split_half_one$cognitive_engagement<-composite_mean_maker(split_half_one, IMPTYOU, IMPTGOAL)
split_half_one$affective_engagement<-composite_mean_maker(split_half_one, ENJOY, INTEREST)
split_half_one$behavioral_engagement<-composite_mean_maker(split_half_one, CONCEN, WRKHARD)
split_half_two$cognitive_engagement<-composite_mean_maker(split_half_two, IMPTYOU, IMPTGOAL)
split_half_two$affective_engagement<-composite_mean_maker(split_half_two, ENJOY, INTEREST)
split_half_two$behavioral_engagement<-composite_mean_maker(split_half_two, CONCEN, WRKHARD)
#imuscle_profs<-compare_models_mplus(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, starts = c(100, 20))
#compare_models_lpa(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement)
#compare_models_lpa(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, prior_control = T)
#imuscle_profs[[1]] %>%
  #gather(key,val, -n_profiles) %>%
  #mutate(val = as.numeric(val)) %>%
  #ggplot(aes(x=n_profiles, y=val, color=key, group=key)) +
  #geom_line() +
  #geom_point()
m1<-create_profiles_mplus(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=6, model=1, return_save_data = T)

plot_profiles_mplus(m1, to_center=TRUE, to_scale = TRUE)

m7<-create_profiles_lpa(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=6, model=1)

plot_profiles_lpa(m7, to_center=TRUE, to_scale = TRUE)

m2<-create_profiles_mplus(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=7, model=1, return_save_data = T)

plot_profiles_mplus(m2, to_center=TRUE, to_scale = TRUE)

m8<-create_profiles_lpa(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=7, model=1)

plot_profiles_lpa(m8, to_center=TRUE, to_scale = TRUE)

m3<-create_profiles_mplus(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=5, model=1, return_save_data = T)

plot_profiles_mplus(m3, to_center=TRUE, to_scale = TRUE)

m8<-create_profiles_lpa(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=5, model=1)

plot_profiles_lpa(m8, to_center=TRUE, to_scale = TRUE)

m4<-create_profiles_mplus(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=4, model=1, return_save_data = T)

plot_profiles_mplus(m4, to_center=TRUE, to_scale = TRUE)

m9<-create_profiles_lpa(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=4, model=1)

plot_profiles_lpa(m9, to_center=TRUE, to_scale = TRUE)

m5<-create_profiles_mplus(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=5, model=2, return_save_data = T)

plot_profiles_mplus(m5, to_center=TRUE, to_scale = TRUE)

Quick interpretation of full esm file profiles: high affective, low all, low cognitive, high all, high cognitive

m5 %>% 
  pluck(2) %>%
  select(COGNITIV: AFFECTIV, C) %>%
  group_by(C) %>%
  summarise_all(mean, na.rm=T)
## # A tibble: 5 x 4
##       C COGNITIV BEHAVIOR AFFECTIV
##   <dbl>    <dbl>    <dbl>    <dbl>
## 1  1.00    0.630    2.26     0.848
## 2  2.00    0.541    0.850    0.705
## 3  3.00    0.715    2.39     2.34 
## 4  4.00    2.30     2.44     2.33 
## 5  5.00    2.32     2.09     0.776
r<-m5 %>% 
  pluck(2)
fit<-manova(cbind(COGNITIV, BEHAVIOR, AFFECTIV) ~ as.factor(r$C), data=r)
summary(fit, test="Pillai")
##                  Df Pillai approx F num Df den Df    Pr(>F)    
## as.factor(r$C)    4 1.7306   2245.2     12  19761 < 2.2e-16 ***
## Residuals      6587                                            
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary.aov(fit)
##  Response COGNITIV :
##                  Df Sum Sq Mean Sq F value    Pr(>F)    
## as.factor(r$C)    4 4684.9 1171.23  4182.2 < 2.2e-16 ***
## Residuals      6587 1844.7    0.28                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##  Response BEHAVIOR :
##                  Df Sum Sq Mean Sq F value    Pr(>F)    
## as.factor(r$C)    4 2359.2  589.80  1809.1 < 2.2e-16 ***
## Residuals      6587 2147.5    0.33                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##  Response AFFECTIV :
##                  Df Sum Sq Mean Sq F value    Pr(>F)    
## as.factor(r$C)    4 3931.6  982.91  3468.6 < 2.2e-16 ***
## Residuals      6587 1866.6    0.28                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
aov1<-aov(COGNITIV ~ C, data = r)
summary(aov1)
##               Df Sum Sq Mean Sq F value Pr(>F)    
## C              1   3727    3727    8763 <2e-16 ***
## Residuals   6590   2803       0                   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
TukeyHSD(aov(r$COGNITIV ~ as.factor(r$C)))
##   Tukey multiple comparisons of means
##     95% family-wise confidence level
## 
## Fit: aov(formula = r$COGNITIV ~ as.factor(r$C))
## 
## $`as.factor(r$C)`
##            diff         lwr         upr     p adj
## 2-1 -0.08868734 -0.15025448 -0.02712020 0.0008149
## 3-1  0.08516891  0.01726861  0.15306921 0.0056378
## 4-1  1.67235761  1.61951124  1.72520397 0.0000000
## 5-1  1.69101713  1.61420510  1.76782916 0.0000000
## 3-2  0.17385625  0.10897093  0.23874157 0.0000000
## 4-2  1.76104495  1.71213286  1.80995703 0.0000000
## 5-2  1.77970447  1.70554423  1.85386471 0.0000000
## 4-3  1.58718870  1.53051128  1.64386612 0.0000000
## 5-3  1.60584822  1.52635181  1.68534463 0.0000000
## 5-4  0.01865952 -0.04843717  0.08575622 0.9422519
aov2<-aov(BEHAVIOR ~ C, data = r)
summary(aov2)
##               Df Sum Sq Mean Sq F value Pr(>F)    
## C              1    489   489.2   802.5 <2e-16 ***
## Residuals   6590   4017     0.6                   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
TukeyHSD(aov(r$BEHAVIOR ~ as.factor(r$C)))
##   Tukey multiple comparisons of means
##     95% family-wise confidence level
## 
## Fit: aov(formula = r$BEHAVIOR ~ as.factor(r$C))
## 
## $`as.factor(r$C)`
##           diff         lwr         upr     p adj
## 2-1 -1.4092005 -1.47562873 -1.34277222 0.0000000
## 3-1  0.1338478  0.06058631  0.20710923 0.0000063
## 4-1  0.1796676  0.12264865  0.23668649 0.0000000
## 5-1 -0.1666407 -0.24951751 -0.08376386 0.0000004
## 3-2  1.5430482  1.47303982  1.61305666 0.0000000
## 4-2  1.5888680  1.53609404  1.64164204 0.0000000
## 5-2  1.2425598  1.16254413  1.32257545 0.0000000
## 4-3  0.0458198 -0.01533266  0.10697226 0.2449271
## 5-3 -0.3004885 -0.38626161 -0.21471530 0.0000000
## 5-4 -0.3463083 -0.41870266 -0.27391385 0.0000000
aov3<-aov(AFFECTIV ~ C, data = r)
summary(aov3)
##               Df Sum Sq Mean Sq F value Pr(>F)    
## C              1   1415  1415.4    2128 <2e-16 ***
## Residuals   6590   4383     0.7                   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
TukeyHSD(aov(r$AFFECTIV ~ as.factor(r$C)))
##   Tukey multiple comparisons of means
##     95% family-wise confidence level
## 
## Fit: aov(formula = r$AFFECTIV ~ as.factor(r$C))
## 
## $`as.factor(r$C)`
##             diff          lwr          upr     p adj
## 2-1 -0.143363157 -0.205294159 -0.081432155 0.0000000
## 3-1  1.488911991  1.420610402  1.557213579 0.0000000
## 4-1  1.482862473  1.429703789  1.536021157 0.0000000
## 5-1 -0.072693761 -0.149959745  0.004572223 0.0766603
## 3-2  1.632275148  1.567006360  1.697543936 0.0000000
## 4-2  1.626225630  1.577024478  1.675426782 0.0000000
## 5-2  0.070669396 -0.003929128  0.145267921 0.0732552
## 4-3 -0.006049518 -0.063061897  0.050962861 0.9984686
## 5-3 -1.561605751 -1.641571981 -1.481639522 0.0000000
## 5-4 -1.555556234 -1.623049467 -1.488063000 0.0000000
m51<-create_profiles_mplus(split_half_one, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=5, model=2, return_save_data = T)

plot_profiles_mplus(m51, to_center=TRUE, to_scale = TRUE)

Quick interpretation of split half profiles: very low all, low all, high all, high affective, high cognitive

m52<-create_profiles_mplus(split_half_two, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=5, model=2, return_save_data = T)

plot_profiles_mplus(m52, to_center=TRUE, to_scale = TRUE)

Quick interpretation of split half two profiles: high affective, low all, low cognitive, high all, high cognitive

m10<-create_profiles_lpa(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=5, model=2)

plot_profiles_lpa(m10, to_center=TRUE, to_scale = TRUE)

m6<-create_profiles_mplus(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=7, model=2, return_save_data = T)

plot_profiles_mplus(m6, to_center=TRUE, to_scale = TRUE)

m11<-create_profiles_lpa(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=7, model=2)

plot_profiles_lpa(m11, to_center=TRUE, to_scale = TRUE)

m12<-create_profiles_mplus(iMUScLE_esm, cognitive_engagement, behavioral_engagement, affective_engagement, n_profiles=4, model=2, return_save_data = T)

plot_profiles_mplus(m12, to_center=TRUE, to_scale = TRUE)