Skip to main content

An IERI – International Educational Research Institute Journal

Table 5 Multinomial regression of number of lessons per week on PL

From: The standard setting process: validating interpretations of stakeholders

Criterion

Predictors

Model A

Model B

Model C

Model D

OR

SE log odds

OR

SE log odds

OR

SE log odds

OR

SE log odds

0 vs. 1

Science

0.65

0.31

0.72

0.30

0.72

0.30

0.77

0.31

Mathematics

  

0.83

0.11

0.84

0.11

0.93

0.07

German

    

0.98

0.07

1.00

0.06

School type

      

1.49

0.35

1 vs. 2

Science

0.58*

0.15

0.62*

0.15

0.63*

0.15

0.66*

0.15

Mathematics

  

0.73*

0.05

0.73*

0.04

0.80*

0.04

German

    

0.95

0.04

1.01

0.04

School type

      

1.54*

0.16

2 vs. 3

Science

0.55*

0.08

0.55*

0.08

0.56*

0.09

0.58*

0.09

Mathematics

  

0.85*

0.04

0.81*

0.04

0.84*

0.05

German

    

0.91

0.06

1.05

0.06

School type

      

1.65*

0.19

3 vs. 4

Science

0.53*

0.12

0.52*

0.12

0.53*

0.13

0.53*

0.14

Mathematics

  

0.86

0.08

0.72

0.23

0.66

0.25

German

    

0.81

0.28

1.00

0.39

School type

      

2.00

0.75

  1. For each cut score, the reference category of the dependent variable PL in science is the consecutive cut score. Mplus does not provide Pseudo R2 for multinomial logistic regressions. The Nagelkerke R2 for the four models calculating manifest regressions in SPSS for all five PVs are: \(R^{2}_{{{\text{Model}}\;{\text{A}}}}\) = 0.17–0.20, \(R^{2}_{{{\text{Model}}\;{\text{B}}}}\) = 0.21–0.23, \(R^{2}_{{{\text{Model}}\;{\text{C}}}}\) = 0.21–0.23, \(R^{2}_{{{\text{Model}}\;{\text{D}}}}\) = 0.22–0.24. OR = odds ratio, SElog odds = standard error of the log odds
  2. p < 0.05