Measurement equivalence across racial/ethnic groups of the mood and feelings questionnaire for childhood depression

Abstract

As research continues to document differences in the prevalence of mental health problems such as depression across racial/ethnic groups, the issue of measurement equivalence becomes increasingly important to address. The Mood and Feelings Questionnaire (MFQ) is a widely used screening tool for child and adolescent depression. This study applied a differential item functioning (DIF) framework to data from a sample of 6th and 8th grade students in the Seattle Public School District (N = 3,593) to investigate the measurement equivalence of the MFQ. Several items in the MFQ were found to have DIF, but this DIF was associated with negligible individual- or group-level impact. These results suggest that differences in MFQ scores across groups are unlikely to be caused by measurement non-equivalence.

Fig. 1

Item category characteristic curves for two items that have DIF presence across racial/ethnic groups. The y-axis represents the probability of symptom endorsement and the x-axis represents the theta, θ, level. The solid black curves represent the item difficulty curves for whites at “not true” to “sometimes” theta level. The dashed black curves represent the item difficulty curve for whites at “sometimes” to “true” theta level. The solid gray curves represent the item category characteristic curves (ICCC) for Asian Americans, African Americans and Hispanic Americans at “not true” to “sometimes” theta level. The dashed gray curves represent the ICCCs for Asian Americans, African Americans and Hispanic Americans at “sometimes” to “true” theta level. In panel A, top left, ICCCs show a significant DIF presence with a change in β coefficient of 14% when comparing Asian Americans to whites for the item, “Felt grumpy and cross with my parents.” The vertical straight gray reference line illustrates that at 0 theta, there is 60% probability that whites endorsed the items versus 31% of Asian Americans. At 20% probability of symptom endorsement when endorsing “sometimes” or “true” on the item, theta was 0.6 for whites but 1.8 for Asian Americans (cf. horizontal gray line). The middle top panel shows significant DIF presence with change in β coefficient of 10% when comparing African Americans to whites. The right top panel illustrates negligible DIF presence change in β coefficient of 3% when comparing Hispanic Americans to whites. All three bottom panels illustrate relatively negligible DIF presence for the item, “Found it difficult to think properly or concentrate.” A White vs. Asian American youth. B White vs. African American youth. C White vs. Hispanic youth

Fig. 2

Individual-level DIF impact for each covariate and for all covariates. The first six box-and-whisker plots delineate individual-level DIF impact associated with each of the six covariates evaluated in turn, while the last plot delineates individual-level DIF impact associated with all the covariates considered here. The values plotted are the differences between the unadjusted IRT score and IRT scores that accounted for DIF associated with each covariate (first six plots) or with multiple covariates (last plot). A difference of 0 (the middle reference line) would mean that DIF made no difference for that person. Large positive values indicate that scores accounting for DIF were higher than scores that ignored DIF, which means that ignoring DIF resulted in underestimates of depression severity. Large negative values indicate that scores accounting for DIF were lower than scores that ignored DIF; thus ignoring DIF resulted in overestimates of depression severity. These box-and-whisker plots are indexed by 1x the median standard error of measurement (SEM) of the MFQ among these participants. Observations outside of ±.3 SEM indicate that a covariate has salient individual-level DIF impact (first six plots) or that the covariates evaluated for multiple sources of DIF considered together have salient individual-level DIF impact (last plot)

Fig. 3

Group-level DIF impact for multiple covariates presented separately by Asian American versus white subgroup. Group-level DIF impact for multiple covariates presented separately by Asian American versus white subgroup. Difference scores were obtained as described in the note to Fig. 2; these are plotted separately for subgroups as indicated in the figure. Vertical reference lines are drawn at 1 standard deviation of the unadjusted IRT score for the largest subgroup. For example, for Asian American vs. white, the standard deviation for whites was 1.0, so vertical lines are drawn in Fig. 3 at 1.0 and −1.0