Skip to main content

Table 9 Significance tests for the means of r WG(J) and AD M(J) for each international non-governmental organization (study four)

From: Organizational readiness for implementing change: a psychometric assessment of a new measure

   

Change commitment (CC)

Change efficacy (CE)

     

Simulation-based

  

Simulation-based

   

Sample values

Percentiles

Sample values

Percentiles

Vignette

CC mean

CE mean

r WG(J)

AD M(J)

0.95 r WG(J)

0.05 AD M(J)

r WG(J)

AD M(J)

0.95 r WG(J)

0.05 AD M(J)

INGO 1

4.54

4.00

0.92

0.60

0.62

0.94

0.83

0.72

0.67

0.95

INGO 2

4.30

4.30

0.90

0.67

0.77

0.79

0.87

0.72

0.75

0.86

INGO 3

4.02

4.35

0.57

0.90

0.79

0.74

0.78

0.82

0.81

0.76

INGO 4

3.37

3.41

0.85

0.69

0.76

0.80

0.91

0.60

0.81

0.79

INGO 5

4.03

4.15

0.81

0.76

0.57

1.00

0.82

0.82

0.61

1.00

INGO 6

3.94

4.39

0.96

0.45

0.70

0.85

0.92

0.52

0.70

0.88

INGO 7

4.08

4.37

0.84

0.69

0.42

1.08

0.89

0.65

0.45

1.09

INGO 8

3.48

3.50

0.67

0.94

0.69

0.89

0.55

1.07

0.70

0.92

INGO 9

3.72

3.98

0.88

0.69

0.73

0.84

0.83

0.81

0.60

0.91

INGO 10

3.94

4.20

0.80

0.79

0.61

0.96

0.82

0.80

0.61

0.99

  1. Note: CC = change commitment scale (4 items); CE = change efficacy scale (5 items). For eight INGOs, the sample values for r WG(J) exceed the 95th percentile values of the empirical distributions derived from 100,000 simulated random samples. Likewise, the sample values for AD M(J) were smaller than the 5th percentile values of the empirical distributions derived from 100,000 simulated random samples. Therefore, these INGOs, we reject the null hypothesis of no agreement based on the uniform (rectangular) distribution. For two INGOs (INGO 3 and INGO 8), sample values for r WG(J) did not exceed the 95th percentile values of the empirical distributions derived from 100,000 simulated random samples. Likewise, the sample values for AD M(J) were not smaller than the 5th percentile values of the empirical distributions derived from 100,000 simulated random samples. Therefore, for these two INGOs, we do not reject the null hypothesis of no agreement based on the uniform (rectangular) distribution.