Testing Associations with Baseline Affect

Prediction ~ Baseline Affect

# baseline affect vs prediction
summary(prediction_PA.lmer)
## Linear mixed model fit by maximum likelihood . t-tests use
##   Satterthwaite's method [lmerModLmerTest]
## Formula: prediction ~ PA_bl_vec + (1 + PA_bl_vec | cohort/ID)
##    Data: denseSum
## 
##      AIC      BIC   logLik deviance df.resid 
##   2774.5   2814.8  -1378.3   2756.5      639 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -4.3218 -0.4213  0.1163  0.5406  2.2139 
## 
## Random effects:
##  Groups    Name        Variance  Std.Dev. Corr 
##  ID:cohort (Intercept) 4.038e+00 2.00956       
##            PA_bl_vec   1.953e-05 0.00442  -1.00
##  cohort    (Intercept) 2.021e+00 1.42165       
##            PA_bl_vec   3.248e-04 0.01802  -1.00
##  Residual              2.341e+00 1.53006       
## Number of obs: 648, groups:  ID:cohort, 247; cohort, 5
## 
## Fixed effects:
##             Estimate Std. Error      df t value Pr(>|t|)    
## (Intercept)  7.10485    0.97545 5.68929   7.284 0.000434 ***
## PA_bl_vec    0.02332    0.01535 7.38811   1.519 0.170220    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Correlation of Fixed Effects:
##           (Intr)
## PA_bl_vec -0.977
## convergence code: 0
## Model failed to converge with max|grad| = 0.00204639 (tol = 0.002, component 1)
summary(prediction_NA.lmer)
## Linear mixed model fit by maximum likelihood . t-tests use
##   Satterthwaite's method [lmerModLmerTest]
## Formula: prediction ~ NA_bl_vec + (1 + NA_bl_vec | cohort/ID)
##    Data: denseSum
## 
##      AIC      BIC   logLik deviance df.resid 
##   2776.2   2816.4  -1379.1   2758.2      639 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -4.3142 -0.4163  0.1014  0.5374  2.2177 
## 
## Random effects:
##  Groups    Name        Variance  Std.Dev. Corr 
##  ID:cohort (Intercept) 2.858e+00 1.690628      
##            NA_bl_vec   4.931e-06 0.002221 1.00 
##  cohort    (Intercept) 1.405e-02 0.118544      
##            NA_bl_vec   2.119e-04 0.014556 -1.00
##  Residual              2.344e+00 1.530947      
## Number of obs: 648, groups:  ID:cohort, 247; cohort, 5
## 
## Fixed effects:
##             Estimate Std. Error       df t value Pr(>|t|)    
## (Intercept)  8.79322    0.36054 46.52093  24.389   <2e-16 ***
## NA_bl_vec   -0.01075    0.01125  6.48805  -0.955    0.374    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Correlation of Fixed Effects:
##           (Intr)
## NA_bl_vec -0.831

Outcome ~ Baseline Affect

# baseline affect vs outcome
summary(outcome_PA.lmer)
## Linear mixed model fit by maximum likelihood . t-tests use
##   Satterthwaite's method [lmerModLmerTest]
## Formula: outcome ~ PA_bl_vec + (1 + PA_bl_vec | cohort/ID)
##    Data: denseSum
## 
##      AIC      BIC   logLik deviance df.resid 
##   2986.5   3026.8  -1484.3   2968.5      639 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -3.9648 -0.4532  0.1407  0.5344  2.3899 
## 
## Random effects:
##  Groups    Name        Variance  Std.Dev. Corr 
##  ID:cohort (Intercept) 2.309e+01 4.80539       
##            PA_bl_vec   4.868e-03 0.06977  -0.88
##  cohort    (Intercept) 2.258e+00 1.50266       
##            PA_bl_vec   3.626e-04 0.01904  -1.00
##  Residual              2.864e+00 1.69237       
## Number of obs: 648, groups:  ID:cohort, 247; cohort, 5
## 
## Fixed effects:
##             Estimate Std. Error      df t value Pr(>|t|)    
## (Intercept)  7.31074    1.23800 6.25526   5.905 0.000902 ***
## PA_bl_vec    0.02104    0.02032 9.07212   1.035 0.327304    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Correlation of Fixed Effects:
##           (Intr)
## PA_bl_vec -0.979
summary(outcome_NA.lmer)
## Linear mixed model fit by maximum likelihood . t-tests use
##   Satterthwaite's method [lmerModLmerTest]
## Formula: outcome ~ NA_bl_vec + (1 + NA_bl_vec | cohort/ID)
##    Data: denseSum
## 
##      AIC      BIC   logLik deviance df.resid 
##   2989.0   3029.2  -1485.5   2971.0      639 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -3.9693 -0.4683  0.1459  0.5446  2.4139 
## 
## Random effects:
##  Groups    Name        Variance  Std.Dev. Corr
##  ID:cohort (Intercept) 4.929e+00 2.220081     
##            NA_bl_vec   3.238e-05 0.005691 1.00
##  cohort    (Intercept) 6.162e-02 0.248224     
##            NA_bl_vec   1.711e-05 0.004137 1.00
##  Residual              2.872e+00 1.694806     
## Number of obs: 648, groups:  ID:cohort, 247; cohort, 5
## 
## Fixed effects:
##             Estimate Std. Error       df t value Pr(>|t|)    
## (Intercept)  8.88876    0.47772 25.74872  18.607   <2e-16 ***
## NA_bl_vec   -0.01248    0.01202 63.49931  -1.038    0.303    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Correlation of Fixed Effects:
##           (Intr)
## NA_bl_vec -0.853

PE ~ Baseline Affect

# baseline affect vs PE
summary(PE_PA.lmer)
## Linear mixed model fit by maximum likelihood . t-tests use
##   Satterthwaite's method [lmerModLmerTest]
## Formula: PE ~ PA_bl_vec + (1 + PA_bl_vec | cohort/ID)
##    Data: denseSum
## 
##      AIC      BIC   logLik deviance df.resid 
##   2892.9   2933.2  -1437.5   2874.9      639 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -3.4341 -0.4620  0.0697  0.5784  2.6101 
## 
## Random effects:
##  Groups    Name        Variance  Std.Dev. Corr 
##  ID:cohort (Intercept) 1.265e+01 3.55616       
##            PA_bl_vec   4.051e-03 0.06365  -0.94
##  cohort    (Intercept) 5.619e-01 0.74959       
##            PA_bl_vec   1.034e-04 0.01017  -1.00
##  Residual              3.583e+00 1.89277       
## Number of obs: 648, groups:  ID:cohort, 247; cohort, 5
## 
## Fixed effects:
##              Estimate Std. Error        df t value Pr(>|t|)
## (Intercept)  0.486863   0.809545  7.658190   0.601    0.565
## PA_bl_vec   -0.006578   0.014101 11.308952  -0.467    0.650
## 
## Correlation of Fixed Effects:
##           (Intr)
## PA_bl_vec -0.984
## convergence code: 0
## Model failed to converge with max|grad| = 0.00201958 (tol = 0.002, component 1)
## Model is nearly unidentifiable: very large eigenvalue
##  - Rescale variables?
summary(PE_NA.lmer)
## Linear mixed model fit by maximum likelihood . t-tests use
##   Satterthwaite's method [lmerModLmerTest]
## Formula: PE ~ NA_bl_vec + (1 + NA_bl_vec | cohort/ID)
##    Data: denseSum
## 
##      AIC      BIC   logLik deviance df.resid 
##   2898.5   2938.7  -1440.2   2880.5      639 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -3.3876 -0.4660  0.0716  0.5756  2.6926 
## 
## Random effects:
##  Groups    Name        Variance  Std.Dev. Corr 
##  ID:cohort (Intercept) 1.941e+00 1.393220      
##            NA_bl_vec   8.585e-08 0.000293 1.00 
##  cohort    (Intercept) 6.890e-03 0.083006      
##            NA_bl_vec   5.529e-05 0.007436 -1.00
##  Residual              3.566e+00 1.888275      
## Number of obs: 648, groups:  ID:cohort, 247; cohort, 5
## 
## Fixed effects:
##              Estimate Std. Error        df t value Pr(>|t|)
## (Intercept)  0.185158   0.328627 64.103379   0.563    0.575
## NA_bl_vec   -0.002163   0.008917 12.373408  -0.243    0.812
## 
## Correlation of Fixed Effects:
##           (Intr)
## NA_bl_vec -0.898
## convergence code: 0
## Model failed to converge with max|grad| = 0.00577369 (tol = 0.002, component 1)

Positive Affect x PE Direction

plot(df.PA.PESplit.gg)

Positive Affect x Letter Grade

plot(df.PA.outcomeSplit.gg)

Negative Affect x PE Direction

plot(df.NA.PESplit.gg)

Negative Affect x Letter Grade

plot(df.NA.outcomeSplit.gg)