klein2000
Huang En-Li
2022-10-24
1 Introduction
Column 1:GRPID Group Identifier Column 2:JOBSAT Job Satisfaction (DV) Column 3:COHES Cohesion Column 4:POSAFF Positive Affect Column 5:PAY Pay Column 6:NEGLEAD Negative Leadership Column 7:WLOAD Workload Column 8:TASKSIG Task Significance Column 9:PHYSEN Physical Environment
2 Data management
In this section, we load the data set, select variables of interest, and examine the first 6 lines of the data frame object.
# load the data from the package
data(klein2000, package="multilevel")3 Problem 01
#Explore how much variance in job satisfaction can be attributed to their difference in pay (ignoring the other predictors) among individuals belonging to the same group.
# save a copy with only selected variables
dta <- klein2000[, c("GRPID", "JOBSAT", "PAY", "NEGLEAD")]
str(dta)'data.frame': 750 obs. of 4 variables:
$ GRPID : num 1 1 1 1 1 1 1 1 1 1 ...
$ JOBSAT : num -1.717 0.237 -2.098 2.933 0.145 ...
$ PAY : num 0.511 -1.967 -2.749 1.355 -1.185 ...
$ NEGLEAD: num 0.3128 -1.9047 -0.1243 -1.2984 0.0646 ...# show first 6 lines
head(dta) GRPID JOBSAT PAY NEGLEAD
1 1 -1.717411 0.510543 0.3127974
2 1 0.237316 -1.967199 -1.9046836
3 1 -2.097958 -2.748838 -0.1243344
4 1 2.932552 1.355283 -1.2983974
5 1 0.144981 -1.185332 0.0646287
6 1 -1.995481 -0.399117 0.38705733.1 Visualization 01
We draw a scatter diagram of the job satisfaction against pay and add the regression line.
ggplot(dta, aes(PAY, JOBSAT))+
geom_point(alpha=.5)+
stat_smooth(method='lm', formula=y~x, se=TRUE)+
labs(x="Pay",
y="Job Satisfaction")+
theme_minimal()
3.2 Model 01
m0 <- lm(JOBSAT ~1+ PAY, data=dta)
summary(m0)
Call:
lm(formula = JOBSAT ~ 1 + PAY, data = dta)
Residuals:
Min 1Q Median 3Q Max
-6.590 -1.474 0.096 1.442 9.573
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.0382 0.0843 0.45 0.65
PAY 0.9037 0.0837 10.80 <2e-16
Residual standard error: 2.31 on 748 degrees of freedom
Multiple R-squared: 0.135, Adjusted R-squared: 0.134
F-statistic: 117 on 1 and 748 DF, p-value: <2e-16每個人的平均工作滿意度為0.0382，每增加1單位薪資，工作滿意度會增加0.9037
4 Problem 02
#Determine how much variation in job satisfaction between groups can be attributed to differences in (group-level) negative leadership between groups.
library(tidyverse)
klein2000 %>% group_by(GRPID) %>% mutate(mneglead=mean(NEGLEAD)) -> dta# save a copy with only selected variables
dta <- klein2000[, c("GRPID", "JOBSAT", "PAY", "NEGLEAD")]
str(dta)'data.frame': 750 obs. of 4 variables:
$ GRPID : num 1 1 1 1 1 1 1 1 1 1 ...
$ JOBSAT : num -1.717 0.237 -2.098 2.933 0.145 ...
$ PAY : num 0.511 -1.967 -2.749 1.355 -1.185 ...
$ NEGLEAD: num 0.3128 -1.9047 -0.1243 -1.2984 0.0646 ...# show first 6 lines
head(dta) GRPID JOBSAT PAY NEGLEAD
1 1 -1.717411 0.510543 0.3127974
2 1 0.237316 -1.967199 -1.9046836
3 1 -2.097958 -2.748838 -0.1243344
4 1 2.932552 1.355283 -1.2983974
5 1 0.144981 -1.185332 0.0646287
6 1 -1.995481 -0.399117 0.38705734.1 Visualization 02
We draw a scatter diagram of the job satisfaction against pay and add the regression line by group identifier.
ggplot(dta, aes(PAY, JOBSAT, group=GRPID))+
geom_point(alpha=.5)+
stat_smooth(method='lm', formula=y~x, se=FALSE,
col=1, size=rel(.5))+
labs(x="Pay",
y="Job Satisfaction",
subtitle='Group Identifier')+
theme_minimal()
4.2 Model 02
m1<-lme4::lmer(JOBSAT ~1+ PAY+ (1|GRPID), data=dta)
summary(m1)Linear mixed model fit by REML ['lmerMod']
Formula: JOBSAT ~ 1 + PAY + (1 | GRPID)
Data: dta
REML criterion at convergence: 3327.1
Scaled residuals:
Min 1Q Median 3Q Max
-2.919 -0.651 0.034 0.643 3.637
Random effects:
Groups Name Variance Std.Dev.
GRPID (Intercept) 0.835 0.914
Residual 4.505 2.123
Number of obs: 750, groups: GRPID, 50
Fixed effects:
Estimate Std. Error t value
(Intercept) 0.0360 0.1507 0.24
PAY 0.9544 0.0785 12.16
Correlation of Fixed Effects:
(Intr)
PAY -0.022在Groups層次下，每個人的平均工作滿意度為0.036，每增加1單位薪資，工作滿意度會增加0.9544