Interceptless PA model
head(df.pa[,c("id","exam_num","bc_pa","PE","pos_PE","neg_PE")])
## id exam_num bc_pa PE pos_PE neg_PE
## 1 5002 1 33.648438 1.000000 1 0
## 2 5002 1 19.898438 0.940000 1 0
## 3 5002 1 NA 0.883600 1 0
## 4 5002 1 1.898438 0.830584 1 0
## 5 5002 1 NA 0.780749 1 0
## 6 5002 1 NA 0.733904 1 0
summary(df.pa.lmer)
##
## Call:
## lm(formula = bc_pa ~ 0 + pos_PE + neg_PE, data = df.pa)
##
## Residuals:
## Min 1Q Median 3Q Max
## -68.625 -8.931 0.290 9.409 62.125
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## pos_PE 3.0354 0.3636 8.348 <2e-16 ***
## neg_PE -5.7915 0.4690 -12.348 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 15.33 on 2843 degrees of freedom
## (4403 observations deleted due to missingness)
## Multiple R-squared: 0.07248, Adjusted R-squared: 0.07183
## F-statistic: 111.1 on 2 and 2843 DF, p-value: < 2.2e-16
Contrast: Negative PE - Positive PE = 2.7561259
Interceptless NA model
head(df.na[,c("id","exam_num","bc_na","PE","pos_PE","neg_PE")])
## id exam_num bc_na PE pos_PE neg_PE
## 1 5002 1 -10.77083 1.0000000 1 0
## 2 5002 1 6.56250 0.9600000 1 0
## 3 5002 1 NA 0.9216000 1 0
## 4 5002 1 -12.43750 0.8847360 1 0
## 5 5002 1 NA 0.8493466 1 0
## 6 5002 1 NA 0.8153727 1 0
summary(df.na.lmer)
##
## Call:
## lm(formula = bc_na ~ 0 + pos_PE + neg_PE, data = df.na)
##
## Residuals:
## Min 1Q Median 3Q Max
## -54.366 -11.761 -1.160 9.855 68.840
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## pos_PE -1.3959 0.4081 -3.42 0.000634 ***
## neg_PE 8.4845 0.5264 16.12 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 17.2 on 2843 degrees of freedom
## (4403 observations deleted due to missingness)
## Multiple R-squared: 0.08717, Adjusted R-squared: 0.08652
## F-statistic: 135.7 on 2 and 2843 DF, p-value: < 2.2e-16
Contrast: Negative PE - Positive PE = 7.0886945