Assessing person‐centered outcomes in practice research: a latent transition profile framework

Abstract

Advances in statistics provide new methods for analyzing practice data. These advances include person-centered methods (PCMs) that identify subgroups of research participants with similar characteristics. PCMs derive from a frame of reference that is similar to the risk factor perspective in practice. In practice, the delivery of services is often contingent on identifying at-risk populations and then providing interventions to groups based on shared risk profiles. PCMs use this perspective. Moreover, PCMs provide a means for identifying high-risk groups with a precision rarely afforded by routine variable-centered methods. This article describes a latent profile transition analysis (LPTA), one of several PCMs. To demonstrate LPTA, we estimate risk profiles and treatment effects using data from a cohort study of a school-based social skills training program. We define four steps in PCMs analysis, describe key statistical tests, and conclude with a discussion of the strengths and limitations of PCMs for practice research. C

Sensory subtypes in children with autism spectrum disorder: latent profile transition analysis using a national survey of sensory features

“…Fit statistics were examined for two-to five-profile solutions. Five commonly used measures were considered in choosing the appropriate solution: the Akaike Information Criterion (AIC), the Bayesian Information Criterion (BIC), Entropy, a Lo-Mendell Rubin Test (LMR), and a Bootstrap Likelihood Ratio Test (BLRT) (Lo, Mendell, & Rubin, 2001;Nylund, Asparouhov, & Muth en, 2007;Thompson, Macy, & Fraser, 2011). These methods enable the researcher to test the changes in model fit as the number of profiles (e.g.…”

Understanding School Truancy: Risk–Need Latent Profiles of Adolescents

“…Using LPA to identify unobservable subgroups of individuals with distinct patterns of risk behavior expands on previous work (e.g., Maynard, Salas-Wright, Vaughn, & Peters, 2012; Thompson, Macy, & Fraser, 2011) in two key ways. First, the WARNS enables development of profiles from self-reported risk and needs rather than teacher reports (Thompson et al, 2011). Therefore, perspective on the patterns of risks and needs is depicted by the students themselves, rather than a third party.…”

Campus Sexual Assault: Future Directions for Research

“…Recent work by Moynihan et al (2015) demonstrated that a combination of strategies (social marketing and in-person bystander program) sustained better outcomes over a longer period of time than a solo strategy. Further studies are needed to determine how to best sustain positive gains in changing attitudes and behaviors and more advanced analytic designs, to capture the impact of prevention programs over time, such as latent growth curve modeling (Shaw & Janulis, 2016) and person-centered analysis strategies (Swartout & Swartout, 2012; A. M. Thompson, Macy, & Fraser, 2011).…”