R Notebook

Load Libraries

library(foreign)
library(tidyLPA)

## Warning: package 'tidyLPA' was built under R version 3.4.4

## tidyLPA provides the functionality to carry out Latent Profile Analysis. Note that tidyLPA is still at the beta stage! 
## Please report any bugs at https://github.com/jrosen48/tidyLPA or send an email to jrosen@msu.edu.

library(dplyr)

## Warning: package 'dplyr' was built under R version 3.4.4

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

Load Data

data <- read.spss("data.sav", to.data.frame = T)

## Warning in read.spss("data.sav", to.data.frame = T): data.sav: Very long
## string record(s) found (record type 7, subtype 14), each will be imported
## in consecutive separate variables

#data <- data %>% filter(., Gender == "גבר")   #
        
#"גבר"

#"אישה"

data <- select(data, serial, Abuse, Gender, RTA_factor_1, RTA_factor_2,
               RTA_factor_3, RTA_factor_4, SexVictimization_no1,
               Physical_victimization, SexVictimization_agg_no1,
               Physical_Aggression)

data <- filter(data, Abuse == "reported an abuse")

## Warning: package 'bindrcpp' was built under R version 3.4.4

data <- select(data, RTA_factor_1, RTA_factor_2, RTA_factor_3, RTA_factor_4)

Find Best Fitting Model

compare_solutions(data, RTA_factor_1, RTA_factor_2, RTA_factor_3, RTA_factor_4)

It appears as though model 2 with 2 profiles fits best (as indicated by low BIC). Model 3 with 3 profiles has the lowest BIC but I’m suspicious of it due to large variance.

Look at all stats

model <- estimate_profiles(data,  
                        RTA_factor_1, RTA_factor_2, RTA_factor_3, RTA_factor_4,
                        model = 2,
                        n_profiles = 2,
                        return_orig_df = TRUE)

## Fit varying means, equal variances and covariances (Model 2) model with 2 profiles.

## LogLik is 5493.057

## BIC is 11094.924

## Entropy is 0.925

Model 2 with 2 Profiles

model2 <- estimate_profiles(data, 
                        RTA_factor_1, RTA_factor_2, RTA_factor_3, RTA_factor_4,
                        model = 2,
                        n_profiles = 2,
                        return_orig_df = TRUE)

## Fit varying means, equal variances and covariances (Model 2) model with 2 profiles.

## LogLik is 5493.057

## BIC is 11094.924

## Entropy is 0.925

#write.csv(model6, "dataWithProfiles_maleIncluded.csv", row.names = F)

Entropy is above .8 suggesting distinct profiles.

Estimated Means and Variance for Model 2

y-axis = mean

plot_profiles(model2) #to_center = TRUE, to_scale = TRUE

y-axis = mean centered

plot_profiles(model2, to_center = T) #to_center = TRUE, to_scale = TRUE

y-axis = standard deviation

plot_profiles(model2, to_center = T, to_scale = T) #to_center = TRUE, to_scale = TRUE