An attempt to lower sources of systematic measurement error using Hierarchical Generalized Linear Modeling (HGLM).

The purpose of the present studies was to test the effects of systematic sources of measurement error on the parameter estimates of scales using the Rasch model. Studies 1 and 2 tested the effects of mood and affectivity. Study 3 evaluated the effects of fatigue. Last, studies 4 and 5 tested the effects of motivation on a number of parameters of the Rasch model (e.g., ability estimates). Results indicated that (a) the parameters of interest and the psychometric properties of the scales were substantially distorted in the presence of all systematic sources of error, and, (b) the use of HGLM provides a way of adjusting the parameter estimates in the presence of these sources of error. It is concluded that validity in measurement requires a thorough evaluation of potential sources of error and appropriate adjustments based on each occasion.

Discrimination, marginalization, belonging, and mental health among Somali immigrants in North America.

In this study, we examined the relationships among discrimination and mental health for Somali young adults, a group at risk for an unfavorable context of reception, and the way in which individual- and community-level factors explain these associations. The present study drew upon data collected during the first wave of the Somali Youth Longitudinal Study, a community-based participatory research project focused on understanding and supporting the healthy development of Somali young adults in four different regions in North America: Boston, MA, Minneapolis, MN, and Portland/Lewiston, ME in the United States and Toronto, Canada. Somali men and women aged 18-30 participated in quantitative interviews that included questions about their health, their neighborhoods, and their thoughts and feelings about their resettlement communities (N = 439). Results indicate that discrimination has a direct effect on worse mental health; this effect was mediated through both individual (marginalized acculturation style) and community-level (sense of belonging) factors. These findings suggest that factors associated with a receiving society's attitudes and behaviors, in addition to its structural supports and constraints, may be particularly important in understanding immigrant mental health. (PsycInfo Database Record (c) 2021 APA, all rights reserved).