Romantic Relationships and Sexual Experiences Study 2
2023-09-01
#Loading and installing packages
pacman::p_load(tidyverse, MASS,
magrittr, robustHD,
tibble, psych,
kableExtra, e1071,
knitr, tidyr,
lavaan, semPlot,
jtools, car,
lmtest, ggpubr,
FSA, rstatix,
writexl, readxl,
rcompanion, coin, lm.beta,
rstatix, Hmisc, pwr)#Making these data numeric so they filter correctly
mergedData$`Duration (in seconds).x`<- as.numeric(mergedData$`Duration (in seconds).x`)
mergedData$Consent.x<- as.numeric(mergedData$Consent.x)
mergedData$Q_RelevantIDDuplicateScore.x<- as.numeric(mergedData$Q_RelevantIDDuplicateScore.x)
mergedData$Relationship_length_1<- as.numeric(mergedData$Relationship_length_1)
mergedData$Q_RelevantIDFraudScore.x<- as.numeric(mergedData$Q_RelevantIDFraudScore.x)
median(mergedData$`Duration (in seconds).x`) - (3*(sd(mergedData$`Duration (in seconds).x`, na.rm=T))) #median response time is 529 (or 8.82 minutes)## [1] -15909.42sd(mergedData$`Duration (in seconds).x`)## [1] 5479.475length(which(mergedData$Consent.x==2))## [1] 0length(which(mergedData$Q_RelevantIDDuplicateScore.x>=75))## [1] 2length(which(mergedData$Relationship_length_1 < 1))## [1] 5length(which(mergedData$Q_RelevantIDFraudScore.x>=30))## [1] 14length(which(mergedData$`Duration (in seconds).x` <180))## [1] 13clean.merged.data <- mergedData %>%
dplyr::filter(Consent.x == 1, #(n=0)
`Duration (in seconds).x`>=180, #participants that took survey in less than 3 minutes are removed (n=13)
Q_RelevantIDDuplicateScore.x<75, #getting rid of duplicate participants (n=2)
Relationship_length_1>0,#getting rid of participants that have been in relationship under 1 year (n=5)
Q_RelevantIDFraudScore.x< 30) #A score greater than or equal to 30 means the response is likely fraudulent and a bot (n=13)
data<- clean.merged.data %>%
dplyr::filter(!is.na(SV_1.x),
!is.na(SV_2.x),
!is.na(SV_3),
!is.na(SV_4.x),
!is.na(ACE_1.x),
!is.na(ACE_2),
!is.na(ACE_3)) #getting rid of people who didn't complete all the asa/csa items (n=50)
data %<>% dplyr::mutate_if(is.factor,as.character)
which(colnames(data)=="SV_perp_8") # identifies the column number of different variables## [1] 127data[,c(13:15, 42:127, 148:154)] %<>% dplyr::mutate_if(is.character,as.numeric) #making only select columns numericSexual Violence Variables
data <- data %>%
rename(SV_1 = SV_1.x,#renaming the SV variables
SV_2 = SV_2.x,
SV_4=SV_4.x,
SV_1.pre=SV_1.y,
SV_2.pre=SV_2.y,
SV_3.pre=SV_4.y,
SV_4.pre=SV_5,
ACE.pre=ACE_1.y,
ACE_1=ACE_1.x)
data[,c(131:133)] %<>% dplyr::mutate_if(is.character,as.numeric) #making only select columns numeric
# First creating true false variables for each ASA item
data$SV_1_log<- data$SV_1==1 | data$SV_1==2 | data$SV_1==3 | data$SV_1==4
data$SV_2_log<- data$SV_2==1 | data$SV_2==2 | data$SV_2==3 | data$SV_2==4
data$SV_3_log<- data$SV_3==1 | data$SV_3==2 | data$SV_3==3 | data$SV_3==4
data$SV_4_log<- data$SV_4==1 | data$SV_4==2 | data$SV_4==3 | data$SV_4==4
# ASA item
data$someASA<- data$SV_1_log==TRUE | data$SV_2_log==TRUE | data$SV_3_log==TRUE | data$SV_4_log==TRUE
# CSA items
data$someCSA<- data$ACE_1==1 | data$ACE_1==2 | data$ACE_2==1 | data$ACE_2==2 |data$ACE_3==1 | data$ACE_3==2
# ASA ONLY item
data$ASAonly<- data$someASA==TRUE & data$someCSA==FALSE
# CSA Only item
data$CSAonly<- data$someASA==FALSE & data$someCSA==TRUE
# Revictimized
data$revictimized<- data$someASA==TRUE & data$someCSA==TRUE
#Neither ASA nor CSA
data$NeitherASAnorCSA<- data$someASA==FALSE & data$someCSA==FALSE
################################# getting rid of CSA only group (n=27)##########################
data <- data %>%
dplyr::filter(CSAonly==FALSE)
# SV 1
kableExtra::kable(table(data$SV_1), booktabs = TRUE, col.names = c("Rape: Physical Force", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| Rape: Physical Force | Frequency |
|---|---|
| 0 | 340 |
| 1 | 102 |
| 2 | 65 |
| 3 | 27 |
| 4 | 32 |
# SV 2
kableExtra::kable(table(data$SV_2), booktabs = TRUE, col.names = c("Attempted Rape: Physical Force", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| Attempted Rape: Physical Force | Frequency |
|---|---|
| 0 | 358 |
| 1 | 105 |
| 2 | 58 |
| 3 | 13 |
| 4 | 32 |
# SV 3
kableExtra::kable(table(data$SV_3), booktabs = TRUE, col.names = c("Rape: Drugs or alcohol", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| Rape: Drugs or alcohol | Frequency |
|---|---|
| 0 | 374 |
| 1 | 108 |
| 2 | 46 |
| 3 | 11 |
| 4 | 27 |
# SV 4
kableExtra::kable(table(data$SV_4), booktabs = TRUE, col.names = c("Attempted Rape: Drugs or alcohol", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| Attempted Rape: Drugs or alcohol | Frequency |
|---|---|
| 0 | 412 |
| 1 | 90 |
| 2 | 39 |
| 3 | 9 |
| 4 | 16 |
# SV 5
kableExtra::kable(table(data$SV_Rape), booktabs = TRUE, col.names = c("Self-Reported Rape", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| Self-Reported Rape | Frequency |
|---|---|
| 0 | 361 |
| 1 | 205 |
# ACE 1
# This question reads, "How often did anyone at least 5 years older than you or an adult ever touch you sexually?
kableExtra::kable(table(data$ACE_1), booktabs = TRUE, col.names = c("CSA 1", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| CSA 1 | Frequency |
|---|---|
| 0 | 417 |
| 1 | 48 |
| 2 | 101 |
# ACE 2
# "How often did anyone at least 5 years older than you or an adult try to make you touch them sexually?"
kableExtra::kable(table(data$ACE_2), booktabs = TRUE, col.names = c("CSA 2", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| CSA 2 | Frequency |
|---|---|
| 0 | 442 |
| 1 | 41 |
| 2 | 83 |
# ACE 3
#"How often did anyone at least 5 years older than you or an adult force you to have sex?"
kableExtra::kable(table(data$ACE_3), booktabs = TRUE, col.names = c("CSA 3", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| CSA 3 | Frequency |
|---|---|
| 0 | 491 |
| 1 | 31 |
| 2 | 44 |
# ASA Only
kableExtra::kable(table(data$ASAonly), booktabs = TRUE, col.names = c("ASA Only", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| ASA Only | Frequency |
|---|---|
| FALSE | 368 |
| TRUE | 198 |
# Revictimized Group
kableExtra::kable(table(data$revictimized), booktabs = TRUE, col.names = c("Revictimized", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| Revictimized | Frequency |
|---|---|
| FALSE | 413 |
| TRUE | 153 |
# Neither ASA nor CSA
kableExtra::kable(table(data$NeitherASAnorCSA), booktabs = TRUE, col.names = c("Neither ASA nor CSA", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE)| Neither ASA nor CSA | Frequency |
|---|---|
| FALSE | 351 |
| TRUE | 215 |
# Assault timing variables
data <- data %>%
dplyr::mutate(SV_1_timing = (SV_1_timing_1 + (SV_1_timing_2/12))) #creating timing variables
data <- data %>%
dplyr::mutate(SV_2_timing = (SV_2_timing_1 + (SV_2_timing_2/12)))
data <- data %>%
dplyr::mutate(SV_3_timing = (SV_3_timing_1 + (SV_3_timing_2/12)))
data <- data %>%
dplyr::mutate(SV_4_timing = (SV_4_timing_1 + (SV_4_timing_2/12)))
data %>%
dplyr::select(SV_1_timing, SV_2_timing, SV_3_timing, SV_4_timing)%>%
psych::describe(na.rm=TRUE) %>%
as.data.frame() %>%
dplyr::select("n", "mean", "sd", "median", "min", "max", "range", "skew", "kurtosis") %>%
kableExtra::kable(caption= "Descriptive Statistics for ASA Timing", digits = 2) %>%
kable_styling(bootstrap_options = "striped", full_width = TRUE)| n | mean | sd | median | min | max | range | skew | kurtosis | |
|---|---|---|---|---|---|---|---|---|---|
| SV_1_timing | 225 | 13.34 | 11.03 | 10.00 | 0 | 56.33 | 56.33 | 1.40 | 1.87 |
| SV_2_timing | 207 | 12.31 | 10.07 | 9.83 | 0 | 56.17 | 56.17 | 1.46 | 2.33 |
| SV_3_timing | 189 | 11.74 | 9.23 | 9.00 | 0 | 56.17 | 56.17 | 1.46 | 3.02 |
| SV_4_timing | 153 | 12.38 | 10.16 | 10.00 | 0 | 56.17 | 56.17 | 1.61 | 3.13 |
# Perp gender
length(which(data$SV_gender_1==1)) # cis men (n = 346)## [1] 346length(which(data$SV_gender_2==1)) # cis women (n = 17)## [1] 17length(which(data$SV_gender_3==1)) # non-binary, genderqueer, agender, or gend fluid (n=3)## [1] 3length(which(data$SV_gender_4==1)) # trans man (n=0)## [1] 0length(which(data$SV_gender_5==1)) # trans woman (n=4)## [1] 4length(which(data$SV_gender_6==1)) #gender not listed above (n = 1)## [1] 1# Perp relationship
#see "other" responses coding doc
######acknowledged victims
data$RapeItems<- data$SV_1_log==TRUE | data$SV_3_log==TRUE | data$ACE_3==1 | data$ACE_3==2 #creating variable such that if they experienced completed rape or forced sex in ACE item then they were considered rape victim
data<- data %>%
mutate(unacknowledged = case_when(
data$RapeItems==TRUE & data$SV_Rape==0 ~ 0,
data$RapeItems==TRUE & data$SV_Rape==1 ~ 1,
data$RapeItems==FALSE & data$SV_Rape==0 ~ 3 # this number includes people who experienced ASA just not completed rape
))
table(data$unacknowledged)##
## 0 1 3
## 108 192 253Demographics
# Relationship length variable
data <- data %>%
dplyr::mutate(relLength = (Relationship_length_1 + (Relationship_length_2/12))) #creating relationship length variable
data$relLength<- round(data$relLength, 2)
data$relLength<- as.numeric(data$relLength)
data$Age_1<- as.numeric(data$Age_1)
# relationship length and age descriptives
data %>%
dplyr::select(relLength, Age_1)%>%
psych::describe(na.rm=TRUE) %>%
as.data.frame() %>%
dplyr::select("n", "mean", "sd", "median", "min", "max", "range", "skew", "kurtosis") %>%
kableExtra::kable(caption= "Age and Relationship Length Descriptives", digits = 2) %>%
kable_styling(bootstrap_options = "striped", full_width = TRUE)| n | mean | sd | median | min | max | range | skew | kurtosis | |
|---|---|---|---|---|---|---|---|---|---|
| relLength | 565 | 8.05 | 7.91 | 5.33 | 1 | 48.92 | 47.92 | 2.24 | 5.94 |
| Age_1 | 566 | 35.31 | 12.50 | 32.00 | 18 | 78.00 | 60.00 | 1.00 | 0.39 |
# Relationship status
data$Relationship_type<-as.numeric(data$Relationship_type)
data$Relationship_type <- factor(data$Relationship_type,
levels=c(1, 2, 3, 4, 5, 6),
labels=c("Friends with benefits", "Casually dating", "Seriously dating", "Living together", "Engaged", "Married"))
table(data$Relationship_type)##
## Friends with benefits Casually dating Seriously dating
## 4 9 150
## Living together Engaged Married
## 262 21 120prop.table(table(data$Relationship_type))##
## Friends with benefits Casually dating Seriously dating
## 0.007067138 0.015901060 0.265017668
## Living together Engaged Married
## 0.462897527 0.037102473 0.212014134#Sexual Orientation
data$Orientation<-as.numeric(data$Orientation)
data$Orientation <- factor(data$Orientation,
levels=c(1, 2, 3, 4),
labels=c("Lesbian or gay", "Bisexual or pansexual", "Sexual or heterosexual", "Other"))
table(data$Orientation)##
## Lesbian or gay Bisexual or pansexual Sexual or heterosexual
## 35 161 356
## Other
## 14#see other responses coding for frequencies, one person recoded from 'other' to 'bisexual or panromantic' (so 1 more in that category and 1 less in other in final table)
#Education
data$Education<-as.numeric(data$Education)
data$Education<- factor(data$Education,
levels=c(1.0, 2.0, 6.0, 3.0, 4.0, 7.0, 8.0, 9.0),
labels=c("Some high school", "High school degree", "Associate's degree", "Some college", "Bachelor's degree", "Some graduate school", "Master's degree", "Doctorate degree"))
table(data$Education)##
## Some high school High school degree Associate's degree
## 5 78 47
## Some college Bachelor's degree Some graduate school
## 141 202 9
## Master's degree Doctorate degree
## 63 21prop.table(table(data$Education))##
## Some high school High school degree Associate's degree
## 0.008833922 0.137809187 0.083038869
## Some college Bachelor's degree Some graduate school
## 0.249116608 0.356890459 0.015901060
## Master's degree Doctorate degree
## 0.111307420 0.037102473#Household Income
data$Income<-as.numeric(data$Income)
data$Income<- factor(data$Income,
levels=c(1, 2, 3, 4, 5),
labels=c("Less than $15,000", "$15,000-34,999", "$35,000-49,999", "$50,000-74,999", "$75,000 or more"))
table(data$Income)##
## Less than $15,000 $15,000-34,999 $35,000-49,999 $50,000-74,999
## 47 98 94 127
## $75,000 or more
## 200prop.table(table(data$Income))##
## Less than $15,000 $15,000-34,999 $35,000-49,999 $50,000-74,999
## 0.08303887 0.17314488 0.16607774 0.22438163
## $75,000 or more
## 0.35335689# Gender
Gender<- c("Cisgender woman", "Nonbinary, genderqueer, agender, or genderfluid", "Transgender man", "Transgender woman", "Questioning", "Other")
#See other responses coding for frequencies
# Race (!!!!!!!!!UPDATE!!!!!!!!!!)
length(which(data$Race_1==1))## [1] 3length(which(data$Race_2==1))## [1] 39length(which(data$Race_3==1))## [1] 53length(which(data$Race_4==1))## [1] 55length(which(data$Race_5==1))## [1] 4length(which(data$Race_6==1))## [1] 8length(which(data$Race_7==1))## [1] 469length(which(data$Race_8==1))## [1] 0#1 and 6 accidentally repeated, fixing it
length(which(data$Race_1==1 & data$Race_6==1)) #so only 1 person aid yes to option 1 and 6## [1] 1#8 people said yes to 6 and 3 to q1, so 8+3=11-1=10 because removing repeat
Race<- c("Asian American or Asian", "Black or African American", "Latinx or Hispanic", "Middle Eastern or North African", "Native American, Hawaiian Native/Pacific Islander, or Alaskan Native", "White or European American", "Other")
RaceFrequency<- c(39, 53, 55, 4, 10, 469, 0)
Racedata<- cbind(Race, RaceFrequency)
kableExtra::kable(Racedata, booktabs = TRUE, col.names = c("Race", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE) | Race | Frequency |
|---|---|
| Asian American or Asian | 39 |
| Black or African American | 53 |
| Latinx or Hispanic | 55 |
| Middle Eastern or North African | 4 |
| Native American, Hawaiian Native/Pacific Islander, or Alaskan Native | 10 |
| White or European American | 469 |
| Other | 0 |
# Employment (!!!!!!!!!UPDATE!!!!!!!!!!)
length(which(data$Employment_1==1))## [1] 83length(which(data$Employment_2==1))## [1] 315length(which(data$Employment_3==1))## [1] 120length(which(data$Employment_4==1))## [1] 89length(which(data$Employment_5==1))## [1] 14length(which(data$Employment_6==1))## [1] 3Employment<- c("Student", "Full-time employed", "Part-time employed", "Not employed", "Retired", "Temporarily laid off")
Employmentfrequency<- c(83, 315, 120, 89, 14, 3)
Employmentdata<- cbind(Employment, Employmentfrequency)
kableExtra::kable(Employmentdata, booktabs = TRUE, col.names = c("Employment", "Frequency")) %>%
kableExtra::kable_styling(bootstrap_options = "striped", full_width = TRUE) | Employment | Frequency |
|---|---|
| Student | 83 |
| Full-time employed | 315 |
| Part-time employed | 120 |
| Not employed | 89 |
| Retired | 14 |
| Temporarily laid off | 3 |
Scale creation
#################PRQC Scale and Subscales#######################
data <- data %>%
rowwise() %>%
mutate(PRQC = mean(c(PRQC_1, PRQC_2, PRQC_3, PRQC_4, PRQC_5, PRQC_6, PRQC_7, PRQC_8, PRQC_9, PRQC_10, PRQC_11, PRQC_12, PRQC_13, PRQC_14, PRQC_15, PRQC_16, PRQC_17, PRQC_18), na.rm = TRUE),
Satisfaction = mean(c(PRQC_1, PRQC_2, PRQC_3), na.rm = TRUE),
Commitment = mean(c(PRQC_4, PRQC_5, PRQC_6), na.rm = TRUE),
Intimacy = mean(c(PRQC_7, PRQC_8, PRQC_9), na.rm = TRUE),
Trust = mean(c(PRQC_10, PRQC_11, PRQC_12), na.rm = TRUE),
Passion = mean(c(PRQC_13, PRQC_14, PRQC_15), na.rm = TRUE),
Love = mean(c(PRQC_16, PRQC_17, PRQC_18), na.rm = TRUE))%>%
ungroup()
#################### Attachment###################
#Reverse coding
data <- data %>%
dplyr::mutate(AAQ_1.r = (8 - AAQ_1))
data <- data %>%
dplyr::mutate(AAQ_3.r = (8 - AAQ_3))
data <- data %>%
dplyr::mutate(AAQ_4.r = (8 - AAQ_4))
data <- data %>%
dplyr::mutate(AAQ_12.r = (8 - AAQ_12))
data <- data %>%
dplyr::mutate(AAQ_14.r = (8 - AAQ_14))
data <- data %>%
dplyr::mutate(AAQ_16.r = (8 - AAQ_16))
data <- data %>%
dplyr::mutate(AAQ_17.r = (8 - AAQ_17))
#Anxiety subscale
data <- data %>%
rowwise() %>%
mutate(AAQ.ANX = mean(c(AAQ_4.r, AAQ_10, AAQ_11, AAQ_12.r, AAQ_13, AAQ_14.r, AAQ_15, AAQ_16.r, AAQ_17.r), na.rm = TRUE))%>%
ungroup()
#Avoidance subscale
data <- data %>%
rowwise() %>%
mutate(AAQ.AV = mean(c(AAQ_1.r, AAQ_2, AAQ_3.r, AAQ_5, AAQ_6, AAQ_7, AAQ_8, AAQ_9), na.rm = TRUE))%>%
ungroup()
############### NSSS-S Scale######################
# No reverse-coded items
data <- data %>%
rowwise() %>%
mutate(NSSS = mean(c(`NSSS-S_1`, `NSSS-S_2`, `NSSS-S_3`, `NSSS-S_4`, `NSSS-S_5`, `NSSS-S_6`, `NSSS-S_7`, `NSSS-S_8`, `NSSS-S_9`, `NSSS-S_10`, `NSSS-S_11`, `NSSS-S_12`), na.rm = TRUE))%>%
ungroup()
############################ Brief PRI-8###########################################
# There was issue with coding in qualtrics where participants saw 0-5 scale but
#coding was as follows (what participants should of seen, what they actually saw):
# 0=1, 1=2, 2=3, 3=4, 4=5, 5=7, need to fix
#Recoding values
data <- data %>%
mutate(
PRI_1 = dplyr::recode(`Brief PRI-8_1`, "1"=0, "2"=1, "3"=2, "4"=3, "5"=4, "7"=5))
data <- data %>%
mutate(
PRI_2 = dplyr::recode(`Brief PRI-8_2`, "1"=0, "2"=1, "3"=2, "4"=3, "5"=4, "7"=5))
data <- data %>%
mutate(
PRI_3 = dplyr::recode(`Brief PRI-8_3`, "1"=0, "2"=1, "3"=2, "4"=3, "5"=4, "7"=5))
data <- data %>%
mutate(
PRI_4 = dplyr::recode(`Brief PRI-8_4`, "1"=0, "2"=1, "3"=2, "4"=3, "5"=4, "7"=5))
data <- data %>%
mutate(
PRI_5 = dplyr::recode(`Brief PRI-8_5`, "1"=0, "2"=1, "3"=2, "4"=3, "5"=4, "7"=5))
data <- data %>%
mutate(
PRI_6 = dplyr::recode(`Brief PRI-8_6`, "1"=0, "2"=1, "3"=2, "4"=3, "5"=4, "7"=5))
data <- data %>%
mutate(
PRI_7 = dplyr::recode(`Brief PRI-8_7`, "1"=0, "2"=1, "3"=2, "4"=3, "5"=4, "7"=5))
data <- data %>%
mutate(
PRI_8 = dplyr::recode(`Brief PRI-8_8`, "1"=0, "2"=1, "3"=2, "4"=3, "5"=4, "7"=5))
# Partner responsiveness
data <- data %>%
rowwise() %>%
mutate(PRI_R = mean(c(PRI_1, PRI_2, PRI_3, PRI_4), na.rm = TRUE))%>%
ungroup()
# Partner Insensitivity
data <- data %>%
rowwise() %>%
mutate(PRI_I = mean(c(PRI_5, PRI_6, PRI_7, PRI_8), na.rm = TRUE))%>%
ungroup()
# Recoding insensitivity items
data <- data %>%
dplyr::mutate(PRI_5.r = (5 - PRI_5))
data <- data %>%
dplyr::mutate(PRI_6.r = (5 - PRI_6))
data <- data %>%
dplyr::mutate(PRI_7.r = (5 - PRI_7))
data <- data %>%
dplyr::mutate(PRI_8.r = (5 - PRI_8))
data <- data %>%
rowwise() %>%
mutate(PRI_total = mean(c(PRI_1, PRI_2, PRI_3, PRI_4, PRI_5.r, PRI_6.r, PRI_7.r, PRI_8.r), na.rm = TRUE))%>%
ungroup()Scale reliabilities
#PRQC
data %>%
select(PRQC_1:PRQC_18) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.96 0.97 0.99 0.63 31 0.0022 5.8 1.1 0.64
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.96 0.96 0.97
## Duhachek 0.96 0.96 0.97
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## PRQC_1 0.96 0.97 0.98 0.62 28 0.0024 0.018 0.63
## PRQC_2 0.96 0.97 0.98 0.63 28 0.0024 0.018 0.62
## PRQC_3 0.96 0.97 0.98 0.62 28 0.0024 0.018 0.62
## PRQC_4 0.96 0.97 0.98 0.64 30 0.0023 0.017 0.65
## PRQC_5 0.96 0.97 0.98 0.63 29 0.0023 0.017 0.65
## PRQC_6 0.96 0.97 0.98 0.63 29 0.0023 0.017 0.65
## PRQC_7 0.96 0.97 0.99 0.64 30 0.0023 0.018 0.65
## PRQC_8 0.96 0.97 0.98 0.63 29 0.0024 0.019 0.63
## PRQC_9 0.96 0.97 0.98 0.63 28 0.0024 0.019 0.62
## PRQC_10 0.96 0.97 0.99 0.64 30 0.0023 0.018 0.65
## PRQC_11 0.96 0.97 0.98 0.63 29 0.0023 0.018 0.65
## PRQC_12 0.96 0.97 0.98 0.64 30 0.0023 0.018 0.65
## PRQC_13 0.96 0.97 0.98 0.63 29 0.0023 0.018 0.65
## PRQC_14 0.96 0.97 0.98 0.65 31 0.0022 0.015 0.66
## PRQC_15 0.97 0.97 0.98 0.65 31 0.0021 0.014 0.66
## PRQC_16 0.96 0.97 0.99 0.64 30 0.0023 0.018 0.65
## PRQC_17 0.96 0.97 0.98 0.63 29 0.0023 0.019 0.64
## PRQC_18 0.96 0.97 0.98 0.63 29 0.0023 0.018 0.63
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## PRQC_1 566 0.89 0.89 0.89 0.87 5.7 1.34
## PRQC_2 566 0.88 0.88 0.88 0.86 5.8 1.35
## PRQC_3 566 0.89 0.90 0.90 0.88 5.7 1.35
## PRQC_4 566 0.73 0.76 0.76 0.71 6.5 0.99
## PRQC_5 566 0.77 0.80 0.80 0.75 6.4 1.03
## PRQC_6 566 0.77 0.80 0.80 0.75 6.4 1.07
## PRQC_7 566 0.81 0.79 0.77 0.78 5.3 1.57
## PRQC_8 566 0.86 0.86 0.85 0.84 6.0 1.33
## PRQC_9 566 0.88 0.88 0.87 0.87 5.9 1.38
## PRQC_10 566 0.77 0.77 0.76 0.74 5.9 1.52
## PRQC_11 566 0.80 0.81 0.81 0.77 6.0 1.39
## PRQC_12 566 0.78 0.79 0.78 0.75 5.9 1.41
## PRQC_13 566 0.83 0.80 0.79 0.80 4.9 1.76
## PRQC_14 566 0.73 0.69 0.69 0.68 4.4 1.90
## PRQC_15 566 0.71 0.67 0.66 0.66 4.3 1.92
## PRQC_16 566 0.76 0.79 0.78 0.74 6.5 0.98
## PRQC_17 566 0.83 0.84 0.83 0.80 6.0 1.41
## PRQC_18 566 0.81 0.83 0.83 0.79 6.2 1.29
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## PRQC_1 0.01 0.02 0.05 0.09 0.19 0.33 0.32 0
## PRQC_2 0.01 0.02 0.04 0.08 0.15 0.33 0.37 0
## PRQC_3 0.01 0.03 0.04 0.10 0.16 0.30 0.37 0
## PRQC_4 0.00 0.00 0.01 0.04 0.07 0.15 0.71 0
## PRQC_5 0.01 0.00 0.02 0.04 0.08 0.18 0.67 0
## PRQC_6 0.01 0.00 0.02 0.05 0.08 0.17 0.67 0
## PRQC_7 0.02 0.05 0.07 0.12 0.20 0.25 0.29 0
## PRQC_8 0.01 0.02 0.04 0.06 0.14 0.25 0.47 0
## PRQC_9 0.01 0.03 0.04 0.08 0.14 0.26 0.43 0
## PRQC_10 0.03 0.03 0.02 0.07 0.11 0.24 0.50 0
## PRQC_11 0.01 0.03 0.04 0.05 0.13 0.24 0.50 0
## PRQC_12 0.01 0.03 0.04 0.05 0.12 0.28 0.46 0
## PRQC_13 0.05 0.07 0.10 0.11 0.22 0.22 0.22 0
## PRQC_14 0.10 0.10 0.12 0.15 0.20 0.16 0.17 0
## PRQC_15 0.10 0.11 0.12 0.14 0.20 0.15 0.16 0
## PRQC_16 0.00 0.01 0.01 0.04 0.07 0.18 0.69 0
## PRQC_17 0.02 0.02 0.03 0.07 0.11 0.19 0.56 0
## PRQC_18 0.01 0.02 0.02 0.05 0.10 0.16 0.63 0# Trust
data %>%
select(PRQC_10:PRQC_12) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.93 0.94 0.92 0.83 15 0.0051 5.9 1.4 0.81
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.92 0.93 0.94
## Duhachek 0.92 0.93 0.94
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## PRQC_10 0.96 0.96 0.92 0.92 23.5 0.0034 NA 0.92
## PRQC_11 0.86 0.86 0.76 0.76 6.3 0.0116 NA 0.76
## PRQC_12 0.89 0.89 0.81 0.81 8.3 0.0091 NA 0.81
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## PRQC_10 566 0.91 0.91 0.82 0.80 5.9 1.5
## PRQC_11 566 0.96 0.97 0.96 0.92 6.0 1.4
## PRQC_12 566 0.95 0.95 0.93 0.88 5.9 1.4
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## PRQC_10 0.03 0.03 0.02 0.07 0.11 0.24 0.50 0
## PRQC_11 0.01 0.03 0.04 0.05 0.13 0.24 0.50 0
## PRQC_12 0.01 0.03 0.04 0.05 0.12 0.28 0.46 0# Attachment Avoidance
data %>%
select(AAQ_1.r, AAQ_2, AAQ_3.r, AAQ_5:AAQ_9) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.88 0.88 0.89 0.48 7.5 0.0073 3.8 1.3 0.48
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.87 0.88 0.9
## Duhachek 0.87 0.88 0.9
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## AAQ_1.r 0.87 0.87 0.88 0.50 6.9 0.0079 0.035 0.49
## AAQ_2 0.88 0.88 0.89 0.51 7.3 0.0077 0.030 0.52
## AAQ_3.r 0.90 0.90 0.90 0.55 8.5 0.0067 0.017 0.52
## AAQ_5 0.86 0.86 0.86 0.46 6.0 0.0089 0.025 0.48
## AAQ_6 0.85 0.85 0.85 0.45 5.6 0.0095 0.022 0.45
## AAQ_7 0.86 0.86 0.87 0.47 6.2 0.0088 0.031 0.45
## AAQ_8 0.85 0.85 0.85 0.45 5.7 0.0095 0.022 0.45
## AAQ_9 0.87 0.87 0.88 0.49 6.7 0.0081 0.031 0.48
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## AAQ_1.r 566 0.70 0.70 0.63 0.59 4.0 1.8
## AAQ_2 566 0.65 0.65 0.57 0.54 4.8 1.8
## AAQ_3.r 566 0.50 0.51 0.40 0.37 3.4 1.7
## AAQ_5 566 0.82 0.82 0.81 0.75 3.4 1.7
## AAQ_6 566 0.87 0.87 0.89 0.82 3.5 1.8
## AAQ_7 566 0.80 0.79 0.76 0.71 4.6 1.9
## AAQ_8 566 0.87 0.87 0.87 0.81 3.7 1.9
## AAQ_9 566 0.72 0.72 0.66 0.62 3.4 1.8
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## AAQ_1.r 0.10 0.14 0.19 0.18 0.15 0.14 0.11 0
## AAQ_2 0.07 0.07 0.10 0.13 0.21 0.23 0.19 0
## AAQ_3.r 0.16 0.19 0.22 0.18 0.11 0.10 0.06 0
## AAQ_5 0.16 0.20 0.17 0.18 0.15 0.08 0.05 0
## AAQ_6 0.16 0.21 0.14 0.16 0.19 0.08 0.06 0
## AAQ_7 0.09 0.09 0.11 0.13 0.20 0.21 0.17 0
## AAQ_8 0.16 0.18 0.13 0.15 0.18 0.11 0.08 0
## AAQ_9 0.20 0.18 0.18 0.16 0.14 0.08 0.07 0# Attachment Anxiety
data %>%
select(AAQ_4.r, AAQ_10, AAQ_11, AAQ_12.r, AAQ_13, AAQ_14.r, AAQ_15, AAQ_16.r, AAQ_17.r) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.83 0.82 0.85 0.34 4.6 0.011 3.4 1.2 0.36
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.8 0.83 0.85
## Duhachek 0.8 0.83 0.85
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## AAQ_4.r 0.80 0.80 0.83 0.34 4.0 0.012 0.031 0.35
## AAQ_10 0.82 0.82 0.84 0.36 4.6 0.011 0.029 0.39
## AAQ_11 0.78 0.78 0.81 0.31 3.6 0.014 0.029 0.24
## AAQ_12.r 0.81 0.81 0.84 0.35 4.2 0.012 0.031 0.34
## AAQ_13 0.82 0.82 0.83 0.36 4.4 0.011 0.029 0.39
## AAQ_14.r 0.80 0.80 0.83 0.33 4.0 0.012 0.026 0.34
## AAQ_15 0.82 0.82 0.83 0.36 4.4 0.011 0.029 0.39
## AAQ_16.r 0.79 0.79 0.81 0.31 3.7 0.013 0.025 0.34
## AAQ_17.r 0.80 0.80 0.82 0.33 3.9 0.013 0.027 0.35
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## AAQ_4.r 566 0.67 0.66 0.60 0.55 4.3 2.0
## AAQ_10 566 0.48 0.52 0.43 0.36 2.7 1.5
## AAQ_11 566 0.79 0.79 0.78 0.71 3.1 2.0
## AAQ_12.r 566 0.63 0.60 0.52 0.48 3.6 2.1
## AAQ_13 566 0.52 0.55 0.49 0.40 2.4 1.6
## AAQ_14.r 566 0.68 0.66 0.61 0.56 4.4 1.9
## AAQ_15 566 0.52 0.55 0.49 0.39 3.1 1.7
## AAQ_16.r 566 0.78 0.76 0.74 0.68 4.3 2.0
## AAQ_17.r 566 0.70 0.69 0.66 0.60 2.7 1.8
##
## Non missing response frequency for each item
## 1 2 3 4 5 6 7 miss
## AAQ_4.r 0.12 0.11 0.12 0.13 0.18 0.17 0.16 0
## AAQ_10 0.25 0.27 0.22 0.13 0.07 0.04 0.02 0
## AAQ_11 0.31 0.19 0.12 0.11 0.10 0.08 0.09 0
## AAQ_12.r 0.21 0.18 0.15 0.08 0.12 0.11 0.14 0
## AAQ_13 0.40 0.22 0.13 0.13 0.07 0.03 0.02 0
## AAQ_14.r 0.08 0.13 0.11 0.18 0.14 0.16 0.20 0
## AAQ_15 0.21 0.22 0.20 0.14 0.13 0.07 0.04 0
## AAQ_16.r 0.11 0.13 0.11 0.12 0.17 0.18 0.17 0
## AAQ_17.r 0.34 0.21 0.15 0.12 0.08 0.06 0.05 0# NSSS
data %>%
select(`NSSS-S_1`:`NSSS-S_12`) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.96 0.96 0.96 0.64 21 0.0028 3.4 1 0.64
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.95 0.96 0.96
## Duhachek 0.95 0.96 0.96
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## NSSS-S_1 0.95 0.95 0.96 0.64 19 0.0031 0.0074 0.63
## NSSS-S_2 0.95 0.95 0.96 0.64 19 0.0031 0.0067 0.64
## NSSS-S_3 0.95 0.95 0.96 0.63 19 0.0032 0.0067 0.63
## NSSS-S_4 0.95 0.95 0.96 0.65 21 0.0029 0.0053 0.65
## NSSS-S_5 0.95 0.95 0.96 0.64 20 0.0030 0.0077 0.64
## NSSS-S_6 0.95 0.95 0.96 0.64 20 0.0030 0.0076 0.65
## NSSS-S_7 0.95 0.95 0.96 0.63 19 0.0031 0.0076 0.63
## NSSS-S_8 0.95 0.95 0.96 0.64 20 0.0031 0.0076 0.65
## NSSS-S_9 0.95 0.95 0.96 0.65 21 0.0029 0.0058 0.65
## NSSS-S_10 0.95 0.95 0.96 0.64 19 0.0031 0.0064 0.64
## NSSS-S_11 0.95 0.95 0.96 0.63 19 0.0031 0.0064 0.63
## NSSS-S_12 0.95 0.95 0.96 0.64 20 0.0030 0.0073 0.65
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## NSSS-S_1 566 0.83 0.83 0.81 0.79 3.4 1.3
## NSSS-S_2 566 0.85 0.85 0.84 0.81 3.1 1.3
## NSSS-S_3 566 0.88 0.88 0.87 0.85 3.3 1.3
## NSSS-S_4 566 0.74 0.74 0.72 0.69 3.2 1.3
## NSSS-S_5 566 0.81 0.82 0.79 0.78 3.6 1.2
## NSSS-S_6 566 0.81 0.81 0.80 0.77 3.7 1.2
## NSSS-S_7 566 0.85 0.85 0.84 0.82 3.4 1.3
## NSSS-S_8 566 0.82 0.82 0.80 0.78 3.4 1.3
## NSSS-S_9 566 0.74 0.75 0.72 0.69 4.0 1.2
## NSSS-S_10 566 0.85 0.85 0.84 0.81 3.3 1.3
## NSSS-S_11 566 0.85 0.85 0.85 0.82 3.2 1.3
## NSSS-S_12 566 0.80 0.79 0.77 0.75 3.0 1.4
##
## Non missing response frequency for each item
## 1 2 3 4 5 miss
## NSSS-S_1 0.11 0.12 0.23 0.31 0.22 0
## NSSS-S_2 0.17 0.16 0.26 0.23 0.19 0
## NSSS-S_3 0.12 0.14 0.25 0.26 0.23 0
## NSSS-S_4 0.14 0.15 0.26 0.25 0.19 0
## NSSS-S_5 0.07 0.09 0.23 0.34 0.27 0
## NSSS-S_6 0.07 0.07 0.21 0.35 0.29 0
## NSSS-S_7 0.13 0.11 0.26 0.25 0.25 0
## NSSS-S_8 0.11 0.11 0.27 0.27 0.23 0
## NSSS-S_9 0.07 0.04 0.15 0.32 0.41 0
## NSSS-S_10 0.13 0.15 0.25 0.25 0.22 0
## NSSS-S_11 0.15 0.15 0.26 0.25 0.19 0
## NSSS-S_12 0.18 0.18 0.23 0.24 0.17 0#PRI-R
data %>%
select(PRI_1:PRI_4) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.95 0.95 0.94 0.83 19 0.0035 3.9 1.2 0.82
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.94 0.95 0.96
## Duhachek 0.94 0.95 0.96
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## PRI_1 0.94 0.94 0.91 0.83 15 0.0047 0.00031 0.82
## PRI_2 0.93 0.93 0.90 0.81 13 0.0052 0.00104 0.80
## PRI_3 0.93 0.94 0.91 0.83 14 0.0048 0.00152 0.82
## PRI_4 0.94 0.94 0.91 0.83 15 0.0047 0.00140 0.82
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## PRI_1 566 0.93 0.93 0.90 0.87 3.8 1.2
## PRI_2 566 0.94 0.94 0.92 0.89 3.9 1.3
## PRI_3 566 0.93 0.93 0.90 0.88 4.0 1.2
## PRI_4 566 0.93 0.93 0.90 0.87 3.9 1.3
##
## Non missing response frequency for each item
## 0 1 2 3 4 5 miss
## PRI_1 0.02 0.05 0.07 0.20 0.33 0.33 0
## PRI_2 0.03 0.04 0.07 0.16 0.27 0.43 0
## PRI_3 0.02 0.04 0.05 0.13 0.29 0.46 0
## PRI_4 0.02 0.04 0.07 0.16 0.26 0.44 0#PRI-I
data %>%
select(PRI_5.r: PRI_8.r) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.96 0.96 0.95 0.85 22 0.003 4 1.3 0.85
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.95 0.96 0.96
## Duhachek 0.95 0.96 0.96
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## PRI_5.r 0.95 0.95 0.93 0.86 18 0.0038 0.00033 0.87
## PRI_6.r 0.94 0.94 0.91 0.84 16 0.0044 0.00070 0.84
## PRI_7.r 0.94 0.94 0.91 0.84 16 0.0045 0.00063 0.84
## PRI_8.r 0.95 0.95 0.92 0.86 18 0.0039 0.00041 0.86
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## PRI_5.r 566 0.93 0.93 0.90 0.88 4.1 1.4
## PRI_6.r 566 0.95 0.95 0.93 0.91 4.1 1.3
## PRI_7.r 566 0.95 0.95 0.93 0.91 3.9 1.5
## PRI_8.r 566 0.94 0.93 0.90 0.88 3.9 1.5
##
## Non missing response frequency for each item
## 0 1 2 3 4 5 miss
## PRI_5.r 0.03 0.06 0.06 0.07 0.18 0.60 0
## PRI_6.r 0.03 0.04 0.07 0.10 0.17 0.59 0
## PRI_7.r 0.04 0.06 0.07 0.12 0.19 0.52 0
## PRI_8.r 0.04 0.06 0.09 0.10 0.16 0.56 0#PRI Total
data %>%
select(PRI_1:PRI_4,PRI_5.r: PRI_8.r) %>%
psych::alpha( ,check.keys = F, na.rm = TRUE)##
## Reliability analysis
## Call: psych::alpha(x = ., na.rm = TRUE, check.keys = F)
##
## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r
## 0.96 0.96 0.97 0.75 24 0.0027 4 1.2 0.71
##
## 95% confidence boundaries
## lower alpha upper
## Feldt 0.95 0.96 0.96
## Duhachek 0.95 0.96 0.96
##
## Reliability if an item is dropped:
## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## PRI_1 0.96 0.96 0.96 0.76 22 0.0029 0.0063 0.71
## PRI_2 0.95 0.95 0.96 0.75 21 0.0030 0.0069 0.71
## PRI_3 0.95 0.95 0.96 0.75 20 0.0031 0.0078 0.70
## PRI_4 0.95 0.95 0.96 0.75 21 0.0030 0.0073 0.71
## PRI_5.r 0.95 0.96 0.96 0.75 21 0.0031 0.0063 0.71
## PRI_6.r 0.95 0.95 0.96 0.74 20 0.0032 0.0064 0.70
## PRI_7.r 0.95 0.95 0.96 0.74 20 0.0032 0.0066 0.70
## PRI_8.r 0.95 0.95 0.96 0.75 21 0.0031 0.0069 0.71
##
## Item statistics
## n raw.r std.r r.cor r.drop mean sd
## PRI_1 566 0.85 0.86 0.84 0.81 3.8 1.2
## PRI_2 566 0.88 0.88 0.87 0.84 3.9 1.3
## PRI_3 566 0.89 0.89 0.88 0.85 4.0 1.2
## PRI_4 566 0.87 0.88 0.86 0.83 3.9 1.3
## PRI_5.r 566 0.88 0.87 0.86 0.84 4.1 1.4
## PRI_6.r 566 0.90 0.90 0.89 0.87 4.1 1.3
## PRI_7.r 566 0.91 0.90 0.89 0.87 3.9 1.5
## PRI_8.r 566 0.89 0.88 0.87 0.85 3.9 1.5
##
## Non missing response frequency for each item
## 0 1 2 3 4 5 miss
## PRI_1 0.02 0.05 0.07 0.20 0.33 0.33 0
## PRI_2 0.03 0.04 0.07 0.16 0.27 0.43 0
## PRI_3 0.02 0.04 0.05 0.13 0.29 0.46 0
## PRI_4 0.02 0.04 0.07 0.16 0.26 0.44 0
## PRI_5.r 0.03 0.06 0.06 0.07 0.18 0.60 0
## PRI_6.r 0.03 0.04 0.07 0.10 0.17 0.59 0
## PRI_7.r 0.04 0.06 0.07 0.12 0.19 0.52 0
## PRI_8.r 0.04 0.06 0.09 0.10 0.16 0.56 0Histograms and QQ Plots
par(mfrow=c(1, 2))
#PRQC
hist(data$PRQC, main="Histogram of PRQC", xlab="PRQC", col='hotpink', sub=paste("Skewness:",
round(e1071::skewness(data$PRQC, na.rm=TRUE), 2)))
qqnorm(data$PRQC, pch = 1, frame = FALSE, main="QQ Plot of PRQC")
qqline(data$PRQC, col = "hotpink", lwd = 2)
#PRQC Trust
hist(data$Trust, main="Histogram of PRQC Trust", xlab="Trust",
col='hotpink', sub=paste("Skewness:",
round(e1071::skewness(data$Trust, na.rm=TRUE), 2)))
qqnorm(data$Trust, pch = 1, frame = FALSE, main="QQ Plot of PRQC Trust")
qqline(data$Trust, col = "hotpink", lwd = 2)
# Attachment Avoidance
hist(data$AAQ.AV, main="Histogram of Attachment Avoidance", xlab="Avoidance",
col='hotpink', sub=paste("Skewness:",
round(e1071::skewness(data$AAQ.AV, na.rm=TRUE), 2)))
qqnorm(data$AAQ.AV, pch = 1, frame = FALSE, main="QQ Plot of Attachment Avoidance")
qqline(data$AAQ.AV, col = "hotpink", lwd = 2)
# Attachment Anxiety
hist(data$AAQ.ANX, main="Histogram of Attachment Anxiety", xlab="Anxiety",
col='hotpink', sub=paste("Skewness:",
round(e1071::skewness(data$AAQ.ANX, na.rm=TRUE), 2)))
qqnorm(data$AAQ.ANX, pch = 1, frame = FALSE, main="QQ Plot of Attachment Anxiety")
qqline(data$AAQ.ANX, col = "hotpink", lwd = 2)
# NSSS
hist(data$NSSS, main="Histogram of Sexual Satisfaction", xlab="NSSS", col='hotpink',
sub=paste("Skewness:",
round(e1071::skewness(data$NSSS, na.rm=TRUE), 2)))
qqnorm(data$NSSS, pch = 1, frame = FALSE, main="QQ Plot of Sexual Satisfaction")
qqline(data$NSSS, col = "hotpink", lwd = 2)
# Perceived Partner Responsiveness
hist(data$PRI_R, main="Histogram of Responsiveness", xlab="Responsiveness",
col='hotpink', sub=paste("Skewness:",
round(e1071::skewness(data$PRI_R, na.rm=TRUE), 2)))
qqnorm(data$PRI_R, pch = 1, frame = FALSE, main="QQ Plot of Partner Responsiveness")
qqline(data$PRI_R, col = "hotpink", lwd = 2)
# Perceived Partner Insensitivity
hist(data$PRI_I, main="Histogram of Insensitivity", xlab="Insensitivity", col='hotpink',
sub=paste("Skewness:",
round(e1071::skewness(data$PRI_I, na.rm=TRUE), 2)))
qqnorm(data$PRI_I, pch = 1, frame = FALSE, main="QQ Plot of Partner Insensitivity")
qqline(data$PRI_I, col = "hotpink", lwd = 2)
Descriptive Statistics
data %>%
dplyr::select(AAQ.ANX, AAQ.AV, PRQC, Trust, NSSS, PRI_total)%>%
psych::describe(na.rm=TRUE) %>%
as.data.frame() %>%
dplyr::select("n", "mean", "sd", "median", "min", "max", "range", "skew", "kurtosis") %>%
kableExtra::kable(caption= "Descriptive Statistics for Study Variables", digits = 2) %>%
kable_styling(bootstrap_options = "striped", full_width = TRUE)| n | mean | sd | median | min | max | range | skew | kurtosis | |
|---|---|---|---|---|---|---|---|---|---|
| AAQ.ANX | 566 | 3.42 | 1.20 | 3.44 | 1.00 | 7 | 6.00 | 0.12 | -0.41 |
| AAQ.AV | 566 | 3.84 | 1.34 | 3.88 | 1.00 | 7 | 6.00 | -0.15 | -0.56 |
| PRQC | 566 | 5.77 | 1.12 | 6.06 | 1.61 | 7 | 5.39 | -1.29 | 1.32 |
| Trust | 566 | 5.94 | 1.35 | 6.33 | 1.00 | 7 | 6.00 | -1.64 | 2.29 |
| NSSS | 566 | 3.39 | 1.04 | 3.58 | 1.00 | 5 | 4.00 | -0.44 | -0.49 |
| PRI_total | 566 | 3.96 | 1.18 | 4.44 | 0.00 | 5 | 5.00 | -1.30 | 0.98 |
sum(data$AAQ.ANX>3.5)/length(data$AAQ.ANX) #percentage of people above midpoints## [1] 0.4734982sum(data$AAQ.AV>3.5)/length(data$AAQ.AV)## [1] 0.5812721sum(data$PRQC>3.5)/length(data$PRQC)## [1] 0.9381625sum(data$Trust>3.5)/length(data$Trust)## [1] 0.9204947sum(data$NSSS>2.5)/length(data$NSSS)## [1] 0.7932862sum(data$PRI_total>2.5)/length(data$PRI_total)## [1] 0.8586572Correlations
which(colnames(data)=="someASA") # identifies the column number of different variables## [1] 159data[,c(49:103)] %<>% dplyr::mutate_if(is.character,as.numeric) #making only select columns numeric
corrdata<- data %>%
select(someASA, PRQC, Trust, NSSS, PRI_total, AAQ.ANX, AAQ.AV)
cor<- cor(corrdata, use= "everything", method=c("spearman"))
round(cor,2)## someASA PRQC Trust NSSS PRI_total AAQ.ANX AAQ.AV
## someASA 1.00 -0.13 -0.09 -0.03 -0.16 0.09 0.14
## PRQC -0.13 1.00 0.78 0.68 0.72 -0.33 -0.28
## Trust -0.09 0.78 1.00 0.44 0.69 -0.40 -0.28
## NSSS -0.03 0.68 0.44 1.00 0.46 -0.29 -0.28
## PRI_total -0.16 0.72 0.69 0.46 1.00 -0.36 -0.25
## AAQ.ANX 0.09 -0.33 -0.40 -0.29 -0.36 1.00 0.36
## AAQ.AV 0.14 -0.28 -0.28 -0.28 -0.25 0.36 1.00rcorr(as.matrix(corrdata), type = c("spearman"))## someASA PRQC Trust NSSS PRI_total AAQ.ANX AAQ.AV
## someASA 1.00 -0.13 -0.09 -0.03 -0.16 0.09 0.14
## PRQC -0.13 1.00 0.78 0.68 0.72 -0.33 -0.28
## Trust -0.09 0.78 1.00 0.44 0.69 -0.40 -0.28
## NSSS -0.03 0.68 0.44 1.00 0.46 -0.29 -0.28
## PRI_total -0.16 0.72 0.69 0.46 1.00 -0.36 -0.25
## AAQ.ANX 0.09 -0.33 -0.40 -0.29 -0.36 1.00 0.36
## AAQ.AV 0.14 -0.28 -0.28 -0.28 -0.25 0.36 1.00
##
## n= 566
##
##
## P
## someASA PRQC Trust NSSS PRI_total AAQ.ANX AAQ.AV
## someASA 0.0019 0.0430 0.4835 0.0000 0.0307 0.0012
## PRQC 0.0019 0.0000 0.0000 0.0000 0.0000 0.0000
## Trust 0.0430 0.0000 0.0000 0.0000 0.0000 0.0000
## NSSS 0.4835 0.0000 0.0000 0.0000 0.0000 0.0000
## PRI_total 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
## AAQ.ANX 0.0307 0.0000 0.0000 0.0000 0.0000 0.0000
## AAQ.AV 0.0012 0.0000 0.0000 0.0000 0.0000 0.0000mean(corrdata$someASA)## [1] 0.6201413sd(corrdata$someASA)## [1] 0.4857808Mean Centering
data$cPRI_total<- data$PRI_total - mean(data$PRI_total, na.rm=TRUE)
data$cANX<- data$AAQ.ANX - mean(data$AAQ.ANX, na.rm=TRUE)
data$cAV<- data$AAQ.AV - mean(data$AAQ.AV, na.rm=TRUE)Detecting Outliers
## PRQC
PRQCmin <- mean(data$PRQC, na.rm=T) - (3*(sd(data$PRQC, na.rm=T)))
PRQCmax <- mean(data$PRQC, na.rm=T) + (3*(sd(data$PRQC, na.rm=T)))
data$PRQC[which(data$PRQC < PRQCmin | data$PRQC > PRQCmax)] # 8 outliers## [1] 2.111111 2.277778 2.166667 2.277778 2.277778 1.944444 1.611111 1.777778# Sexual Satisfaction
NSSSmin <- mean(data$NSSS, na.rm=T) - (3*(sd(data$NSSS, na.rm=T)))
NSSSmax <- mean(data$NSSS, na.rm=T) + (3*(sd(data$NSSS, na.rm=T)))
data$NSSS[which(data$NSSS< NSSSmin | data$NSSS > NSSSmax)] # 0 outliers## numeric(0)#Trust
Trustmin <- mean(data$Trust, na.rm=T) - (3*(sd(data$Trust, na.rm=T)))
Trustmax <- mean(data$Trust, na.rm=T) + (3*(sd(data$Trust, na.rm=T)))
data$Trust[which(data$Trust< Trustmin | data$Trust > Trustmax)] # 11 outliers## [1] 1.333333 1.666667 1.000000 1.666667 1.666667 1.000000 1.666667 1.666667
## [9] 1.000000 1.333333 1.000000# Perceived partner responsiveness
cPRI_totalmin <- mean(data$cPRI_total, na.rm=T) - (3*(sd(data$cPRI_total, na.rm=T)))
cPRI_totalmax <- mean(data$cPRI_total, na.rm=T) + (3*(sd(data$cPRI_total, na.rm=T)))
data$cPRI_total[which(data$cPRI_total< cPRI_totalmin | data$cPRI_total > cPRI_totalmax)] # 7 outliers## [1] -3.958922 -3.958922 -3.958922 -3.958922 -3.708922 -3.708922 -3.958922# Anxiety
cANXmin <- mean(data$cANX, na.rm=T) - (3*(sd(data$cANX, na.rm=T)))
cANXmax <- mean(data$cANX, na.rm=T) + (3*(sd(data$cANX, na.rm=T)))
data$cANX[which(data$cANX< cANXmin | data$cANX > cANXmax)] #0 outliers## numeric(0)# Avoidance
cAVmin <- mean(data$cAV, na.rm=T) - (3*(sd(data$cAV, na.rm=T)))
cAVmax <- mean(data$cAV, na.rm=T) + (3*(sd(data$cAV, na.rm=T)))
data$cAV[which(data$cAV< cAVmin | data$cAV > cAVmax)] # 0 outliers## numeric(0)Hypothesis 1 & 2
Hypothesis 1 and 2: Having experienced adult sexual assault will be associated with poorer (a1) romantic relationship quality, (a2) partner trust, and (b) poorer sexual satisfaction in participants’ current romantic relationship. The association between experiencing adult sexual assault and (a1) romantic relationship quality, (a2) partner trust, and (b) sexual satisfaction will be moderated by perceived partner responsiveness, such that individuals who have experienced ASA and perceive their partners as being less responsive will report worse relationship quality and lower partner trust and sexual satisfaction than individuals who have experienced ASA and perceive their partners as more responsive.
PRQC
logPRQC<- log(data$PRQC) #log transform did not improve distribution
prqc2<- (data$PRQC)^2 #exponential transform did not improve distribution
# Effect coding
data<- data %>%
mutate(AnyASA = case_when(
data$ASAonly==TRUE ~ 1,
data$revictimized==TRUE ~ 1,
data$NeitherASAnorCSA==TRUE ~ -1
))
data$AnyASA<- as.factor(data$AnyASA) # converting to factor
data$InversePRQC<- 8-data$PRQC ### Reverse scoring PRQC so it fits gamma distribution
# descriptives
group_by(data, AnyASA) %>%
summarise(
count = n(),
mean = mean(PRQC, na.rm = TRUE),
sd = sd(PRQC, na.rm = TRUE),
median = median(PRQC, na.rm = TRUE),
IQR = IQR(PRQC, na.rm = TRUE)
)## # A tibble: 2 × 6
## AnyASA count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 5.93 1.08 6.28 1.28
## 2 1 351 5.68 1.13 6 1.39Prqc.step1<- glm(formula= InversePRQC ~ AnyASA, family=Gamma(link="log"),
na.action=na.exclude, data=data)
prqc.step2<- glm(formula= InversePRQC ~ AnyASA+cPRI_total, family=Gamma(link="log"),
na.action=na.exclude, data=data)
prqc.step3<- glm(formula= InversePRQC ~ AnyASA*cPRI_total, family=Gamma(link="log"),
na.action=na.exclude, data=data)
data$PRQC.win<- winsorize(data$PRQC, method="zscore",
threshold=3, robust=FALSE) #removing outliers +-3 sds from mean
data$cPRI.win<- winsorize(data$cPRI_total, method="zscore",
threshold=3, robust=FALSE)
PRQC.gamma.noOutliers<- glm(formula= InversePRQC ~ AnyASA*cPRI.win, family=Gamma(link="log"),
na.action=na.exclude, data=data)
#Trying linear model
prqc.lin.mod1<- lm(PRQC.win ~ AnyASA, data=data)
prqc.lin.mod2<- lm(PRQC.win ~ AnyASA+cPRI.win, data=data)
prqc.lin.mod3<- lm(PRQC.win ~ AnyASA*cPRI.win, data=data)
# Non-parametric t-test
wilcox.test(PRQC ~ AnyASA, data=data, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: PRQC by AnyASA
## W = 43574, p-value = 0.001967
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## 0.05556703 0.38882948
## sample estimates:
## difference in location
## 0.2221814wilcoxonRG(x = data$PRQC, g = data$AnyASA)## rg
## 0.155wilcox_effsize(PRQC ~ AnyASA, data=data)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 PRQC -1 1 0.130 215 351 smallwilcox_test(PRQC ~ AnyASA, data=data, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: PRQC by AnyASA (-1, 1)
## Z = 3.0954, p-value = 0.001966
## alternative hypothesis: true mu is not equal to 0#### Results
summary(prqc.lin.mod1)##
## Call:
## lm(formula = PRQC.win ~ AnyASA, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.7785 -0.5775 0.2130 0.7130 1.2130
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.02194 0.05985 100.621 < 2e-16 ***
## AnyASA1 -0.23492 0.07600 -3.091 0.00209 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8775 on 564 degrees of freedom
## Multiple R-squared: 0.01666, Adjusted R-squared: 0.01492
## F-statistic: 9.555 on 1 and 564 DF, p-value: 0.002093lm.beta(prqc.lin.mod1) # standardized beta##
## Call:
## lm(formula = PRQC.win ~ AnyASA, data = data)
##
## Standardized Coefficients::
## (Intercept) AnyASA1
## NA -0.1290706summary(prqc.lin.mod2)##
## Call:
## lm(formula = PRQC.win ~ AnyASA + cPRI.win, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.21204 -0.44572 0.06546 0.41477 2.15758
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.76254 0.04337 132.863 <2e-16 ***
## AnyASA1 -0.02041 0.05409 -0.377 0.706
## cPRI.win 0.75647 0.03138 24.110 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6161 on 563 degrees of freedom
## Multiple R-squared: 0.5162, Adjusted R-squared: 0.5145
## F-statistic: 300.3 on 2 and 563 DF, p-value: < 2.2e-16lm.beta(prqc.lin.mod2) #stan beta##
## Call:
## lm(formula = PRQC.win ~ AnyASA + cPRI.win, data = data)
##
## Standardized Coefficients::
## (Intercept) AnyASA1 cPRI.win
## NA -0.01121495 0.71653193summary(prqc.lin.mod3)##
## Call:
## lm(formula = PRQC.win ~ AnyASA * cPRI.win, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.21003 -0.44503 0.06508 0.41221 2.15972
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.763742 0.046064 125.124 <2e-16 ***
## AnyASA1 -0.021714 0.056660 -0.383 0.702
## cPRI.win 0.752972 0.054824 13.734 <2e-16 ***
## AnyASA1:cPRI.win 0.005212 0.066884 0.078 0.938
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6166 on 562 degrees of freedom
## Multiple R-squared: 0.5162, Adjusted R-squared: 0.5136
## F-statistic: 199.9 on 3 and 562 DF, p-value: < 2.2e-16lm.beta(prqc.lin.mod3)#stan beta##
## Call:
## lm(formula = PRQC.win ~ AnyASA * cPRI.win, data = data)
##
## Standardized Coefficients::
## (Intercept) AnyASA1 cPRI.win AnyASA1:cPRI.win
## NA -0.011930466 0.713214889 0.003995239confint(prqc.lin.mod3, level=0.95)## 2.5 % 97.5 %
## (Intercept) 5.6732634 5.85422113
## AnyASA1 -0.1330054 0.08957667
## cPRI.win 0.6452884 0.86065652
## AnyASA1:cPRI.win -0.1261605 0.13658486#### AIC values
AIC(prqc.lin.mod1, prqc.lin.mod2, prqc.lin.mod3)## df AIC
## prqc.lin.mod1 3 1462.358
## prqc.lin.mod2 4 1062.918
## prqc.lin.mod3 5 1064.912stepAIC(prqc.lin.mod3, direction="both", trace=F) #this means that best fitting model is one that ONLY has perceived partner responsiveness##
## Call:
## lm(formula = PRQC.win ~ cPRI.win, data = data)
##
## Coefficients:
## (Intercept) cPRI.win
## 5.7496 0.7584Mean relationship quality is slightly higher in the AnyASA group than the neither group. Regression results indicate that ASA, conditioned on perceived partner responsiveness, does not predict relationship quality. When only ASA is in the model, it is sig but with an Adjusted \(R^2\) of .02. The AIC values also suggest that it is not a good fit to the data. The AIC value of the model that only contains perceived partner responsiveness is extremely similar to our full model, which basically means that ASA and the interaction adds almost no information to our model.
Trust
########### Trust ####################
data$InverseTrust<- 8-data$Trust ### Reverse scoring Trust so it fits gamma distribution
group_by(data, AnyASA) %>%
summarise(
count = n(),
mean = mean(Trust, na.rm = TRUE),
sd = sd(Trust, na.rm = TRUE),
median = median(Trust, na.rm = TRUE),
IQR = IQR(Trust, na.rm = TRUE)
)## # A tibble: 2 × 6
## AnyASA count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 6.11 1.22 6.67 1
## 2 1 351 5.84 1.42 6.33 1.67trust.step1<- glm(formula= InverseTrust ~ AnyASA, family=Gamma(link="log"),
na.action=na.exclude, data=data)
trust.step2<- glm(formula= InverseTrust ~ AnyASA + cPRI_total, family=Gamma(link="log"),
na.action=na.exclude, data=data)
trust.step3<- glm(formula= InverseTrust ~ AnyASA * cPRI_total, family=Gamma(link="log"),
na.action=na.exclude, data=data)
#Winorizing outliers
data$InverseTrust.win<- winsorize(data$InverseTrust, method="zscore",
threshold=3, robust=FALSE)
data$Trust.win<- winsorize(data$Trust, method="zscore",
threshold=3, robust=FALSE)
trust.h1.noOutliers<- glm(formula= InverseTrust.win ~ AnyASA * cPRI.win, family=Gamma(link="log"),
na.action=na.exclude, data=data)
#lin models
trust.lin.mod1<- lm(Trust.win ~ AnyASA, data=data)
trust.lin.mod2<- lm(Trust.win ~ AnyASA+cPRI.win, data=data)
trust.lin.mod3<- lm(Trust.win ~ AnyASA*cPRI.win, data=data)
#non-parametric ttest
wilcox.test(Trust ~ AnyASA, data=data, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: Trust by AnyASA
## W = 41445, p-value = 0.04308
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## 1.629236e-06 3.332659e-01
## sample estimates:
## difference in location
## 4.06057e-05wilcoxonRG(x = data$Trust, g = data$AnyASA)## rg
## 0.0984wilcox_effsize(Trust ~ AnyASA, data=data)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 Trust -1 1 0.0850 215 351 smallwilcox_test(Trust ~ AnyASA, data=data, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: Trust by AnyASA (-1, 1)
## Z = 2.0232, p-value = 0.04306
## alternative hypothesis: true mu is not equal to 0### Results
summary(trust.lin.mod1)##
## Call:
## lm(formula = Trust.win ~ AnyASA, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.8842 -0.7026 0.2974 0.7593 0.9641
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.24070 0.06382 97.782 <2e-16 ***
## AnyASA1 -0.20478 0.08105 -2.527 0.0118 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9358 on 564 degrees of freedom
## Multiple R-squared: 0.01119, Adjusted R-squared: 0.00944
## F-statistic: 6.384 on 1 and 564 DF, p-value: 0.01179lm.beta(trust.lin.mod1) #stan beta##
## Call:
## lm(formula = Trust.win ~ AnyASA, data = data)
##
## Standardized Coefficients::
## (Intercept) AnyASA1
## NA -0.1057965summary(trust.lin.mod2)##
## Call:
## lm(formula = Trust.win ~ AnyASA + cPRI.win, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.2436 -0.4894 0.2071 0.3999 1.9407
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.97631 0.04827 123.80 <2e-16 ***
## AnyASA1 0.01386 0.06020 0.23 0.818
## cPRI.win 0.77104 0.03492 22.08 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6857 on 563 degrees of freedom
## Multiple R-squared: 0.4701, Adjusted R-squared: 0.4682
## F-statistic: 249.7 on 2 and 563 DF, p-value: < 2.2e-16lm.beta(trust.lin.mod2) #stan beta##
## Call:
## lm(formula = Trust.win ~ AnyASA + cPRI.win, data = data)
##
## Standardized Coefficients::
## (Intercept) AnyASA1 cPRI.win
## NA 0.007159769 0.686745086summary(trust.lin.mod3)##
## Call:
## lm(formula = Trust.win ~ AnyASA * cPRI.win, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.2553 -0.4783 0.1915 0.3872 1.9468
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.987501 0.051251 116.827 <2e-16 ***
## AnyASA1 0.001726 0.063040 0.027 0.978
## cPRI.win 0.738414 0.060997 12.106 <2e-16 ***
## AnyASA1:cPRI.win 0.048559 0.074415 0.653 0.514
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6861 on 562 degrees of freedom
## Multiple R-squared: 0.4705, Adjusted R-squared: 0.4676
## F-statistic: 166.4 on 3 and 562 DF, p-value: < 2.2e-16lm.beta(trust.lin.mod3) #stan beta##
## Call:
## lm(formula = Trust.win ~ AnyASA * cPRI.win, data = data)
##
## Standardized Coefficients::
## (Intercept) AnyASA1 cPRI.win AnyASA1:cPRI.win
## NA 0.0008914872 0.6576861165 0.0350002737confint(trust.lin.mod3, level=0.95)## 2.5 % 97.5 %
## (Intercept) 5.88683447 6.0881675
## AnyASA1 -0.12209652 0.1255476
## cPRI.win 0.61860454 0.8582225
## AnyASA1:cPRI.win -0.09760601 0.1947237#Model comparison
AIC(trust.lin.mod1)## [1] 1535.148AIC(trust.lin.mod2)## [1] 1184.121AIC(trust.lin.mod3)## [1] 1185.693stepAIC(trust.lin.mod3, direction="both", trace=FALSE)##
## Call:
## lm(formula = Trust.win ~ cPRI.win, data = data)
##
## Coefficients:
## (Intercept) cPRI.win
## 5.9851 0.7697When ASA is the only predictor in the model, it significantly predicts worse relationship quality with an Adjusted \(R^2\) of .01. The AIC values suggest that this is not a good fit for the data (supporting that ASA is not very useful predictor). When perceived partner responsiveness is added to the model, it is no longer significant. The Adjusted \(R^2\) of the full model is .46.
Sexual Satisfaction
######### NSSS-S ########################
group_by(data, AnyASA) %>%
summarise(
count = n(),
mean = mean(NSSS, na.rm = TRUE),
sd = sd(NSSS, na.rm = TRUE),
median = median(NSSS, na.rm = TRUE),
IQR = IQR(NSSS, na.rm = TRUE)
)## # A tibble: 2 × 6
## AnyASA count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 3.45 0.979 3.5 1.25
## 2 1 351 3.35 1.08 3.58 1.58sex.sat.step1<- lm(NSSS ~ AnyASA, data=data)
sex.sat.step2<- lm(NSSS ~ AnyASA + cPRI.win, data=data)
sex.sat.step3<- lm(NSSS ~ AnyASA*cPRI.win, data=data)
#### Results
summary(sex.sat.step1)##
## Call:
## lm(formula = NSSS ~ AnyASA, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.4461 -0.6961 0.1455 0.7268 1.6455
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.44612 0.07118 48.412 <2e-16 ***
## AnyASA1 -0.09166 0.09039 -1.014 0.311
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.044 on 564 degrees of freedom
## Multiple R-squared: 0.00182, Adjusted R-squared: 5.001e-05
## F-statistic: 1.028 on 1 and 564 DF, p-value: 0.311lm.beta(sex.sat.step1) #stan beta##
## Call:
## lm(formula = NSSS ~ AnyASA, data = data)
##
## Standardized Coefficients::
## (Intercept) AnyASA1
## NA -0.04265956summary(sex.sat.step2)##
## Call:
## lm(formula = NSSS ~ AnyASA + cPRI.win, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.91327 -0.62973 0.08673 0.65558 2.42603
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.25095 0.06564 49.523 <2e-16 ***
## AnyASA1 0.06974 0.08187 0.852 0.395
## cPRI.win 0.56920 0.04749 11.986 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9325 on 563 degrees of freedom
## Multiple R-squared: 0.2047, Adjusted R-squared: 0.2019
## F-statistic: 72.48 on 2 and 563 DF, p-value: < 2.2e-16lm.beta(sex.sat.step2) #stan beta##
## Call:
## lm(formula = NSSS ~ AnyASA + cPRI.win, data = data)
##
## Standardized Coefficients::
## (Intercept) AnyASA1 cPRI.win
## NA 0.03245832 0.45669726summary(sex.sat.step3)##
## Call:
## lm(formula = NSSS ~ AnyASA * cPRI.win, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.8886 -0.6338 0.1087 0.6417 2.5048
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.23331 0.06968 46.400 < 2e-16 ***
## AnyASA1 0.08887 0.08571 1.037 0.30
## cPRI.win 0.62063 0.08294 7.483 2.81e-13 ***
## AnyASA1:cPRI.win -0.07655 0.10118 -0.757 0.45
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9328 on 562 degrees of freedom
## Multiple R-squared: 0.2056, Adjusted R-squared: 0.2013
## F-statistic: 48.47 on 3 and 562 DF, p-value: < 2.2e-16lm.beta(sex.sat.step3)#stan beta##
## Call:
## lm(formula = NSSS ~ AnyASA * cPRI.win, data = data)
##
## Standardized Coefficients::
## (Intercept) AnyASA1 cPRI.win AnyASA1:cPRI.win
## NA 0.04135951 0.49796209 -0.04970170confint(sex.sat.step3, level=.95)## 2.5 % 97.5 %
## (Intercept) 3.09643810 3.3701846
## AnyASA1 -0.07948988 0.2572244
## cPRI.win 0.45772783 0.7835291
## AnyASA1:cPRI.win -0.27528203 0.1221899# non-parametric ttest
wilcox.test(NSSS ~ AnyASA, data=data, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: NSSS by AnyASA
## W = 39056, p-value = 0.4831
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## -0.08339813 0.24999952
## sample estimates:
## difference in location
## 0.0832895wilcoxonRG(x = data$NSSS, g = data$AnyASA)## rg
## 0.0351wilcox_effsize(NSSS ~ AnyASA, data=data)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 NSSS -1 1 0.0295 215 351 smallwilcox_test(NSSS ~ AnyASA, data=data, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: NSSS by AnyASA (-1, 1)
## Z = 0.70154, p-value = 0.483
## alternative hypothesis: true mu is not equal to 0#### AIC
AIC(sex.sat.step1)## [1] 1658.705AIC(sex.sat.step2)## [1] 1532.067AIC(sex.sat.step3)## [1] 1533.491stepAIC(sex.sat.step3, direction="both", trace=FALSE)##
## Call:
## lm(formula = NSSS ~ cPRI.win, data = data)
##
## Coefficients:
## (Intercept) cPRI.win
## 3.2953 0.5625Sexual assault does not predict sexual satisfaction. The Adjustwed \(R^2\) of the full model is .20.
EQ: Responsiveness
group_by(data, AnyASA) %>%
summarise(
count = n(),
mean = mean(PRI_total, na.rm = TRUE),
sd = sd(PRI_total, na.rm = TRUE),
median = median(PRI_total, na.rm = TRUE),
IQR = IQR(PRI_total, na.rm = TRUE)
)## # A tibble: 2 × 6
## AnyASA count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 4.20 1.02 4.62 1.19
## 2 1 351 3.81 1.25 4.25 1.81# non-parametric ttest
wilcox.test(PRI_total ~ AnyASA, data=data, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: PRI_total by AnyASA
## W = 45043, p-value = 9.617e-05
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## 0.1249420 0.3749968
## sample estimates:
## difference in location
## 0.249997wilcoxonRG(x = data$PRI_total, g = data$AnyASA)## rg
## 0.194wilcox_effsize(PRI_total ~ AnyASA, data=data)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 PRI_total -1 1 0.164 215 351 smallwilcox_test(PRI_total ~ AnyASA, data=data, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: PRI_total by AnyASA (-1, 1)
## Z = 3.9003, p-value = 9.606e-05
## alternative hypothesis: true mu is not equal to 0Hypothesis 3
Revictimization (Child Sexual Assault and ASA) will be associated with poorer (a1) romantic relationship quality, (a2) partner trust, and (b) lower sexual satisfaction.
3.A1: PRQC
# Effect coding
data<- data %>%
mutate(SVGroup = case_when(
data$ASAonly==TRUE ~ 0,
data$revictimized==TRUE ~ 1,
data$NeitherASAnorCSA==TRUE ~ -1
))
data$SVGroup<- as.factor(data$SVGroup)
contrasts(data$SVGroup) <- contr.sum(3)
contrasts(data$SVGroup)## [,1] [,2]
## -1 1 0
## 0 0 1
## 1 -1 -1group_by(data, SVGroup) %>%
summarise(
count = n(),
mean = mean(PRQC, na.rm = TRUE),
sd = sd(PRQC, na.rm = TRUE),
median = median(PRQC, na.rm = TRUE),
IQR = IQR(PRQC, na.rm = TRUE)
)## # A tibble: 3 × 6
## SVGroup count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 5.93 1.08 6.28 1.28
## 2 0 198 5.68 1.11 5.94 1.33
## 3 1 153 5.67 1.17 6 1.33ggplot(data, aes(SVGroup, PRQC)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Assault History", y="PRQC") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
PRQC.kruskal<- kruskal.test(PRQC ~ SVGroup, data = data)
PRQC.kruskal##
## Kruskal-Wallis rank sum test
##
## data: PRQC by SVGroup
## Kruskal-Wallis chi-squared = 9.6284, df = 2, p-value = 0.008114dunnTest(PRQC ~ SVGroup, data=data, method="bh")## Dunn (1964) Kruskal-Wallis multiple comparison## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 -1 - 0 2.8249725 0.004728467 0.01418540
## 2 -1 - 1 2.4099025 0.015956783 0.02393518
## 3 0 - 1 -0.2170016 0.828207122 0.82820712data %>% kruskal_effsize(PRQC ~ SVGroup)## # A tibble: 1 × 5
## .y. n effsize method magnitude
## * <chr> <int> <dbl> <chr> <ord>
## 1 PRQC 566 0.0135 eta2[H] smallConclusion: There are significant group differences. Specifically, those who have not been assaulted report better relationship quality than those who have experience ASA and who have experienced revictimization. There are no significant group differences comparing ASA Only and revictimized group.
3.A2: Trust
group_by(data, SVGroup) %>%
summarise(
count = n(),
mean = mean(Trust, na.rm = TRUE),
sd = sd(Trust, na.rm = TRUE),
median = median(Trust, na.rm = TRUE),
IQR = IQR(Trust, na.rm = TRUE)
)## # A tibble: 3 × 6
## SVGroup count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 6.11 1.22 6.67 1
## 2 0 198 5.95 1.35 6.33 1.67
## 3 1 153 5.70 1.50 6 2ggplot(data, aes(SVGroup, Trust)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Assault History", y="Trust") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
kruskal.test(Trust ~ SVGroup, data = data)##
## Kruskal-Wallis rank sum test
##
## data: Trust by SVGroup
## Kruskal-Wallis chi-squared = 6.3711, df = 2, p-value = 0.04136dunnTest(Trust ~ SVGroup, data=data, method="bh")## Dunn (1964) Kruskal-Wallis multiple comparison## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 -1 - 0 1.059915 0.28918340 0.28918340
## 2 -1 - 1 2.523008 0.01163559 0.03490677
## 3 0 - 1 1.509265 0.13123107 0.19684661data %>% kruskal_effsize(Trust ~ SVGroup)## # A tibble: 1 × 5
## .y. n effsize method magnitude
## * <chr> <int> <dbl> <chr> <ord>
## 1 Trust 566 0.00776 eta2[H] smallConclusion: There is a significant difference comparing those who have never been assaulted and revictimized individuals.
3.A3: Sexual Satisfaction
group_by(data, SVGroup) %>%
summarise(
count = n(),
mean = mean(NSSS, na.rm = TRUE),
sd = sd(NSSS, na.rm = TRUE),
median = median(NSSS, na.rm = TRUE),
IQR = IQR(NSSS, na.rm = TRUE)
)## # A tibble: 3 × 6
## SVGroup count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 3.45 0.979 3.5 1.25
## 2 0 198 3.39 1.01 3.54 1.25
## 3 1 153 3.31 1.17 3.58 1.75shapiro.test(data$NSSS) #data is not normally distributed##
## Shapiro-Wilk normality test
##
## data: data$NSSS
## W = 0.96429, p-value = 1.703e-10car::leveneTest(NSSS ~ SVGroup, center = median, data = data) #variances are equal## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 2 3.6411 0.02684 *
## 563
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1ggplot(data, aes(SVGroup, NSSS)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Assault History", y="Sexual Satisfaction") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
kruskal.test(NSSS ~ SVGroup, data = data) #Not significant##
## Kruskal-Wallis rank sum test
##
## data: NSSS by SVGroup
## Kruskal-Wallis chi-squared = 0.51117, df = 2, p-value = 0.7745data %>% kruskal_effsize(NSSS ~ SVGroup)## # A tibble: 1 × 5
## .y. n effsize method magnitude
## * <chr> <int> <dbl> <chr> <ord>
## 1 NSSS 566 -0.00264 eta2[H] smallConclusion:There are no significant differences in sexual satisfaction based on sexual assault history.
EQ: PRI
Are there significant group differences in perceived partner responsiveness?
group_by(data, SVGroup) %>%
summarise(
count = n(),
mean = mean(PRI_total, na.rm = TRUE),
sd = sd(PRI_total, na.rm = TRUE),
median = median(PRI_total, na.rm = TRUE),
IQR = IQR(PRI_total, na.rm = TRUE)
)## # A tibble: 3 × 6
## SVGroup count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 4.20 1.02 4.62 1.19
## 2 0 198 3.85 1.20 4.25 1.5
## 3 1 153 3.76 1.31 4.25 2ggplot(data, aes(SVGroup, PRI_total)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Assault History", y="Partner responsiveness") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
kruskal.test(PRI_total ~ SVGroup, data = data) #significant##
## Kruskal-Wallis rank sum test
##
## data: PRI_total by SVGroup
## Kruskal-Wallis chi-squared = 15.258, df = 2, p-value = 0.0004861dunnTest(PRI_total ~ SVGroup, data=data, method="bh")## Dunn (1964) Kruskal-Wallis multiple comparison## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 -1 - 0 3.3276881 0.0008756984 0.002627095
## 2 -1 - 1 3.3161116 0.0009127936 0.001369190
## 3 0 - 1 0.2134405 0.8309833740 0.830983374data %>% kruskal_effsize(PRI_total ~ SVGroup)## # A tibble: 1 × 5
## .y. n effsize method magnitude
## * <chr> <int> <dbl> <chr> <ord>
## 1 PRI_total 566 0.0235 eta2[H] smallBoth individuals in the ASA only group and the revictimized group perceive their partners as being less responsive than individuals in the no ASA group. There are no significant differences between individuals in the ASA only and the revictimized groups.
Hypothesis 4
Multiple victimizations (i.e., more than one ASA experience) will be associated with poorer (a1) romantic relationship quality, (a2) partner trust, and (b) lower sexual satisfaction.
data$multvic<- data$SV_1 + data$SV_2 + data$SV_3 + data$SV_4
SAdata<- subset(data, AnyASA==1)
# Effect coding
SAdata<- SAdata %>%
mutate(multvic.log = case_when(
SAdata$multvic==1 ~ -1,
SAdata$multvic>1 ~ 1
))
SAdata$multvic.log<- as.factor(SAdata$multvic.log)
ggplot(SAdata, aes(multvic.log, PRQC)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Multiple victimizations", y="PRQC") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
ggplot(SAdata, aes(multvic.log, NSSS)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Multiple victimizations", y="NSSS") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
ggplot(SAdata, aes(multvic.log, Trust)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Multiple victimizations", y="Trust") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
ggplot(SAdata, aes(multvic.log, PRI_total)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Multiple victimizations", y="PRI") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
#Mean differences
group_by(SAdata, multvic.log) %>%
summarise(
count = n(),
mean = mean(PRQC, na.rm = TRUE),
sd = sd(PRQC, na.rm = TRUE),
median = median(PRQC, na.rm = TRUE),
IQR = IQR(PRQC, na.rm = TRUE)
)## # A tibble: 2 × 6
## multvic.log count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 85 5.70 1.14 5.89 1.28
## 2 1 266 5.67 1.13 6 1.38group_by(SAdata, multvic.log) %>%
summarise(
count = n(),
mean = mean(NSSS, na.rm = TRUE),
sd = sd(NSSS, na.rm = TRUE),
median = median(NSSS, na.rm = TRUE),
IQR = IQR(NSSS, na.rm = TRUE)
)## # A tibble: 2 × 6
## multvic.log count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 85 3.35 1.06 3.42 1.5
## 2 1 266 3.36 1.09 3.58 1.58group_by(SAdata, multvic.log) %>%
summarise(
count = n(),
mean = mean(Trust, na.rm = TRUE),
sd = sd(Trust, na.rm = TRUE),
median = median(Trust, na.rm = TRUE),
IQR = IQR(Trust, na.rm = TRUE)
)## # A tibble: 2 × 6
## multvic.log count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 85 6.07 1.28 6.33 1.33
## 2 1 266 5.76 1.46 6.33 1.67## Non-parametric ttests
#PRQC
wilcox_effsize(PRQC ~ multvic.log, data=SAdata)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 PRQC -1 1 0.0136 85 266 smallwilcox_test(PRQC ~ multvic.log, data=SAdata, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: PRQC by multvic.log (-1, 1)
## Z = 0.25489, p-value = 0.7988
## alternative hypothesis: true mu is not equal to 0wilcox.test(PRQC ~ multvic.log, data=SAdata, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: PRQC by multvic.log
## W = 11512, p-value = 0.7993
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## -0.1666737 0.2777265
## sample estimates:
## difference in location
## 4.806546e-05# Trust
wilcox_effsize(Trust ~ multvic.log, data=SAdata)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 Trust -1 1 0.0934 85 266 smallwilcox_test(Trust ~ multvic.log, data=SAdata, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: Trust by multvic.log (-1, 1)
## Z = 1.7505, p-value = 0.08003
## alternative hypothesis: true mu is not equal to 0wilcox.test(Trust ~ multvic.log, data=SAdata, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: Trust by multvic.log
## W = 12695, p-value = 0.08014
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## -9.392948e-06 3.334085e-01
## sample estimates:
## difference in location
## 1.858111e-05# Sexual Satisfaction
wilcox_effsize(NSSS ~ multvic.log, data=SAdata)## # A tibble: 1 × 7
## .y. group1 group2 effsize n1 n2 magnitude
## * <chr> <chr> <chr> <dbl> <int> <int> <ord>
## 1 NSSS -1 1 0.00774 85 266 smallwilcox_test(NSSS ~ multvic.log, data=SAdata, conf.level=.95)##
## Asymptotic Wilcoxon-Mann-Whitney Test
##
## data: NSSS by multvic.log (-1, 1)
## Z = -0.14497, p-value = 0.8847
## alternative hypothesis: true mu is not equal to 0wilcox.test(NSSS ~ multvic.log, data=SAdata, alternative=c("two.sided"), conf.int=T, conf.level=.95)##
## Wilcoxon rank sum test with continuity correction
##
## data: NSSS by multvic.log
## W = 11187, p-value = 0.8852
## alternative hypothesis: true location shift is not equal to 0
## 95 percent confidence interval:
## -0.3332543 0.2499712
## sample estimates:
## difference in location
## -2.580545e-05#Correlations dont' really make sense here because we don't know exactly how many incidents each person experience (e.g., if they say more than 4, we don't know how many more than 4), but I wanted to do a sanity check here to make sure we're seeing similar patterns of results using the continuous-ish variable vs. the dichotomized variable
cor.test(SAdata$multvic, SAdata$Trust, method="spearman")## Warning in cor.test.default(SAdata$multvic, SAdata$Trust, method = "spearman"):
## Cannot compute exact p-value with ties##
## Spearman's rank correlation rho
##
## data: SAdata$multvic and SAdata$Trust
## S = 7702145, p-value = 0.1993
## alternative hypothesis: true rho is not equal to 0
## sample estimates:
## rho
## -0.06867367cor.test(SAdata$multvic, SAdata$PRQC, method="spearman")## Warning in cor.test.default(SAdata$multvic, SAdata$PRQC, method = "spearman"):
## Cannot compute exact p-value with ties##
## Spearman's rank correlation rho
##
## data: SAdata$multvic and SAdata$PRQC
## S = 6998467, p-value = 0.5887
## alternative hypothesis: true rho is not equal to 0
## sample estimates:
## rho
## 0.02896168cor.test(SAdata$multvic, SAdata$NSSS, method="spearman")## Warning in cor.test.default(SAdata$multvic, SAdata$NSSS, method = "spearman"):
## Cannot compute exact p-value with ties##
## Spearman's rank correlation rho
##
## data: SAdata$multvic and SAdata$NSSS
## S = 6909446, p-value = 0.4404
## alternative hypothesis: true rho is not equal to 0
## sample estimates:
## rho
## 0.04131336cor.test(SAdata$multvic, SAdata$PRI_total, method="spearman")## Warning in cor.test.default(SAdata$multvic, SAdata$PRI_total, method =
## "spearman"): Cannot compute exact p-value with ties##
## Spearman's rank correlation rho
##
## data: SAdata$multvic and SAdata$PRI_total
## S = 7362997, p-value = 0.6865
## alternative hypothesis: true rho is not equal to 0
## sample estimates:
## rho
## -0.0216169I created a binary variable where someone can be in either the one ASA experience bin or more than once bin (I excluded participants with no history of assault). There were no significant differences between groups in any of the dependent variables.
Hypothesis 5
Time since sexual assault will be associated with (a1) romantic relationship quality, (a2) partner trust, and (b) sexual satisfaction, such that participants who report more recent sexual assault experiences will report poorer romantic relationship quality and partner trust and lower sexual satisfaction.
summary(SAdata$min_t)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 0.00 5.00 9.00 11.72 15.50 56.17 1sd(SAdata$min_t, na.rm=T)## [1] 10.12415hist(SAdata$min_t, main="Histogram of SV Timing", xlab="Most recent SV experience",
col='hotpink', sub=paste("Skewness:",
round(e1071::skewness(SAdata$min_t, na.rm=TRUE), 2)))
#plots
ggplot(SAdata, aes(x = min_t, y = PRQC)) +
geom_point(alpha = 0.6, aes(color = "hotpink")) +
ggtitle("SV timing vs PRQC") +
xlab("SV timing") +
ylab("PRQC")## Warning: Removed 1 rows containing missing values (`geom_point()`).
ggplot(SAdata, aes(x = min_t, y = NSSS)) +
geom_point(alpha = 0.6, aes(color = "hotpink")) +
ggtitle("SV timing vs NSSS") +
xlab("SV timing") +
ylab("NSSS")## Warning: Removed 1 rows containing missing values (`geom_point()`).
ggplot(SAdata, aes(x = min_t, y = PRI_total)) +
geom_point(alpha = 0.6, aes(color = "hotpink")) +
ggtitle("SV timing vs PRI") +
xlab("SV timing") +
ylab("PRI_total")## Warning: Removed 1 rows containing missing values (`geom_point()`).
ggplot(SAdata, aes(x = min_t, y = Trust)) +
geom_point(alpha = 0.6, aes(color = "hotpink")) +
ggtitle("SV timing vs Trust") +
xlab("SV timing") +
ylab("Trust")## Warning: Removed 1 rows containing missing values (`geom_point()`).
#Correlations
timingdata<- SAdata %>%
select(min_t, PRQC, Trust, NSSS, PRI_total, AAQ.AV, AAQ.ANX, relLength, Age_1)
timingdata <- timingdata[complete.cases(timingdata), ]
rcorr(as.matrix(timingdata), type = c("spearman"))## min_t PRQC Trust NSSS PRI_total AAQ.AV AAQ.ANX relLength Age_1
## min_t 1.00 -0.11 0.00 -0.07 -0.08 -0.04 -0.23 0.64 0.74
## PRQC -0.11 1.00 0.78 0.71 0.74 -0.22 -0.35 -0.15 -0.15
## Trust 0.00 0.78 1.00 0.45 0.71 -0.25 -0.39 -0.02 -0.06
## NSSS -0.07 0.71 0.45 1.00 0.46 -0.21 -0.29 -0.10 -0.07
## PRI_total -0.08 0.74 0.71 0.46 1.00 -0.23 -0.40 -0.15 -0.13
## AAQ.AV -0.04 -0.22 -0.25 -0.21 -0.23 1.00 0.34 0.04 -0.10
## AAQ.ANX -0.23 -0.35 -0.39 -0.29 -0.40 0.34 1.00 -0.22 -0.20
## relLength 0.64 -0.15 -0.02 -0.10 -0.15 0.04 -0.22 1.00 0.51
## Age_1 0.74 -0.15 -0.06 -0.07 -0.13 -0.10 -0.20 0.51 1.00
##
## n= 350
##
##
## P
## min_t PRQC Trust NSSS PRI_total AAQ.AV AAQ.ANX relLength Age_1
## min_t 0.0413 0.9455 0.1655 0.1136 0.4683 0.0000 0.0000 0.0000
## PRQC 0.0413 0.0000 0.0000 0.0000 0.0000 0.0000 0.0049 0.0054
## Trust 0.9455 0.0000 0.0000 0.0000 0.0000 0.0000 0.6929 0.2840
## NSSS 0.1655 0.0000 0.0000 0.0000 0.0001 0.0000 0.0686 0.1674
## PRI_total 0.1136 0.0000 0.0000 0.0000 0.0000 0.0000 0.0063 0.0133
## AAQ.AV 0.4683 0.0000 0.0000 0.0001 0.0000 0.0000 0.4809 0.0650
## AAQ.ANX 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0001
## relLength 0.0000 0.0049 0.6929 0.0686 0.0063 0.4809 0.0000 0.0000
## Age_1 0.0000 0.0054 0.2840 0.1674 0.0133 0.0650 0.0001 0.0000# exploratory regressions conditioning on relationship length
timeNSSS<- lm(NSSS ~ Age_1 + relLength + min_t, data=SAdata)
summary(timeNSSS)##
## Call:
## lm(formula = NSSS ~ Age_1 + relLength + min_t, data = SAdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.4625 -0.7239 0.1922 0.8411 1.9464
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.554604 0.225387 15.771 <2e-16 ***
## Age_1 -0.003824 0.008068 -0.474 0.636
## relLength -0.013322 0.010350 -1.287 0.199
## min_t 0.003970 0.010235 0.388 0.698
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.083 on 346 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.008574, Adjusted R-squared: -2.2e-05
## F-statistic: 0.9974 on 3 and 346 DF, p-value: 0.3941lm.beta(timeNSSS)##
## Call:
## lm(formula = NSSS ~ Age_1 + relLength + min_t, data = SAdata)
##
## Standardized Coefficients::
## (Intercept) Age_1 relLength min_t
## NA -0.04197067 -0.08943191 0.03711582confint(timeNSSS, level=.95)## 2.5 % 97.5 %
## (Intercept) 3.11130275 3.997905768
## Age_1 -0.01969217 0.012044038
## relLength -0.03367951 0.007034979
## min_t -0.01616032 0.024101263timePRQC<- lm(PRQC.win ~ Age_1 + relLength + min_t, data=SAdata)
summary(timePRQC)##
## Call:
## lm(formula = PRQC.win ~ Age_1 + relLength + min_t, data = SAdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.7345 -0.6126 0.1288 0.7576 1.4859
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.213540 0.185566 33.484 <2e-16 ***
## Age_1 -0.013572 0.006642 -2.043 0.0418 *
## relLength -0.014718 0.008522 -1.727 0.0850 .
## min_t 0.016109 0.008427 1.912 0.0567 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8917 on 346 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.02365, Adjusted R-squared: 0.01519
## F-statistic: 2.794 on 3 and 346 DF, p-value: 0.0403lm.beta(timePRQC)##
## Call:
## lm(formula = PRQC.win ~ Age_1 + relLength + min_t, data = SAdata)
##
## Standardized Coefficients::
## (Intercept) Age_1 relLength min_t
## NA -0.1795407 -0.1190878 0.1815039confint(timePRQC, level=.95)## 2.5 % 97.5 %
## (Intercept) 5.8485612671 6.5785184796
## Age_1 -0.0266363585 -0.0005073321
## relLength -0.0314785324 0.0020424830
## min_t -0.0004651678 0.0326829600timeTrust<- lm(Trust.win ~ Age_1 + relLength + min_t, data=SAdata)
summary(timeTrust)##
## Call:
## lm(formula = Trust.win ~ Age_1 + relLength + min_t, data = SAdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.8978 -0.7553 0.2662 0.9103 1.2649
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.293237 0.204958 30.705 <2e-16 ***
## Age_1 -0.011039 0.007337 -1.505 0.1333
## relLength -0.006509 0.009412 -0.692 0.4897
## min_t 0.017193 0.009307 1.847 0.0656 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9849 on 346 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.01013, Adjusted R-squared: 0.001549
## F-statistic: 1.18 on 3 and 346 DF, p-value: 0.3171lm.beta(timeTrust)##
## Call:
## lm(formula = Trust.win ~ Age_1 + relLength + min_t, data = SAdata)
##
## Standardized Coefficients::
## (Intercept) Age_1 relLength min_t
## NA -0.13313383 -0.04800964 0.17659987confint(timeTrust, level=.95)## 2.5 % 97.5 %
## (Intercept) 5.890115458 6.69635784
## Age_1 -0.025469245 0.00339043
## relLength -0.025020751 0.01200342
## min_t -0.001113168 0.03549915mean(timingdata$min_t, na.rm=T)## [1] 11.7219sd(timingdata$min_t, na.rm=T)## [1] 10.12415The variable \(min_t\) filters through all of their SV timing variables and selects the most recent experience. The correlations between amount of time since one’s most recent sexual assault experience and PRQC, trust, sexual satisfaction, and perceived partner responsiveness are very small and likely insignificant. None of the exploratory regressions that conditioned on relationship length were significant.
Exploratory Question
qualdatafull<- data[, c("SubjectID", "SVGroup", "SV_partner_response", "SV_relationship")] #creating dataframe that only has subject ID, assault status, and qualitative data
qualdata<- subset(qualdatafull, SVGroup==0 | SVGroup==1) #removing people that haven't experienced SA
write_xlsx(qualdata, "~/Downloads/qualdata.xlsx") #function that exports this dataframe to a xcel fileExploratory Question
Does attachment orientation moderate the association between ASA and (a1) romantic relationship quality, (a2) partner trust, or (b) sexual satisfaction such that individuals who have experienced ASA and are anxiously or avoidantly attached report worse relationship quality and lower partner trust and sexual satisfaction than individuals who have experienced ASA and are securely attached?
group_by(data, SVGroup) %>% #descriptives for attachment anxiety
summarise(
count = n(),
mean = mean(AAQ.ANX, na.rm = TRUE),
sd = sd(AAQ.ANX, na.rm = TRUE),
median = median(AAQ.ANX, na.rm = TRUE),
IQR = IQR(AAQ.ANX, na.rm = TRUE)
)## # A tibble: 3 × 6
## SVGroup count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 3.26 1.16 3.33 1.56
## 2 0 198 3.39 1.19 3.44 1.78
## 3 1 153 3.68 1.23 3.67 1.56group_by(data, SVGroup) %>% #descriptives for attachment avoidance
summarise(
count = n(),
mean = mean(AAQ.AV, na.rm = TRUE),
sd = sd(AAQ.AV, na.rm = TRUE),
median = median(AAQ.AV, na.rm = TRUE),
IQR = IQR(AAQ.AV, na.rm = TRUE)
)## # A tibble: 3 × 6
## SVGroup count mean sd median IQR
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 -1 215 3.63 1.33 3.5 1.88
## 2 0 198 3.89 1.27 4 1.62
## 3 1 153 4.07 1.41 4.12 2.25ggplot(data, aes(AnyASA, AAQ.ANX)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Assault History", y="Attachment Anxiety") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
ggplot(data, aes(AnyASA, AAQ.AV)) +
geom_boxplot(fill = "lightpink", color = "hotpink") +
labs(x="Assault History", y="Attachment Avoidance") +
theme_bw() +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())
AAQ.ANX.kruskal<- kruskal.test(AAQ.ANX ~ SVGroup, data = data)
AAQ.ANX.kruskal##
## Kruskal-Wallis rank sum test
##
## data: AAQ.ANX by SVGroup
## Kruskal-Wallis chi-squared = 8.3054, df = 2, p-value = 0.01572dunnTest(AAQ.ANX ~ SVGroup, data=data, method="bh")## Dunn (1964) Kruskal-Wallis multiple comparison## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 -1 - 0 -0.9900569 0.322146299 0.32214630
## 2 -1 - 1 -2.8638218 0.004185634 0.01255690
## 3 0 - 1 -1.9080782 0.056381104 0.08457166data %>% kruskal_effsize(AAQ.ANX ~ SVGroup)## # A tibble: 1 × 5
## .y. n effsize method magnitude
## * <chr> <int> <dbl> <chr> <ord>
## 1 AAQ.ANX 566 0.0112 eta2[H] smallAAQ.AV.kruskal<- kruskal.test(AAQ.AV ~ SVGroup, data = data)
AAQ.AV.kruskal##
## Kruskal-Wallis rank sum test
##
## data: AAQ.AV by SVGroup
## Kruskal-Wallis chi-squared = 12.192, df = 2, p-value = 0.002252dunnTest(AAQ.AV ~ SVGroup, data=data, method="bh")## Dunn (1964) Kruskal-Wallis multiple comparison
## p-values adjusted with the Benjamini-Hochberg method.## Comparison Z P.unadj P.adj
## 1 -1 - 0 -2.191903 0.0283865227 0.042579784
## 2 -1 - 1 -3.410708 0.0006479446 0.001943834
## 3 0 - 1 -1.345700 0.1783992345 0.178399234data %>% kruskal_effsize(AAQ.AV ~ SVGroup)## # A tibble: 1 × 5
## .y. n effsize method magnitude
## * <chr> <int> <dbl> <chr> <ord>
## 1 AAQ.AV 566 0.0181 eta2[H] smallattach.PRQC.gamma.mod<- glm(formula= InversePRQC ~ SVGroup*cANX + SVGroup*cAV, family=Gamma(link="log"),
na.action=na.exclude, data=data)
attach.prqc.lin.mod<- lm(PRQC.win ~ SVGroup*cANX + SVGroup*cAV, data=data)
summary(attach.prqc.lin.mod)##
## Call:
## lm(formula = PRQC.win ~ SVGroup * cANX + SVGroup * cAV, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.1550 -0.5192 0.1371 0.5993 1.6315
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.88438 0.03487 168.732 < 2e-16 ***
## SVGroup1 0.07959 0.04753 1.674 0.0946 .
## SVGroup2 -0.09722 0.04815 -2.019 0.0440 *
## cANX -0.21150 0.03085 -6.856 1.88e-11 ***
## cAV -0.11879 0.02752 -4.316 1.88e-05 ***
## SVGroup1:cANX 0.06538 0.04324 1.512 0.1311
## SVGroup2:cANX 0.05807 0.04282 1.356 0.1756
## SVGroup1:cAV -0.04724 0.03812 -1.239 0.2158
## SVGroup2:cAV 0.05176 0.03919 1.321 0.1872
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8079 on 557 degrees of freedom
## Multiple R-squared: 0.1769, Adjusted R-squared: 0.1651
## F-statistic: 14.97 on 8 and 557 DF, p-value: < 2.2e-16lm.beta(attach.prqc.lin.mod)##
## Call:
## lm(formula = PRQC.win ~ SVGroup * cANX + SVGroup * cAV, data = data)
##
## Standardized Coefficients::
## (Intercept) SVGroup1 SVGroup2 cANX cAV
## NA 0.07197322 -0.08622093 -0.28716247 -0.18040574
## SVGroup1:cANX SVGroup2:cANX SVGroup1:cAV SVGroup2:cAV
## 0.07124407 0.06274857 -0.05905346 0.06165583confint(attach.prqc.lin.mod, level=.95)## 2.5 % 97.5 %
## (Intercept) 5.81587794 5.95287975
## SVGroup1 -0.01377401 0.17294947
## SVGroup2 -0.19180273 -0.00262939
## cANX -0.27208983 -0.15090758
## cAV -0.17285095 -0.06473450
## SVGroup1:cANX -0.01955814 0.15032196
## SVGroup2:cANX -0.02603448 0.14216736
## SVGroup1:cAV -0.12212683 0.02764478
## SVGroup2:cAV -0.02522158 0.12873287EQ2.A2
attach.trust.gamma.mod<- glm(formula= InverseTrust.win ~ SVGroup*cANX + SVGroup*cAV, family=Gamma(link="log"),
na.action=na.exclude, data=data)
attach.trust.lin.mod<- lm(Trust.win ~ SVGroup*cANX + SVGroup*cAV, data=data)
summary(attach.trust.lin.mod)##
## Call:
## lm(formula = Trust.win ~ SVGroup * cANX + SVGroup * cAV, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.3671 -0.6304 0.1532 0.6336 1.8995
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.113145 0.036317 168.326 < 2e-16 ***
## SVGroup1 0.062344 0.049498 1.260 0.208362
## SVGroup2 -0.008754 0.050147 -0.175 0.861485
## cANX -0.284023 0.032124 -8.842 < 2e-16 ***
## cAV -0.112132 0.028660 -3.912 0.000103 ***
## SVGroup1:cANX 0.035184 0.045033 0.781 0.434959
## SVGroup2:cANX 0.011576 0.044588 0.260 0.795247
## SVGroup1:cAV -0.010089 0.039702 -0.254 0.799491
## SVGroup2:cAV 0.031098 0.040811 0.762 0.446388
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8413 on 557 degrees of freedom
## Multiple R-squared: 0.2107, Adjusted R-squared: 0.1994
## F-statistic: 18.59 on 8 and 557 DF, p-value: < 2.2e-16lm.beta(attach.trust.lin.mod)##
## Call:
## lm(formula = Trust.win ~ SVGroup * cANX + SVGroup * cAV, data = data)
##
## Standardized Coefficients::
## (Intercept) SVGroup1 SVGroup2 cANX cAV
## NA 0.053015185 -0.007300566 -0.362619300 -0.160128247
## SVGroup1:cANX SVGroup2:cANX SVGroup1:cAV SVGroup2:cAV
## 0.036050703 0.011763150 -0.011859639 0.034835533confint(attach.trust.lin.mod, level=.95)## 2.5 % 97.5 %
## (Intercept) 6.04180956 6.18448035
## SVGroup1 -0.03488050 0.15956938
## SVGroup2 -0.10725451 0.08974661
## cANX -0.34712113 -0.22092450
## cAV -0.16842713 -0.05583695
## SVGroup1:cANX -0.05327080 0.12363874
## SVGroup2:cANX -0.07600468 0.09915715
## SVGroup1:cAV -0.08807395 0.06789504
## SVGroup2:cAV -0.04906475 0.11126015EQ2.A3
attach.sexsat.mod<- lm(NSSS ~ SVGroup*cANX + SVGroup*cAV, data=data)
summary(attach.sexsat.mod)##
## Call:
## lm(formula = NSSS ~ SVGroup * cANX + SVGroup * cAV, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.7985 -0.5951 0.1209 0.6568 1.9521
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.39736 0.04203 80.841 < 2e-16 ***
## SVGroup1 -0.02322 0.05728 -0.405 0.685
## SVGroup2 -0.01174 0.05803 -0.202 0.840
## cANX -0.20288 0.03717 -5.458 7.26e-08 ***
## cAV -0.15925 0.03316 -4.802 2.02e-06 ***
## SVGroup1:cANX 0.05828 0.05211 1.118 0.264
## SVGroup2:cANX 0.04950 0.05160 0.959 0.338
## SVGroup1:cAV -0.07506 0.04594 -1.634 0.103
## SVGroup2:cAV 0.07598 0.04723 1.609 0.108
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9735 on 557 degrees of freedom
## Multiple R-squared: 0.1424, Adjusted R-squared: 0.13
## F-statistic: 11.56 on 8 and 557 DF, p-value: 3.023e-15lm.beta(attach.sexsat.mod)##
## Call:
## lm(formula = NSSS ~ SVGroup * cANX + SVGroup * cAV, data = data)
##
## Standardized Coefficients::
## (Intercept) SVGroup1 SVGroup2 cANX cAV
## NA -0.017783564 -0.008819981 -0.233340737 -0.204858666
## SVGroup1:cANX SVGroup2:cANX SVGroup1:cAV SVGroup2:cAV
## 0.053793628 0.045315418 -0.079477365 0.076673853confint(attach.sexsat.mod, level=.95)## 2.5 % 97.5 %
## (Intercept) 3.31481355 3.47990886
## SVGroup1 -0.13572154 0.08929133
## SVGroup2 -0.12572259 0.10224251
## cANX -0.27589968 -0.12986788
## cAV -0.22439011 -0.09410338
## SVGroup1:cANX -0.04407821 0.16063738
## SVGroup2:cANX -0.05184216 0.15085103
## SVGroup1:cAV -0.16529935 0.01518433
## SVGroup2:cAV -0.01678067 0.16874357Conclusion: There are group differences in attachment such that revictimized individuals report more attachment anxiety and avoidance but there were no moderation effects.
Prescreener descriptives
SV.check <- data[, c("SubjectID", "SV_1", "SV_2", "SV_3", "SV_4", "ACE_1", "ACE_2", "ACE_3", "ACE.pre", "someASA", "someCSA", "ASAonly", "revictimized", "NeitherASAnorCSA", "SV_1_log", "SV_1.pre", "SV_2_log", "SV_2.pre", "SV_3_log", "SV_3.pre", "SV_4_log", "SV_4.pre" )]
SV.check<- SV.check %>%
mutate(SV1.check = case_when(
SV.check$SV_1_log==TRUE & SV.check$SV_1.pre==1~ 0,
SV.check$SV_1_log==FALSE & SV.check$SV_1.pre==0~ 0,
SV.check$SV_1_log==TRUE & SV.check$SV_1.pre==0~ 1,
SV.check$SV_1_log==FALSE & SV.check$SV_1.pre==1~ 1
))
SV.check<- SV.check %>%
mutate(SV2.check = case_when(
SV.check$SV_2_log==TRUE & SV.check$SV_2.pre==1~ 0,
SV.check$SV_2_log==FALSE & SV.check$SV_2.pre==0~ 0,
SV.check$SV_2_log==TRUE & SV.check$SV_2.pre==0~ 1,
SV.check$SV_2_log==FALSE & SV.check$SV_2.pre==1~ 1
))
sum(SV.check$SV2.check)## [1] 74SV.check$someASA.pre<- SV.check$SV_1.pre | SV.check$SV_2.pre | SV.check$SV_3.pre | SV.check$SV_4.pre
SV.check<- SV.check %>%
mutate(ASA.check = case_when(
SV.check$someASA==TRUE & SV.check$someASA.pre==TRUE~ 0,
SV.check$someASA==FALSE & SV.check$someASA.pre==FALSE~ 0,
SV.check$someASA==TRUE & SV.check$someASA.pre==FALSE~ 1,
SV.check$someASA==FALSE & SV.check$someASA.pre==TRUE~ 2
))
table(SV.check$ASA.check)##
## 0 1 2
## 533 14 19SV.check<- SV.check %>%
mutate(CSA.check = case_when(
SV.check$someCSA==TRUE & SV.check$ACE.pre==1~ 0,
SV.check$someCSA==FALSE & SV.check$ACE.pre==0~ 0,
SV.check$someCSA==TRUE & SV.check$ACE.pre==0~ 1,
SV.check$someCSA==FALSE & SV.check$ACE.pre==1~ 2
))
table(SV.check$CSA.check)##
## 0 1 2
## 538 14 14sum(SV.check$ASA.check)## [1] 52sum(SV.check$CSA.check)## [1] 42data<- cbind(data, SV.check$ASA.check, SV.check$CSA.check)
discrepant<- subset(data, SV.check$ASA.check==1 | SV.check$CSA.check==1) #no glaring oddities for people with inconsistent responses between surveysCoding “Other” Responses
Other.responses <- data[, c("SubjectID", "Gender_2", "Gender_3", "Gender_4", "Gender_5", "Gender_6", "Gender_7", "Gender_7_TEXT", "Orientation", "Orientation_4_TEXT", "SV_perp_1", "SV_perp_2", "SV_perp_3", "SV_perp_4", "SV_perp_5", "SV_perp_6", "SV_perp_7", "SV_perp_8", "SV_perp_8_TEXT")]
write_xlsx(Other.responses, "~/Downloads/RRSE.Other.responses.xlsx") #function that exports this dataframe to a xcel file