library(knitr)
library(tidyverse)
library(kableExtra) #table report
library(janitor) #totals
library(apaTables) #tables ANOVA

Please see below the ‘updated’ analyses for BabyNet Epals project. I would like you to review the code, if possible, check the dataset, and give me feedback. This analysis was checked several times during the last two years, but now I could not give the appropriate focus to interpret all results. I also inform that the dataset being used here produces virtually equal results than the ones published by Ed’s. Small differences come because I’ve got the SPSS file and what SPSS call as “pooled” is almost the same thing than getting the average of the multiple imputation datasets, but not entirely the same.

1 MI20 (April) dataset

1.1 Participants’ race

data_april %>% filter(M_5R1003 == "1") %>% count(M_5R104A)

## # A tibble: 2 x 2
##   M_5R104A     n
##      <dbl> <int>
## 1        1     1
## 2      NaN    60

data_april %>% filter(M_5R1003 == "1") %>% count(M_5R104B)

## # A tibble: 1 x 2
##   M_5R104B     n
##      <dbl> <int>
## 1      NaN    61

data_april %>% filter(M_5R1003 == "1") %>% count(M_5R104C)

## # A tibble: 1 x 2
##   M_5R104C     n
##      <dbl> <int>
## 1      NaN    61

data_april %>% filter(M_5R1003 == "1") %>% count(M_5R104D)

## # A tibble: 2 x 2
##   M_5R104D     n
##      <dbl> <int>
## 1        1    30
## 2      NaN    31

data_april %>% filter(M_5R1003 == "1") %>% count(M_5R104E)

## # A tibble: 1 x 2
##   M_5R104E     n
##      <dbl> <int>
## 1      NaN    61

data_april %>% filter(M_5R1003 == "1") %>% count(M_5R104F)

## # A tibble: 2 x 2
##   M_5R104F     n
##      <dbl> <int>
## 1        1     5
## 2      NaN    56

data_april %>% filter(M_5R1003 == "0") %>% count(M_5R104A)

## # A tibble: 2 x 2
##   M_5R104A     n
##      <dbl> <int>
## 1        1     5
## 2      NaN    92

data_april %>% filter(M_5R1003 == "0") %>% count(M_5R104B)

## # A tibble: 2 x 2
##   M_5R104B     n
##      <dbl> <int>
## 1        1     4
## 2      NaN    93

data_april %>% filter(M_5R1003 == "0") %>% count(M_5R104C)

## # A tibble: 2 x 2
##   M_5R104C     n
##      <dbl> <int>
## 1        1    26
## 2      NaN    71

data_april %>% filter(M_5R1003 == "0") %>% count(M_5R104D)

## # A tibble: 2 x 2
##   M_5R104D     n
##      <dbl> <int>
## 1        1    72
## 2      NaN    25

data_april %>% filter(M_5R1003 == "0") %>% count(M_5R104E)

## # A tibble: 2 x 2
##   M_5R104E     n
##      <dbl> <int>
## 1        1     1
## 2      NaN    96

data_april %>% filter(M_5R1003 == "0") %>% count(M_5R104F)

## # A tibble: 1 x 2
##   M_5R104F     n
##      <dbl> <int>
## 1      NaN    97

data_april %>% 
  filter(M_5R1003 == "0" & M_5R104D == "1" & M_5R104C == "1")

## # A tibble: 5 x 138
##      ID Imputation_ CONDITION.x OM4G1315 OM4G1316 OM4G1317 OM4G1318
##   <dbl>       <dbl>       <dbl>    <dbl>    <dbl>    <dbl>    <dbl>
## 1 10043        10.5           1        2        3        5        1
## 2 10065        10.5           0        4        4        4        3
## 3 10081        10.5           0        3        2        5        2
## 4 20236        10.5           1        2        1        1        1
## 5 20285        10.5           0        4        2        5        3
## # … with 131 more variables: OK4G1319 <dbl>, OK4G1320 <dbl>,
## #   OK4G1321 <dbl>, OK4G1322 <dbl>, OK4G1323 <dbl>, OM4G2315 <dbl>,
## #   OM4G2316 <dbl>, OM4G2317 <dbl>, OM4G2318 <dbl>, OK4G2319 <dbl>,
## #   OK4G2320 <dbl>, OK4G2321 <dbl>, OK4G2322 <dbl>, OK4G2323 <dbl>,
## #   M_5O1ABS <dbl>, M_5O2ABS <dbl>, M_4D1TO <dbl>, M_4D2TO <dbl>,
## #   M_4I1TOT <dbl>, M_4I2TOT <dbl>, LAST_LEVEL <dbl>, LAST_SECTION <dbl>,
## #   LAST_PAGE <dbl>, SECT_PERCENT <dbl>, HLS <dbl>, HL <dbl>, hlsr <dbl>,
## #   zmenth1 <dbl>, zmflex1 <dbl>, zmbr1 <dbl>, zmenth2 <dbl>,
## #   zmflex2 <dbl>, zmbr2 <dbl>, mlang1 <dbl>, mlang2 <dbl>,
## #   zbbkeng1 <dbl>, zbrecl1 <dbl>, zbexpl1 <dbl>, zbbkeng2 <dbl>,
## #   zbrecl2 <dbl>, zbexpl2 <dbl>, blang1 <dbl>, blang2 <dbl>, zhls <dbl>,
## #   zcond <dbl>, doscond <dbl>, zmlang1 <dbl>, zmlang2 <dbl>,
## #   zblang1 <dbl>, zblang2 <dbl>, zmknow1 <dbl>, zmknow2 <dbl>,
## #   zdoscond <dbl>, mlangc <dbl>, blangc <dbl>, mombeh <dbl>,
## #   kidbeh <dbl>, momknow <dbl>, momchg <dbl>, babychg <dbl>, dcond <dbl>,
## #   dcond_d <dbl>, CONDITION.y <dbl>, ATTRITION <dbl>, M_5R1002 <dbl>,
## #   M_5R1003 <dbl>, M_5R104A <dbl>, M_5R104B <dbl>, M_5R104C <dbl>,
## #   M_5R104D <dbl>, M_5R104E <dbl>, M_5R104F <dbl>, M_5R1005 <dbl>,
## #   M_5R1006 <dbl>, M_5R1007 <dbl>, M_5R1008 <dbl>, M_5R1009 <dbl>,
## #   M_5R1010 <dbl>, M_5R1011 <dbl>, M_5R1012 <dbl>, M_5R1013 <dbl>,
## #   M_5R1014 <dbl>, M_5R1015 <dbl>, M_5R1016 <dbl>, M_5R1017 <dbl>,
## #   M_5R1018 <dbl>, M_5R1019 <dbl>, M_5R120A <dbl>, M_5R120B <dbl>,
## #   M_5R120C <dbl>, M_5R120D <dbl>, M_5R120E <dbl>, M_5R120F <dbl>,
## #   M_5R120G <dbl>, M_5R1022 <dbl>, M_5R1023 <dbl>, M_5R124A <dbl>,
## #   M_5R124B <dbl>, M_5R124C <dbl>, M_5R124D <dbl>, …

1.2 Infant behavioral

data_april %>% 
  group_by(CONDITION.x) %>%             
  summarise(mean(blang1), 
            sd(blang1), 
            mean(blang2), 
            sd(blang2),
            n()) %>% 
  mutate(CONDITION.x = if_else(CONDITION.x == "1","Experimental", "Attention")) %>% 
  adorn_totals() %>% 
  kable(digits = 2, caption = "Data name = MI20") %>%
  kable_styling(bootstrap_options=c('striped','hover', 'condensed','responsive'),position='center',
                font_size = 12,full_width=F)#check table page 24

Data name = MI20
CONDITION.x	mean(blang1)	sd(blang1)	mean(blang2)	sd(blang2)	n()
Attention	0.44	2.3	-0.16	2.7	76
Experimental	-0.40	2.1	0.15	2.0	83
Total	0.04	4.4	-0.01	4.7	159

include_graphics("ed table 2 - blang.PNG")

1.3 Maternal behavioral

data_april %>% 
  group_by(CONDITION.x) %>%             
  summarise(mean(mlang1), 
            sd(mlang1), 
            mean(mlang2), 
            sd(mlang2),
            n()) %>% 
  mutate(CONDITION.x = if_else(CONDITION.x == "1","Experimental", "Attention")) %>% 
  adorn_totals() %>% 
  kable(digits = 2, caption = "Data name = MI20") %>%
  kable_styling(bootstrap_options=c('striped','hover', 'condensed','responsive'),position='center',
                font_size = 12,full_width=F)#check table page 24

Data name = MI20
CONDITION.x	mean(mlang1)	sd(mlang1)	mean(mlang2)	sd(mlang2)	n()
Attention	0.47	2.6	-0.44	2.5	76
Experimental	-0.43	2.4	0.40	2.4	83
Total	0.04	5.0	-0.04	4.9	159

From these results, I am assuming this dataset is virtually equal as Ed’s dataset (I think now it was published).

include_graphics("ed table 2 - mlang.PNG")

Almost ok. SD slightly different.

1.4 Maternal knowledge

data_april %>% 
  group_by(CONDITION.x) %>%             
  summarise(mean(M_4I1TOT), 
            sd(M_4I1TOT), 
            mean(M_4I2TOT), 
            sd(M_4I2TOT),
            n()) %>% 
  mutate(CONDITION.x = if_else(CONDITION.x == "1","Experimental", "Attention")) %>% 
  adorn_totals() %>% 
  kable(digits = 2, caption = "Data name = MI20") %>%
  kable_styling(bootstrap_options=c('striped','hover', 'condensed','responsive'),position='center',
                font_size = 12,full_width=F)#check table page 24

Data name = MI20
CONDITION.x	mean(M_4I1TOT)	sd(M_4I1TOT)	mean(M_4I2TOT)	sd(M_4I2TOT)	n()
Attention	6.0	2.9	6.4	2.8	76
Experimental	6.3	2.7	9.6	3.0	83
Total	12.2	5.6	16.0	5.8	159

include_graphics("ed table 3 - epals total.PNG")

2 Demographic information

data_april %>% 
  count(M_5R1003)

## # A tibble: 3 x 2
##   M_5R1003     n
##      <dbl> <int>
## 1        0    97
## 2        1    61
## 3      NaN     1

2.1 Mother and her children age

Table 1 here

desc_ttest(dependent = "M1AGE")

## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by race
## t = 2, df = 35, p-value = 0.1
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -1.4 14.9
## sample estimates:
## mean in group Latinx  mean in group White 
##                   35                   29 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by race
## t = 0.8, df = 47, p-value = 0.4
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -3.7  8.9
## sample estimates:
## mean in group Latinx  mean in group White 
##                   31                   29 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = 0.9, df = 54, p-value = 0.4
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -5.3 13.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              35                              31 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = -0.003, df = 87, p-value = 1
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -3.7  3.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              29                              29

race	group	mean	sd	n
Latinx	Attention-control	35	22.0	32
Latinx	Experimental	31	14.6	29
White	Attention-control	29	6.5	43
White	Experimental	29	11.4	54

desc_ttest(dependent = "C1AGE")

## 
##  Welch Two Sample t-test
## 
## data:  C1AGE by race
## t = -2, df = 73, p-value = 0.1
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -1.64  0.22
## sample estimates:
## mean in group Latinx  mean in group White 
##                  3.7                  4.4 
## 
## 
##  Welch Two Sample t-test
## 
## data:  C1AGE by race
## t = -1, df = 54, p-value = 0.2
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -1.63  0.26
## sample estimates:
## mean in group Latinx  mean in group White 
##                  4.2                  4.9 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = 0.9, df = 54, p-value = 0.4
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -5.3 13.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              35                              31 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = -0.003, df = 87, p-value = 1
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -3.7  3.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              29                              29

race	group	mean	sd	n
Latinx	Attention-control	3.7	1.7	32
Latinx	Experimental	4.2	2.1	29
White	Attention-control	4.4	2.4	43
White	Experimental	4.9	1.9	54

2.2 Biological sex

The percentage of boys and girls in each group was also computed.

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  rename(sex = M_5R1022) %>% 
  mutate(sex = if_else(sex == 1, "Male","Female")) %>% 
  filter(!is.na(race) & !is.na(sex)) %>% 
  tabyl(sex, race, group) %>% #apresentar tabela
  adorn_totals(c("row", "col")) %>% 
  adorn_percentages("col") %>% 
  adorn_pct_formatting(rounding = "half up", digits = 0) %>% 
  adorn_ns() %>% 
  kable() %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

sex	Latinx	White	Total
Female	53% (17)	56% (24)	55% (41)
Male	47% (15)	44% (19)	45% (34)
Total	100% (32)	100% (43)	100% (75)

sex	Latinx	White	Total
Female	62% (18)	53% (28)	56% (46)
Male	38% (11)	47% (25)	44% (36)
Total	100% (29)	100% (53)	100% (82)

The association between race and sex was computed for Attention-control group.

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  rename(sex = M_5R1022) %>% 
  mutate(sex = if_else(sex == 1, "Male","Female")) %>% 
  filter(!is.na(race) & !is.na(sex)) %>% 
  filter(group == "Attention-control") %>%
  {descr::crosstab(.$sex,.$race, chisq = TRUE)}

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |-------------------------|
## 
## ================================
##           .$race
## .$sex     Latinx   White   Total
## --------------------------------
## Female        17      24      41
## --------------------------------
## Male          15      19      34
## --------------------------------
## Total         32      43      75
## ================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 0.054      d.f. = 1      p = 0.817 
## 
## Pearson's Chi-squared test with Yates' continuity correction 
## ------------------------------------------------------------
## Chi^2 = 2.2e-31      d.f. = 1      p = 1 
##         Minimum expected frequency: 15

The association between race and sex was computed for Experimental group.

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  rename(sex = M_5R1022) %>% 
  mutate(sex = if_else(sex == 1, "Male","Female")) %>% 
  filter(!is.na(race) & !is.na(sex)) %>% 
  filter(group == "Experimental") %>%
  {descr::crosstab(.$sex,.$race, chisq = TRUE)}

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |-------------------------|
## 
## ================================
##           .$race
## .$sex     Latinx   White   Total
## --------------------------------
## Female        18      28      46
## --------------------------------
## Male          11      25      36
## --------------------------------
## Total         29      53      82
## ================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 0.65      d.f. = 1      p = 0.42 
## 
## Pearson's Chi-squared test with Yates' continuity correction 
## ------------------------------------------------------------
## Chi^2 = 0.33      d.f. = 1      p = 0.566 
##         Minimum expected frequency: 13

2.3 Primary language

The primary language spoken at home was compared between whites and latinxs in both conditions.

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 

  filter(!is.na(race) & !is.na(M_5R1005)) %>% 
  tabyl(M_5R1005, race, group) %>% #apresentar tabela
  adorn_totals(c("row", "col")) %>% 
  adorn_percentages("col") %>% 
  adorn_pct_formatting(rounding = "half up", digits = 0) %>% 
  adorn_ns() %>% 
  kable() %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

M_5R1005	Latinx	White	Total
1	33% (10)	100% (43)	73% (53)
2	67% (20)	0% (0)	27% (20)
3	0% (0)	0% (0)	0% (0)
Total	100% (30)	100% (43)	100% (73)

M_5R1005	Latinx	White	Total
1	32% (9)	98% (51)	75% (60)
2	64% (18)	0% (0)	23% (18)
3	4% (1)	2% (1)	3% (2)
Total	100% (28)	100% (52)	100% (80)

In Attention-controle condition, these percentages were compared.

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 

  filter(!is.na(race) & !is.na(M_5R1005)) %>% 
  filter(group == "Attention-control") %>%
  {descr::crosstab(.$M_5R1005,.$race, prop.c = TRUE,chisq = TRUE)}

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |          Column Percent | 
## |-------------------------|
## 
## ====================================
##               .$race
## .$M_5R1005    Latinx   White   Total
## ------------------------------------
## 1                10      43      53 
##                  33%    100%        
## ------------------------------------
## 2                20       0      20 
##                  67%      0%        
## ------------------------------------
## Total            30      43      73 
##                  41%     59%        
## ====================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 39      d.f. = 1      p = 3.31e-10 
## 
## Pearson's Chi-squared test with Yates' continuity correction 
## ------------------------------------------------------------
## Chi^2 = 36      d.f. = 1      p = 1.78e-09 
##         Minimum expected frequency: 8.2

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1005)) %>% 
  filter(group == "Experimental") %>%
  {descr::crosstab(.$M_5R1005,.$race, prop.c = TRUE,chisq = TRUE)}

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |          Column Percent | 
## |-------------------------|
## 
## ====================================
##               .$race
## .$M_5R1005    Latinx   White   Total
## ------------------------------------
## 1                 9      51      60 
##                32.1%   98.1%        
## ------------------------------------
## 2                18       0      18 
##                64.3%    0.0%        
## ------------------------------------
## 3                 1       1       2 
##                 3.6%    1.9%        
## ------------------------------------
## Total            28      52      80 
##                  35%     65%        
## ====================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 44      d.f. = 2      p = 2.55e-10 
## 
##         Minimum expected frequency: 0.7 
## Cells with Expected Frequency < 5: 2 of 6 (33%)

2.4 Maternal depression

desc_ttest(dependent = "M_4D1TO")

## 
##  Welch Two Sample t-test
## 
## data:  M_4D1TO by race
## t = -0.6, df = 72, p-value = 0.6
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -12.9   7.2
## sample estimates:
## mean in group Latinx  mean in group White 
##                   59                   62 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M_4D1TO by race
## t = -0.2, df = 40, p-value = 0.8
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -16  13
## sample estimates:
## mean in group Latinx  mean in group White 
##                   60                   61 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = 0.9, df = 54, p-value = 0.4
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -5.3 13.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              35                              31 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = -0.003, df = 87, p-value = 1
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -3.7  3.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              29                              29

race	group	mean	sd	n
Latinx	Attention-control	59	20	32
Latinx	Experimental	60	34	29
White	Attention-control	62	23	43
White	Experimental	61	21	54

2.5 Computer comfort

desc_ttest(dependent = "M_5R1027")

## 
##  Welch Two Sample t-test
## 
## data:  M_5R1027 by race
## t = -2, df = 64, p-value = 0.04
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.745 -0.028
## sample estimates:
## mean in group Latinx  mean in group White 
##                  3.1                  3.5 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M_5R1027 by race
## t = -4, df = 41, p-value = 0.0003
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -1.05 -0.34
## sample estimates:
## mean in group Latinx  mean in group White 
##                  2.9                  3.6 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = 0.9, df = 54, p-value = 0.4
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -5.3 13.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              35                              31 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = -0.003, df = 87, p-value = 1
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -3.7  3.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              29                              29

race	group	mean	sd	n
Latinx	Attention-control	3.1	0.79	32
Latinx	Experimental	2.9	0.86	29
White	Attention-control	3.5	0.74	43
White	Experimental	3.6	0.57	54

2.6 Computer ownership

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1029)) %>% # <-- CHANGE 
  tabyl(M_5R1029, race, group) %>% #apresentar tabela <-- CHANGE 
  adorn_totals(c("row", "col")) %>% 
  adorn_percentages("col") %>% 
  adorn_pct_formatting(rounding = "half up", digits = 0) %>% 
  adorn_ns() %>% 
  kable() %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

M_5R1029	Latinx	White	Total
0	68% (21)	30% (13)	46% (34)
1	32% (10)	70% (30)	54% (40)
Total	100% (31)	100% (43)	100% (74)

M_5R1029	Latinx	White	Total
0	59% (17)	46% (25)	51% (42)
1	41% (12)	54% (29)	49% (41)
Total	100% (29)	100% (54)	100% (83)

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1029)) %>% # <-- change
  filter(group == "Attention-control") %>%
  {descr::crosstab(.$M_5R1029,.$race, prop.c = TRUE,chisq = TRUE)} # <-- change

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |          Column Percent | 
## |-------------------------|
## 
## ====================================
##               .$race
## .$M_5R1029    Latinx   White   Total
## ------------------------------------
## 0                21      13      34 
##                  68%     30%        
## ------------------------------------
## 1                10      30      40 
##                  32%     70%        
## ------------------------------------
## Total            31      43      74 
##                  42%     58%        
## ====================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 10      d.f. = 1      p = 0.0014 
## 
## Pearson's Chi-squared test with Yates' continuity correction 
## ------------------------------------------------------------
## Chi^2 = 8.8      d.f. = 1      p = 0.0031 
##         Minimum expected frequency: 14

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1029)) %>% # <-- change
  filter(group == "Experimental") %>%
  {descr::crosstab(.$M_5R1029,.$race, prop.c = TRUE,chisq = TRUE)} # <-- change

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |          Column Percent | 
## |-------------------------|
## 
## ====================================
##               .$race
## .$M_5R1029    Latinx   White   Total
## ------------------------------------
## 0                17      25      42 
##                  59%     46%        
## ------------------------------------
## 1                12      29      41 
##                  41%     54%        
## ------------------------------------
## Total            29      54      83 
##                  35%     65%        
## ====================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 1.1      d.f. = 1      p = 0.284 
## 
## Pearson's Chi-squared test with Yates' continuity correction 
## ------------------------------------------------------------
## Chi^2 = 0.71      d.f. = 1      p = 0.401 
##         Minimum expected frequency: 14

2.7 Living together

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1002)) %>% # <-- CHANGE 
  mutate(married = if_else(M_5R1002 == "4" | M_5R1002 == "5", 1,0)) %>% # <-- married
  tabyl(married, race, group) %>% #apresentar tabela <-- CHANGE 
  adorn_totals(c("row", "col")) %>% 
  adorn_percentages("col") %>% 
  adorn_pct_formatting(rounding = "half up", digits = 0) %>% 
  adorn_ns() %>% 
  kable() %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

married	Latinx	White	Total
0	22% (7)	28% (12)	25% (19)
1	78% (25)	72% (31)	75% (56)
Total	100% (32)	100% (43)	100% (75)

married	Latinx	White	Total
0	31% (9)	39% (21)	36% (30)
1	69% (20)	61% (33)	64% (53)
Total	100% (29)	100% (54)	100% (83)

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1002)) %>% # <-- change
  filter(!is.na(race) & !is.na(M_5R1002)) %>% # <-- CHANGE 
  mutate(married = if_else(M_5R1002 == "4" | M_5R1002 == "5", 1,0)) %>% # <-- married
  
  
  filter(group == "Attention-control") %>%
  {descr::crosstab(.$married,.$race, prop.c = TRUE,chisq = TRUE)} # <-- change

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |          Column Percent | 
## |-------------------------|
## 
## ===================================
##              .$race
## .$married    Latinx   White   Total
## -----------------------------------
## 0                7      12      19 
##                 22%     28%        
## -----------------------------------
## 1               25      31      56 
##                 78%     72%        
## -----------------------------------
## Total           32      43      75 
##                 43%     57%        
## ===================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 0.35      d.f. = 1      p = 0.552 
## 
## Pearson's Chi-squared test with Yates' continuity correction 
## ------------------------------------------------------------
## Chi^2 = 0.11      d.f. = 1      p = 0.745 
##         Minimum expected frequency: 8.1

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1002)) %>% # <-- change
  filter(!is.na(race) & !is.na(M_5R1002)) %>% # <-- CHANGE 
  mutate(married = if_else(M_5R1002 == "4" | M_5R1002 == "5", 1,0)) %>% # <-- married
  filter(group == "Experimental") %>%
  {descr::crosstab(.$married,.$race, prop.c = TRUE,chisq = TRUE)} # <-- change

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |          Column Percent | 
## |-------------------------|
## 
## ===================================
##              .$race
## .$married    Latinx   White   Total
## -----------------------------------
## 0                9      21      30 
##                 31%     39%        
## -----------------------------------
## 1               20      33      53 
##                 69%     61%        
## -----------------------------------
## Total           29      54      83 
##                 35%     65%        
## ===================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 0.5      d.f. = 1      p = 0.478 
## 
## Pearson's Chi-squared test with Yates' continuity correction 
## ------------------------------------------------------------
## Chi^2 = 0.22      d.f. = 1      p = 0.638 
##         Minimum expected frequency: 10

2.8 Financial stress

desc_ttest(dependent = "M_5R1FST")

## 
##  Welch Two Sample t-test
## 
## data:  M_5R1FST by race
## t = -0.6, df = 65, p-value = 0.6
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -3.4  1.9
## sample estimates:
## mean in group Latinx  mean in group White 
##                   16                   17 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M_5R1FST by race
## t = 2, df = 60, p-value = 0.1
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.5  4.1
## sample estimates:
## mean in group Latinx  mean in group White 
##                   17                   15 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = 0.9, df = 54, p-value = 0.4
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -5.3 13.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              35                              31 
## 
## 
##  Welch Two Sample t-test
## 
## data:  M1AGE by group
## t = -0.003, df = 87, p-value = 1
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -3.7  3.7
## sample estimates:
## mean in group Attention-control      mean in group Experimental 
##                              29                              29

race	group	mean	sd	n
Latinx	Attention-control	16	5.8	32
Latinx	Experimental	17	4.9	29
White	Attention-control	17	5.5	43
White	Experimental	15	5.2	54

2.9 Surgency

we’ll need to ask Craig

2.10 Education

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1012)) %>% # <-- CHANGE 
  tabyl(M_5R1012, race, group) %>% #apresentar tabela <-- CHANGE 
  adorn_totals(c("row", "col")) %>% 
  adorn_percentages("col") %>% 
  adorn_pct_formatting(rounding = "half up", digits = 0) %>% 
  adorn_ns() %>% 
  kable() %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

M_5R1012	Latinx	White	Total
1	22% (7)	14% (6)	17% (13)
2	44% (14)	14% (6)	27% (20)
3	13% (4)	7% (3)	9% (7)
4	13% (4)	44% (19)	31% (23)
5	9% (3)	21% (9)	16% (12)
Total	100% (32)	100% (43)	100% (75)

M_5R1012	Latinx	White	Total
1	24% (7)	11% (6)	16% (13)
2	38% (11)	30% (16)	33% (27)
3	21% (6)	4% (2)	10% (8)
4	10% (3)	44% (24)	33% (27)
5	7% (2)	11% (6)	10% (8)
Total	100% (29)	100% (54)	100% (83)

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1012)) %>% # <-- change
  filter(group == "Attention-control") %>%
  {descr::crosstab(.$M_5R1012,.$race, prop.c = TRUE,chisq = TRUE)} # <-- change

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |          Column Percent | 
## |-------------------------|
## 
## ====================================
##               .$race
## .$M_5R1012    Latinx   White   Total
## ------------------------------------
## 1                 7       6      13 
##                21.9%   14.0%        
## ------------------------------------
## 2                14       6      20 
##                43.8%   14.0%        
## ------------------------------------
## 3                 4       3       7 
##                12.5%    7.0%        
## ------------------------------------
## 4                 4      19      23 
##                12.5%   44.2%        
## ------------------------------------
## 5                 3       9      12 
##                 9.4%   20.9%        
## ------------------------------------
## Total            32      43      75 
##                  43%     57%        
## ====================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 15      d.f. = 4      p = 0.00489 
## 
##         Minimum expected frequency: 3 
## Cells with Expected Frequency < 5: 2 of 10 (20%)

data_april %>% 
  mutate(group = if_else(CONDITION.x == 0,"Attention-control","Experimental")) %>% ## GROUP INSTEAD OF CONDITION.X
  rename(race = M_5R1003) %>%  ##RENAME RACE
  mutate(race = if_else(race == 0, "White","Latinx")) %>% 
  
  filter(!is.na(race) & !is.na(M_5R1012)) %>% # <-- change
  filter(group == "Experimental") %>%
  {descr::crosstab(.$M_5R1012,.$race, prop.c = TRUE,chisq = TRUE)} # <-- change

##    Cell Contents 
## |-------------------------|
## |                   Count | 
## |          Column Percent | 
## |-------------------------|
## 
## ====================================
##               .$race
## .$M_5R1012    Latinx   White   Total
## ------------------------------------
## 1                 7       6      13 
##                24.1%   11.1%        
## ------------------------------------
## 2                11      16      27 
##                37.9%   29.6%        
## ------------------------------------
## 3                 6       2       8 
##                20.7%    3.7%        
## ------------------------------------
## 4                 3      24      27 
##                10.3%   44.4%        
## ------------------------------------
## 5                 2       6       8 
##                 6.9%   11.1%        
## ------------------------------------
## Total            29      54      83 
##                  35%     65%        
## ====================================
## 
## Statistics for All Table Factors
## 
## Pearson's Chi-squared test 
## ------------------------------------------------------------
## Chi^2 = 15      d.f. = 4      p = 0.00434 
## 
##         Minimum expected frequency: 2.8 
## Cells with Expected Frequency < 5: 3 of 10 (30%)

3 Main checks

Using all dataset (n = 159), we can check (and compare) mean differences between latinos and non-latinos. Let’s just remember that we have 61 latinos, and 97 non latinos.

To everything becomes transparent, one participant was not used in this comparison. In the original dataset, the information about his “race” was empty.

include_graphics("mean comparisons missing.PNG")

3.1 Infant Behavior1 (blang variable)

data_april %>% 
  filter(!is.na(M_5R1003)) %>% #remove who is missing
    group_by(CONDITION.x) %>% #group by experimental condition
    select(ID, CONDITION.x, M_5R1003, blang1,blang2) %>% #get focal variables
    gather(key, value, blang1,blang2) %>%  #turn to long
  tbl_df() %>% #tidy table
    setNames(c("ID","condition","hispanic", "time","result")) %>% #change names to make understandable
    mutate(time = ifelse(time == "blang1", "Time 1", "Time 2"),
           hispanic = recode_factor(hispanic, "0" = "White", "1" = "Latinx"),
           condition = recode_factor(condition, "0" = "Control", "1" = "Intervention")) %>%  #same thing
  ggplot(.,
         aes(x=time, y=result,group = interaction(hispanic,condition), color=hispanic)) +
  stat_summary(geom="line", size=1.5, aes(linetype=condition)) +
  stat_summary(geom="errorbar", size=0.2, width = .2) +
  scale_linetype_manual(values=c("solid", "dotted")) +
  scale_color_manual(values=c('#999999','#E69F00')) +
  labs(title = "Infant social behaviors by condition between Latinx and White children.", y = "Results", x = "Time", color = "Race", linetype = "Group")

Due the graphic, I checked the interaction between ethinicity (Hispanic or White) and Condition (Attention or Intervention group) at the baseline. Despite the marginal main effect, no interaction was found (F(1, 154) = 0.03, p = 0.867). Full results are below.

lm(blang1 ~ CONDITION.x * M_5R1003, data = data_april) %>% apa.aov.table(.)

## 
## 
## ANOVA results using blang1 as the dependent variable
##  
## 
##               Predictor     SS  df    MS    F    p partial_eta2
##             (Intercept)  11.88   1 11.88 2.41 .122             
##             CONDITION.x  19.04   1 19.04 3.87 .051          .02
##                M_5R1003   0.87   1  0.87 0.18 .674          .00
##  CONDITION.x x M_5R1003   0.14   1  0.14 0.03 .867          .00
##                   Error 757.82 154  4.92                       
##  CI_90_partial_eta2
##                    
##          [.00, .08]
##          [.00, .02]
##          [.00, .01]
##                    
## 
## Note: Values in square brackets indicate the bounds of the 90% confidence interval for partial eta-squared

3.2 Maternal Behavior (mlang variable)

data_april %>% 
  filter(!is.na(M_5R1003)) %>% #remove who is missing
    group_by(CONDITION.x) %>% #group by experimental condition
    select(ID, CONDITION.x, M_5R1003, mlang1,mlang2) %>% #get focal variables
    gather(key, value, mlang1,mlang2) %>%  #turn to long
  tbl_df() %>% #tidy table
    setNames(c("ID","condition","hispanic", "time","result")) %>% #change names to make understandable
    mutate(time = ifelse(time == "mlang1", "Time 1", "Time 2"),
           hispanic = recode_factor(hispanic, "0" = "White", "1" = "Latinx"),
           condition = recode_factor(condition, "0" = "Control", "1" = "Intervention")) %>%  #same thing
  ggplot(.,
         aes(x=time, y=result,group = interaction(hispanic,condition), color=hispanic)) +
  stat_summary(geom="line", size=1.5, aes(linetype=condition)) +
  stat_summary(geom="errorbar", size=0.2, width = .2) +
  scale_linetype_manual(values=c("solid", "dotted")) +
  scale_color_manual(values=c('#999999','#E69F00')) +
  labs(title = "Maternal Responsive behaviors by condition between Latinx and White mothers.", y = "Results", x = "Time", color = "Race", linetype = "Group")

3.3 Maternal knowledge Epals (M_4I1TOT variable)

data_april %>% 
  filter(!is.na(M_5R1003)) %>% #remove who is missing
    group_by(CONDITION.x) %>% #group by experimental condition
    select(ID, CONDITION.x, M_5R1003, M_4I1TOT,M_4I2TOT) %>% #get focal variables
    gather(key, value, M_4I1TOT,M_4I2TOT) %>%  #turn to long
  tbl_df() %>% #tidy table
    setNames(c("ID","condition","hispanic", "time","result")) %>% #change names to make understandable
    mutate(time = ifelse(time == "M_4I1TOT", "Time 1", "Time 2"),
           hispanic = recode_factor(hispanic, "0" = "White", "1" = "Latinx"),
           condition = recode_factor(condition, "0" = "Control", "1" = "Intervention")) %>%  #same thing
  ggplot(.,
         aes(x=time, y=result,group = interaction(hispanic,condition), color=hispanic)) +
  stat_summary(geom="line", size=1.5, aes(linetype=condition)) +
  stat_summary(geom="errorbar", size=0.2, width = .2) +
  scale_linetype_manual(values=c("solid", "dotted")) +
  scale_color_manual(values=c('#999999','#E69F00')) +
  labs(title = "Maternal knowledge by condition between Latinx and White mothers.", y = "Results", x = "Time", color = "Race", linetype = "Group")

data_april %>% 
  filter(!is.na(M_5R1003)) %>% #remove who is missing
  group_by(CONDITION.x) %>% #group by experimental condition
  select(ID, CONDITION.x, M_5R1003, M_4I1TOT,M_4I2TOT) %>% #get focal variables
  gather(key, value, M_4I1TOT,M_4I2TOT) %>%  #turn to long
  tbl_df() %>% #tidy table
  setNames(c("ID","condition","hispanic", "time","result")) %>% #change names to make understandable
  mutate(time = ifelse(time == "M_4I1TOT", "Time 1", "Time 2"),
         hispanic = recode_factor(hispanic, "0" = "White", "1" = "Latinx"),
         condition = recode_factor(condition, "0" = "Control", "1" = "Intervention")) %>%  #same thing
  ggplot(.,
         aes(x=time, y=result,group = condition, color=condition)) +
  stat_summary(geom="line", size=1.5, aes(linetype=condition)) +
  stat_summary(geom="errorbar", size=0.2, width = .2) +
  scale_linetype_manual(values=c("solid", "dotted")) +
  scale_color_manual(values=c('#999999','#E69F00')) +
  labs(title = "Maternal knowledge by condition", y = "Results", x = "Time")

data_april %>% 
  filter(!is.na(M_5R1003)) %>% #remove who is missing
  group_by(CONDITION.x) %>% #group by experimental condition
  select(ID, CONDITION.x, M_5R1003, M_4I1TOT,M_4I2TOT) %>% #get focal variables
  gather(key, value, M_4I1TOT,M_4I2TOT) %>%  #turn to long
  tbl_df() %>% #tidy table
  setNames(c("ID","condition","hispanic", "time","result")) %>% #change names to make understandable
  mutate(time = ifelse(time == "M_4I1TOT", "Time 1", "Time 2"),
         hispanic = recode_factor(hispanic, "0" = "White", "1" = "Latinx"),
         condition = recode_factor(condition, "0" = "Control", "1" = "Intervention")) %>%  #same thing
  ggplot(.,
         aes(x=time, y=result,group = hispanic, color=hispanic)) +
  stat_summary(geom="line", size=1.5) +
  stat_summary(geom="errorbar", size=0.2, width = .2) +
  scale_linetype_manual(values=c("solid", "dotted")) +
  scale_color_manual(values=c('#999999','#E69F00')) +
  labs(title = "Maternal knowledge by Race", y = "Results", x = "Time")

lm(M_4I1TOT ~ CONDITION.x * M_5R1003, data = data_april) %>% apa.aov.table(.)

## 
## 
## ANOVA results using M_4I1TOT as the dependent variable
##  
## 
##               Predictor      SS  df      MS      F    p partial_eta2
##             (Intercept) 2010.14   1 2010.14 275.33 .000             
##             CONDITION.x    0.01   1    0.01   0.00 .979          .00
##                M_5R1003   75.21   1   75.21  10.30 .002          .06
##  CONDITION.x x M_5R1003    1.34   1    1.34   0.18 .669          .00
##                   Error 1124.31 154    7.30                         
##  CI_90_partial_eta2
##                    
##         [.00, 1.00]
##          [.01, .13]
##          [.00, .02]
##                    
## 
## Note: Values in square brackets indicate the bounds of the 90% confidence interval for partial eta-squared

stargazer::stargazer(lm(M_4I1TOT ~ CONDITION.x * M_5R1003, data = data_april), type="text")

## 
## ================================================
##                          Dependent variable:    
##                      ---------------------------
##                               M_4I1TOT          
## ------------------------------------------------
## CONDITION.x                     0.015           
##                                (0.550)          
##                                                 
## M_5R1003                      -2.000***         
##                                (0.630)          
##                                                 
## CONDITION.x:M_5R1003            0.380           
##                                (0.890)          
##                                                 
## Constant                      6.800***          
##                                (0.410)          
##                                                 
## ------------------------------------------------
## Observations                     158            
## R2                              0.100           
## Adjusted R2                     0.086           
## Residual Std. Error       2.700 (df = 154)      
## F Statistic            5.900*** (df = 3; 154)   
## ================================================
## Note:                *p<0.1; **p<0.05; ***p<0.01

Despite the difference at the baseline (seemed by the graph), both groups increase its knowledge.

3.4 Dosage difference

t.test(doscond ~ M_5R1003, data = data_april)

## 
##  Welch Two Sample t-test
## 
## data:  doscond by M_5R1003
## t = -0.7, df = 147, p-value = 0.5
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.41  0.19
## sample estimates:
## mean in group 0 mean in group 1 
##          -0.199          -0.091

4 all comparions (table 2)

mod_repmeasure <- function(x,y) {
  
  x <- enquo(x)
  y <- enquo(y)
  
  ## SUMMARY TABLE
  data_april %>% 
  mutate(race = if_else(M_5R1003 == "1","Hispanic","Not Hispanic")) %>% 
  mutate(Condition = ifelse(CONDITION.x == "1", "Experimental", "Attention")) %>% 
  select(ID,race, Condition, !!x, !!y) %>% 
  gather(key = "Time",value="results", !!x,!!y) %>% 
  mutate(Time = ifelse(Time == !!x, "Time 1", "Time 2")) %>% 
  arrange(ID) %>%  
  group_by(race,Condition, Time) %>% 
  summarise(mean=mean(results), sd=sd(results), n=n()) %>% 
    kable(digits = 2) %>%
    kable_styling(bootstrap_options = c("striped", "hover"),position='center',
                  font_size = 12,full_width=F) %>% 
    print()

  
  ## PLOT
  plot <-  data_april %>% 
    mutate(race = if_else(M_5R1003 == "1","Hispanic","Not Hispanic")) %>% 
    mutate(Condition = ifelse(CONDITION.x == "1", "Experimental", "Attention")) %>% 
    select(ID,race, Condition, !!x, !!y) %>% 
    gather(key = "Time",value="results", !!x, !!y) %>% 
    mutate(Time = ifelse(Time == !!x, "Time 1", "Time 2")) %>% 
    ggplot(aes(x = Time, y = results, group = interaction(race,Condition))) +
    stat_summary(fun.y = mean, geom = "line",
                 aes(color = Condition, linetype = race),size=1.0) +
    stat_summary(fun.data = mean_se, geom = "errorbar", width=.1, 
                 aes(color = Condition,linetype = race)) +
  labs(title = as_label(x))
    
  ## INFERENTIAL TEST
  data_april %>% 
  mutate(race = if_else(M_5R1003 == "1","Hispanic","Not Hispanic")) %>% 
  mutate(Condition = ifelse(CONDITION.x == "1", "Experimental", "Attention")) %>% 
  select(ID,race, Condition, !!x, !!y) %>% 
  gather(key = "Time",value="results", !!x, !!y) %>% 
  mutate(Time = ifelse(Time == !!x, "Time 1", "Time 2")) %>% 
  lm(results ~ Time * Condition * race, data = .) ->> linear_result
  apaTables::apa.aov.table(linear_result) %>% print()

  
    plot

}

mod_repmeasure(x = "M_4I1TOT", y = "M_4I2TOT")

## <table class="table table-striped table-hover" style="font-size: 12px; width: auto !important; margin-left: auto; margin-right: auto;">
##  <thead>
##   <tr>
##    <th style="text-align:left;"> race </th>
##    <th style="text-align:left;"> Condition </th>
##    <th style="text-align:left;"> Time </th>
##    <th style="text-align:right;"> mean </th>
##    <th style="text-align:right;"> sd </th>
##    <th style="text-align:right;"> n </th>
##   </tr>
##  </thead>
## <tbody>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> 4.8 </td>
##    <td style="text-align:right;"> 2.3 </td>
##    <td style="text-align:right;"> 32 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> 5.7 </td>
##    <td style="text-align:right;"> 2.9 </td>
##    <td style="text-align:right;"> 32 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> 5.2 </td>
##    <td style="text-align:right;"> 2.5 </td>
##    <td style="text-align:right;"> 29 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> 9.0 </td>
##    <td style="text-align:right;"> 3.0 </td>
##    <td style="text-align:right;"> 29 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> 6.8 </td>
##    <td style="text-align:right;"> 3.1 </td>
##    <td style="text-align:right;"> 44 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> 6.9 </td>
##    <td style="text-align:right;"> 2.6 </td>
##    <td style="text-align:right;"> 44 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> 6.8 </td>
##    <td style="text-align:right;"> 2.6 </td>
##    <td style="text-align:right;"> 54 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> 9.9 </td>
##    <td style="text-align:right;"> 3.0 </td>
##    <td style="text-align:right;"> 54 </td>
##   </tr>
## </tbody>
## </table>
## 
## 
## ANOVA results using results as the dependent variable
##  
## 
##                Predictor      SS  df     MS     F    p partial_eta2
##              (Intercept)  741.12   1 741.12 95.85 .000             
##                     Time   12.25   1  12.25  1.58 .209          .01
##                Condition    2.37   1   2.37  0.31 .581          .00
##                     race   74.53   1  74.53  9.64 .002          .03
##         Time x Condition   66.76   1  66.76  8.63 .004          .03
##              Time x race    5.21   1   5.21  0.67 .412          .00
##         Condition x race    1.22   1   1.22  0.16 .692          .00
##  Time x Condition x race    0.03   1   0.03  0.00 .949          .00
##                    Error 2396.88 310   7.73                        
##  CI_90_partial_eta2
##                    
##          [.00, .03]
##          [.00, .01]
##          [.01, .07]
##          [.01, .06]
##          [.00, .02]
##          [.00, .01]
##         [.00, 1.00]
##                    
## 
## Note: Values in square brackets indicate the bounds of the 90% confidence interval for partial eta-squared

mod_repmeasure(x = "mlang1", y = "mlang2")

## <table class="table table-striped table-hover" style="font-size: 12px; width: auto !important; margin-left: auto; margin-right: auto;">
##  <thead>
##   <tr>
##    <th style="text-align:left;"> race </th>
##    <th style="text-align:left;"> Condition </th>
##    <th style="text-align:left;"> Time </th>
##    <th style="text-align:right;"> mean </th>
##    <th style="text-align:right;"> sd </th>
##    <th style="text-align:right;"> n </th>
##   </tr>
##  </thead>
## <tbody>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> 0.47 </td>
##    <td style="text-align:right;"> 2.9 </td>
##    <td style="text-align:right;"> 32 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> -1.23 </td>
##    <td style="text-align:right;"> 2.6 </td>
##    <td style="text-align:right;"> 32 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> -0.24 </td>
##    <td style="text-align:right;"> 2.7 </td>
##    <td style="text-align:right;"> 29 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> -0.35 </td>
##    <td style="text-align:right;"> 2.3 </td>
##    <td style="text-align:right;"> 29 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> 0.48 </td>
##    <td style="text-align:right;"> 2.4 </td>
##    <td style="text-align:right;"> 44 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> 0.13 </td>
##    <td style="text-align:right;"> 2.3 </td>
##    <td style="text-align:right;"> 44 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> -0.54 </td>
##    <td style="text-align:right;"> 2.2 </td>
##    <td style="text-align:right;"> 54 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> 0.80 </td>
##    <td style="text-align:right;"> 2.4 </td>
##    <td style="text-align:right;"> 54 </td>
##   </tr>
## </tbody>
## </table>
## 
## 
## ANOVA results using results as the dependent variable
##  
## 
##                Predictor      SS  df    MS    F    p partial_eta2
##              (Intercept)    6.96   1  6.96 1.16 .282             
##                     Time   45.83   1 45.83 7.65 .006          .02
##                Condition    7.60   1  7.60 1.27 .261          .00
##                     race    0.00   1  0.00 0.00 .985          .00
##         Time x Condition   19.17   1 19.17 3.20 .075          .01
##              Time x race   16.86   1 16.86 2.81 .094          .01
##         Condition x race    0.88   1  0.88 0.15 .703          .00
##  Time x Condition x race    0.04   1  0.04 0.01 .933          .00
##                    Error 1857.78 310  5.99                       
##  CI_90_partial_eta2
##                    
##          [.00, .06]
##          [.00, .02]
##         [.00, 1.00]
##          [.00, .04]
##          [.00, .03]
##          [.00, .01]
##          [.00, .00]
##                    
## 
## Note: Values in square brackets indicate the bounds of the 90% confidence interval for partial eta-squared

mod_repmeasure(x = "blang1", y = "blang2")

## <table class="table table-striped table-hover" style="font-size: 12px; width: auto !important; margin-left: auto; margin-right: auto;">
##  <thead>
##   <tr>
##    <th style="text-align:left;"> race </th>
##    <th style="text-align:left;"> Condition </th>
##    <th style="text-align:left;"> Time </th>
##    <th style="text-align:right;"> mean </th>
##    <th style="text-align:right;"> sd </th>
##    <th style="text-align:right;"> n </th>
##   </tr>
##  </thead>
## <tbody>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> 0.31 </td>
##    <td style="text-align:right;"> 2.0 </td>
##    <td style="text-align:right;"> 32 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> -1.03 </td>
##    <td style="text-align:right;"> 2.3 </td>
##    <td style="text-align:right;"> 32 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> -0.46 </td>
##    <td style="text-align:right;"> 2.1 </td>
##    <td style="text-align:right;"> 29 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> -0.68 </td>
##    <td style="text-align:right;"> 2.0 </td>
##    <td style="text-align:right;"> 29 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> 0.53 </td>
##    <td style="text-align:right;"> 2.5 </td>
##    <td style="text-align:right;"> 44 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Attention </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> 0.47 </td>
##    <td style="text-align:right;"> 2.8 </td>
##    <td style="text-align:right;"> 44 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 1 </td>
##    <td style="text-align:right;"> -0.37 </td>
##    <td style="text-align:right;"> 2.1 </td>
##    <td style="text-align:right;"> 54 </td>
##   </tr>
##   <tr>
##    <td style="text-align:left;"> Not Hispanic </td>
##    <td style="text-align:left;"> Experimental </td>
##    <td style="text-align:left;"> Time 2 </td>
##    <td style="text-align:right;"> 0.59 </td>
##    <td style="text-align:right;"> 1.9 </td>
##    <td style="text-align:right;"> 54 </td>
##   </tr>
## </tbody>
## </table>
## 
## 
## ANOVA results using results as the dependent variable
##  
## 
##                Predictor      SS  df    MS    F    p partial_eta2
##              (Intercept)    3.02   1  3.02 0.60 .439             
##                     Time   28.45   1 28.45 5.65 .018          .02
##                Condition    9.02   1  9.02 1.79 .182          .01
##                     race    0.93   1  0.93 0.18 .668          .00
##         Time x Condition    9.51   1  9.51 1.89 .170          .01
##              Time x race   15.01   1 15.01 2.98 .085          .01
##         Condition x race    0.15   1  0.15 0.03 .862          .00
##  Time x Condition x race    0.05   1  0.05 0.01 .921          .00
##                    Error 1560.66 310  5.03                       
##  CI_90_partial_eta2
##                    
##          [.00, .05]
##          [.00, .03]
##          [.00, .01]
##          [.00, .03]
##          [.00, .04]
##          [.00, .01]
##          [.00, .00]
##                    
## 
## Note: Values in square brackets indicate the bounds of the 90% confidence interval for partial eta-squared

library(emmeans)
emmeans(linear_result, pairwise ~ Time | Condition, adjust = "bonferroni")
emmeans(linear_result, pairwise ~ race, adjust = "bonferroni")

emmeans(linear_result, pairwise ~ Time,  adjust = "bonferroni")
emmeans(linear_result, pairwise ~ Time,  adjust = "bonferroni")

5 New analyses

First, I’ll assign the groups in the proper way. Therefore, the first group is formed of latinxs (M_5R1002 = 1) of color (all but M_5R104D).

include_graphics("latinxs of color.PNG")

In contrast, the second group is composed of white-only (M_5R104D = 1) non-latixs participants (M_5R1002 = 0).

include_graphics("white only.PNG")

After this division, this dataset results 31 latinxs dyads (mother/child) and 72 white dyads. Once this study gathered on two groups (intervention vs control), the following table describes the sample within each condition. I’ve added to dataset a variable (dummy coding) to make easier to find these participants just in case we need that later.

data_april <- data_april %>% 
  mutate(latino_group = case_when(
    M_5R1003 == "1" & M_5R104D != "1" ~ "latino",
    M_5R1003 == "0" & M_5R104D == "1" ~ "white"))

data_april <- data_april %>% 
  mutate(condition = if_else(CONDITION.x == 0,"attention-control","experimental"))

data_april %>% 
  filter(!is.na(latino_group) & !is.na(condition)) %>% 
  tabyl(condition, latino_group) %>% #apresentar tabela
  adorn_totals(c("row", "col")) %>% 
  adorn_percentages("col") %>% 
  adorn_pct_formatting(rounding = "half up", digits = 0) %>% 
  adorn_ns() %>% 
  kable() %>% 
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

condition	latino	white	Total
attention-control	55% (17)	44% (32)	48% (49)
experimental	45% (14)	56% (40)	52% (54)
Total	100% (31)	100% (72)	100% (103)

Three main variables are targets in this study: Infant Behavior (blang), Maternal Behavior (mlang), and Maternal knowledge Epals (M_4I1TOT).

Each variable will be tested in the following analyses.

mod_repmeasure_new_analyses <- function(x,y) {
  
  x <- enquo(x)
  y <- enquo(y)
  
  #diplay statistics summary table
  
  stats_table <- data_april %>% 
    filter(!is.na(latino_group)) %>% 
    select(ID,latino_group, condition, !!x, !!y) %>% 
    pivot_longer(cols = !!x:!!y, #variables that have results
                 names_to = "time",
                 values_to = "results") %>% 
    mutate(time = recode_factor(time, `!!x` = "baseline", `!!y` = "post")) %>% #change value coding
    group_by(latino_group,condition, time) %>% 
    summarise(mean=mean(results), sd=sd(results), n=n()) %>% 
    kable(digits = 2) %>%
    kable_styling(bootstrap_options = c("striped", "hover"),position='center',
                  font_size = 12,full_width=F) 
  
    # plot
  
  plot <- data_april %>% 
    filter(!is.na(latino_group)) %>% 
    select(ID,latino_group, condition, !!x, !!y) %>% 
    pivot_longer(cols = !!x:!!y, #variables that have results
                 names_to = "time",
                 values_to = "results") %>% 
    mutate(time = recode_factor(time, `!!x` = "baseline", `!!y` = "post")) %>% 
    ggplot(aes(x = time, y = results, group = interaction(latino_group,condition))) +
    stat_summary(fun.y = mean, geom = "line", aes(color = condition, linetype = latino_group),size=1.5) +
    stat_summary(fun.data = mean_se, geom = "errorbar", width=.2, size=1, aes(color = condition,linetype = latino_group)) +
    scale_linetype_manual(values=c("solid", "dotted")) +
    scale_color_manual(values=c('black','#E69F00')) +
    labs(title = as_label(x), y = "Results", x = "Time")

    ## INFERENTIAL TEST
  inferential_test <- data_april %>% 
    filter(!is.na(latino_group)) %>% 
    select(ID,latino_group, condition, !!x, !!y) %>% 
    pivot_longer(cols = !!x:!!y, #variables that have results
                 names_to = "time",
                 values_to = "results") %>% 
    mutate(time = recode_factor(time, `!!x` = "baseline", `!!y` = "post")) %>% 
    lm(results ~ time * condition * latino_group, data = .) %>% 
    apaTables::apa.aov.table(.)
  
  return(list(stats_table, plot, inferential_test))
  
}

5.1 Mother knowledge (M_4I1TOT)

mod_repmeasure_new_analyses(x = "M_4I1TOT", y = "M_4I2TOT")

[[1]]

latino_group	condition	time	mean	sd	n
latino	attention-control	M_4I1TOT	4.1	2.0	17
latino	attention-control	M_4I2TOT	4.9	2.7	17
latino	experimental	M_4I1TOT	5.6	2.9	14
latino	experimental	M_4I2TOT	9.0	2.6	14
white	attention-control	M_4I1TOT	7.4	3.0	32
white	attention-control	M_4I2TOT	7.5	2.6	32
white	experimental	M_4I1TOT	7.4	2.6	40
white	experimental	M_4I2TOT	10.0	3.0	40

[[2]] [[3]]

ANOVA results using results as the dependent variable

                Predictor      SS  df     MS     F    p
                 (Intercept)  280.06   1 280.06 37.40 .000
                        time    5.76   1   5.76  0.77 .381
                   condition   17.57   1  17.57  2.35 .127
                latino_group  126.73   1 126.73 16.92 .000
            time x condition   26.05   1  26.05  3.48 .064
         time x latino_group    2.85   1   2.85  0.38 .538
    condition x latino_group   12.88   1  12.88  1.72 .191

time x condition x latino_group 0.05 1 0.05 0.01 .935 Error 1482.76 198 7.49
partial_eta2 CI_90_partial_eta2

   .00         [.00, .03]
      .01         [.00, .05]
      .08         [.03, .14]
      .02         [.00, .06]
      .00         [.00, .02]
      .01         [.00, .04]
      .00         [.00, .00]

Note: Values in square brackets indicate the bounds of the 90% confidence interval for partial eta-squared

5.2 Infant Behavior (blang)

mod_repmeasure_new_analyses(x = "blang1", y = "blang2")

[[1]]

latino_group	condition	time	mean	sd	n
latino	attention-control	blang1	0.32	1.7	17
latino	attention-control	blang2	-1.60	2.1	17
latino	experimental	blang1	-0.47	1.4	14
latino	experimental	blang2	-0.54	2.3	14
white	attention-control	blang1	0.52	2.7	32
white	attention-control	blang2	0.62	2.9	32
white	experimental	blang1	-0.45	2.3	40
white	experimental	blang2	0.63	2.1	40

[[2]] [[3]]

ANOVA results using results as the dependent variable

                Predictor      SS  df    MS    F    p partial_eta2
                 (Intercept)    1.76   1  1.76 0.32 .570             
                        time   31.41   1 31.41 5.78 .017          .03
                   condition    4.77   1  4.77 0.88 .350          .00
                latino_group    0.46   1  0.46 0.08 .773          .00
            time x condition   13.19   1 13.19 2.43 .121          .01
         time x latino_group   22.68   1 22.68 4.18 .042          .02
    condition x latino_group    0.18   1  0.18 0.03 .855          .00

time x condition x latino_group 2.05 1 2.05 0.38 .540 .00 Error 1075.13 198 5.43
CI_90_partial_eta2

  [.00, .08]
     [.00, .03]
     [.00, .02]
     [.00, .05]
     [.00, .06]
     [.00, .01]
     [.00, .02]

Note: Values in square brackets indicate the bounds of the 90% confidence interval for partial eta-squared

5.3 Maternal Behavior (mlang)

mod_repmeasure_new_analyses(x = "mlang1", y = "mlang2")

[[1]]

latino_group	condition	time	mean	sd	n
latino	attention-control	mlang1	0.53	2.8	17
latino	attention-control	mlang2	-1.71	2.2	17
latino	experimental	mlang1	0.26	3.0	14
latino	experimental	mlang2	-0.44	2.4	14
white	attention-control	mlang1	0.66	2.5	32
white	attention-control	mlang2	0.27	2.4	32
white	experimental	mlang1	-0.59	2.4	40
white	experimental	mlang2	0.68	2.5	40

[[2]] [[3]]

ANOVA results using results as the dependent variable

                Predictor      SS  df    MS    F    p partial_eta2
                 (Intercept)    4.83   1  4.83 0.77 .381             
                        time   42.95   1 42.95 6.86 .009          .03
                   condition    0.56   1  0.56 0.09 .766          .00
                latino_group    0.19   1  0.19 0.03 .861          .00
            time x condition    9.17   1  9.17 1.47 .227          .01
         time x latino_group   19.00   1 19.00 3.04 .083          .02
    condition x latino_group    5.15   1  5.15 0.82 .365          .00

time x condition x latino_group 0.03 1 0.03 0.01 .941 .00 Error 1238.87 198 6.26
CI_90_partial_eta2

  [.00, .08]
     [.00, .02]
     [.00, .01]
     [.00, .04]
     [.00, .05]
     [.00, .03]
     [.00, .00]

Note: Values in square brackets indicate the bounds of the 90% confidence interval for partial eta-squared

Babynet some analysis

Luis Anunciacao

09 February, 2020

1 MI20 (April) dataset

1.1 Participants’ race

1.2 Infant behavioral

1.3 Maternal behavioral

1.4 Maternal knowledge

2 Demographic information

2.1 Mother and her children age

2.2 Biological sex

2.3 Primary language

2.4 Maternal depression

2.5 Computer comfort

2.6 Computer ownership

2.7 Living together

2.8 Financial stress

2.9 Surgency

2.10 Education

3 Main checks

3.1 Infant Behavior1 (blang variable)

3.2 Maternal Behavior (mlang variable)

3.3 Maternal knowledge Epals (M_4I1TOT variable)

3.4 Dosage difference

4 all comparions (table 2)

5 New analyses

5.1 Mother knowledge (M_4I1TOT)

5.2 Infant Behavior (blang)

5.3 Maternal Behavior (mlang)