Latent Profile Analysis - Amira Usage - MAP Growth
Yanwen Xu
2023-11-20
Getting Packages Ready
library(tidyLPA)## Warning: package 'tidyLPA' was built under R version 4.3.2## You can use the function citation('tidyLPA') to create a citation for the use of {tidyLPA}.
## Mplus is not installed. Use only package = 'mclust' when calling estimate_profiles().library(dplyr)## Warning: package 'dplyr' was built under R version 4.3.2##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(mclust)## Warning: package 'mclust' was built under R version 4.4.0## Package 'mclust' version 6.0.1
## Type 'citation("mclust")' for citing this R package in publications.Import Data
Data <- read.csv("G:\\Shared Data\\Research\\Administration\\Evaluation\\Assessment\\2023 Amira Study\\Data\\Exploratory Analysis\\Latent Profile Analysis - For R.csv")Descriptive on Data
summary(Data)## StudentNumber Gender RaceEthnicity EBFEligibility
## Min. :533644 Min. :0.000 Min. :0.000 Min. :0.0000
## 1st Qu.:591942 1st Qu.:0.000 1st Qu.:1.000 1st Qu.:0.0000
## Median :602289 Median :0.000 Median :1.000 Median :1.0000
## Mean :599838 Mean :0.495 Mean :1.517 Mean :0.7319
## 3rd Qu.:612370 3rd Qu.:1.000 3rd Qu.:2.000 3rd Qu.:1.0000
## Max. :628036 Max. :1.000 Max. :6.000 Max. :1.0000
##
## SpecialEd HighAbilityLearner ConvertedELStatus Grade
## Min. :0.0000 Min. :0.00000 Min. :0.0000 Min. :0.00
## 1st Qu.:0.0000 1st Qu.:0.00000 1st Qu.:0.0000 1st Qu.:0.00
## Median :0.0000 Median :0.00000 Median :0.0000 Median :1.00
## Mean :0.1816 Mean :0.05433 Mean :0.4484 Mean :1.48
## 3rd Qu.:0.0000 3rd Qu.:0.00000 3rd Qu.:1.0000 3rd Qu.:2.00
## Max. :1.0000 Max. :2.00000 Max. :2.0000 Max. :3.00
##
## MAP.School.change..Fall.2022...Spring.2023. MAP.Fall.RIT.Score
## Min. :0.00000 Min. :104.0
## 1st Qu.:0.00000 1st Qu.:143.0
## Median :0.00000 Median :157.0
## Mean :0.01864 Mean :159.7
## 3rd Qu.:0.00000 3rd Qu.:175.0
## Max. :1.00000 Max. :231.0
## NA's :296 NA's :296
## MAP.Winter.RIT.Score MAP.Spring.RIT.Score MAP.Growth
## Min. :114.0 Min. :113.0 Min. :-30.00
## 1st Qu.:150.0 1st Qu.:157.0 1st Qu.: 7.00
## Median :164.0 Median :172.0 Median : 13.00
## Mean :166.5 Mean :173.2 Mean : 13.23
## 3rd Qu.:183.0 3rd Qu.:189.0 3rd Qu.: 19.00
## Max. :236.0 Max. :239.0 Max. : 78.00
## NA's :352 NA's :296
## MAP.Fall.To.Spring.Projected.Growth MAP.Fall.To.Spring.Met.Projected.Growth
## Min. : 4.0 Min. :0.0000
## 1st Qu.:13.0 1st Qu.:0.0000
## Median :15.0 Median :0.0000
## Mean :14.3 Mean :0.4783
## 3rd Qu.:16.0 3rd Qu.:1.0000
## Max. :26.0 Max. :1.0000
## NA's :316 NA's :316
## BOY.Dyslexia.Risk.Category MOY.Dyslexia.Risk.Category
## Min. :0.0000 Min. :0.0000
## 1st Qu.:0.0000 1st Qu.:0.0000
## Median :0.0000 Median :0.0000
## Mean :0.1591 Mean :0.1141
## 3rd Qu.:0.0000 3rd Qu.:0.0000
## Max. :1.0000 Max. :1.0000
## NA's :1771 NA's :2300
## EOY.Dyslexia.Risk.Category EOY.Growth.Rate BOY.ARM
## Min. :0.000 Min. :-2.2714 Min. : -3.593
## 1st Qu.:0.000 1st Qu.: 0.1370 1st Qu.: 4.503
## Median :0.000 Median : 0.4349 Median : 25.187
## Mean :0.099 Mean : 0.4234 Mean : 39.993
## 3rd Qu.:0.000 3rd Qu.: 0.6505 3rd Qu.: 69.204
## Max. :1.000 Max. : 1.8351 Max. :219.558
## NA's :2039 NA's :1735 NA's :1735
## MOY.ARM EOY.ARM ELPA21.Summative.Overall.Scale.Score
## Min. : 0.00 Min. : 0.00 Min. :2934
## 1st Qu.: 14.06 1st Qu.: 16.00 1st Qu.:4947
## Median : 46.51 Median : 57.03 Median :5357
## Mean : 56.04 Mean : 61.85 Mean :5327
## 3rd Qu.: 92.88 3rd Qu.: 99.87 3rd Qu.:5734
## Max. :247.96 Max. :223.39 Max. :7032
## NA's :1735 NA's :1735 NA's :9265
## ELPA.Reading.Scale.Score Amira.Weeks.Lapsed Amira.Average.Weekly.Practice
## Length:14724 Min. : 8.00 Min. : 0.05263
## Class :character 1st Qu.:33.29 1st Qu.: 0.87022
## Mode :character Median :35.43 Median : 2.00000
## Mean :34.27 Mean : 2.69495
## 3rd Qu.:37.00 3rd Qu.: 3.84299
## Max. :40.71 Max. :34.76923
## NA's :143 NA's :143
## Amira.Usage.Met X
## Min. :0.00000 Mode:logical
## 1st Qu.:0.00000 NA's:14724
## Median :0.00000
## Mean :0.09671
## 3rd Qu.:0.00000
## Max. :1.00000
##
## percentage.of.students.who.met.the.district.requirement
## Length:14724
## Class :character
## Mode :character
##
##
##
## Estimation of profiles
prof1 <- Data %>%
select(BOY.ARM, Grade, Amira.Weeks.Lapsed, Amira.Average.Weekly.Practice) %>%
estimate_profiles(1:8)## Warning in estimate_profiles_mclust(df_full, n_profiles, model_numbers, : The mclust algorithm does not allow for missing data. Some rows were omitted from analysis. Consider using Mplus, which accounts for cases with partially missing data, or use a non-parametric single imputation technique prior to analysis, such as the R-package 'missForest'.## Warning:
## One or more analyses resulted in warnings! Examine these analyses carefully: model_1_class_8prof1## tidyLPA analysis using mclust:
##
## Model Classes AIC BIC Entropy prob_min prob_max n_min n_max BLRT_p
## 1 1 306806.85 306866.63 1.00 1.00 1.00 1.00 1.00
## 1 2 299135.37 299232.50 0.85 0.95 0.97 0.50 0.50 0.01
## 1 3 296432.37 296566.86 0.87 0.83 0.97 0.05 0.49 0.01
## 1 4 293005.30 293177.15 0.82 0.83 0.94 0.04 0.43 0.01
## 1 5 286188.17 286397.38 0.82 0.80 0.97 0.04 0.40 0.01
## 1 6 284998.42 285245.00 0.80 0.75 0.97 0.02 0.31 0.01
## 1 7 283527.51 283811.44 0.80 0.76 0.99 0.02 0.26 1.00
## 1 8 282668.07 282989.36 0.82 0.79 0.97 0.01 0.28 0.01Plot of the profiles
plot_profiles(prof1, rawdata=FALSE)
## Decided to look at group 2, 3, and 6
plot_profiles(prof1[[2]], rawdata=FALSE)
plot_profiles(prof1[[3]], rawdata=FALSE)
plot_profiles(prof1[[6]], rawdata=FALSE)
prof2 <- Data %>%
select(BOY.ARM, Grade, Amira.Weeks.Lapsed, Amira.Average.Weekly.Practice) %>%
estimate_profiles(2)## Warning in estimate_profiles_mclust(df_full, n_profiles, model_numbers, : The mclust algorithm does not allow for missing data. Some rows were omitted from analysis. Consider using Mplus, which accounts for cases with partially missing data, or use a non-parametric single imputation technique prior to analysis, such as the R-package 'missForest'.prof2## tidyLPA analysis using mclust:
##
## Model Classes AIC BIC Entropy prob_min prob_max n_min n_max BLRT_p
## 1 2 299134.68 299231.81 0.85 0.95 0.97 0.50 0.50 0.01prof3 <- Data %>%
select( BOY.ARM, Grade, Amira.Weeks.Lapsed, Amira.Average.Weekly.Practice) %>%
estimate_profiles(3)## Warning in estimate_profiles_mclust(df_full, n_profiles, model_numbers, : The mclust algorithm does not allow for missing data. Some rows were omitted from analysis. Consider using Mplus, which accounts for cases with partially missing data, or use a non-parametric single imputation technique prior to analysis, such as the R-package 'missForest'.prof3## tidyLPA analysis using mclust:
##
## Model Classes AIC BIC Entropy prob_min prob_max n_min n_max BLRT_p
## 1 3 291035.55 291170.04 0.91 0.93 0.97 0.04 0.65 0.01prof6 <- Data %>%
select(BOY.ARM, Grade, Amira.Weeks.Lapsed, Amira.Average.Weekly.Practice) %>%
estimate_profiles(6)## Warning in estimate_profiles_mclust(df_full, n_profiles, model_numbers, : The mclust algorithm does not allow for missing data. Some rows were omitted from analysis. Consider using Mplus, which accounts for cases with partially missing data, or use a non-parametric single imputation technique prior to analysis, such as the R-package 'missForest'.prof6## tidyLPA analysis using mclust:
##
## Model Classes AIC BIC Entropy prob_min prob_max n_min n_max BLRT_p
## 1 6 284999.05 285245.63 0.80 0.75 0.97 0.02 0.31 0.01Estimates for the variables
get_estimates(prof2) ## # A tibble: 16 × 8
## Category Parameter Estimate se p Class Model Classes
## <chr> <chr> <dbl> <dbl> <dbl> <int> <dbl> <dbl>
## 1 Means BOY.ARM 64.4 5.04e-1 0 1 1 2
## 2 Means Grade 2.45 7.76e-3 0 1 1 2
## 3 Means Amira.Weeks.Lapsed 35.2 5.24e-2 0 1 1 2
## 4 Means Amira.Average.Weekl… 2.48 2.64e-2 0 1 1 2
## 5 Variances BOY.ARM 1071. 1.68e+1 0 1 1 2
## 6 Variances Grade 0.325 1.75e-3 0 1 1 2
## 7 Variances Amira.Weeks.Lapsed 16.8 4.64e-1 5.35e-287 1 1 2
## 8 Variances Amira.Average.Weekl… 6.23 1.67e-1 3.44e-304 1 1 2
## 9 Means BOY.ARM 16.5 2.62e-1 0 2 1 2
## 10 Means Grade 0.566 7.90e-3 0 2 1 2
## 11 Means Amira.Weeks.Lapsed 34.3 4.82e-2 0 2 1 2
## 12 Means Amira.Average.Weekl… 3.02 3.41e-2 0 2 1 2
## 13 Variances BOY.ARM 1071. 1.68e+1 0 2 1 2
## 14 Variances Grade 0.325 1.75e-3 0 2 1 2
## 15 Variances Amira.Weeks.Lapsed 16.8 4.64e-1 5.35e-287 2 1 2
## 16 Variances Amira.Average.Weekl… 6.23 1.67e-1 3.44e-304 2 1 2estimate3 <- get_estimates(prof3)
print(estimate3, n = 24) ## # A tibble: 24 × 8
## Category Parameter Estimate se p Class Model Classes
## <chr> <chr> <dbl> <dbl> <dbl> <int> <dbl> <dbl>
## 1 Means BOY.ARM 90.6 0.460 0 1 1 3
## 2 Means Grade 2.39 0.0125 0 1 1 3
## 3 Means Amira.Weeks.Lapsed 36.1 0.0379 0 1 1 3
## 4 Means Amira.Average.Weekly… 2.68 0.0327 0 1 1 3
## 5 Variances BOY.ARM 480. 9.70 0 1 1 3
## 6 Variances Grade 0.837 0.0107 0 1 1 3
## 7 Variances Amira.Weeks.Lapsed 6.55 0.0956 0 1 1 3
## 8 Variances Amira.Average.Weekly… 6.22 0.189 1.78e-238 1 1 3
## 9 Means BOY.ARM 16.5 0.208 0 2 1 3
## 10 Means Grade 1.06 0.0126 0 2 1 3
## 11 Means Amira.Weeks.Lapsed 35.1 0.0285 0 2 1 3
## 12 Means Amira.Average.Weekly… 2.88 0.0299 0 2 1 3
## 13 Variances BOY.ARM 480. 9.70 0 2 1 3
## 14 Variances Grade 0.837 0.0107 0 2 1 3
## 15 Variances Amira.Weeks.Lapsed 6.55 0.0956 0 2 1 3
## 16 Variances Amira.Average.Weekly… 6.22 0.189 1.78e-238 2 1 3
## 17 Means BOY.ARM 27.0 1.55 6.44e- 68 3 1 3
## 18 Means Grade 1.46 0.0542 5.21e-159 3 1 3
## 19 Means Amira.Weeks.Lapsed 19.2 0.210 0 3 1 3
## 20 Means Amira.Average.Weekly… 1.46 0.117 3.93e- 36 3 1 3
## 21 Variances BOY.ARM 480. 9.70 0 3 1 3
## 22 Variances Grade 0.837 0.0107 0 3 1 3
## 23 Variances Amira.Weeks.Lapsed 6.55 0.0956 0 3 1 3
## 24 Variances Amira.Average.Weekly… 6.22 0.189 1.78e-238 3 1 3estimate6 <- get_estimates(prof6)
print(estimate6, n = 48) ## # A tibble: 48 × 8
## Category Parameter Estimate se p Class Model Classes
## <chr> <chr> <dbl> <dbl> <dbl> <int> <dbl> <dbl>
## 1 Means BOY.ARM 21.0 8.08e-1 8.96e-149 1 1 6
## 2 Means Grade 1.01 2.66e-2 0 1 1 6
## 3 Means Amira.Weeks.Lapsed 36.7 4.05e-2 0 1 1 6
## 4 Means Amira.Average.Weekl… 5.14 2.67e-1 1.83e- 82 1 1 6
## 5 Variances BOY.ARM 458. 1.12e+1 0 1 1 6
## 6 Variances Grade 0.473 9.08e-3 0 1 1 6
## 7 Variances Amira.Weeks.Lapsed 5.93 1.33e-1 0 1 1 6
## 8 Variances Amira.Average.Weekl… 2.91 1.26e-1 2.90e-117 1 1 6
## 9 Means BOY.ARM 21.0 2.81e+0 7.51e- 14 2 1 6
## 10 Means Grade 0.943 8.85e-2 1.63e- 26 2 1 6
## 11 Means Amira.Weeks.Lapsed 36.4 3.14e-1 0 2 1 6
## 12 Means Amira.Average.Weekl… 12.7 8.67e-1 5.91e- 49 2 1 6
## 13 Variances BOY.ARM 458. 1.12e+1 0 2 1 6
## 14 Variances Grade 0.473 9.08e-3 0 2 1 6
## 15 Variances Amira.Weeks.Lapsed 5.93 1.33e-1 0 2 1 6
## 16 Variances Amira.Average.Weekl… 2.91 1.26e-1 2.90e-117 2 1 6
## 17 Means BOY.ARM 36.0 8.13e-1 0 3 1 6
## 18 Means Grade 2.31 1.94e-2 0 3 1 6
## 19 Means Amira.Weeks.Lapsed 35.2 8.34e-2 0 3 1 6
## 20 Means Amira.Average.Weekl… 1.77 4.36e-2 0 3 1 6
## 21 Variances BOY.ARM 458. 1.12e+1 0 3 1 6
## 22 Variances Grade 0.473 9.08e-3 0 3 1 6
## 23 Variances Amira.Weeks.Lapsed 5.93 1.33e-1 0 3 1 6
## 24 Variances Amira.Average.Weekl… 2.91 1.26e-1 2.90e-117 3 1 6
## 25 Means BOY.ARM 9.95 2.62e-1 0 4 1 6
## 26 Means Grade 0.439 1.49e-2 2.05e-191 4 1 6
## 27 Means Amira.Weeks.Lapsed 34.1 6.99e-2 0 4 1 6
## 28 Means Amira.Average.Weekl… 1.79 4.81e-2 7.71e-304 4 1 6
## 29 Variances BOY.ARM 458. 1.12e+1 0 4 1 6
## 30 Variances Grade 0.473 9.08e-3 0 4 1 6
## 31 Variances Amira.Weeks.Lapsed 5.93 1.33e-1 0 4 1 6
## 32 Variances Amira.Average.Weekl… 2.91 1.26e-1 2.90e-117 4 1 6
## 33 Means BOY.ARM 27.3 1.51e+0 8.01e- 73 5 1 6
## 34 Means Grade 1.46 5.76e-2 2.81e-142 5 1 6
## 35 Means Amira.Weeks.Lapsed 18.9 2.10e-1 0 5 1 6
## 36 Means Amira.Average.Weekl… 1.30 8.93e-2 3.19e- 48 5 1 6
## 37 Variances BOY.ARM 458. 1.12e+1 0 5 1 6
## 38 Variances Grade 0.473 9.08e-3 0 5 1 6
## 39 Variances Amira.Weeks.Lapsed 5.93 1.33e-1 0 5 1 6
## 40 Variances Amira.Average.Weekl… 2.91 1.26e-1 2.90e-117 5 1 6
## 41 Means BOY.ARM 99.4 7.81e-1 0 6 1 6
## 42 Means Grade 2.40 1.62e-2 0 6 1 6
## 43 Means Amira.Weeks.Lapsed 36.2 4.93e-2 0 6 1 6
## 44 Means Amira.Average.Weekl… 2.78 5.05e-2 0 6 1 6
## 45 Variances BOY.ARM 458. 1.12e+1 0 6 1 6
## 46 Variances Grade 0.473 9.08e-3 0 6 1 6
## 47 Variances Amira.Weeks.Lapsed 5.93 1.33e-1 0 6 1 6
## 48 Variances Amira.Average.Weekl… 2.91 1.26e-1 2.90e-117 6 1 6Decided on model 3 - 3 groups
Print out data for subsequent research
Model3 <- get_data(prof3)
#Exporting the Model3 file
write.csv(Model3, "G:\\Shared Data\\Research\\Administration\\Evaluation\\Assessment\\2023 Amira Study\\Data\\Exploratory Analysis\\Latent Group.csv")