Table 5 Self-Efficacy (SE) and Outcome Expectancy (OE) Rating Scale Performance.

From: Rasch analysis in developing a questionnaire to measure self-efficacy beliefs of Omani preservice science teachers for teaching through engineering design

 

Category Label

Observed Count

Observed%

ObservedAverage (Logits)

Infit MNSQ

Outfit MNSQ

Andrich Threshold

Gap (Logits)

Self-Efficacy (SE)

Strongly Disagree

218

2

−0.82

1.44

1.66

NONE

 

Disagree

644

6

−0.15

1.14

1.18

−2.18

0.90

Neutral

2958

28

0.59

0.86

0.86

−1.28

2.05

Agree

4231

40

1.61

0.86

0.85

0.77

1.92

Strongly Agree

2575

24

2.84

1.04

1.03

2.69

 

Outcome Expectancy (OE)

Strongly Disagree

167

2

−0.55

1.39

1.47

NONE

 

Disagree

695

7

−0.20

1.02

1.05

−2.47

1.26

Neutral

2995

31

0.58

0.89

0.92

−1.21

1.99

Agree

3886

40

1.51

0.85

0.85

0.78

2.13

Strongly Agree

1857

19

2.79

1.10

1.08

2.91

 
  1. Table 5 details a range of statistical measures for survey responses: “Observed Count” indicates the number of respondents who selected each category, while “Observed %“ reflects the percentage of total respondents choosing each specific category. The “Observed Average” is described by Linacre (2023) as the mean of the differences between person measures and item difficulties, which are used to model the observed responses in each category. Additionally, “Infit MNSQ” is defined as the average of the INFIT mean-squares linked to the responses in each category, and “Outfit MNSQ” is the average of the OUTFIT mean-squares for responses in each category, both according to Linacre (2023). The “Andrich Threshold” is the calibrated measurement of the transition from one category to the next, and “Gap” signifies the difference between adjacent Andrich thresholds.
Back to article page