Some background

Dataset

1 Street Safety DataSet

1.1 Descriptive information

a. Summary

summary(safe)
##      street        person         sex             age           agediv10    
##  1      : 10   1      :100   Min.   :0.000   Min.   :20.00   Min.   :2.000  
##  2      : 10   2      :100   1st Qu.:0.000   1st Qu.:35.00   1st Qu.:3.500  
##  3      : 10   3      :100   Median :1.000   Median :48.00   Median :4.800  
##  4      : 10   4      :100   Mean   :0.519   Mean   :47.15   Mean   :4.715  
##  5      : 10   5      :100   3rd Qu.:1.000   3rd Qu.:60.00   3rd Qu.:6.000  
##  6      : 10   6      :100   Max.   :1.000   Max.   :72.00   Max.   :7.200  
##  (Other):940   (Other):400                                                  
##     economic          crowded         unsafe     
##  Min.   :-2.4743   Min.   :1.00   Min.   :1.000  
##  1st Qu.:-0.7567   1st Qu.:2.00   1st Qu.:1.000  
##  Median : 0.0923   Median :4.00   Median :1.000  
##  Mean   : 0.0000   Mean   :3.96   Mean   :1.684  
##  3rd Qu.: 0.7191   3rd Qu.:6.00   3rd Qu.:2.000  
##  Max.   : 2.2048   Max.   :7.00   Max.   :3.000  
## 

b. Correlation

1.2 Null Model

a. Model Estimation

## Linear mixed model fit by maximum likelihood . t-tests use Satterthwaite's
##   method [lmerModLmerTest]
## Formula: unsafe ~ 1 + (1 | street)
##    Data: safe
## 
##      AIC      BIC   logLik deviance df.resid 
##   2241.1   2255.8  -1117.6   2235.1      997 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -1.7825 -0.7988 -0.3070  0.6337  2.3612 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  street   (Intercept) 0.1068   0.3268  
##  Residual             0.4873   0.6981  
## Number of obs: 1000, groups:  street, 100
## 
## Fixed effects:
##              Estimate Std. Error        df t value Pr(>|t|)    
## (Intercept)   1.68400    0.03944 100.00000    42.7   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

b. Intraclass Correlation Coefficient

## [1] 17.97724

1.3 Hypothesis Testing

a. Comparison Null model vs Linear Model

## Linear mixed model fit by maximum likelihood . t-tests use Satterthwaite's
##   method [lmerModLmerTest]
## Formula: unsafe ~ 1 + (1 | street)
##    Data: safe
## 
##      AIC      BIC   logLik deviance df.resid 
##   2241.1   2255.8  -1117.6   2235.1      997 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -1.7825 -0.7988 -0.3070  0.6337  2.3612 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  street   (Intercept) 0.1068   0.3268  
##  Residual             0.4873   0.6981  
## Number of obs: 1000, groups:  street, 100
## 
## Fixed effects:
##              Estimate Std. Error        df t value Pr(>|t|)    
## (Intercept)   1.68400    0.03944 100.00000    42.7   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

b. Confidence Interval

## Computing profile confidence intervals ...
##                 2.5 %    97.5 %
## .sig01      0.2650816 0.3992783
## .sigma      0.6670436 0.7316272
## (Intercept) 1.6059524 1.7620476

2. Popularity DataSet

2.1 Descriptive information

a. Summary

summary(popul)
##     student         class          gender           extrav      
##  2      : 100   17     :  26   Min.   :0.0000   Min.   : 1.000  
##  4      : 100   63     :  25   1st Qu.:0.0000   1st Qu.: 4.000  
##  5      : 100   10     :  24   Median :1.0000   Median : 5.000  
##  6      : 100   15     :  24   Mean   :0.5055   Mean   : 5.215  
##  7      : 100   4      :  23   3rd Qu.:1.0000   3rd Qu.: 6.000  
##  8      : 100   21     :  23   Max.   :1.0000   Max.   :10.000  
##  (Other):1400   (Other):1855                                    
##       texp          popular     
##  Min.   : 2.00   Min.   :0.000  
##  1st Qu.: 8.00   1st Qu.:4.100  
##  Median :15.00   Median :5.100  
##  Mean   :14.26   Mean   :5.076  
##  3rd Qu.:20.00   3rd Qu.:6.000  
##  Max.   :25.00   Max.   :9.500  
## 

b. Correlation

table04

2.2 Null Model

a. Model Estimation

## Linear mixed model fit by maximum likelihood . t-tests use Satterthwaite's
##   method [lmerModLmerTest]
## Formula: popular ~ 1 + (1 | class)
##    Data: popul
## 
##      AIC      BIC   logLik deviance df.resid 
##   6333.5   6350.3  -3163.7   6327.5     1997 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -3.5662 -0.6983  0.0021  0.6758  3.3173 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  class    (Intercept) 0.6945   0.8333  
##  Residual             1.2218   1.1053  
## Number of obs: 2000, groups:  class, 100
## 
## Fixed effects:
##             Estimate Std. Error       df t value Pr(>|t|)    
## (Intercept)  5.07786    0.08696 99.90774    58.4   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

b. Intraclass Correlation Coefficient

## [1] 36.24051

2.3 Hypothesis Testing

a. Comparison Null model vs Linear Model

## Linear mixed model fit by maximum likelihood . t-tests use Satterthwaite's
##   method [lmerModLmerTest]
## Formula: popular ~ 1 + (1 | class)
##    Data: popul
## 
##      AIC      BIC   logLik deviance df.resid 
##   6333.5   6350.3  -3163.7   6327.5     1997 
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -3.5662 -0.6983  0.0021  0.6758  3.3173 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  class    (Intercept) 0.6945   0.8333  
##  Residual             1.2218   1.1053  
## Number of obs: 2000, groups:  class, 100
## 
## Fixed effects:
##             Estimate Std. Error       df t value Pr(>|t|)    
## (Intercept)  5.07786    0.08696 99.90774    58.4   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

b. Confidence Interval

## Computing profile confidence intervals ...
##                2.5 %    97.5 %
## .sig01      0.719963 0.9744636
## .sigma      1.071113 1.1414463
## (Intercept) 4.905780 5.2499439