Skip to main content

Table 1 Summary fit statistics for individual subscales of the PRWEa

From: Rasch analysis of the patient-rated wrist evaluation questionnaire

Analysis

Item fits residual

Person Fit residual

Item-trait interaction

Unidimensionality

PSIa

Mean

SD

Mean

SDa

Chi-square (df)a

P

Per C < 5%a

PAIN SUBSCALE

 Initial

−0.41

1.72

−0.45

0.94

23 (20)

0.49

6.9%

0.89

 Final

−0.53

1.17

−0.44

0.91

18 (20)

0.54

2.5%

0.84

SPECIFIC ACTIVITIES SUBSCALE

 Initial

−0.60

2.13

−0.45

1.07

54 (30)

0.00

4.7%

0.81

 Final

−0.47

1.87

−0.37

1.03

32 (25)

0.04

- (since items were split for DIF)

0.80

USUAL ACTIVITIES SUBSCALE

 Initial

−0.41

1.72

−0.45

0.94

23 (20)

0.28

1.5%

0.78

 Final

−0.43

1.58

−0.45

1.00

35 (25)

0.07

- (since items were split for DIF)

0.81

  1. aSource of misfit to the Rasch model; SD = Standard deviation; df = Degrees of freedom; per C < 5% = proportion of t-tests that were significant at level of significance of 0.05; 95% CI = 95% confidence interval; PSI = Person separation index; PRWE – Patient-Rated Wrist Evaluation
  2. For the data to satisfy Rasch model requirements:
  3. The mean is expected to be approx. Around zero (Can range between ± 2.5);
  4. S.D. should be approx. 1;
  5. Chi-square value is expected to be small and statistically non-significant;
  6. For a measure to be unidimensional per C < 5% should be less than 0.05; if it is higher than 0.05 then look into the lower limit the 95% confidence interval if it is less than 0.05 then the measure is unidimensional
  7. PSI (Person separation index) should be greater than 0.70 for the summary statistics to be reliable;