Table 4.
Evaluation of psychometric properties of the HHI total scale, reduced scale and deleted items (N = 167)
| Step | Original HHI Scale | Reduced HHI Scale | Deleted items |
|---|---|---|---|
| (12 items) | (7 items) | (5 items) | |
| (N = 167) | (N = 167) | (N = 167) | |
| Differential item functioning (DIF): Are item difficulty calibrations stable in relation to key demographic variables? | Gender. Item #6 Cohabitation: Item #3 Age: Items #5 and #6 |
None | Gender. Item #6 Cohabitation: Item #3 Age: Items #5 and #6 |
| Rating scale functioning: Does the rating scale function consistently across items? | Acceptable | Acceptable | Acceptable |
| Internal scale validity: Item misfit: How well do the actual item responses match the expected responses from the Rasch model? |
Items #3 to #7 | None | None |
| Unidimensionality (i.e., does the scale measure a single construct?): Variance explained by 1st dimension %: | 48.3% | 52.6% | 52.7% |
| Person-response validity: How well do the individual responses match expected responses from the Rasch model? n (%) | 21 (12.4%) | 9 (5.3%) | 8 (4.7%) |
| Person misfit, n (%) | 4 (2.4%) | 15 (8.9%) | 8 (4.7%) |
| Maximum score, n (%) | None | None | None |
| Minimum score, n (%) | |||
| Person-separation reliability: Can the scale distinguish ≥3 distinct groups of depression in the sample tested? Person-separation index (without extremes) |
1.84 | 1.72 | 0.82 |
| Person reliability: Cronbach’s alpha equivalent | 0.77 | 0.75 | 0.40 |