final_report
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.0 ──## ✓ ggplot2 3.3.2 ✓ purrr 0.3.4
## ✓ tibble 3.0.4 ✓ dplyr 1.0.2
## ✓ tidyr 1.1.2 ✓ stringr 1.4.0
## ✓ readr 1.4.0 ✓ forcats 0.5.0## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()require(dplyr)
library(lavaan)## This is lavaan 0.6-7## lavaan is BETA software! Please report any bugs.library(psych)##
## Attaching package: 'psych'## The following object is masked from 'package:lavaan':
##
## cor2cov## The following objects are masked from 'package:ggplot2':
##
## %+%, alpha# reading in the data
df_all <- read_csv("/home/asma/Desktop/CFA_FINAL/Data_new2.csv")##
## ── Column specification ────────────────────────────────────────────────────────
## cols(
## .default = col_double(),
## Majors = col_character(),
## GPA = col_number(),
## `sector work` = col_character(),
## `sector name` = col_character(),
## current_job = col_character(),
## Cur_salar = col_character(),
## Career_day = col_character(),
## Career_day_help = col_character(),
## `start studying again KAU` = col_character(),
## `recommend KAU` = col_character()
## )
## ℹ Use `spec()` for the full column specifications.# taking a quick look
glimpse(df)## function (x, df1, df2, ncp, log = FALSE)In our Analysis we will need the follwoing columns:
#sector_work_class Career_day Career_day_help Relation Cur_salar PA1 PA2 PA3 PA4 PA5 PA6 PA7 PA8 PA9 PA10 PA11 PA12 PA13 PA14 SA1 SA2 SA3 SA4 SA5 SA6 SA7 SA8 SA9 SA10 SA11 SA12 SA13 IS1 IS2 IS3 IS4 IS5 IS6 IS7 IS8 IS9 IS10 IS11 IS12 IS13 IS14 GA1 GA2 GA3 GA4 #Relation is the cuurent job related to your speciality # Let’s change our data to numurical
df_all$sector_work_class<- as.numeric(as.factor(df_all$sector_work_class))
df_all$Career_day<- as.numeric(as.factor(df_all$Career_day))
df_all$Career_day_help<- as.numeric(as.factor(df_all$Career_day_help))
df_all$current_job<- as.numeric(as.factor(df_all$current_job))
df_all$Cur_salar<- as.numeric(as.factor(df_all$Cur_salar)) #Change all the NA values by the median
df_all$sector_work_class[is.na(df_all$sector_work_class)] <- median(df_all$sector_work_class, na.rm=TRUE)
df_all$Career_day[is.na(df_all$Career_day)] <- median(df_all$Career_day, na.rm=TRUE)
df_all$Career_day_help[is.na(df_all$Career_day_help)] <- median(df_all$Career_day_help, na.rm=TRUE)
df_all$current_job[is.na(df_all$current_job)] <- median(df_all$current_job, na.rm=TRUE)
df_all$Cur_salar[is.na(df_all$Cur_salar)] <- median(df_all$Cur_salar, na.rm=TRUE)
df_all## # A tibble: 336 x 66
## Gender Age Nationality Majors `Study program` `year of enroll…
## <dbl> <dbl> <dbl> <chr> <dbl> <dbl>
## 1 2 25 1 Gener… 1 2013
## 2 2 23 1 Gener… 1 2015
## 3 2 24 1 Gener… 1 2015
## 4 2 23 1 Gener… 1 2016
## 5 2 25 1 Gener… 1 2016
## 6 2 24 1 Gener… 1 2015
## 7 2 22 1 Micro… 1 2016
## 8 2 26 2 Gener… 1 2013
## 9 2 24 2 Gener… 1 2016
## 10 2 22 1 Gener… 1 2016
## # … with 326 more rows, and 60 more variables: `graduation year` <dbl>, `study
## # years` <dbl>, `graduated before` <dbl>, GPA <dbl>, Num_courses <dbl>,
## # sector_work_class <dbl>, `sector work` <chr>, `sector name` <chr>,
## # YearJop <dbl>, current_job <dbl>, Cur_salar <dbl>, Career_day <dbl>,
## # Career_day_help <dbl>, PA1 <dbl>, PA2 <dbl>, PA3 <dbl>, PA4 <dbl>,
## # PA5 <dbl>, PA6 <dbl>, PA7 <dbl>, PA8 <dbl>, PA9 <dbl>, PA10 <dbl>,
## # PA11 <dbl>, PA12 <dbl>, PA13 <dbl>, PA14 <dbl>, SA1 <dbl>, SA2 <dbl>,
## # SA3 <dbl>, SA4 <dbl>, SA5 <dbl>, SA6 <dbl>, SA7 <dbl>, SA8 <dbl>,
## # SA9 <dbl>, SA10 <dbl>, SA11 <dbl>, SA12 <dbl>, SA13 <dbl>, IS1 <dbl>,
## # IS2 <dbl>, IS3 <dbl>, IS4 <dbl>, IS5 <dbl>, IS6 <dbl>, IS7 <dbl>,
## # IS8 <dbl>, IS9 <dbl>, IS10 <dbl>, IS11 <dbl>, IS12 <dbl>, IS13 <dbl>,
## # IS14 <dbl>, GA1 <dbl>, GA2 <dbl>, GA3 <dbl>, GA4 <dbl>, `start studying
## # again KAU` <chr>, `recommend KAU` <chr>#Now our Data is clean and we have only the items of our CFA in the dataframe.
2. Confirmatory Factor Analysis CFA
Model 0
#Let’s read only the Data that we need for factor analysis
names(df_all)[names(df_all) == "current_job"] <- "Relation"
df<- df_all
###########################################################################
# Experiment with the factors that have the lowest measure of uniqueness
###########################################################################
m1a<- 'Program Quality =~ PA1+PA2+PA3+PA4+PA5+PA6+PA7+PA8+PA9+PA10+PA11+PA12+PA13+PA14+GA1+GA2+GA3+GA4
Faculty Assistance =~ SA1+SA2+SA3+SA4+SA5+SA6+SA7+SA8+SA9+SA10+SA11+SA12+SA13
Student Satisfaction =~ IS1+IS2+IS3+IS4+IS5+IS6+IS7+IS8+IS9+IS10+IS11+IS12
Current Employmentn=~ sector_work_class+Career_day+Career_day_help+Relation+Cur_salar'
fourfac_a<- cfa(m1a, data=df, std.lv=TRUE)
summary(fourfac_a, fit.measures=TRUE)## lavaan 0.6-7 ended normally after 39 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 102
##
## Used Total
## Number of observations 320 336
##
## Model Test User Model:
##
## Test statistic 3122.535
## Degrees of freedom 1074
## P-value (Chi-square) 0.000
##
## Model Test Baseline Model:
##
## Test statistic 9765.482
## Degrees of freedom 1128
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.763
## Tucker-Lewis Index (TLI) 0.751
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -17808.119
## Loglikelihood unrestricted model (H1) -16246.852
##
## Akaike (AIC) 35820.238
## Bayesian (BIC) 36204.607
## Sample-size adjusted Bayesian (BIC) 35881.081
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.077
## 90 Percent confidence interval - lower 0.074
## 90 Percent confidence interval - upper 0.080
## P-value RMSEA <= 0.05 0.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.075
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality =~
## PA1 0.702 0.048 14.573 0.000
## PA2 0.716 0.047 15.344 0.000
## PA3 0.694 0.047 14.647 0.000
## PA4 0.684 0.048 14.167 0.000
## PA5 0.807 0.053 15.207 0.000
## PA6 0.727 0.052 14.096 0.000
## PA7 0.755 0.049 15.308 0.000
## PA8 0.771 0.059 13.013 0.000
## PA9 0.765 0.050 15.158 0.000
## PA10 0.808 0.058 13.996 0.000
## PA11 0.758 0.053 14.216 0.000
## PA12 0.848 0.058 14.551 0.000
## PA13 0.722 0.058 12.409 0.000
## PA14 0.741 0.078 9.485 0.000
## GA1 0.619 0.048 12.934 0.000
## GA2 0.630 0.056 11.314 0.000
## GA3 0.575 0.051 11.252 0.000
## GA4 0.635 0.065 9.788 0.000
## FacultyAssistance =~
## SA1 0.775 0.048 16.256 0.000
## SA2 0.810 0.046 17.486 0.000
## SA3 0.841 0.047 17.822 0.000
## SA4 0.711 0.047 15.225 0.000
## SA5 0.723 0.048 15.072 0.000
## SA6 0.695 0.047 14.916 0.000
## SA7 0.699 0.045 15.607 0.000
## SA8 0.744 0.053 14.167 0.000
## SA9 0.698 0.049 14.112 0.000
## SA10 0.651 0.050 13.068 0.000
## SA11 0.663 0.052 12.846 0.000
## SA12 0.773 0.053 14.633 0.000
## SA13 0.802 0.049 16.430 0.000
## StudentSatisfaction =~
## IS1 0.671 0.064 10.446 0.000
## IS2 0.734 0.053 13.898 0.000
## IS3 0.792 0.059 13.461 0.000
## IS4 0.765 0.059 13.027 0.000
## IS5 0.648 0.049 13.110 0.000
## IS6 0.589 0.055 10.709 0.000
## IS7 0.672 0.051 13.078 0.000
## IS8 0.836 0.070 11.920 0.000
## IS9 0.428 0.048 8.833 0.000
## IS10 0.569 0.061 9.382 0.000
## IS11 0.618 0.057 10.876 0.000
## IS12 0.683 0.080 8.540 0.000
## CurrentEmploymentn =~
## sectr_wrk_clss 0.053 0.027 1.938 0.053
## Career_day 0.291 0.050 5.841 0.000
## Career_day_hlp -0.227 0.043 -5.252 0.000
## Relation -0.014 0.028 -0.503 0.615
## Cur_salar -0.042 0.029 -1.448 0.148
##
## Covariances:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality ~~
## FacultyAssstnc 0.721 0.030 23.675 0.000
## StudentStsfctn 0.551 0.044 12.397 0.000
## CrrntEmplymntn 0.264 0.075 3.504 0.000
## FacultyAssistance ~~
## StudentStsfctn 0.664 0.037 17.986 0.000
## CrrntEmplymntn 0.275 0.075 3.657 0.000
## StudentSatisfaction ~~
## CrrntEmplymntn 0.277 0.078 3.573 0.000
##
## Variances:
## Estimate Std.Err z-value P(>|z|)
## .PA1 0.466 0.039 11.856 0.000
## .PA2 0.413 0.035 11.718 0.000
## .PA3 0.448 0.038 11.844 0.000
## .PA4 0.481 0.040 11.920 0.000
## .PA5 0.539 0.046 11.744 0.000
## .PA6 0.552 0.046 11.930 0.000
## .PA7 0.463 0.039 11.725 0.000
## .PA8 0.778 0.064 12.076 0.000
## .PA9 0.490 0.042 11.754 0.000
## .PA10 0.696 0.058 11.945 0.000
## .PA11 0.584 0.049 11.912 0.000
## .PA12 0.683 0.058 11.859 0.000
## .PA13 0.773 0.064 12.145 0.000
## .PA14 1.580 0.128 12.389 0.000
## .GA1 0.509 0.042 12.085 0.000
## .GA2 0.747 0.061 12.251 0.000
## .GA3 0.632 0.052 12.257 0.000
## .GA4 1.080 0.087 12.369 0.000
## .SA1 0.401 0.035 11.585 0.000
## .SA2 0.336 0.030 11.260 0.000
## .SA3 0.336 0.030 11.151 0.000
## .SA4 0.419 0.035 11.792 0.000
## .SA5 0.447 0.038 11.819 0.000
## .SA6 0.427 0.036 11.845 0.000
## .SA7 0.374 0.032 11.721 0.000
## .SA8 0.571 0.048 11.961 0.000
## .SA9 0.508 0.042 11.968 0.000
## .SA10 0.550 0.045 12.102 0.000
## .SA11 0.596 0.049 12.127 0.000
## .SA12 0.560 0.047 11.891 0.000
## .SA13 0.414 0.036 11.545 0.000
## .IS1 0.968 0.081 11.934 0.000
## .IS2 0.543 0.049 11.121 0.000
## .IS3 0.694 0.062 11.257 0.000
## .IS4 0.711 0.062 11.380 0.000
## .IS5 0.500 0.044 11.357 0.000
## .IS6 0.702 0.059 11.889 0.000
## .IS7 0.542 0.048 11.366 0.000
## .IS8 1.076 0.092 11.651 0.000
## .IS9 0.585 0.048 12.168 0.000
## .IS10 0.900 0.074 12.096 0.000
## .IS11 0.744 0.063 11.859 0.000
## .IS12 1.609 0.132 12.204 0.000
## .sectr_wrk_clss 0.137 0.011 12.468 0.000
## .Career_day 0.083 0.027 3.021 0.003
## .Career_day_hlp 0.157 0.020 7.666 0.000
## .Relation 0.150 0.012 12.638 0.000
## .Cur_salar 0.157 0.013 12.552 0.000
## ProgramQuality 1.000
## FacultyAssstnc 1.000
## StudentStsfctn 1.000
## CrrntEmplymntn 1.000Model 1 Estimation and results
m1<- 'Program Quality =~ PA2+PA5+PA7+PA12
Faculty Assistance =~ SA1+SA2+SA3+SA13
Student Satisfaction =~ IS2+IS3+IS4
Current Employment =~ Career_day_help+Relation+Cur_salar'
fourfactor<- cfa(m1, data=df, std.lv=TRUE)
summary(fourfactor, fit.measures=TRUE)## lavaan 0.6-7 ended normally after 35 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 34
##
## Used Total
## Number of observations 320 336
##
## Model Test User Model:
##
## Test statistic 136.042
## Degrees of freedom 71
## P-value (Chi-square) 0.000
##
## Model Test Baseline Model:
##
## Test statistic 1782.022
## Degrees of freedom 91
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.962
## Tucker-Lewis Index (TLI) 0.951
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -4873.445
## Loglikelihood unrestricted model (H1) -4805.424
##
## Akaike (AIC) 9814.890
## Bayesian (BIC) 9943.013
## Sample-size adjusted Bayesian (BIC) 9835.171
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.054
## 90 Percent confidence interval - lower 0.040
## 90 Percent confidence interval - upper 0.067
## P-value RMSEA <= 0.05 0.321
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.049
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality =~
## PA2 0.667 0.050 13.237 0.000
## PA5 0.874 0.054 16.070 0.000
## PA7 0.790 0.051 15.403 0.000
## PA12 0.851 0.061 13.868 0.000
## FacultyAssistance =~
## SA1 0.821 0.048 17.263 0.000
## SA2 0.843 0.047 18.122 0.000
## SA3 0.860 0.048 17.945 0.000
## SA13 0.759 0.051 14.828 0.000
## StudentSatisfaction =~
## IS2 0.792 0.056 14.120 0.000
## IS3 0.859 0.062 13.827 0.000
## IS4 0.833 0.062 13.486 0.000
## CurrentEmployment =~
## Career_day_hlp 0.194 0.086 2.267 0.023
## Relation 0.073 0.041 1.779 0.075
## Cur_salar 0.046 0.041 1.129 0.259
##
## Covariances:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality ~~
## FacultyAssstnc 0.614 0.044 13.992 0.000
## StudentStsfctn 0.359 0.062 5.836 0.000
## CurrntEmplymnt -0.301 0.162 -1.863 0.062
## FacultyAssistance ~~
## StudentStsfctn 0.520 0.052 10.002 0.000
## CurrntEmplymnt -0.393 0.177 -2.219 0.027
## StudentSatisfaction ~~
## CurrntEmplymnt -0.247 0.156 -1.584 0.113
##
## Variances:
## Estimate Std.Err z-value P(>|z|)
## .PA2 0.480 0.045 10.577 0.000
## .PA5 0.426 0.050 8.594 0.000
## .PA7 0.410 0.045 9.198 0.000
## .PA12 0.678 0.066 10.254 0.000
## .SA1 0.328 0.034 9.624 0.000
## .SA2 0.281 0.032 8.902 0.000
## .SA3 0.305 0.034 9.066 0.000
## .SA13 0.481 0.044 10.914 0.000
## .IS2 0.455 0.056 8.103 0.000
## .IS3 0.583 0.069 8.467 0.000
## .IS4 0.603 0.068 8.860 0.000
## .Career_day_hlp 0.171 0.035 4.915 0.000
## .Relation 0.145 0.012 11.575 0.000
## .Cur_salar 0.157 0.013 12.262 0.000
## ProgramQuality 1.000
## FacultyAssstnc 1.000
## StudentStsfctn 1.000
## CurrntEmplymnt 1.000# Model 2 Estimation and results
m2<- 'Program Quality =~ PA1+PA2+PA3+PA4+PA5+PA6+PA7+PA9+PA11+PA12
Faculty Assistance =~ SA1+SA2+SA3+SA13
Student Satisfaction =~ IS1+IS2++IS4+IS5+IS6
Current Employment =~ Career_day_help+Relation+Cur_salar'
fourfactor_1<- cfa(m2, data=df, std.lv=TRUE)
summary(fourfactor_1, fit.measures=TRUE)## lavaan 0.6-7 ended normally after 40 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 50
##
## Used Total
## Number of observations 320 336
##
## Model Test User Model:
##
## Test statistic 555.113
## Degrees of freedom 203
## P-value (Chi-square) 0.000
##
## Model Test Baseline Model:
##
## Test statistic 3607.040
## Degrees of freedom 231
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.896
## Tucker-Lewis Index (TLI) 0.881
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -7829.449
## Loglikelihood unrestricted model (H1) -7551.893
##
## Akaike (AIC) 15758.899
## Bayesian (BIC) 15947.315
## Sample-size adjusted Bayesian (BIC) 15788.724
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.074
## 90 Percent confidence interval - lower 0.066
## 90 Percent confidence interval - upper 0.081
## P-value RMSEA <= 0.05 0.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.056
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality =~
## PA1 0.717 0.048 14.889 0.000
## PA2 0.727 0.047 15.556 0.000
## PA3 0.710 0.047 15.004 0.000
## PA4 0.720 0.048 15.064 0.000
## PA5 0.837 0.053 15.896 0.000
## PA6 0.753 0.051 14.665 0.000
## PA7 0.787 0.049 16.117 0.000
## PA9 0.753 0.051 14.703 0.000
## PA11 0.732 0.054 13.479 0.000
## PA12 0.841 0.059 14.274 0.000
## FacultyAssistance =~
## SA1 0.821 0.047 17.304 0.000
## SA2 0.841 0.047 18.085 0.000
## SA3 0.861 0.048 18.011 0.000
## SA13 0.758 0.051 14.831 0.000
## StudentSatisfaction =~
## IS1 0.627 0.068 9.269 0.000
## IS2 0.637 0.057 11.087 0.000
## IS4 0.720 0.062 11.550 0.000
## IS5 0.750 0.049 15.170 0.000
## IS6 0.719 0.055 13.181 0.000
## CurrentEmployment =~
## Career_day_hlp 0.182 0.079 2.312 0.021
## Relation 0.078 0.042 1.866 0.062
## Cur_salar 0.049 0.041 1.179 0.238
##
## Covariances:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality ~~
## FacultyAssstnc 0.640 0.039 16.426 0.000
## StudentStsfctn 0.562 0.047 11.839 0.000
## CurrntEmplymnt -0.313 0.158 -1.979 0.048
## FacultyAssistance ~~
## StudentStsfctn 0.604 0.046 13.116 0.000
## CurrntEmplymnt -0.402 0.176 -2.292 0.022
## StudentSatisfaction ~~
## CurrntEmplymnt -0.242 0.156 -1.552 0.121
##
## Variances:
## Estimate Std.Err z-value P(>|z|)
## .PA1 0.444 0.039 11.483 0.000
## .PA2 0.397 0.035 11.308 0.000
## .PA3 0.425 0.037 11.455 0.000
## .PA4 0.432 0.038 11.440 0.000
## .PA5 0.490 0.044 11.208 0.000
## .PA6 0.514 0.045 11.537 0.000
## .PA7 0.414 0.037 11.138 0.000
## .PA9 0.509 0.044 11.528 0.000
## .PA11 0.622 0.053 11.781 0.000
## .PA12 0.695 0.060 11.624 0.000
## .SA1 0.327 0.034 9.658 0.000
## .SA2 0.284 0.031 9.019 0.000
## .SA3 0.303 0.033 9.086 0.000
## .SA13 0.481 0.044 10.944 0.000
## .IS1 1.025 0.088 11.633 0.000
## .IS2 0.677 0.061 11.051 0.000
## .IS4 0.778 0.072 10.864 0.000
## .IS5 0.358 0.043 8.409 0.000
## .IS6 0.532 0.053 10.017 0.000
## .Career_day_hlp 0.175 0.031 5.749 0.000
## .Relation 0.144 0.013 11.349 0.000
## .Cur_salar 0.157 0.013 12.193 0.000
## ProgramQuality 1.000
## FacultyAssstnc 1.000
## StudentStsfctn 1.000
## CurrntEmplymnt 1.000# Model 3 Estimation and results
m3<- 'Program Quality =~ PA2+PA5+PA7+PA12
Faculty Assistance =~ SA1+SA2+SA3+SA13
Student Satisfaction =~ IS2+IS4
Current Employment =~ Career_day_help+Relation+Cur_salar'
fourfactor_2<- cfa(m3, data=df, std.lv=TRUE)
summary(fourfactor_2, fit.measures=TRUE)## lavaan 0.6-7 ended normally after 40 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 32
##
## Used Total
## Number of observations 320 336
##
## Model Test User Model:
##
## Test statistic 100.077
## Degrees of freedom 59
## P-value (Chi-square) 0.001
##
## Model Test Baseline Model:
##
## Test statistic 1577.059
## Degrees of freedom 78
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.973
## Tucker-Lewis Index (TLI) 0.964
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -4459.237
## Loglikelihood unrestricted model (H1) -4409.198
##
## Akaike (AIC) 8982.473
## Bayesian (BIC) 9103.060
## Sample-size adjusted Bayesian (BIC) 9001.561
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.047
## 90 Percent confidence interval - lower 0.030
## 90 Percent confidence interval - upper 0.062
## P-value RMSEA <= 0.05 0.620
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.042
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality =~
## PA2 0.668 0.050 13.249 0.000
## PA5 0.875 0.054 16.086 0.000
## PA7 0.790 0.051 15.397 0.000
## PA12 0.850 0.061 13.849 0.000
## FacultyAssistance =~
## SA1 0.820 0.048 17.253 0.000
## SA2 0.843 0.046 18.144 0.000
## SA3 0.861 0.048 17.984 0.000
## SA13 0.757 0.051 14.787 0.000
## StudentSatisfaction =~
## IS2 0.780 0.068 11.402 0.000
## IS4 0.804 0.074 10.915 0.000
## CurrentEmployment =~
## Career_day_hlp 0.189 0.081 2.344 0.019
## Relation 0.072 0.040 1.774 0.076
## Cur_salar 0.054 0.041 1.314 0.189
##
## Covariances:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality ~~
## FacultyAssstnc 0.614 0.044 13.986 0.000
## StudentStsfctn 0.407 0.065 6.284 0.000
## CurrntEmplymnt -0.309 0.161 -1.918 0.055
## FacultyAssistance ~~
## StudentStsfctn 0.579 0.054 10.668 0.000
## CurrntEmplymnt -0.394 0.173 -2.272 0.023
## StudentSatisfaction ~~
## CurrntEmplymnt -0.326 0.175 -1.861 0.063
##
## Variances:
## Estimate Std.Err z-value P(>|z|)
## .PA2 0.479 0.045 10.574 0.000
## .PA5 0.425 0.050 8.582 0.000
## .PA7 0.410 0.045 9.207 0.000
## .PA12 0.680 0.066 10.267 0.000
## .SA1 0.329 0.034 9.645 0.000
## .SA2 0.280 0.031 8.900 0.000
## .SA3 0.303 0.034 9.047 0.000
## .SA13 0.483 0.044 10.936 0.000
## .IS2 0.474 0.083 5.712 0.000
## .IS4 0.649 0.094 6.911 0.000
## .Career_day_hlp 0.173 0.032 5.389 0.000
## .Relation 0.145 0.012 11.616 0.000
## .Cur_salar 0.156 0.013 12.124 0.000
## ProgramQuality 1.000
## FacultyAssstnc 1.000
## StudentStsfctn 1.000
## CurrntEmplymnt 1.000Model 4 Estimation and results
m4<- 'Program Quality =~ PA2+PA5+PA7+PA12
Faculty Assistance =~ SA1+SA2+SA3+SA13
Student Satisfaction =~ IS2+IS3+IS4
Current Employment =~ Career_day+Cur_salar'
fourfactor_3<- cfa(m4, data=df, std.lv=TRUE)
summary(fourfactor_3, fit.measures=TRUE)## lavaan 0.6-7 ended normally after 44 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 32
##
## Used Total
## Number of observations 320 336
##
## Model Test User Model:
##
## Test statistic 134.045
## Degrees of freedom 59
## P-value (Chi-square) 0.000
##
## Model Test Baseline Model:
##
## Test statistic 1789.315
## Degrees of freedom 78
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.956
## Tucker-Lewis Index (TLI) 0.942
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -4683.259
## Loglikelihood unrestricted model (H1) -4616.236
##
## Akaike (AIC) 9430.517
## Bayesian (BIC) 9551.104
## Sample-size adjusted Bayesian (BIC) 9449.605
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.063
## 90 Percent confidence interval - lower 0.049
## 90 Percent confidence interval - upper 0.077
## P-value RMSEA <= 0.05 0.063
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.052
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality =~
## PA2 0.664 0.050 13.166 0.000
## PA5 0.877 0.054 16.166 0.000
## PA7 0.787 0.051 15.326 0.000
## PA12 0.854 0.061 13.932 0.000
## FacultyAssistance =~
## SA1 0.821 0.048 17.253 0.000
## SA2 0.841 0.047 18.054 0.000
## SA3 0.861 0.048 17.983 0.000
## SA13 0.760 0.051 14.862 0.000
## StudentSatisfaction =~
## IS2 0.796 0.056 14.245 0.000
## IS3 0.856 0.062 13.793 0.000
## IS4 0.832 0.062 13.497 0.000
## CurrentEmployment =~
## Career_day 0.217 0.141 1.536 0.125
## Cur_salar -0.042 0.035 -1.204 0.229
##
## Covariances:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality ~~
## FacultyAssstnc 0.614 0.044 13.988 0.000
## StudentStsfctn 0.361 0.061 5.865 0.000
## CurrntEmplymnt 0.374 0.256 1.464 0.143
## FacultyAssistance ~~
## StudentStsfctn 0.520 0.052 10.029 0.000
## CurrntEmplymnt 0.334 0.232 1.437 0.151
## StudentSatisfaction ~~
## CurrntEmplymnt 0.467 0.309 1.512 0.131
##
## Variances:
## Estimate Std.Err z-value P(>|z|)
## .PA2 0.483 0.046 10.621 0.000
## .PA5 0.420 0.049 8.521 0.000
## .PA7 0.415 0.045 9.281 0.000
## .PA12 0.674 0.066 10.231 0.000
## .SA1 0.329 0.034 9.617 0.000
## .SA2 0.284 0.032 8.950 0.000
## .SA3 0.303 0.034 9.015 0.000
## .SA13 0.479 0.044 10.893 0.000
## .IS2 0.449 0.056 8.062 0.000
## .IS3 0.590 0.069 8.607 0.000
## .IS4 0.605 0.068 8.936 0.000
## .Career_day 0.120 0.061 1.965 0.049
## .Cur_salar 0.157 0.013 12.437 0.000
## ProgramQuality 1.000
## FacultyAssstnc 1.000
## StudentStsfctn 1.000
## CurrntEmplymnt 1.000Model 5 Estimation and results
m5<- 'Program Quality =~ PA2+PA5+PA7+PA12
Faculty Assistance =~ SA1+SA2+SA3+SA13
Student Satisfaction =~ IS2+IS3+IS4
Current Employment =~ Relation+Cur_salar'
fourfactor_4<- cfa(m5, data=df, std.lv=TRUE)
summary(fourfactor_4, fit.measures=TRUE)## lavaan 0.6-7 ended normally after 57 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 32
##
## Used Total
## Number of observations 320 336
##
## Model Test User Model:
##
## Test statistic 124.278
## Degrees of freedom 59
## P-value (Chi-square) 0.000
##
## Model Test Baseline Model:
##
## Test statistic 1762.310
## Degrees of freedom 78
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.961
## Tucker-Lewis Index (TLI) 0.949
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -4674.205
## Loglikelihood unrestricted model (H1) -4612.066
##
## Akaike (AIC) 9412.409
## Bayesian (BIC) 9532.996
## Sample-size adjusted Bayesian (BIC) 9431.497
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.059
## 90 Percent confidence interval - lower 0.044
## 90 Percent confidence interval - upper 0.073
## P-value RMSEA <= 0.05 0.151
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.050
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality =~
## PA2 0.667 0.050 13.242 0.000
## PA5 0.874 0.054 16.085 0.000
## PA7 0.791 0.051 15.450 0.000
## PA12 0.849 0.061 13.826 0.000
## FacultyAssistance =~
## SA1 0.821 0.048 17.253 0.000
## SA2 0.841 0.047 18.042 0.000
## SA3 0.861 0.048 17.999 0.000
## SA13 0.760 0.051 14.858 0.000
## StudentSatisfaction =~
## IS2 0.791 0.056 14.094 0.000
## IS3 0.860 0.062 13.840 0.000
## IS4 0.834 0.062 13.498 0.000
## CurrentEmployment =~
## Relation 0.116 0.088 1.326 0.185
## Cur_salar 0.065 0.053 1.240 0.215
##
## Covariances:
## Estimate Std.Err z-value P(>|z|)
## ProgramQuality ~~
## FacultyAssstnc 0.614 0.044 13.974 0.000
## StudentStsfctn 0.358 0.062 5.823 0.000
## CurrntEmplymnt -0.385 0.287 -1.341 0.180
## FacultyAssistance ~~
## StudentStsfctn 0.519 0.052 9.996 0.000
## CurrntEmplymnt -0.173 0.203 -0.855 0.392
## StudentSatisfaction ~~
## CurrntEmplymnt -0.112 0.198 -0.564 0.573
##
## Variances:
## Estimate Std.Err z-value P(>|z|)
## .PA2 0.480 0.045 10.586 0.000
## .PA5 0.426 0.049 8.603 0.000
## .PA7 0.407 0.044 9.179 0.000
## .PA12 0.682 0.066 10.290 0.000
## .SA1 0.329 0.034 9.617 0.000
## .SA2 0.285 0.032 8.960 0.000
## .SA3 0.302 0.034 8.999 0.000
## .SA13 0.479 0.044 10.895 0.000
## .IS2 0.457 0.056 8.132 0.000
## .IS3 0.582 0.069 8.448 0.000
## .IS4 0.601 0.068 8.845 0.000
## .Relation 0.136 0.023 6.039 0.000
## .Cur_salar 0.155 0.014 11.257 0.000
## ProgramQuality 1.000
## FacultyAssstnc 1.000
## StudentStsfctn 1.000
## CurrntEmplymnt 1.000Generate Plot of selected model
#1. Conduct a Path Analysis for all the variables
# create a correlation matrix
#We will use a Factor_Analysis Data with contains only the needed items from our Dataframe
library(semPlot)## Registered S3 methods overwritten by 'huge':
## method from
## plot.sim BDgraph
## print.sim BDgraphdf <- read_csv("/home/asma/Desktop/CFA_FINAL/Factor_Analysis.csv")##
## ── Column specification ────────────────────────────────────────────────────────
## cols(
## .default = col_double()
## )
## ℹ Use `spec()` for the full column specifications.covariance_matrix <- round(cov(df[,]),2)
model <- 'PA1~ PA2+PA3+PA4+PA5+PA6+PA7+PA8+PA9+PA10+PA11+PA12+PA13+PA14+GA1+GA2+GA3+GA4
SA1~ SA2+SA3+SA4+SA5+SA6+SA7+SA8+SA9+SA10+SA11+SA12+SA13
IS1~ IS2+IS3+IS4+IS5+IS6+IS7+IS8+IS9+IS10+IS11+IS12
Cur_salar~ sector_work_class+Career_day+Career_day_help+Relation+Cur_salar'
fit <- cfa(model, data=df, std.lv=TRUE)## Warning in lav_model_vcov(lavmodel = lavmodel, lavsamplestats = lavsamplestats, : lavaan WARNING:
## Could not compute standard errors! The information matrix could
## not be inverted. This may be a symptom that the model is not
## identified.semPaths(fit, intercept = FALSE, whatLabel = "est", residuals = FALSE, exoCov = FALSE, layout = 'tree')
#1. Conduct a Path Analysis for the selected best model
# create a correlation matrix
library(semPlot)
df <- read_csv("/home/asma/Desktop/CFA_FINAL/Factor_Analysis.csv")##
## ── Column specification ────────────────────────────────────────────────────────
## cols(
## .default = col_double()
## )
## ℹ Use `spec()` for the full column specifications.covariance_matrix <- round(cov(df[,]),2)
model <-'PA1~ PA2+PA5+PA7+PA12
SA1~ SA2+SA3+SA13
IS2~ IS3+IS4
Cur_salar~ Career_day+Cur_salar'
fit <- cfa(model, data=df, std.lv=TRUE)## Warning in lav_model_vcov(lavmodel = lavmodel, lavsamplestats = lavsamplestats, : lavaan WARNING:
## Could not compute standard errors! The information matrix could
## not be inverted. This may be a symptom that the model is not
## identified.summary(fit)## lavaan 0.6-7 ended normally after 22 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 18
##
## Number of observations 336
##
## Model Test User Model:
##
## Test statistic 74.607
## Degrees of freedom 32
## P-value (Chi-square) 0.000
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Regressions:
## Estimate Std.Err z-value P(>|z|)
## PA1 ~
## PA2 0.239 NA
## PA5 0.057 NA
## PA7 0.359 NA
## PA12 0.101 NA
## SA1 ~
## SA2 0.318 NA
## SA3 0.290 NA
## SA13 0.233 NA
## IS2 ~
## IS3 0.368 NA
## IS4 0.267 NA
## Cur_salar ~
## Career_day 0.261 NA
## Cur_salar 0.962 NA
##
## Covariances:
## Estimate Std.Err z-value P(>|z|)
## .PA1 ~~
## .SA1 0.047 NA
## .IS2 0.052 NA
## .SA1 ~~
## .IS2 0.000 NA
##
## Variances:
## Estimate Std.Err z-value P(>|z|)
## .PA1 0.483 NA
## .SA1 0.407 NA
## .IS2 0.623 NA
## .Cur_salar 1.201 NAsemPaths(fit, intercept = FALSE, whatLabel = "est", residuals = FALSE, exoCov = FALSE, layout = 'tree')
#2. Conduct a Structure Equation Analysis for the selected best model
library(semPlot)
library(lavaanPlot)
semPaths(fourfactor_3, 'std', layout = 'tree')