Skip to main content

Table 3 PROMIS 28 Thresholds ordering, item location estimates, item fit statistics and item dependency

From: Psychometric performance of the PROMIS® depression item bank: a comparison of the 28- and 51-item versions using Rasch measurement theory

Item labelDisordered ThresholdsaLocationbFit StatisticsRes. re
EstimateSEFit ResidualChi-Square
26I felt disappointed in myself −0.680.06−0.1816.89≤0.30
31I felt discouraged about the future −0.640.061.3611.02≤0.30
36I felt unhappy −0.620.06−4.8626.40≤0.30
46I felt pessimistic −0.580.062.1511.32≤0.30
17I felt sad −0.570.06−1.0515.03≤0.30
54I felt emotionally exhausted −0.520.06−0.26.74≤0.30
23I had trouble feeling close to people −0.460.061.869.27≤0.30
28I felt lonely −0.440.062.369.85≤0.30
29I felt depressed −0.230.06−3.6316.81≤0.30
7I withdrew from other people −≤0.30
35I found that things in my life −≤0.30
42I felt ignored by people −≤0.30
30I had trouble with decisions −≤0.30
50I felt guilty −≤0.30
21I felt that I was to blame for −0.010.06−0.1910.5≤0.30
48I felt that my life was emptyX−0.010.06−0.5915.51≤0.30
14I felt that I was not as good as other people≤0.30
22I felt like a failure 0.060.06−3.716.23≤0.30
27I felt that I was not needed 0.190.06−0.265.08≤0.30
5I felt that I had nothing to look forward to 0.310.06−1.5112.54≤0.30
45I felt that nothing was interesting 0.310.061.215.06≤0.30
4I felt worthless 0.460.07−2.4422.97≤0.30
6I felt helpless 0.470.06−1.6320.05≤0.30
9I felt that nothing could cheer me up 0.540.07−3.4721.44≤0.30
41I felt hopeless 0.560.07−4.7435.32≤0.30
19I felt that I wanted to give up on everything 0.600.07−1.058.23≤0.30
44I felt upset for no reason 0.700.07−0.2912.75≤0.30
39I felt I had no reason for living 1.160.08−1.2914.79≤0.30
  1. a√: ordered thresholds; X: disordered thresholds; bitem location estimates and standard error (SE); cFit residuals, numbers outside recommended range of −2.5 to + 2.5 are printed in bold; dchi-square value; when statistically significant at ≤0.01 after Bonferroni correction number is printed in bold; emaximum residual correlation (Res. r) for each item when value is > 0.30 rule of thumb the actual number is reported