Skip to main content
Large-scale Assessments in Education

An IERI – International Educational Research Institute Journal

Table 1 Correlation of variables

From: Instructional quality: catalyst or pitfall in educational systems’ aim for high achievement and equity? An answer based on multilevel SEM analyses of TIMSS 2015 data in Flanders (Belgium), Germany, and Norway

 

Flanders

Germany

Norway

(1)

(2)

(3)

(4)

MATH

(1)

(2)

(3)

(4)

MATH

(1)

(2)

(3)

(4)

MATH

Within

 SES

.11***

.02*

− .01

− .13***

.21***

.13***

.06**

.03*

− .15***

.33***

.13***

.03

.02

− .08**

.24***

 LANG (1)

 

.10***

.07**

− .08***

.14***

 

.09***

.02

− .04*

.12***

 

.04

.02

− .06*

.11***

 STUD CM (2)

  

.18***

− .26***

.16***

  

.16***

− .16***

.16***

  

.15***

.05

.17***

 STUD SC (3)

   

.22***

.01

   

.54**

.10***

   

.73***

.00

 STUD CA (4)

    

− .42***

    

− .25***

    

− .16***

 

(5)

(6)

MATH

  

(5)

(6)

MATH

  

(5)

(6)

MATH

  

Between

 CM

.34***

.08

.30***

  

.11

.39***

.38***

  

.53***

.54***

.49***

  

 SC (5)

 

.90***

.03

   

.97***

− .03

   

.83***

.07

  

 CA (6)

  

− .50***

    

.06

    

.08

  
  1. Correlations were estimated in MPlus by means of a model without any confounders, only including the two variables for which a correlation was estimated. In estimating correlations of between level variables, the variables of interest were also modelled at the within level, as suggested by Marsh et al. (2012)
  2. *** p < .001; ** p < .01; * p < .05
Back to article page