.1 ## Examined Neither## [1] 72.14.2+3.3+4.1+16.3+72.1 ## Sums to 100## [1] 100tabyl(remir_us, inter_benefit12_ses_cat, subg_signif12_ses_cat)tabyl(remir_us, inter_benefit12_ses_cat, subg_signif12_ses_cat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## e. Relative Benefit in Interaction
remir_us %>% filter(inter_benefit12_ses_cat2 == 'Low' &
is.na(subg_signif12_ses_cat)) %>% count()/240## f. Absolute Benefit in Subgroup
remir_us %>% filter(subg_signif12_ses_cat == 'Low' &
is.na(inter_benefit12_ses_cat2)) %>% count()/240## g. Benefit in Interaction & Subgroup
remir_us %>% filter( (subg_signif12_ses_cat == 'Low' &
inter_benefit12_ses_cat2 == 'Low') |
(subg_signif12_ses_cat == 'Low' &
inter_benefit12_ses_cat2 == 'High') |
(inter_benefit12_ses_cat2 == 'Low' &
subg_signif12_ses_cat == 'Other') ) %>% count()/240## h. Homogeneous Sample
remir_us %>% filter(inter_benefit12_ses_cat2 == 'Homog') %>% count()/240## i. No Benefit
# Benefit Other
remir_us %>% filter(
(subg_signif12_ses_cat == 'Other' &
inter_benefit12_ses_cat2 == 'High') |
(subg_signif12_ses_cat == 'Other' &
is.na(inter_benefit12_ses_cat2)) |
(inter_benefit12_ses_cat2 == 'High' &
is.na(subg_signif12_ses_cat))) %>% count()/240## Manually check Percentages
3.8 ## Benefitted Low SES Interaction only## [1] 3.83.3 ## Benefitted Low SES Subgroup only## [1] 3.32.9+0.4+0.4## Benefitted Low SES Interaction and Subgroup## [1] 3.716.3 ## No tests But Homogeneous Sample## [1] 16.30.4+0.4 ## Did Not Benefit Low SES ## [1] 0.872.1 ## Did Not Examine Low SES ## [1] 72.13.8+3.3+3.7+16.3+0.8+72.1 ## Sums to 100## [1] 100#######################################################################
############## PREPARE SES CODES FOR TABLE 2 ########################
#######################################################################
############## What SES Interaction Group Were Examined ########################
## COMBINED INTERACTION AND SUBGROUP SES EXAMINED
table(remir_us$inter_examine12_ses_cat, remir_us$subg_examine12_ses,
useNA = "always")##
## Low Low;High <NA>
## Homog 0 0 39
## Low/High 2 8 0
## Yes 0 0 10
## <NA> 2 6 173tabyl(remir_us, inter_examine12_ses_cat, subg_examine12_ses) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## Compute Percentages for Four Categories
0.0+4.2 ## Examined Interaction only## [1] 4.20.8+2.5 ## Examined Subgroup only## [1] 3.30.8+3.3 ## Examined Interaction and Subgroup## [1] 4.116.3 ## No Tests but Homogeneous Sample## [1] 16.372.1 ## Examined Neither## [1] 72.14.2+3.3+4.1+16.3+72.1 ## Sums to 100## [1] 100####################### Table 2. Which Groups Benefitted ###############
## INTERACTION SES BENEFITTED
tabyl(remir_us$inter_benefit12_ses)## Simplify Codes
tabyl(remir_us, inter_benefit12_ses, inter_examine12_ses)remir_us$inter_benefit12_ses_cat <-
ifelse(remir_us$inter_benefit12_ses == 'None' &
remir_us$main_effect == 'Yes', 'Both',
ifelse(remir_us$inter_benefit12_ses == 'None' &
remir_us$main_effect == 'No', 'Neither',
remir_us$inter_benefit12_ses))
tabyl(remir_us, inter_benefit12_ses, inter_benefit12_ses_cat)## SUBGROUP SES BENEFITTED
tabyl(remir_us$subg_signif12_ses)## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_us$inter_benefit12_ses_cat <- ifelse(
(remir_us$econ_all > .75) &
is.na(remir_us$inter_benefit12_ses_cat) &
is.na(remir_us$subg_signif12_ses),
"Homog", remir_us$inter_benefit12_ses_cat)
tabyl(remir_us$inter_benefit12_ses_cat)## COMBINE INTERACTION AND SUBGROUP SES BENEFITTED
tabyl(remir_us, inter_benefit12_ses_cat, subg_signif12_ses)tabyl(remir_us, inter_benefit12_ses_cat, subg_signif12_ses) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)2.1+1.7 ## Benefitted Low SES Interaction only## [1] 3.82.1+0.8+0.4 ## Benefitted Low SES Subgroup only## [1] 3.30.4+0.8+1.7 ## Benefitted Low SES Interaction and Subgroup## [1] 2.916.3 ## No tests But Homogeneous Sample## [1] 16.30.4+0.4+0.4+0.4 ## Did Not Benefit Low SES ## [1] 1.672.1 ## Did Not Examine Low SES ## [1] 72.13.8+2.9+3.3+16.3+1.6+72.1 ## Sums to 100## [1] 100# Compute examined proportions for location for US
table(remir_us$inter_examine12_loc, remir_us$subg_examine12_loc, useNA = "always")##
## Rural Urban;Non-Urban <NA>
## Urban/Non-Urban 0 2 0
## <NA> 2 0 236tabyl(remir_us, inter_examine12_loc, subg_examine12_loc) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)# Generate summary tables for examine12_loc_short for US
tabyl(remir_us$examine12_loc_short)## INTERACTION LOCATION EXAMINED
tabyl(remir_us$inter_examine12_loc)## SUBGROUP LOCATION EXAMINED
tabyl(remir_us$subg_examine12_loc)## SEPARATE OUT HOMOGENEOUS SAMPLE
## No measure of sample proportion for urban or rural
## COMBINE INTERACTION AND SUBGROUP LOCATION EXAMINED
table(remir_us$inter_examine12_loc, remir_us$subg_examine12_loc,
useNA = "always")##
## Rural Urban;Non-Urban <NA>
## Urban/Non-Urban 0 2 0
## <NA> 2 0 236tabyl(remir_us, inter_examine12_loc, subg_examine12_loc) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## Compute Percentages for Four Categories
0 ## Examined Interaction only## [1] 00.8 ## Examined Subgroup only## [1] 0.80.8 ## Examined Interaction and Subgroup## [1] 0.898.3 ## Examined Neither## [1] 98.30.8+0.8+98.3 ## Sums to 99.9## [1] 99.9## INTERACTION LOCATION BENEFITTED
tabyl(remir_us$inter_benefit12_loc)## SUBGROUP LOCATION BENEFITTED
tabyl(remir_us$subg_signif12_loc)## COMBINE INTERACTION AND SUBGROUP LOCATION BENEFITTED
tabyl(remir_us, inter_benefit12_loc, subg_signif12_loc) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)## Compute Percentages Benefit Rural or Urban
0.0 ## Interaction only## [1] 00.8 ## Subgroup Only## [1] 0.80.4 ## Interaction and Subgroup## [1] 0.40.4 ## Neither## [1] 0.498.3 ## Not examained## [1] 98.3remir_us$subg_signif12_nat <- ifelse(is.na(remir_us$subg_signif1_nat),
remir_us$subg_signif2_nat, remir_us$subg_signif1_nat)
tabyl(remir_us$subg_signif12_nat)tabyl(remir_all, inter_examine12_nat, subg_examine12_nat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_all$examine12_nat_short <-
ifelse(is.na(remir_all$inter_examine12_nat) &
is.na(remir_all$subg_examine12_nat), 'None',
ifelse(remir_all$inter_examine12_nat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_all$examine12_nat_short)## INTERACTION NATIVITY EXAMINED
tabyl(remir_us$inter_examine12_nat)## SUBGROUP NATIVITY EXAMINED
tabyl(remir_us$subg_examine12_nat)## COMBINE INTERACTION AND SUBGROUP NATIVITY EXAMINED
table(remir_us$inter_examine12_nat, remir_us$subg_examine12_nat,
useNA = "always")##
## <NA>
## Immigrant/Nonimmigrant 1
## <NA> 239tabyl(remir_us, inter_examine12_nat, subg_examine12_nat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 0)## Compute Percentages for Four Categories
0 ## Examined Interaction only## [1] 00 ## Examined Subgroup only## [1] 00 ## Examined Interaction and Subgroup## [1] 0100 ## Examined Neither## [1] 100## INTERACTION NATIVITY BENEFITTED
tabyl(remir_us$inter_benefit12_nat)## SUBGROUP NATIVITY BENEFITTED
tabyl(remir_us$subg_signif12_nat)## COMBINE INTERACTION AND SUBGROUP NATIVITY BENEFITTED
tabyl(remir_us, inter_benefit12_nat, subg_signif12_nat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 0)## No culturally tailored programs
## Create Short Variable for Examined with Three categories
tabyl(remir_us, inter_examine12_nat, subg_examine12_nat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_us$examine12_nat_short <-
ifelse(is.na(remir_us$inter_examine12_nat) &
is.na(remir_us$subg_examine12_nat), 'None',
ifelse(remir_us$inter_examine12_nat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_us$examine12_nat_short)# About Online Supplemental Table 1: Ns and Proportions for Characteristics of Sample EBPI Evaluation Reports
## Note: EBPI – Evidence-Based Preventive Intervention
## a Tested for one or more of the following subgroups: race, ethnicity, gender, sexual identity, economic disadvantage, location (rural, urban, suburban), nativity status (foreign-born – yes/no).
## b Proportions add to more than 1.0, as EBPIs may target multiple groups and multiple outcomes.# Design (1 = Cluster randomized control trial (c-RCT), 2 = Quasi-experimental design (QED), 3 = Randomized control trial (RCT))
remir_all$design <- factor(remir_all$design, levels = c(1, 2, 3),
labels = c("Cluster randomized control trial (c-RCT)",
"Quasi-experimental design (QED)",
"Randomized control trial (RCT)"))
tabyl(remir_all, design)# Article published in an academic journal
tabyl(remir_all, "published") # Is this program certified by Blueprints Healthy Youth Development?
tabyl(remir_all, "Certified.")# Primary age group targeted
age_group_table <- data.frame(
Age_Group = c(
"Infant (ages 0-2 years)",
"Preschool (ages 3-4 years)",
"Elementary school (ages 5-11 years)",
"Middle school (ages 12-14 years)",
"High school (ages 15-18 years)",
"Young adult (ages 19-24 years)",
"Adult"
),
Frequency = c(
sum(remir_all$age_infant, na.rm = TRUE),
sum(remir_all$age_preschool, na.rm = TRUE),
sum(remir_all$age_elementary, na.rm = TRUE),
sum(remir_all$age_middle, na.rm = TRUE),
sum(remir_all$age_high, na.rm = TRUE),
sum(remir_all$age_youngadult, na.rm = TRUE),
sum(remir_all$age_adult, na.rm = TRUE)
)
)
total_participants <- sum(age_group_table$Frequency)
age_group_table <- age_group_table %>%
mutate(Proportion = round(Frequency / 292, 2))
print(age_group_table)## Age_Group Frequency Proportion
## 1 Infant (ages 0-2 years) 12 0.04
## 2 Preschool (ages 3-4 years) 35 0.12
## 3 Elementary school (ages 5-11 years) 98 0.34
## 4 Middle school (ages 12-14 years) 114 0.39
## 5 High school (ages 15-18 years) 138 0.47
## 6 Young adult (ages 19-24 years) 73 0.25
## 7 Adult 15 0.05## Place
place_table <- data.frame(
Place = c(
"Community",
"Correctional facility",
"Home",
"Medical setting",
"Online",
"School",
"Service setting"
),
Frequency = c(
sum(remir_all$pl_comm, na.rm = TRUE),
sum(remir_all$pl_corr, na.rm = TRUE),
sum(remir_all$pl_home, na.rm = TRUE),
sum(remir_all$pl_med, na.rm = TRUE),
sum(remir_all$pl_online, na.rm = TRUE),
sum(remir_all$pl_school, na.rm = TRUE),
sum(remir_all$pl_service, na.rm = TRUE)
)
)
place_table <- place_table %>%
mutate(Proportion = round(Frequency / 292, 2))
print(place_table)## Place Frequency Proportion
## 1 Community 40 0.14
## 2 Correctional facility 12 0.04
## 3 Home 27 0.09
## 4 Medical setting 19 0.07
## 5 Online 12 0.04
## 6 School 174 0.60
## 7 Service setting 8 0.03## Outcome
outcome_table <- data.frame(
Outcome = c(
"Adult outcomes",
"Educational outcomes",
"Emotional outcomes",
"Physical outcomes",
"Positive relationship outcomes",
"Problem behavior outcomes"
),
Frequency = c(
sum(remir_all$out_adult, na.rm = TRUE),
sum(remir_all$out_educ, na.rm = TRUE),
sum(remir_all$out_emot, na.rm = TRUE),
sum(remir_all$out_phys, na.rm = TRUE),
sum(remir_all$out_posrel, na.rm = TRUE),
sum(remir_all$out_problem, na.rm = TRUE)
)
)
outcome_table <- outcome_table %>%
mutate(Proportion = round(Frequency / 292, 2))
print(outcome_table)## Outcome Frequency Proportion
## 1 Adult outcomes 12 0.04
## 2 Educational outcomes 105 0.36
## 3 Emotional outcomes 43 0.15
## 4 Physical outcomes 7 0.02
## 5 Positive relationship outcomes 7 0.02
## 6 Problem behavior outcomes 156 0.53# Design (1 = Cluster randomized control trial (c-RCT), 2 = Quasi-experimental design (QED), 3 = Randomized control trial (RCT))
remir_us$design <- factor(remir_us$design, levels = c(1, 2, 3),
labels = c("Cluster randomized control trial (c-RCT)",
"Quasi-experimental design (QED)",
"Randomized control trial (RCT)"))
tabyl(remir_us, design)# Article published in an academic journal
tabyl(remir_us, "published") # Is this program certified by Blueprints Healthy Youth Development?
tabyl(remir_us, "Certified.")# Primary age group targeted
age_group_table_us <- data.frame(
Age_Group = c(
"Infant (ages 0-2 years)",
"Preschool (ages 3-4 years)",
"Elementary school (ages 5-11 years)",
"Middle school (ages 12-14 years)",
"High school (ages 15-18 years)",
"Young adult (ages 19-24 years)",
"Adult"
),
Frequency = c(
sum(remir_us$age_infant, na.rm = TRUE),
sum(remir_us$age_preschool, na.rm = TRUE),
sum(remir_us$age_elementary, na.rm = TRUE),
sum(remir_us$age_middle, na.rm = TRUE),
sum(remir_us$age_high, na.rm = TRUE),
sum(remir_us$age_youngadult, na.rm = TRUE),
sum(remir_us$age_adult, na.rm = TRUE)
)
)
age_group_table_us <- age_group_table_us %>%
mutate(Proportion = round(Frequency / 240, 2))
print(age_group_table_us)## Age_Group Frequency Proportion
## 1 Infant (ages 0-2 years) 12 0.05
## 2 Preschool (ages 3-4 years) 20 0.08
## 3 Elementary school (ages 5-11 years) 72 0.30
## 4 Middle school (ages 12-14 years) 88 0.37
## 5 High school (ages 15-18 years) 124 0.52
## 6 Young adult (ages 19-24 years) 65 0.27
## 7 Adult 13 0.05# Place
place_table_us <- data.frame(
Place = c(
"Community",
"Correctional facility",
"Home",
"Medical setting",
"Online",
"School",
"Service setting"
),
Frequency = c(
sum(remir_us$pl_comm, na.rm = TRUE),
sum(remir_us$pl_corr, na.rm = TRUE),
sum(remir_us$pl_home, na.rm = TRUE),
sum(remir_us$pl_med, na.rm = TRUE),
sum(remir_us$pl_online, na.rm = TRUE),
sum(remir_us$pl_school, na.rm = TRUE),
sum(remir_us$pl_service, na.rm = TRUE)
)
)
place_table_us <- place_table_us %>%
mutate(Proportion = round(Frequency / 240, 2))
print(place_table_us)## Place Frequency Proportion
## 1 Community 33 0.14
## 2 Correctional facility 10 0.04
## 3 Home 17 0.07
## 4 Medical setting 19 0.08
## 5 Online 11 0.05
## 6 School 142 0.59
## 7 Service setting 8 0.03# Outcome
outcome_table_us <- data.frame(
Outcome = c(
"Adult outcomes",
"Educational outcomes",
"Emotional outcomes",
"Physical outcomes",
"Positive relationship outcomes",
"Problem behavior outcomes"
),
Frequency = c(
sum(remir_us$out_adult, na.rm = TRUE),
sum(remir_us$out_educ, na.rm = TRUE),
sum(remir_us$out_emot, na.rm = TRUE),
sum(remir_us$out_phys, na.rm = TRUE),
sum(remir_us$out_posrel, na.rm = TRUE),
sum(remir_us$out_problem, na.rm = TRUE)
)
)
outcome_table_us <- outcome_table_us %>%
mutate(Proportion = round(Frequency / 240, 2))
print(outcome_table_us)## Outcome Frequency Proportion
## 1 Adult outcomes 10 0.04
## 2 Educational outcomes 98 0.41
## 3 Emotional outcomes 38 0.16
## 4 Physical outcomes 7 0.03
## 5 Positive relationship outcomes 6 0.03
## 6 Problem behavior outcomes 114 0.48# Design (1 = Cluster randomized control trial (c-RCT), 2 = Quasi-experimental design (QED), 3 = Randomized control trial (RCT))
remir_us_subgroup$design <- dplyr::recode(remir_us_subgroup$design,
`1` = "Cluster randomized control trial (c-RCT)",
`2` = "Quasi-experimental design (QED)",
`3` = "Randomized control trial (RCT)"
)
tabyl(remir_us_subgroup, design)# Article published in an academic journal
tabyl(remir_us_subgroup, "published")# Is this program certified by Blueprints Healthy Youth Development?
tabyl(remir_us_subgroup, "Certified.")# Primary age group targeted
age_group_table_us_subgroup <- data.frame(
Age_Group = c(
"Infant (ages 0-2 years)",
"Preschool (ages 3-4 years)",
"Elementary school (ages 5-11 years)",
"Middle school (ages 12-14 years)",
"High school (ages 15-18 years)",
"Young adult (ages 19-24 years)",
"Adult"
),
Frequency = c(
sum(remir_us_subgroup$age_infant, na.rm = TRUE),
sum(remir_us_subgroup$age_preschool, na.rm = TRUE),
sum(remir_us_subgroup$age_elementary, na.rm = TRUE),
sum(remir_us_subgroup$age_middle, na.rm = TRUE),
sum(remir_us_subgroup$age_high, na.rm = TRUE),
sum(remir_us_subgroup$age_youngadult, na.rm = TRUE),
sum(remir_us_subgroup$age_adult, na.rm = TRUE)
)
)
age_group_table_us_subgroup <- age_group_table_us_subgroup %>%
mutate(Proportion = round(Frequency / 100, 2))
print(age_group_table_us_subgroup)## Age_Group Frequency Proportion
## 1 Infant (ages 0-2 years) 6 0.06
## 2 Preschool (ages 3-4 years) 7 0.07
## 3 Elementary school (ages 5-11 years) 26 0.26
## 4 Middle school (ages 12-14 years) 33 0.33
## 5 High school (ages 15-18 years) 57 0.57
## 6 Young adult (ages 19-24 years) 33 0.33
## 7 Adult 1 0.01# Place
place_table_us_subgroup <- data.frame(
Place = c(
"Community",
"Correctional facility",
"Home",
"Medical setting",
"Online",
"School",
"Service setting"
),
Frequency = c(
sum(remir_us_subgroup$pl_comm, na.rm = TRUE),
sum(remir_us_subgroup$pl_corr, na.rm = TRUE),
sum(remir_us_subgroup$pl_home, na.rm = TRUE),
sum(remir_us_subgroup$pl_med, na.rm = TRUE),
sum(remir_us_subgroup$pl_online, na.rm = TRUE),
sum(remir_us_subgroup$pl_school, na.rm = TRUE),
sum(remir_us_subgroup$pl_service, na.rm = TRUE)
)
)
place_table_us_subgroup <- place_table_us_subgroup %>%
mutate(Proportion = round(Frequency / 100, 2))
print(place_table_us_subgroup)## Place Frequency Proportion
## 1 Community 14 0.14
## 2 Correctional facility 0 0.00
## 3 Home 4 0.04
## 4 Medical setting 7 0.07
## 5 Online 4 0.04
## 6 School 68 0.68
## 7 Service setting 3 0.03# Outcome
outcome_table_us_subgroup <- data.frame(
Outcome = c(
"Adult outcomes",
"Educational outcomes",
"Emotional outcomes",
"Physical outcomes",
"Positive relationship outcomes",
"Problem behavior outcomes"
),
Frequency = c(
sum(remir_us_subgroup$out_adult, na.rm = TRUE),
sum(remir_us_subgroup$out_educ, na.rm = TRUE),
sum(remir_us_subgroup$out_emot, na.rm = TRUE),
sum(remir_us_subgroup$out_phys, na.rm = TRUE),
sum(remir_us_subgroup$out_posrel, na.rm = TRUE),
sum(remir_us_subgroup$out_problem, na.rm = TRUE)
)
)
outcome_table_us_subgroup <- outcome_table_us_subgroup %>%
mutate(Proportion = round(Frequency / 100, 2))
print(outcome_table_us_subgroup)## Outcome Frequency Proportion
## 1 Adult outcomes 0 0.00
## 2 Educational outcomes 59 0.59
## 3 Emotional outcomes 9 0.09
## 4 Physical outcomes 3 0.03
## 5 Positive relationship outcomes 4 0.04
## 6 Problem behavior outcomes 34 0.34# About Online Supplemental Table 2: Ns, Proportions, and Descriptive Statistics for Characteristics of Sample EBPI Evaluation Reports
## Note. Remir 1.0 measured if program targeted location (rural and urban), but did not measure if sample distribution was reported, or percentage of sample in rural and urban locations.
## Note: a Tested for one or more of the following subgroups: race, ethnicity, gender, sexual identity, economic disadvantage, location (rural, urban, suburban), nativity status (foreign-born – yes/no).
## b No studies reported another category for persons of nonbinary gender.# Reported sample distribution of:
## Race ( 0 = did not report, 1 = reported)
tabyl(remir_all, ReportRace)## Ethnicity ( 0 = did not report, 1 = reported)
tabyl(remir_all, ReportHisp)## Gender ( 0 = did not report, 1 = reported)
tabyl(remir_all, ReportGend)## Economic Disadvantage ( 0 = did not report, 1 = reported)
tabyl(remir_all, ReportEcon)remir_all$ReportRace <- factor(remir_all$ReportRace, levels = c(0, 1), labels = c("No", "Yes"))
remir_all$ReportHisp <- factor(remir_all$ReportHisp, levels = c(0, 1), labels = c("No", "Yes"))
remir_all$ReportGend <- factor(remir_all$ReportGend, levels = c(0, 1), labels = c("No", "Yes"))
remir_all$ReportEcon <- factor(remir_all$ReportEcon, levels = c(0, 1), labels = c("No", "Yes"))
tab_race <- tabyl(remir_all, ReportRace) %>%
mutate(Proportion = round(n / sum(n), 2))
tab_hisp <- tabyl(remir_all, ReportHisp) %>%
mutate(Proportion = round(n / sum(n), 2))
tab_gend <- tabyl(remir_all, ReportGend) %>%
mutate(Proportion = round(n / sum(n), 2))
tab_econ <- tabyl(remir_all, ReportEcon) %>%
mutate(Proportion = round(n / sum(n), 2))
## Race
print(tab_race)## ReportRace n percent Proportion
## No 65 0.2226027 0.22
## Yes 227 0.7773973 0.78## Ethnicity
print(tab_hisp)## ReportHisp n percent Proportion
## No 114 0.390411 0.39
## Yes 178 0.609589 0.61## Gender
print(tab_gend)## ReportGend n percent Proportion
## No 27 0.09246575 0.09
## Yes 265 0.90753425 0.91## Economic disadt
print(tab_econ)## ReportEcon n percent Proportion
## No 192 0.6575342 0.66
## Yes 100 0.3424658 0.34# Sample statistics for reported Race (n = 227)
demographic_table <- data.frame(
Demographic = c(
"Asian or Asian American",
"Black or African American",
"Native American or American Indian or Alaska Native",
"Native Hawaiian or Pacific Islander",
"White",
"Multi-racial/Biracial",
"Not specified"
),
Mean = c(
mean(remir_all$asian_pc, na.rm = TRUE),
mean(remir_all$black_pc, na.rm = TRUE),
mean(remir_all$native_pc, na.rm = TRUE),
mean(remir_all$pacif_pc, na.rm = TRUE),
mean(remir_all$white_pc, na.rm = TRUE),
mean(remir_all$multi_pc, na.rm = TRUE),
mean(remir_all$none_pc, na.rm = TRUE)
),
SD = c(
sd(remir_all$asian_pc, na.rm = TRUE),
sd(remir_all$black_pc, na.rm = TRUE),
sd(remir_all$native_pc, na.rm = TRUE),
sd(remir_all$pacif_pc, na.rm = TRUE),
sd(remir_all$white_pc, na.rm = TRUE),
sd(remir_all$multi_pc, na.rm = TRUE),
sd(remir_all$none_pc, na.rm = TRUE)
)
)
demographic_table <- demographic_table %>%
mutate(Mean = round(Mean, 2), SD = round(SD, 2))
print(demographic_table)## Demographic Mean SD
## 1 Asian or Asian American 0.03 0.08
## 2 Black or African American 0.30 0.30
## 3 Native American or American Indian or Alaska Native 0.01 0.02
## 4 Native Hawaiian or Pacific Islander 0.00 0.00
## 5 White 0.38 0.31
## 6 Multi-racial/Biracial 0.02 0.04
## 7 Not specified 0.26 0.22# Sample statistics for reported Ethnicity (n = 178), Gender (n = 265), and Economic Disadvantage (n = 100)
statistics_table <- data.frame(
Variable = c(
"Ethnicity (n = 178)",
"Gender (n = 265)",
"Economic Disadvantage (n = 100)"
),
Mean = c(
mean(remir_all$hisp_pc, na.rm = TRUE),
mean(remir_all$female_pc, na.rm = TRUE),
mean(remir_all$econ_pc, na.rm = TRUE)
),
SD = c(
sd(remir_all$hisp_pc, na.rm = TRUE),
sd(remir_all$female_pc, na.rm = TRUE),
sd(remir_all$econ_pc, na.rm = TRUE)
)
)
statistics_table <- statistics_table %>%
mutate(Mean = round(Mean, 2), SD = round(SD, 2))
print(statistics_table)## Variable Mean SD
## 1 Ethnicity (n = 178) 0.24 0.20
## 2 Gender (n = 265) 0.56 0.23
## 3 Economic Disadvantage (n = 100) 0.68 0.23## Studies That Reported Sample Distribution
remir_us$ReportRace <- factor(remir_us$ReportRace, levels = c(0, 1), labels = c("No", "Yes"))
remir_us$ReportHisp <- factor(remir_us$ReportHisp, levels = c(0, 1), labels = c("No", "Yes"))
remir_us$ReportGend <- factor(remir_us$ReportGend, levels = c(0, 1), labels = c("No", "Yes"))
remir_us$ReportEcon <- factor(remir_us$ReportEcon, levels = c(0, 1), labels = c("No", "Yes"))
tab_race_us <- tabyl(remir_us, ReportRace) %>%
mutate(Proportion = round(n / sum(n), 2))
tab_hisp_us <- tabyl(remir_us, ReportHisp) %>%
mutate(Proportion = round(n / sum(n), 2))
tab_gend_us <- tabyl(remir_us, ReportGend) %>%
mutate(Proportion = round(n / sum(n), 2))
tab_econ_us <- tabyl(remir_us, ReportEcon) %>%
mutate(Proportion = round(n / sum(n), 2))
## Race
print(tab_race_us)## ReportRace n percent Proportion
## No 30 0.125 0.12
## Yes 210 0.875 0.88## Ethnicity
print(tab_hisp_us)## ReportHisp n percent Proportion
## No 62 0.2583333 0.26
## Yes 178 0.7416667 0.74## Gender
print(tab_gend_us)## ReportGend n percent Proportion
## No 23 0.09583333 0.1
## Yes 217 0.90416667 0.9## Economic disadvantage
print(tab_econ_us)## ReportEcon n percent Proportion
## No 152 0.6333333 0.63
## Yes 88 0.3666667 0.37# Sample statistics for studies that report characteristic
demographic_table_us <- data.frame(
Demographic = c(
"Asian or Asian American",
"Black or African American",
"Native American or American Indian or Alaska Native",
"Native Hawaiian or Pacific Islander",
"White",
"Multi-racial/Biracial",
"Not specified"
),
Mean = c(
mean(remir_us$asian_pc, na.rm = TRUE),
mean(remir_us$black_pc, na.rm = TRUE),
mean(remir_us$native_pc, na.rm = TRUE),
mean(remir_us$pacif_pc, na.rm = TRUE),
mean(remir_us$white_pc, na.rm = TRUE),
mean(remir_us$multi_pc, na.rm = TRUE),
mean(remir_us$none_pc, na.rm = TRUE)
),
SD = c(
sd(remir_us$asian_pc, na.rm = TRUE),
sd(remir_us$black_pc, na.rm = TRUE),
sd(remir_us$native_pc, na.rm = TRUE),
sd(remir_us$pacif_pc, na.rm = TRUE),
sd(remir_us$white_pc, na.rm = TRUE),
sd(remir_us$multi_pc, na.rm = TRUE),
sd(remir_us$none_pc, na.rm = TRUE)
)
)
demographic_table_us <- demographic_table_us %>%
mutate(Mean = round(Mean, 2), SD = round(SD, 2))
print(demographic_table_us)## Demographic Mean SD
## 1 Asian or Asian American 0.03 0.06
## 2 Black or African American 0.31 0.30
## 3 Native American or American Indian or Alaska Native 0.01 0.02
## 4 Native Hawaiian or Pacific Islander 0.00 0.00
## 5 White 0.36 0.31
## 6 Multi-racial/Biracial 0.02 0.04
## 7 Not specified 0.27 0.22# Sample statistics for reported Ethnicity, Gender, and Economic Disadvantage
statistics_table_us <- data.frame(
Variable = c(
"Ethnicity (n = 178)",
"Gender (n = 265)",
"Economic Disadvantage (n = 100)"
),
Mean = c(
mean(remir_us$hisp_pc, na.rm = TRUE),
mean(remir_us$female_pc, na.rm = TRUE),
mean(remir_us$econ_pc, na.rm = TRUE)
),
SD = c(
sd(remir_us$hisp_pc, na.rm = TRUE),
sd(remir_us$female_pc, na.rm = TRUE),
sd(remir_us$econ_pc, na.rm = TRUE)
)
)
statistics_table_us <- statistics_table_us %>%
mutate(Mean = round(Mean, 2), SD = round(SD, 2))
print(statistics_table_us)## Variable Mean SD
## 1 Ethnicity (n = 178) 0.24 0.20
## 2 Gender (n = 265) 0.57 0.23
## 3 Economic Disadvantage (n = 100) 0.70 0.23########## SUB SAMPLE ##############################
## Studies That Reported Sample Distribution
remir_us_subgroup$ReportRace <- factor(remir_us_subgroup$ReportRace, levels = c(0, 1), labels = c("No", "Yes"))
remir_us_subgroup$ReportHisp <- factor(remir_us_subgroup$ReportHisp, levels = c(0, 1), labels = c("No", "Yes"))
remir_us_subgroup$ReportGend <- factor(remir_us_subgroup$ReportGend, levels = c(0, 1), labels = c("No", "Yes"))
remir_us_subgroup$ReportEcon <- factor(remir_us_subgroup$ReportEcon, levels = c(0, 1), labels = c("No", "Yes"))
## Studies That Reported Sample Distribution
tabyl(remir_us_subgroup, ReportRace)tabyl(remir_us_subgroup, ReportHisp)tabyl(remir_us_subgroup, ReportGend)tabyl(remir_us_subgroup, ReportEcon)## Sample statistics for studies that report characteristic:
describe(remir_us_subgroup$asian_pc)describe(remir_us_subgroup$black_pc)describe(remir_us_subgroup$native_pc)describe(remir_us_subgroup$pacif_pc)describe(remir_us_subgroup$white_pc)describe(remir_us_subgroup$multi_pc)describe(remir_us_subgroup$none_pc)describe(remir_us_subgroup$hisp_pc)describe(remir_us_subgroup$female_pc)describe(remir_us_subgroup$econ_pc)tab_race_subgroup <- tabyl(remir_us_subgroup, ReportRace) %>%
mutate(Proportion = round(n / sum(n), 2))
tab_hisp_subgroup <- tabyl(remir_us_subgroup, ReportHisp) %>%
mutate(Proportion = round(n / sum(n), 2))
tab_gend_subgroup <- tabyl(remir_us_subgroup, ReportGend) %>%
mutate(Proportion = round(n / sum(n), 2))
tab_econ_subgroup <- tabyl(remir_us_subgroup, ReportEcon) %>%
mutate(Proportion = round(n / sum(n), 2))
## Race
print(tab_race_subgroup)## ReportRace n percent Proportion
## No 10 0.1 0.1
## Yes 90 0.9 0.9## Ethnicity
print(tab_hisp_subgroup)## ReportHisp n percent Proportion
## No 18 0.18 0.18
## Yes 82 0.82 0.82## Gender
print(tab_gend_subgroup)## ReportGend n percent Proportion
## No 9 0.09 0.09
## Yes 91 0.91 0.91## Economic disadvantage
print(tab_econ_subgroup)## ReportEcon n percent Proportion
## No 56 0.56 0.56
## Yes 44 0.44 0.44# Sample statistics for studies that report characteristic
demographic_table_subgroup <- data.frame(
Demographic = c(
"Asian or Asian American",
"Black or African American",
"Native American or American Indian or Alaska Native",
"Native Hawaiian or Pacific Islander",
"White",
"Multi-racial/Biracial",
"Not specified"
),
Mean = c(
mean(remir_us_subgroup$asian_pc, na.rm = TRUE),
mean(remir_us_subgroup$black_pc, na.rm = TRUE),
mean(remir_us_subgroup$native_pc, na.rm = TRUE),
mean(remir_us_subgroup$pacif_pc, na.rm = TRUE),
mean(remir_us_subgroup$white_pc, na.rm = TRUE),
mean(remir_us_subgroup$multi_pc, na.rm = TRUE),
mean(remir_us_subgroup$none_pc, na.rm = TRUE)
),
SD = c(
sd(remir_us_subgroup$asian_pc, na.rm = TRUE),
sd(remir_us_subgroup$black_pc, na.rm = TRUE),
sd(remir_us_subgroup$native_pc, na.rm = TRUE),
sd(remir_us_subgroup$pacif_pc, na.rm = TRUE),
sd(remir_us_subgroup$white_pc, na.rm = TRUE),
sd(remir_us_subgroup$multi_pc, na.rm = TRUE),
sd(remir_us_subgroup$none_pc, na.rm = TRUE)
)
)
demographic_table_subgroup <- demographic_table_subgroup %>%
mutate(Mean = round(Mean, 2), SD = round(SD, 2))
print(demographic_table_subgroup)## Demographic Mean SD
## 1 Asian or Asian American 0.02 0.06
## 2 Black or African American 0.30 0.25
## 3 Native American or American Indian or Alaska Native 0.01 0.02
## 4 Native Hawaiian or Pacific Islander 0.00 0.00
## 5 White 0.35 0.30
## 6 Multi-racial/Biracial 0.01 0.04
## 7 Not specified 0.31 0.21# Sample statistics for reported Ethnicity, Gender, and Economic Disadvantage
statistics_table_subgroup <- data.frame(
Variable = c(
"Ethnicity (n = 178)",
"Gender (n = 265)",
"Economic Disadvantage (n = 100)"
),
Mean = c(
mean(remir_us_subgroup$hisp_pc, na.rm = TRUE),
mean(remir_us_subgroup$female_pc, na.rm = TRUE),
mean(remir_us_subgroup$econ_pc, na.rm = TRUE)
),
SD = c(
sd(remir_us_subgroup$hisp_pc, na.rm = TRUE),
sd(remir_us_subgroup$female_pc, na.rm = TRUE),
sd(remir_us_subgroup$econ_pc, na.rm = TRUE)
)
)
statistics_table_subgroup <- statistics_table_subgroup %>%
mutate(Mean = round(Mean, 2), SD = round(SD, 2))
print(statistics_table_subgroup)## Variable Mean SD
## 1 Ethnicity (n = 178) 0.24 0.18
## 2 Gender (n = 265) 0.53 0.13
## 3 Economic Disadvantage (n = 100) 0.66 0.23# About Supplemental Table 3
#Note: We created a dichotomous variable for reports evaluating culturally tailored or non-culturally tailored programs. Then we collapsed the categories to three: examined subgroup differences, used a homogenous sample, or did not examine subgroup differences. This table lists the proportions that examined subgroup differences within the culturally tailored categories.
#Notes: * Chi-square p < .05 ## Test for Relationship Between Examined and Culturally Tailored
tabyl(remir_us$target_race)tabyl(remir_us,examine12_race_short, target_race ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_race_short, remir_us$target_race, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_race_short and remir_us$target_race
## X-squared = 59.018, df = 2, p-value = 1.529e-13## Test for Relationship Between Examined and Culturally Tailored
tabyl(remir_us$target_ethnic)tabyl(remir_us,examine12_ethnic_short, target_ethnic ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_ethnic_short, remir_us$target_ethnic, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_ethnic_short and remir_us$target_ethnic
## X-squared = 15.904, df = 2, p-value = 0.000352## Test for Relationship Between Examined and Culturally Tailored
tabyl(remir_us$target_gender)tabyl(remir_us,examine12_gender_short, target_gender ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_gender_short, remir_us$target_gender, correct=FALSE)##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_gender_short and remir_us$target_gender
## X-squared = 209.74, df = 2, p-value < 2.2e-16## No Culturally Tailored Program for Sex
## Test for Relationship Between Examined and Culturally Tailored
## Assume all citations are not culturally tailed
tabyl(remir_us$examine12_sex_short)## Test for Relationship Between Examined and Culturally Tailored
tabyl(remir_us$target_ses)tabyl(remir_us,examine12_ses_short, target_ses ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_ses_short, remir_us$target_ses, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_ses_short and remir_us$target_ses
## X-squared = 1.5917, df = 1, p-value = 0.2071## Test for Relationship Between Examined/Benefitted and Culturally Tailored
tabyl(remir_us,examine12_loc_short, target_loc ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_loc_short, remir_us$target_loc, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_loc_short and remir_us$target_loc
## X-squared = 0.25199, df = 1, p-value = 0.6157## There were no culturally tailored programs#About Online Supplemental Table 4:
# Note: We created a dichotomous variable for reports published before 2016 (early period) and after 2017 (late period). Then we collapsed the categories to three: examined subgroup differences, used a homogenous sample, or did not examine subgroup differences. This table lists the proportions that examined subgroup differences within “period of time”.
#* Chi-square p < .05# NOTE TO TEAM (August 2, 2024): I was having trouble getting Online Supplemental Table 4 to run without redefining the variables again. I'm not sure why this is, but am guessing there is a variable somewhere in these tables that were not pre-defined in the script above. I will troubleshoot this more later, once we settle on a general organizational structure for the code.
########### RECREATE LONG TEST VARIABLES FOR FULL SAMPLE ############
### RACE
## ## Change Interaction Examined Variable Names
names(remir_all)[13] <- "inter_examine1_race"
names(remir_all)[76] <- "inter_examine2_race"
table(remir_all$inter_examine1_race, useNA = "always")##
## Black/Not Black
## 2
## Minority/majority
## 2
## Minority/Majority
## 2
## W/B (White/African American or Black)
## 2
## W/B (White/African American or Black);W/As (White/Asian or Asian American);W/PI (White/Native Hawaiian or Pacific Islander);W/NA (White/Native American or American Indian or Native Alaskan);B/As (Black or African American/Asian or Asian American);B/PI (Black or African American/Native Hawaiian or Pacific Islander);B/NA (Black or African American/Native American or American Indian or Native Alaskan);As/NA (Asian or Asian American/Native American or American Indian or Native Alaskan);PI/NA (Native Hawaiian or Pacific Islander/Native American or American Indian or Native Alaskan)
## 1
## W/B (White/African American or Black);W/As (White/Asian or Asian American);W/PI (White/Native Hawaiian or Pacific Islander);W/NA (White/Native American or American Indian or Native Alaskan);B/As (Black or African American/Asian or Asian American);B/PI (Black or African American/Native Hawaiian or Pacific Islander);B/NA (Black or African American/Native American or American Indian or Native Alaskan);As/NA (Asian or Asian American/Native American or American Indian or Native Alaskan);PI/NA (Native Hawaiian or Pacific Islander/Native American or American Indian or Native Alaskan);White/Multiracial; Black/Multiracial; Asian or Pacific Islander/Multiracial; Native American/Multiracial
## 1
## W/B (White/African American or Black);White/Other or Multiracial
## 2
## White/Minority
## 4
## White/Non-White
## 2
## White/Not White
## 1
## <NA>
## 273table(remir_all$inter_examine2_race, useNA = "always")##
## Black/Not Black
## 5
## Minority/Majority
## 5
## None
## 1
## W/As (White/Asian or Asian American);White/Minority
## 1
## W/B (White/African American or Black)
## 6
## W/B (White/African American or Black);White/Multiracial; Black/Multiracial
## 1
## White + Another Race/Black
## 3
## White/Other
## 1
## <NA>
## 269## Combine Two Tnteractions Examined
remir_all$inter_examine12_race <- ifelse(is.na(remir_all$inter_examine1_race),
remir_all$inter_examine2_race, remir_all$inter_examine1_race)
tabyl(remir_all$inter_examine12_race)############## What Race Interactions Were Significant ################
## Change Interaction Significant Variable Names
names(remir_all)[14] <- "inter_signif1_race"
names(remir_all)[77] <- "inter_signif2_race"
table(remir_all$inter_signif1_race, useNA = "always")##
## None
## 14
## W/B (White/African American or Black)
## 1
## W/B (White/African American or Black);White/Other or Multiracial
## 1
## White/Minority
## 3
## <NA>
## 273table(remir_all$inter_signif2_race, useNA = "always")##
## Black/Not Black
## 3
## Minority/Majority
## 2
## None
## 14
## W/As (White/Asian or Asian American);White/Minority
## 1
## W/B (White/African American or Black)
## 1
## W/B (White/African American or Black);White/Multiracial; Black/Multiracial
## 1
## White + Another Race/Black
## 1
## <NA>
## 269## Combine Two Tnteractions Significant
remir_all$inter_signif12_race <- ifelse(is.na(remir_all$inter_signif1_race),
remir_all$inter_signif2_race, remir_all$inter_signif1_race)
tabyl(remir_all$inter_signif12_race)############## What Race Interaction Groups Benefitted More ##############
## Change Interaction Benefitted Variable Names
names(remir_all)[16] <- "inter_benefit1_race"
names(remir_all)[79] <- "inter_benefit2_race"
table(remir_all$inter_benefit1_race, useNA = "always")##
## Black or African American;Other or Multiracial
## 1
## Minority
## 3
## None
## 14
## White
## 1
## <NA>
## 273table(remir_all$inter_benefit2_race, useNA = "always")##
## Asian or Asian American;Minority Black or African American
## 1 2
## Black or African American;White Majority
## 1 2
## None Not Black
## 14 2
## White + Another Race <NA>
## 1 269## Combine Two Tnteractions Benefitted
remir_all$inter_benefit12_race <- ifelse(is.na(remir_all$inter_benefit1_race),
remir_all$inter_benefit2_race, remir_all$inter_benefit1_race)
tabyl(remir_all$inter_benefit12_race)############# What Race Subgroup Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[47] <- "subg_examine1_race"
names(remir_all)[80] <- "subg_examine2_race"
table(remir_all$subg_examine1_race, useNA = "always")##
## Asian or Asian American;Black or African American
## 1
## Black or African American
## 3
## Black or African American;White
## 3
## White;Non-White
## 2
## <NA>
## 283table(remir_all$subg_examine2_race, useNA = "always")##
## Asian or Asian American;Black or African American;White
## 1
## Asian or Asian American;White;Minority
## 1
## Black or African American
## 3
## Black or African American;Not Black
## 1
## Black or African American;White
## 4
## Black or African American;White + Another Race
## 3
## Black or African American;White;Multiracial
## 1
## Minority
## 1
## Minority; Majority
## 2
## None
## 5
## White
## 1
## <NA>
## 269## Combine Two Subgroups Examined
remir_all$subg_examine12_race <- ifelse(is.na(remir_all$subg_examine1_race),
remir_all$subg_examine2_race, remir_all$subg_examine1_race)
tabyl(remir_all$subg_examine12_race)############## What Race Subgroups Were Significant ######################
## Change Subgroup Significant Variable Names
names(remir_all)[49] <- "subg_signif1_race"
names(remir_all)[82] <- "subg_signif2_race"
table(remir_all$subg_signif1_race, useNA = "always")##
## Asian or Asian American;Black or African American
## 1
## Black or African American
## 5
## None
## 1
## White;Non-White
## 2
## <NA>
## 283table(remir_all$subg_signif2_race, useNA = "always")##
## Asian or Asian American;Minority
## 1
## Black or African American
## 5
## Black or African American;Not Black
## 1
## Black or African American;White
## 3
## Black or African American;White + Another Race
## 3
## Black or African American;White;Multiracial
## 1
## Minority
## 1
## Minority; Majority
## 2
## None
## 5
## White
## 1
## <NA>
## 269## Combine Two Subgroup Significant
remir_all$subg_signif12_race <- ifelse(is.na(remir_all$subg_signif1_race),
remir_all$subg_signif2_race, remir_all$subg_signif1_race)
tabyl(remir_all$subg_signif12_race)## INTERACTION RACE EXAMINED
tabyl(remir_all$inter_examine12_race)## Simplify Codes
## Define as mixed if combined Black, Asian, etc into broad category
## Define as exact if referred specifically to Black, Asian, etc
## None = NA, as it means that race subgroups not examined
## Recode to Mixed and None
remir_all$inter_examine12_race_cat <-
car::recode(remir_all$inter_examine12_race, "
'Minority/Majority' = 'Mixed';
'Minority/majority' = 'Mixed';
'White/Minority' = 'Mixed';
'White/Not White' = 'Mixed';
'White/Non-White' = 'Mixed';
'White/Other' = 'Mixed';
'None' = NA ")
## Code remaining to Exact
remir_all$inter_examine12_race_cat <-
ifelse(remir_all$inter_examine12_race_cat != "Mixed", "Exact",
remir_all$inter_examine12_race_cat)
## Check -- Looks OK
tabyl(remir_all$inter_examine12_race_cat)tabyl(remir_all, inter_examine12_race, inter_examine12_race_cat)## SUBGROUP RACE EXAMINED
tabyl(remir_all$subg_examine12_race)## Simplify codes
## Define as mixed if combined Black, Asian, etcs into broad category
## Define as exact if referred specifically to Black, Asian, etc
## None = NA, as it means that race subgroups not examined
remir_all$subg_examine12_race_cat <-
ifelse(remir_all$subg_examine12_race == "White;Non-White" |
remir_all$subg_examine12_race == "Minority" |
remir_all$subg_examine12_race == "Minority; Majority", "Mixed",
ifelse(remir_all$subg_examine12_race == "None", NA,"Exact"))
## Check -- Looks OK
tabyl(remir_all, subg_examine12_race, subg_examine12_race_cat)tabyl(remir_all$subg_examine12_race_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE -- define over .75 as Homogeneous
tabyl(remir_all$inter_examine12_race_cat)remir_all$inter_examine12_race_cat <- ifelse(
(remir_all$black_all > .75 |
remir_all$asian_all > .75 |
remir_all$native_all > .75 |
remir_all$pacif_all > .75 ) &
is.na(remir_all$inter_examine12_race_cat) &
is.na(remir_all$subg_examine12_race_cat),
"Homog", remir_all$inter_examine12_race_cat)
tabyl(remir_all$inter_examine12_race_cat)## Check -- Mixed and Exact unchanged but many NAs now homogeneous
remir_all %>% filter(remir_all$inter_examine12_race_cat == 'Homog') %>%
select(black_all, asian_all, native_all, pacif_all, white_all) %>%
print(n=26)## # A tibble: 26 × 5
## black_all asian_all native_all pacif_all white_all
## <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 0.8 0 0 0 0
## 2 0.8 0 0 0 0
## 3 0.87 0 0 0 0
## 4 0.84 0 0 0 0
## 5 0.9 0.01 0 0 0.07
## 6 0.83 0 0 0 0.01
## 7 0.83 0 0 0 0.01
## 8 0.85 0.05 0 0 0.02
## 9 0.95 0 0 0 0
## 10 1 0 0 0 0
## 11 1 0 0 0 0
## 12 1 0 0 0 0
## 13 1 0 0 0 0
## 14 1 0 0 0 0
## 15 0.85 0 0 0 0
## 16 0.85 0 0 0 0
## 17 0.85 0 0 0 0
## 18 0.82 0 0 0 0.15
## 19 0.82 0 0 0 0.15
## 20 0.8 0 0 0 0
## 21 0.8 0 0 0 0
## 22 1 0 0 0 0
## 23 1 0 0 0 0
## 24 1 0 0 0 0
## 25 1 0 0 0 0
## 26 0.87 0 0 0 0## All Homogeneous samples are African American
## INTERACTION RACE BENEFIT
tabyl(remir_all$inter_benefit12_race)## Simplify Codes
## Need to Use Three Variables to Simplify (benefit, examine, main effect)
tabyl(remir_all, inter_benefit12_race, inter_examine12_race_cat)tabyl(remir_all, inter_benefit12_race, main_effect)## Code as Exact if benefit any specific group of Blacks, Asians, etc
## Code as Mixed if benefit a combined group of Blacks, Asians, etc
## Code as Other if benefit whites only or no group
## If None (i.e., equal effects), count as Exact or Mixed benefit if main effect
## If None (i.e., equal effects), count as no benefit or other if No main effect
## If None (i.e., equal effects) but examined is NA, then None means NA
remir_all$inter_benefit12_race_cat <-
ifelse(remir_all$inter_benefit12_race == 'Asian or Asian American;Minority' |
remir_all$inter_benefit12_race == 'Black or African American;Other or Multiracial' |
remir_all$inter_benefit12_race == 'Black or African American' |
remir_all$inter_benefit12_race == 'Black or African American;White', 'Exact',
ifelse(remir_all$inter_benefit12_race == 'Minority' |
remir_all$inter_benefit12_race == 'White + Another Race', 'Mixed',
ifelse(remir_all$inter_benefit12_race == 'Not Black' |
remir_all$inter_benefit12_race == 'White' |
remir_all$inter_benefit12_race == 'Majority', 'Other',
ifelse(remir_all$inter_benefit12_race == 'None' &
remir_all$main_effect == 'Yes' &
remir_all$inter_examine12_race_cat == 'Exact', 'Exact',
ifelse(remir_all$inter_benefit12_race == 'None' &
remir_all$main_effect == 'Yes' &
remir_all$inter_examine12_race_cat == 'Mixed', 'Mixed',
ifelse(remir_all$inter_benefit12_race == 'None' &
remir_all$main_effect == 'No' &
!is.na(remir_all$inter_examine12_race_cat), 'Other', NA))))))
## Check -- Looks OK
tabyl(remir_all$inter_benefit12_race_cat)tabyl(remir_all, inter_benefit12_race, inter_benefit12_race_cat)## SUBGROUP RACE BENEFIT
tabyl(remir_all$subg_signif12_race)## Simplify codes
## None is NA if examined subgroups is NA
## None means no subgroup effects if examined is not NA (i.e., other or no benefit)
remir_all$subg_signif12_race_cat <-
ifelse(is.na(remir_all$subg_examine12_race_cat), NA,
ifelse(remir_all$subg_signif12_race == "White;Non-White" |
remir_all$subg_signif12_race == 'Minority' |
remir_all$subg_signif12_race == 'Minority; Majority', "Mixed",
ifelse(remir_all$subg_signif12_race == "None" |
remir_all$subg_signif12_race == "White", "Other", "Exact")))
tabyl(remir_all$subg_signif12_race_cat)## Check Looks OK
tabyl(remir_all$subg_signif12_race_cat)tabyl(remir_all, subg_signif12_race, subg_signif12_race_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_all$inter_benefit12_race_cat <- ifelse(
(remir_all$black_all > .75 |
remir_all$asian_all > .75 |
remir_all$native_all > .75 |
remir_all$pacif_all > .75 ) &
is.na(remir_all$inter_benefit12_race_cat) &
is.na(remir_all$subg_signif12_race_cat),
"Homog", remir_all$inter_benefit12_race_cat)
tabyl(remir_all$inter_benefit12_race_cat)## ETHNICITY
## Change Inteaction Examined Variable Names
names(remir_all)[18] <- "inter_examine1_ethnic"
names(remir_all)[84] <- "inter_examine2_ethnic"
table(remir_all$inter_examine1_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/majority Minority/Majority
## 5 2 1
## White/Minority White/Non-White <NA>
## 3 1 280table(remir_all$inter_examine2_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/Majority None
## 14 5 2
## White/Minority White/Other <NA>
## 1 1 269## Combine Two Tnteractions Examined
remir_all$inter_examine12_ethnic <- ifelse(is.na(remir_all$inter_examine1_ethnic),
remir_all$inter_examine2_ethnic, remir_all$inter_examine1_ethnic)
tabyl(remir_all$inter_examine12_ethnic)############## What Ethnic Interactions Were Significant ###############
## Change Interaction Significant Variable Names
names(remir_all)[19] <- "inter_signif1_ethnic"
names(remir_all)[85] <- "inter_signif2_ethnic"
table(remir_all$inter_signif1_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic None White/Minority
## 2 11 3
## <NA>
## 276table(remir_all$inter_signif2_ethnic, useNA = "always")##
## Hispanic/Non-Hispanic Minority/Majority None
## 6 2 14
## White/Minority <NA>
## 1 269## Combine Two Tnteractions Significant
remir_all$inter_signif12_ethnic <- ifelse(is.na(remir_all$inter_signif1_ethnic),
remir_all$inter_signif2_ethnic, remir_all$inter_signif1_ethnic)
tabyl(remir_all$inter_signif12_ethnic)############## What Ethnic Interaction Groups Benefited More ############
## Change Interaction Benefitted Variable Names
names(remir_all)[21] <- "inter_benefit1_ethnic"
names(remir_all)[87] <- "inter_benefit2_ethnic"
table(remir_all$inter_benefit1_ethnic, useNA = "always")##
## Hispanic Minority None <NA>
## 2 3 11 276table(remir_all$inter_benefit2_ethnic, useNA = "always")##
## Hispanic Majority Minority Non-Hispanic None <NA>
## 3 2 1 3 14 269## Combine Two Tnteractions Benefitted
remir_all$inter_benefit12_ethnic <- ifelse(is.na(remir_all$inter_benefit1_ethnic),
remir_all$inter_benefit2_ethnic, remir_all$inter_benefit1_ethnic)
tabyl(remir_all$inter_benefit12_ethnic)############# What Ethnic Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[51] <- "subg_examine1_ethnic"
names(remir_all)[88] <- "subg_examine2_ethnic"
table(remir_all$subg_examine1_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic <NA>
## 6 1 285table(remir_all$subg_examine2_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic Minority
## 12 3 2
## Minority; Majority None <NA>
## 2 4 269## Combine Two Subgroups Examined
remir_all$subg_examine12_ethnic <- ifelse(is.na(remir_all$subg_examine1_ethnic),
remir_all$subg_examine2_ethnic, remir_all$subg_examine1_ethnic)
tabyl(remir_all$subg_examine12_ethnic)############## What Ethnic Subgroups Were Significant ##################
## Change Subgroup Significant Variable Names
names(remir_all)[53] <- "subg_signif1_ethnic"
names(remir_all)[90] <- "subg_signif2_ethnic"
table(remir_all$subg_signif1_ethnic, useNA = "always")##
## Hispanic None <NA>
## 5 2 285table(remir_all$subg_signif2_ethnic, useNA = "always")##
## Hispanic Hispanic;Non-Hispanic Minority
## 10 3 2
## Minority; Majority None <NA>
## 2 6 269## Combine Two Tnteractions Significant
remir_all$subg_signif12_ethnic <- ifelse(is.na(remir_all$subg_signif1_ethnic),
remir_all$subg_signif2_ethnic, remir_all$subg_signif1_ethnic)
tabyl(remir_all$subg_signif12_ethnic)## INTERACTION ETHNIC EXAMINED
tabyl(remir_all$inter_examine12_ethnic)## Simplify Codes
## Define as mixed if combined Hispanics into broad category
## Define as exact if referred specifically to Hispanic, Non-Hispanic
## None = NA for examined, as it means that Hispanic subgroups not examined
## First, assign exact and none
remir_all$inter_examine12_ethnic_cat <-
car::recode(remir_all$inter_examine12_ethnic, "
'Hispanic/Non-Hispanic' = 'Exact';
'None' = NA ")
tabyl(remir_all$inter_examine12_ethnic_cat)## Second, remaining categores are assigned as mixed
remir_all$inter_examine12_ethnic_cat <-
ifelse(remir_all$inter_examine12_ethnic_cat != "Exact", "Mixed",
remir_all$inter_examine12_ethnic_cat)
tabyl(remir_all$inter_examine12_ethnic_cat)## Check -- Looks OK
tabyl(remir_all, inter_examine12_ethnic, inter_examine12_ethnic_cat)## SUBGROUP ETHNIC EXAMINED
tabyl(remir_all$subg_examine12_ethnic)## Simplify Codes
## Define as mixed if combined Hispanics into broad category
## Define as exact if referred specifically to Hispanic, Non-HispaniceBlack, Asian, etc
## None = NA for examined, as it means that Hispanic subgroups not examined
remir_all$subg_examine12_ethnic_cat <-
ifelse(remir_all$subg_examine12_ethnic == 'Minority'|
remir_all$subg_examine12_ethnic == 'Minority; Majority',
'Mixed',
ifelse(remir_all$subg_examine12_ethnic == 'None', NA,
'Exact'))
## Check -- Looks OK
tabyl(remir_all$subg_examine12_ethnic_cat)tabyl(remir_all, subg_examine12_ethnic, subg_examine12_ethnic_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE OF HISPANICS
tabyl(remir_all$inter_examine12_ethnic_cat)remir_all$inter_examine12_ethnic_cat <- ifelse(
(remir_all$hisp_all > .75) &
is.na(remir_all$inter_examine12_ethnic_cat) &
is.na(remir_all$subg_examine12_ethnic_cat),
"Homog", remir_all$inter_examine12_ethnic_cat)
tabyl(remir_all$inter_examine12_ethnic_cat)tabyl(remir_all, inter_examine12_ethnic, inter_examine12_ethnic_cat)## INTERACTION ETHNIC BENEFIT
tabyl(remir_all$inter_benefit12_ethnic)## Simplify Codes
## Code as Exact Benefit if Hispanic
## Code as Other if Non-Hispanic or Majority
## Code as Mixed if Minority
## If None (i.e., equal effects) count as benefit if main effect
## If None (i.e., equal effects) count as other if if no main effect
## If None and ethnic not examined, code as NA as meaning of None for interaction is unclear
## Code Unclear as NA
remir_all$inter_benefit12_ethnic_cat <-
ifelse(remir_all$inter_benefit12_ethnic == 'Hispanic', 'Exact',
ifelse(remir_all$inter_benefit12_ethnic == 'Minority', 'Mixed',
ifelse(remir_all$inter_benefit12_ethnic == 'Non-Hispanic' |
remir_all$inter_benefit12_ethnic == 'Majority', 'Other',
ifelse(remir_all$inter_benefit12_ethnic == 'None' &
remir_all$main_effect == 'Yes' &
remir_all$inter_examine12_ethnic_cat == 'Exact', 'Exact',
ifelse(remir_all$inter_benefit12_ethnic == 'None' &
remir_all$main_effect == 'Yes' &
remir_all$inter_examine12_ethnic_cat == 'Mixed', 'Mixed',
ifelse(remir_all$inter_benefit12_ethnic == 'None' &
remir_all$main_effect == 'No' &
!is.na(remir_all$inter_examine12_ethnic_cat), 'Other', NA))))))
## Check -- Looks OK
tabyl(remir_all$inter_benefit12_ethnic_cat)tabyl(remir_all, inter_benefit12_ethnic, inter_benefit12_ethnic_cat)## SUBGROUP ETHNIC BENEFIT
tabyl(remir_all$subg_signif12_ethnic)## Simplify codes
## None here means no subgroup effects, should be other or no benefit
remir_all$subg_signif12_ethnic_cat <-
ifelse(remir_all$subg_signif12_ethnic == "Hispanic" |
remir_all$subg_signif12_ethnic == "Hispanic;Non-Hispanic" , "Exact",
ifelse(remir_all$subg_signif12_ethnic == "None", "Other",
ifelse(remir_all$subg_signif12_ethnic == "Minority" |
remir_all$subg_signif12_ethnic == "Minority; Majority", "Mixed",
ifelse(is.na(remir_all$subg_examine12_ethnic), NA, 'Check'))))
## Check
tabyl(remir_all$subg_signif12_ethnic_cat)tabyl(remir_all, subg_signif12_ethnic, subg_signif12_ethnic_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE
tabyl(remir_all$inter_benefit12_ethnic_cat)remir_all$inter_benefit12_ethnic_cat <- ifelse(
(remir_all$hisp_all > .75) &
is.na(remir_all$inter_benefit12_ethnic_cat) &
is.na(remir_all$subg_signif12_ethnic_cat),
"Homog", remir_all$inter_benefit12_ethnic_cat)
tabyl(remir_all$inter_benefit12_ethnic_cat)## Problem NA n = 200 for examined n = 198 for benefited
## How can cases that don't examine ethicity show a benefit
remir_all %>% filter(is.na(inter_examine12_ethnic_cat) &
is.na(subg_examine12_ethnic_cat) &
!is.na(inter_benefit12_ethnic_cat) &
is.na(subg_signif12_ethnic_cat)) %>%
select(Citation.ID, Program.Name, inter_examine12_ethnic_cat,
subg_examine12_ethnic_cat) remir_all %>% filter(is.na(inter_examine12_ethnic_cat) &
is.na(subg_examine12_ethnic_cat) &
!is.na(inter_benefit12_ethnic_cat) &
is.na(subg_signif12_ethnic_cat)) %>%
select(Citation.ID, Program.Name, inter_benefit12_ethnic_cat,
subg_signif12_ethnic_cat)## Change 3980 and 3981 to NA so that examined and benefit are consistent
tabyl(remir_all$inter_benefit12_ethnic_cat)remir_all$inter_benefit12_ethnic_cat <-
ifelse(remir_all$Citation.ID == 3980, NA, remir_all$inter_benefit12_ethnic_cat)
tabyl(remir_all$inter_benefit12_ethnic_cat)remir_all$inter_benefit12_ethnic_cat <-
ifelse(remir_all$Citation.ID == 3981, NA, remir_all$inter_benefit12_ethnic_cat)
tabyl(remir_all$inter_benefit12_ethnic_cat)## GENDER
############## What Gender Interactions Were Examined
## Change Interaction Examined Gender Variable Names
names(remir_all)[23] <- "inter_examine1_gender"
names(remir_all)[92] <- "inter_examine2_gender"
table(remir_all$inter_examine1_gender, useNA = "always")##
## Male/Female <NA>
## 41 251table(remir_all$inter_examine2_gender, useNA = "always")##
## Male/Female <NA>
## 37 255## Combine Two Gender Interactions Examined
remir_all$inter_examine12_gender <- ifelse(is.na(remir_all$inter_examine1_gender),
remir_all$inter_examine2_gender, remir_all$inter_examine1_gender)
tabyl(remir_all$inter_examine12_gender)############## What Gender interactions Were Significant ###############
## Change Interaction Significant Variable Names
names(remir_all)[24] <- "inter_signif1_gender"
names(remir_all)[93] <- "inter_signif2_gender"
table(remir_all$inter_signif1_gender, useNA = "always")##
## Male/Female None <NA>
## 6 35 251table(remir_all$inter_signif2_gender, useNA = "always")##
## Male/Female None <NA>
## 18 19 255## Combine Two Tnteractions Significant
remir_all$inter_signif12_gender <- ifelse(is.na(remir_all$inter_signif1_gender),
remir_all$inter_signif2_gender, remir_all$inter_signif1_gender)
tabyl(remir_all$inter_signif12_gender)########## What Gender Interaction Groups Benefited More ##################
## Change Interaction Benefitted Variable Names
names(remir_all)[26] <- "inter_benefit1_gender"
names(remir_all)[95] <- "inter_benefit2_gender"
table(remir_all$inter_benefit1_gender, useNA = "always")##
## Female Male None Unclear <NA>
## 4 1 35 1 251table(remir_all$inter_benefit2_gender, useNA = "always")##
## Female Male None Unclear <NA>
## 7 7 19 4 255## Combine Two Tnteractions
remir_all$inter_benefit12_gender <- ifelse(is.na(remir_all$inter_benefit1_gender),
remir_all$inter_benefit2_gender, remir_all$inter_benefit1_gender)
tabyl(remir_all$inter_benefit12_gender)############# What Gender Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[55] <- "subg_examine1_gender"
names(remir_all)[96] <- "subg_examine2_gender"
table(remir_all$subg_examine1_gender, useNA = "always")##
## Male;Female <NA>
## 13 279table(remir_all$subg_examine2_gender, useNA = "always")##
## Female Male;Female None <NA>
## 1 34 2 255## Combine Two Subgroups Examined
remir_all$subg_examine12_gender <- ifelse(is.na(remir_all$subg_examine1_gender),
remir_all$subg_examine2_gender, remir_all$subg_examine1_gender)
tabyl(remir_all$subg_examine12_gender)############## What Gender Subgroups Were Significant #####################
## Change Subgroup Significant Variable Names
names(remir_all)[57] <- "subg_signif1_gender"
names(remir_all)[98] <- "subg_signif2_gender"
table(remir_all$subg_signif1_gender, useNA = "always")##
## Female Male Male;Female None <NA>
## 2 3 6 2 279table(remir_all$subg_signif2_gender, useNA = "always")##
## Female Male Male;Female None <NA>
## 3 2 25 7 255## Combine Two Tnteractions Significant
remir_all$subg_signif12_gender <- ifelse(is.na(remir_all$subg_signif1_gender),
remir_all$subg_signif2_gender, remir_all$subg_signif1_gender)
tabyl(remir_all$subg_signif12_gender)## INTERACTION GENDER EXAMINED
tabyl(remir_all$inter_examine12_gender)## SUBGROUP GENDER EXAMINED
tabyl(remir_all$subg_examine12_gender)## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_all$inter_examine12_gender <- ifelse(
(remir_all$female_all > .75 ) &
is.na(remir_all$inter_examine12_gender) &
is.na(remir_all$subg_examine12_gender),
"Homog", remir_all$inter_examine12_gender)
tabyl(remir_all$inter_examine12_gender)## INTERACTION GENDER BENEFITTTED
tabyl(remir_all$inter_benefit12_gender)## Simplify Codes
remir_all$inter_benefit12_gender_cat <-
ifelse(remir_all$inter_benefit12_gender == 'None' &
remir_all$main_effect == 'Yes', 'Both',
ifelse(remir_all$inter_benefit12_gender == 'None' &
remir_all$main_effect == 'No', 'Neither',
remir_all$inter_benefit12_gender))
## Check -- Look OK
tabyl(remir_all, inter_benefit12_gender, inter_benefit12_gender_cat)tabyl(remir_all$inter_benefit12_gender_cat)## SUBGROUP GENDER BENEFITTED
tabyl(remir_all$subg_signif12_gender)## SEPARATE OUT HOMOGENOUS SAMPLE
remir_all$inter_benefit12_gender_cat <- ifelse(
(remir_all$female_all > .75) &
is.na(remir_all$inter_benefit12_gender_cat) &
is.na(remir_all$subg_signif12_gender),
"Homog", remir_all$inter_benefit12_gender_cat)
tabyl(remir_all$inter_benefit12_gender_cat)############## SEX ########################
############## What Sex Interaction Groups Were Examined ####################
## Change Interaction Examined Variable Names
names(remir_all)[27] <- "inter_examine1_sex"
names(remir_all)[100] <- "inter_examine2_sex"
table(remir_all$inter_examine1_sex, useNA = "always")##
## No Yes <NA>
## 51 2 239table(remir_all$inter_examine2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Examined
remir_all$inter_examine12_sex <- ifelse(is.na(remir_all$inter_examine1_sex),
remir_all$inter_examine2_sex, remir_all$inter_examine1_sex)
tabyl(remir_all$inter_examine12_sex)############## What Sex interactions Were Significant ###################
## Change Interaction Significant Variable Names
names(remir_all)[28] <- "inter_signif1_sex"
names(remir_all)[101] <- "inter_signif2_sex"
table(remir_all$inter_signif1_sex, useNA = "always")##
## None <NA>
## 2 290table(remir_all$inter_signif2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Significant
remir_all$inter_signif12_sex <- ifelse(is.na(remir_all$inter_signif1_sex),
remir_all$inter_signif2_sex, remir_all$inter_signif1_sex)
tabyl(remir_all$inter_signif12_sex)############ What Sex Interaction Groups Benefited More ####################
## Change Interactions Benefitted Variable Names
names(remir_all)[30] <- "inter_benefit1_sex"
names(remir_all)[103] <- "inter_benefit2_sex"
table(remir_all$inter_benefit1_sex, useNA = "always")##
## None <NA>
## 2 290table(remir_all$inter_benefit2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Benefitted
remir_all$inter_benefit12_sex <- ifelse(is.na(remir_all$inter_benefit1_sex),
remir_all$inter_benefit2_sex, remir_all$inter_benefit1_sex)
tabyl(remir_all$inter_benefit12_sex)############# What Sex Subgroup Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[59] <- "subg_examine1_sex"
names(remir_all)[104] <- "subg_examine2_sex"
table(remir_all$subg_examine1_sex, useNA = "always")##
## <NA>
## 292table(remir_all$subg_examine2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Subgroups Examined
remir_all$subg_examine12_sex <- ifelse(is.na(remir_all$subg_examine1_sex),
remir_all$subg_examine2_sex, remir_all$subg_examine1_sex)
tabyl(remir_all$subg_examine12_sex)############## What Sex Subgroups Were Significant #######################
## Change Subgroup Significant Variable Names
names(remir_all)[61] <- "subg_signif1_sex"
names(remir_all)[106] <- "subg_signif2_sex"
table(remir_all$subg_signif1_sex, useNA = "always")##
## <NA>
## 292table(remir_all$subg_signif2_sex, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Significant
remir_all$subg_signif12_sex <- ifelse(is.na(remir_all$subg_signif1_sex),
remir_all$subg_signif2_sex, remir_all$subg_signif1_sex)
tabyl(remir_all$subg_signif12_sex)## INTERACTION SEX EXAMINED
## Codebook didn't ask which groups examined in interaction
## Used were any groups examined
tabyl(remir_all$inter_examine12_sex)## SUBGROUP SEX EXAMINED
tabyl(remir_all$subg_examine12_sex)## SEPARATE OUT HOMOGENEOUS SAMPLE
## Not possible, no measure of lgbt sample in Remir 1
## INTERACTION SEX BENEFITTED
tabyl(remir_all$inter_benefit12_sex)tabyl(remir_all, inter_benefit12_sex, main_effect)## None with main effect means benefit
remir_all$inter_benefit12_sex_cat <-
ifelse(remir_all$inter_benefit12_sex == 'None' &
remir_all$main_effect == 'Yes', 'Yes', remir_all$inter_benefit12_sex)
tabyl(remir_all$inter_benefit12_sex_cat)## SUBGROUP SEX BENEFITTED
tabyl(remir_all$subg_signif12_sex)############## What SES Interaction Group Were Examined ########################
## Change Interaction SES Varialbe Names
names(remir_all)[31] <- "inter_examine1_ses"
names(remir_all)[108] <- "inter_examine2_ses"
table(remir_all$inter_examine1_ses, useNA = "always")##
## No Yes <NA>
## 40 13 239table(remir_all$inter_examine2_ses, useNA = "always")##
## Low/High <NA>
## 13 279## Combine Two Tnteractions Examined
remir_all$inter_examine12_ses <- ifelse(is.na(remir_all$inter_examine1_ses),
remir_all$inter_examine2_ses, remir_all$inter_examine1_ses)
tabyl(remir_all$inter_examine12_ses)############## What SES Interactions Were Significant ####################
## Change Interaction Significant Variable Names
names(remir_all)[32] <- "inter_signif1_ses"
names(remir_all)[109] <- "inter_signif2_ses"
table(remir_all$inter_signif1_ses, useNA = "always")##
## Low/High None <NA>
## 5 8 279table(remir_all$inter_signif2_ses, useNA = "always")##
## Low/High None <NA>
## 4 9 279## Combine Two Tnteractions Significant
remir_all$inter_signif12_ses <- ifelse(is.na(remir_all$inter_signif1_ses),
remir_all$inter_signif2_ses, remir_all$inter_signif1_ses)
tabyl(remir_all$inter_signif12_ses)############ What SES Interaction Groups Benefited More ##################
## Change Interaction Benefitted Variable Names
names(remir_all)[34] <- "inter_benefit1_ses"
names(remir_all)[111] <- "inter_benefit2_ses"
table(remir_all$inter_benefit1_ses, useNA = "always")##
## High Low None <NA>
## 1 4 8 279table(remir_all$inter_benefit2_ses, useNA = "always")##
## High Low None Unclear <NA>
## 1 2 9 1 279## Combine Two Tnteractions
remir_all$inter_benefit12_ses <- ifelse(is.na(remir_all$inter_benefit1_ses),
remir_all$inter_benefit2_ses, remir_all$inter_benefit1_ses)
tabyl(remir_all$inter_benefit12_ses)############# What SES Subgroups Were Examined ####################
## Change Subgroups Examined Variable Names
names(remir_all)[63] <- "subg_examine1_ses"
names(remir_all)[112] <- "subg_examine2_ses"
table(remir_all$subg_examine1_ses, useNA = "always")##
## Low Low;High <NA>
## 3 6 283table(remir_all$subg_examine2_ses, useNA = "always")##
## Low Low;High <NA>
## 4 9 279## Combine Two Subgroups Examined
remir_all$subg_examine12_ses <- ifelse(is.na(remir_all$subg_examine1_ses),
remir_all$subg_examine2_ses, remir_all$subg_examine1_ses)
tabyl(remir_all$subg_examine12_ses)############## What SES Subgroups Were Significant #########################
## Change Subgroups Significant Variable Names
names(remir_all)[65] <- "subg_signif1_ses"
names(remir_all)[114] <- "subg_signif2_ses"
table(remir_all$subg_signif1_ses, useNA = "always")##
## Low Low;High None <NA>
## 5 2 2 283table(remir_all$subg_signif2_ses, useNA = "always")##
## High Low Low;High <NA>
## 1 5 7 279## Combine Two Tnteractions Significant
remir_all$subg_signif12_ses <- ifelse(is.na(remir_all$subg_signif1_ses),
remir_all$subg_signif2_ses, remir_all$subg_signif1_ses)
tabyl(remir_all$subg_signif12_ses)remir_us$subg_signif12_ses <- ifelse(is.na(remir_us$subg_signif1_ses),
remir_us$subg_signif2_ses, remir_us$subg_signif1_ses)
tabyl(remir_us$subg_signif12_ses)## INTERACTION SES EXAMINED
tabyl(remir_all$inter_examine12_ses)## Change No to NA
remir_all$inter_examine12_ses_cat <- ifelse(
remir_all$inter_examine12_ses == 'No', NA, remir_all$inter_examine12_ses)
tabyl(remir_all$inter_examine12_ses_cat)## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_all$inter_examine12_ses_cat <- ifelse(
(remir_all$econ_all > .75) &
is.na(remir_all$inter_examine12_ses_cat) &
is.na(remir_all$subg_examine12_ses),
"Homog", remir_all$inter_examine12_ses_cat)
tabyl(remir_all$inter_examine12_ses_cat)## INTERACTION SES BENEFITTED
tabyl(remir_all$inter_benefit12_ses)## Simplify Codes
tabyl(remir_all, inter_benefit12_ses, inter_examine12_ses)remir_all$inter_benefit12_ses_cat <-
ifelse(remir_all$inter_benefit12_ses == 'None' &
remir_all$main_effect == 'Yes', 'Both',
ifelse(remir_all$inter_benefit12_ses == 'None' &
remir_all$main_effect == 'No', 'Neither',
remir_all$inter_benefit12_ses))
tabyl(remir_all, inter_benefit12_ses, inter_benefit12_ses_cat)## SUBGROUP SES BENEFITTED
tabyl(remir_all$subg_signif12_ses)## SEPARATE OUT HOMOGENEOUS SAMPLE
remir_all$inter_benefit12_ses_cat <- ifelse(
(remir_all$econ_all > .75) &
is.na(remir_all$inter_benefit12_ses_cat) &
is.na(remir_all$subg_signif12_ses),
"Homog", remir_all$inter_benefit12_ses_cat)
tabyl(remir_all$inter_benefit12_ses_cat)remir_all$examine12_ses_short <-
ifelse(is.na(remir_all$inter_examine12_ses_cat) &
is.na(remir_all$subg_examine12_ses), 'None',
ifelse(remir_all$inter_examine12_ses_cat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_all$examine12_ses_short)## Fix remaining NAs -- Should be yes since they are missing for one test only
remir_all$examine12_ses_short <- ifelse(
is.na(remir_all$examine12_ses_short), 'Yes', remir_all$examine12_ses_short)
tabyl(remir_all$examine12_ses_short)############## LOCATION ###################
## Change Interaction Examined Variable Names
names(remir_all)[36] <- "inter_examine1_loc"
names(remir_all)[116] <- "inter_examine2_loc"
table(remir_all$inter_examine1_loc, useNA = "always")##
## <NA>
## 292table(remir_all$inter_examine2_loc, useNA = "always")##
## Urban/Non-Urban <NA>
## 2 290## Combine Two Tnteractions Examined
remir_all$inter_examine12_loc <- ifelse(is.na(remir_all$inter_examine1_loc),
remir_all$inter_examine2_loc, remir_all$inter_examine1_loc)
tabyl(remir_all$inter_examine12_loc)############ What Location Interactions Were Significant ################
## Change Interaction Significant Variable Names
names(remir_all)[37] <- "inter_signif1_loc"
names(remir_all)[117] <- "inter_signif2_loc"
table(remir_all$inter_signif1_loc, useNA = "always")##
## <NA>
## 292table(remir_all$inter_signif2_loc, useNA = "always")##
## None <NA>
## 2 290## Combine Two Tnteractions Significant
remir_all$inter_signif12_loc <- ifelse(is.na(remir_all$inter_signif1_loc),
remir_all$inter_signif2_loc, remir_all$inter_signif1_loc)
tabyl(remir_all$inter_signif12_loc)############ What Location Interaction Groups Benefited More #############
## Change Interaction Benefitted Variable Names
names(remir_all)[39] <- "inter_benefit1_loc"
names(remir_all)[119] <- "inter_benefit2_loc"
table(remir_all$inter_benefit1_loc, useNA = "always")##
## <NA>
## 292table(remir_all$inter_benefit2_loc, useNA = "always")##
## None <NA>
## 2 290## Combine Two Tnteractions
remir_all$inter_benefit12_loc <- ifelse(is.na(remir_all$inter_benefit1_loc),
remir_all$inter_benefit2_loc, remir_all$inter_benefit1_loc)
tabyl(remir_all$inter_benefit12_loc)############# What Location Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[67] <- "subg_examine1_loc"
names(remir_all)[120] <- "subg_examine2_loc"
table(remir_all$subg_examine1_loc, useNA = "always")##
## Rural <NA>
## 2 290table(remir_all$subg_examine2_loc, useNA = "always")##
## Urban;Non-Urban <NA>
## 2 290## Combine Two Subgroups Examined
remir_all$subg_examine12_loc <- ifelse(is.na(remir_all$subg_examine1_loc),
remir_all$subg_examine2_loc, remir_all$subg_examine1_loc)
tabyl(remir_all$subg_examine12_loc)############## What Location Subgroups Were Significant ####################
## Change Subgroup Significant Variable Names
names(remir_all)[69] <- "subg_signif1_loc"
names(remir_all)[122] <- "subg_signif2_loc"
table(remir_all$subg_signif1_loc, useNA = "always")##
## Rural <NA>
## 2 290table(remir_all$subg_signif2_loc, useNA = "always")##
## None Urban;Non-Urban <NA>
## 1 1 290## Combine Two Tnteractions Significant
remir_all$subg_signif12_loc <- ifelse(is.na(remir_all$subg_signif1_loc),
remir_all$subg_signif2_loc, remir_all$subg_signif1_loc)
tabyl(remir_all$subg_signif12_loc)## INTERACTION LOCATION EXAMINED
tabyl(remir_all$inter_examine12_loc)## SUBGROUP LOCATION EXAMINED
tabyl(remir_all$subg_examine12_loc)## SEPARATE OUT HOMOGENEOUS SAMPLE
## No measure of sample proportion for urban or rural
## INTERACTION LOCATION BENEFITTED
tabyl(remir_all$inter_benefit12_loc)## SUBGROUP LOCATION BENEFITTED
tabyl(remir_all$subg_signif12_loc)############## NATIVITY ################################
## Change Interaction Nativity Variable Names
names(remir_all)[41] <- "inter_examine1_nat"
names(remir_all)[124] <- "inter_examine2_nat"
table(remir_all$inter_examine1_nat, useNA = "always")##
## Immigrant/Nonimmigrant <NA>
## 1 291table(remir_all$inter_examine2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Examined
remir_all$inter_examine12_nat <- ifelse(is.na(remir_all$inter_examine1_nat),
remir_all$inter_examine2_nat, remir_all$inter_examine1_nat)
tabyl(remir_all$inter_examine12_nat)############## What Nativity Interactions Were Significant ##############################
## Change Interaction Significant Variable Names
names(remir_all)[42] <- "inter_signif1_nat"
names(remir_all)[125] <- "inter_signif2_nat"
table(remir_all$inter_signif1_nat, useNA = "always")##
## None <NA>
## 1 291table(remir_all$inter_signif2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Significant
remir_all$inter_signif12_nat <- ifelse(is.na(remir_all$inter_signif1_nat),
remir_all$inter_signif2_nat, remir_all$inter_signif1_nat)
tabyl(remir_all$inter_signif12_nat)############## What Nativity Interaction Groups Benefited More ##############################
## Change Interaction Benfitted Variable Names
names(remir_all)[44] <- "inter_benefit1_nat"
names(remir_all)[127] <- "inter_benefit2_nat"
table(remir_all$inter_benefit1_nat, useNA = "always")##
## None <NA>
## 1 291table(remir_all$inter_benefit2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions
remir_all$inter_benefit12_nat <- ifelse(is.na(remir_all$inter_benefit1_nat),
remir_all$inter_benefit2_nat, remir_all$inter_benefit1_nat)
tabyl(remir_all$inter_benefit12_nat)############# What Nativity Subgroups Were Examined ####################
## Change Subgroup Examined Variable Names
names(remir_all)[71] <- "subg_examine1_nat"
names(remir_all)[128] <- "subg_examine2_nat"
table(remir_all$subg_examine1_nat, useNA = "always")##
## <NA>
## 292table(remir_all$subg_examine2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Subgroups Examined
remir_all$subg_examine12_nat <- ifelse(is.na(remir_all$subg_examine1_nat),
remir_all$subg_examine2_nat, remir_all$subg_examine1_nat)
tabyl(remir_all$subg_examine12_nat)############## What Nativity Subgroups Were Significant ##############################
## Change Subgroup Significant Variable Names
names(remir_all)[73] <- "subg_signif1_nat"
names(remir_all)[130] <- "subg_signif2_nat"
table(remir_all$subg_signif1_nat, useNA = "always")##
## <NA>
## 292table(remir_all$subg_signif2_nat, useNA = "always")##
## <NA>
## 292## Combine Two Tnteractions Significant
remir_all$subg_signif12_nat <- ifelse(is.na(remir_all$subg_signif1_nat),
remir_all$subg_signif2_nat, remir_all$subg_signif1_nat)
tabyl(remir_all$subg_signif12_nat)## INTERACTION NATIVITY EXAMINED
tabyl(remir_all$inter_examine12_nat)## SUBGROUP NATIVITY EXAMINED
tabyl(remir_all$subg_examine12_nat)## INTERACTION NATIVITY BENEFITTED
tabyl(remir_all$inter_benefit12_nat)## SUBGROUP NATIVITY BENEFITTED
tabyl(remir_all$subg_signif12_nat)# Nativity
tabyl(remir_all, inter_examine12_nat, subg_examine12_nat) %>%
adorn_percentages("all") %>% adorn_pct_formatting(rounding = "half up", digits = 1)remir_all$examine12_nat_short <-
ifelse(is.na(remir_all$inter_examine12_nat) &
is.na(remir_all$subg_examine12_nat), 'None',
ifelse(remir_all$inter_examine12_nat == 'Homog', 'Homog', 'Yes'))
tabyl(remir_all$examine12_nat_short)# Ns by time period for Full (N = 292), Analysis (N = 240) and Sub-Analysis samples (N = 100)
## Full Sample (N = 292)
tabyl(remir_all$Citation.Year)remir_all$dyear <- ifelse(remir_all$Citation.Year <= 2016,0,1)
tabyl(remir_all, Citation.Year, dyear)tabyl(remir_all, dyear)# 2010-16 (n = 179)
# 2017-2023 (n = 113)
## Analysis Sample (N = 240)
tabyl(remir_us$Citation.Year)remir_us$dyear <- ifelse(remir_us$Citation.Year <= 2016,0,1)
tabyl(remir_us, Citation.Year, dyear)tabyl(remir_us, dyear)# # 2010-16 (n = 147)
# 2017-2023 (n = 93)
## Sub-Analysis Sample (N = 100)
tabyl(remir_us_subgroup$Citation.Year)remir_us_subgroup$dyear <- ifelse(remir_us_subgroup$Citation.Year <= 2016,0,1)
tabyl(remir_us_subgroup, Citation.Year, dyear)tabyl(remir_us_subgroup, dyear)# # 2010-16 (n = 57)
# 2017-2023 (n = 43)################ RACE ##################
## Test for Relationship Between Examined and Year
tabyl(remir_all,examine12_race_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_all$examine12_race_short, remir_all$dyear, correct=FALSE)##
## Pearson's Chi-squared test
##
## data: remir_all$examine12_race_short and remir_all$dyear
## X-squared = 7.425, df = 2, p-value = 0.02442################ ETHNICTY ##################
## Test for Relationship Between Examined and Year
tabyl(remir_all,examine12_ethnic_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_all$examine12_ethnic_short, remir_all$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_all$examine12_ethnic_short and remir_all$dyear
## X-squared = 16.151, df = 2, p-value = 0.0003111################ GENDER ##################
## Test for Relationship Between Examined and Year
tabyl(remir_all,examine12_gender_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_all$examine12_gender_short, remir_all$dyear, correct=FALSE)##
## Pearson's Chi-squared test
##
## data: remir_all$examine12_gender_short and remir_all$dyear
## X-squared = 2.9677, df = 2, p-value = 0.2268################ SEX ##################
## Test for Relationship Between Examined and Year
tabyl(remir_all,examine12_sex_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_all$examine12_sex_short, remir_all$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_all$examine12_sex_short and remir_all$dyear
## X-squared = 3.19, df = 1, p-value = 0.07409################ SES ##################
## Test for Relationship Between Examined and Year
tabyl(remir_all,examine12_ses_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_all$examine12_ses_short, remir_all$dyear, correct=FALSE)##
## Pearson's Chi-squared test
##
## data: remir_all$examine12_ses_short and remir_all$dyear
## X-squared = 0.19515, df = 2, p-value = 0.907################ Location ##################
## Test for Relationship Between Examined and Year
tabyl(remir_all,examine12_loc_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_all$examine12_loc_short, remir_all$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_all$examine12_loc_short and remir_all$dyear
## X-squared = 2.5602, df = 1, p-value = 0.1096################ Nativity ##################
## Test for Relationship Between Examined and Year
tabyl(remir_all, examine12_nat_short, dyear) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_all$examine12_nat_short, remir_all$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_all$examine12_nat_short and remir_all$dyear
## X-squared = 0.63345, df = 1, p-value = 0.4261################ RACE ##################
## Test for Relationship Between Examined and Year
tabyl(remir_us,examine12_race_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_race_short, remir_us$dyear, correct=FALSE)##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_race_short and remir_us$dyear
## X-squared = 7.4228, df = 2, p-value = 0.02444################ ETHNICTY ##################
## Test for Relationship Between Examined and Year
tabyl(remir_us,examine12_ethnic_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_ethnic_short, remir_us$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_ethnic_short and remir_us$dyear
## X-squared = 16.564, df = 2, p-value = 0.000253################ GENDER ##################
## Test for Relationship Between Examined and Year
tabyl(remir_us,examine12_gender_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_gender_short, remir_us$dyear, correct=FALSE)##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_gender_short and remir_us$dyear
## X-squared = 1.5242, df = 2, p-value = 0.4667################ SEX ##################
## Test for Relationship Between Examined and Year
tabyl(remir_us,examine12_sex_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_sex_short, remir_us$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_sex_short and remir_us$dyear
## X-squared = 3.1879, df = 1, p-value = 0.07419################ SES ##################
## Test for Relationship Between Examined and Year
tabyl(remir_us,examine12_ses_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_ses_short, remir_us$dyear, correct=FALSE)##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_ses_short and remir_us$dyear
## X-squared = 0.12308, df = 1, p-value = 0.7257################ Location ##################
## Test for Relationship Between Examined and Year
tabyl(remir_us,examine12_loc_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_loc_short, remir_us$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_loc_short and remir_us$dyear
## X-squared = 2.5735, df = 1, p-value = 0.1087################ Nativity ##################
## Test for Relationship Between Examined and Year
tabyl(remir_us, examine12_nat_short, dyear) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us$examine12_nat_short, remir_us$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us$examine12_nat_short and remir_us$dyear
## X-squared = 0.6353, df = 1, p-value = 0.4254## RACE
## Test for Relationship Between Examined and Year
tabyl(remir_us_subgroup,examine12_race_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us_subgroup$examine12_race_short, remir_us_subgroup$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us_subgroup$examine12_race_short and remir_us_subgroup$dyear
## X-squared = 6.0621, df = 2, p-value = 0.04826## ETHNICTY
## Test for Relationship Between Examined and Year
tabyl(remir_us_subgroup,examine12_ethnic_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us_subgroup$examine12_ethnic_short, remir_us_subgroup$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us_subgroup$examine12_ethnic_short and remir_us_subgroup$dyear
## X-squared = 4.2895, df = 2, p-value = 0.1171## GENDER
## Test for Relationship Between Examined and Year
tabyl(remir_us_subgroup,examine12_gender_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us_subgroup$examine12_gender_short, remir_us_subgroup$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us_subgroup$examine12_gender_short and remir_us_subgroup$dyear
## X-squared = 4.7829, df = 2, p-value = 0.0915## SEX
## Test for Relationship Between Examined and Year
tabyl(remir_us_subgroup,examine12_sex_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us_subgroup$examine12_sex_short, remir_us_subgroup$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us_subgroup$examine12_sex_short and remir_us_subgroup$dyear
## X-squared = 2.7053, df = 1, p-value = 0.1## SES
## Test for Relationship Between Examined and Year
tabyl(remir_us_subgroup,examine12_ses_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us_subgroup$examine12_ses_short, remir_us_subgroup$dyear, correct=FALSE)##
## Pearson's Chi-squared test
##
## data: remir_us_subgroup$examine12_ses_short and remir_us_subgroup$dyear
## X-squared = 0.84222, df = 1, p-value = 0.3588## Location
## Test for Relationship Between Examined and Year
tabyl(remir_us_subgroup,examine12_loc_short, dyear ) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us_subgroup$examine12_loc_short, remir_us_subgroup$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us_subgroup$examine12_loc_short and remir_us_subgroup$dyear
## X-squared = 3.1433, df = 1, p-value = 0.07624## Nativity
## Test for Relationship Between Examined and Year
tabyl(remir_us_subgroup, examine12_nat_short, dyear) %>%
adorn_percentages("col") %>% adorn_pct_formatting(rounding = "half up", digits = 1)chisq.test(remir_us_subgroup$examine12_nat_short, remir_us_subgroup$dyear, correct=FALSE)## Warning in stats::chisq.test(x, y, ...): Chi-squared approximation may be
## incorrect##
## Pearson's Chi-squared test
##
## data: remir_us_subgroup$examine12_nat_short and remir_us_subgroup$dyear
## X-squared = 0.76201, df = 1, p-value = 0.3827# About Online Supplemental Table 5: Proportions Registering Prospectively and Retrospectively with and without Subgroup Registration for Sub-Analysis Sample
## a, c Tested for one or more of the following subgroups: race, ethnicity, gender, sexual identity, economic disadvantage, location (rural, urban, suburban), nativity status (foreign-born – yes/no).
## b Registered prospectively means that study participants were enrolled and the study was documented in a public registry before the outcomes relevant to the study’s objectives were known or analyzed. This practice enhances the reliability and credibility of research findings by ensuring transparency and reducing the potential for bias.
## d Registered retrospectively means that the study was entered into a public registry after data collection had started or after the study was completed. While retrospective registration provides transparency post-facto, it is generally considered less desirable than prospective registration because it can introduce bias. For example, researchers might selectively report outcomes or change their analysis plan based on the data collected, potentially compromising the study's scientific integrity. ## Use method variable for studies examining subgroup effects
## Subgroup sample (N = 100)
tabyl(remir_us_subgroup, reg, prereg)tabyl(remir_us_subgroup, prereg, prereg_subg)## Recode unclear to No for subg
tabyl(remir_us_subgroup, prereg_subg)remir_us_subgroup$prereg_subg <- ifelse(remir_us_subgroup$prereg_subg == 'Unclear', 'No', remir_us_subgroup$prereg_subg)
tabyl(remir_us_subgroup, prereg_subg)## Add new categories to preregister
remir_us_subgroup$prereg_all <- ifelse(remir_us_subgroup$reg == 'Yes' &
remir_us_subgroup$prereg == 'Yes' & remir_us_subgroup$prereg_subg == 'Yes', 1,
ifelse(remir_us_subgroup$reg == 'Yes' &
remir_us_subgroup$prereg == 'Yes' & remir_us_subgroup$prereg_subg == 'No', 2,
ifelse(remir_us_subgroup$reg == 'Yes' &
remir_us_subgroup$prereg == 'No' & remir_us_subgroup$prereg_subg == 'Yes', 3,
ifelse(remir_us_subgroup$reg == 'Yes' &
remir_us_subgroup$prereg == 'No' & remir_us_subgroup$prereg_subg == 'No', 4,
ifelse(remir_us_subgroup$reg == 'No', 5, NA)))))
tabyl(remir_us_subgroup$prereg_all)# Assigning labels to the 'prereg_all' variable
remir_us_subgroup$prereg_all <- factor(remir_us_subgroup$prereg_all, levels = c(1, 2, 3, 4, 5),
labels = c("Registered Prospectively with Subgroup",
"Registered Prospectively without Subgroup",
"Registered Retrospectively with Subgroup",
"Registered Retrospectively without Subgroup",
"Not registered"))
# Generate the table
tab_prereg <- remir_us_subgroup %>%
filter(!is.na(prereg_all)) %>%
group_by(prereg_all) %>%
summarise(Frequency = n()) %>%
mutate(Proportion = round(Frequency / sum(Frequency), 3))
tab_prereg <- bind_rows(tab_prereg, tibble(prereg_all = "Total", Frequency = sum(tab_prereg$Frequency), Proportion = round(sum(tab_prereg$Proportion), 3)))
print(tab_prereg)## # A tibble: 6 × 3
## prereg_all Frequency Proportion
## <chr> <int> <dbl>
## 1 Registered Prospectively with Subgroup 1 0.01
## 2 Registered Prospectively without Subgroup 1 0.01
## 3 Registered Retrospectively with Subgroup 6 0.06
## 4 Registered Retrospectively without Subgroup 15 0.15
## 5 Not registered 77 0.77
## 6 Total 100 1################### FOR PRISMA DIAGRAM ONLINE FIGURE 1 ################
# Full sample (N = 240)
## Number of Programs
length(unique(remir_all$Program.ID))## [1] 111## Number of Studies
length(unique(remir_all$Study.ID))## [1] 172## Number of Citations
length(unique(remir_all$Citation.ID))## [1] 292# Not Eligible
1598-111## [1] 14872955-172## [1] 27834176-292## [1] 3884# US Studies
## Number of Programs
length(unique(remir_us$Program.ID))## [1] 97## Number of Studies
length(unique(remir_us$Study.ID))## [1] 140## Number of Citations
length(unique(remir_us$Citation.ID))## [1] 240## Non-US Studies
111-97## [1] 14172-140## [1] 32292-240## [1] 52# US Studies with Subgroup Effects
## Number of Programs
length(unique(remir_us_subgroup$Program.ID))## [1] 49## Number of Studies
length(unique(remir_us_subgroup$Study.ID))## [1] 63## Number of Citations
length(unique(remir_us_subgroup$Citation.ID))## [1] 10097-52## [1] 45140-49## [1] 91240-63## [1] 177############################### END ##################################