Kyle M. Lang

Principled Missing Data Treatments.

We review a number of issues regarding missing data treatments for intervention and prevention researchers. Many of the common missing data practices in prevention research are still, unfortunately, ill-advised (e.g., use of listwise and pairwise deletion, insufficient use of auxiliary variables). Our goal is to promote better practice in the handling of missing data. We review the current state of missing data methodology and recent missing data reporting in prevention research. We describe antiquated, ad hoc missing data treatments and discuss their limitations. We discuss two modern, principled missing data treatments: multiple imputation and full information maximum likelihood, and we offer practical tips on how to best employ these methods in prevention research. The principled missing data treatments that we discuss are couched in terms of how they improve causal and statistical inference in the prevention sciences. Our recommendations are firmly grounded in missing data theory and well-validated statistical principles for handling the missing data issues that are ubiquitous in biosocial and prevention research. We augment our broad survey of missing data analysis with references to more exhaustive resources.

On the Benefits of Latent Variable Modeling for Norming Scales: The Case of the "Supports Intensity Scale-Children's Version".

Structural equation modeling (SEM) is a powerful and flexible analytic tool to model latent constructs and their relations with observed variables and other constructs. SEM applications offer advantages over classical models in dealing with statistical assumptions and in adjusting for measurement error. So far, however, SEM has not been fully used to develop norms of assessments in educational or psychological fields. In this article, we highlighted the norming process of the Supports Intensity Scale – Children’s Version (SIS-C) within the SEM framework, using a recently developed method of identification (i.e., effects-coding method) that estimates latent means and variances in the metric of the observed indicators. The SIS-C norming process involved (a) creating parcels, (b) estimating latent means and standard deviations, (c) computing T scores using obtained latent means and standard deviations, and (d) reporting percentile ranks.

The benefits of latent variable modeling to develop norms for a translated version of a standardized scale

This article demonstrates the use of structural equation modeling to develop norms for a translated version of a standardized scale, the Supports Intensity Scale – Children’s Version (SIS-C). The latent variable norming method proposed is useful when the standardization sample for a translated version is relatively small to derive norms independently but the original standardization sample is larger and more robust. Specifically, we leveraged a large, representative US standardization sample (n = 4,015) to add power and stability to a smaller Spanish (n = 405) standardization sample. Using a series of multiple-group mean and covariance structures confirmatory factor analyses using effects-coded scaling constraints, measurement invariance was tested across (a) Spanish only and (b) both US and Spanish age bands (5–6, 7–8, 9–10, 11–12, 13–14, and 15–16). After establishing measurement invariance across the US and Spain, tests for latent means and variance differences within age-bands were only performed for Spanish data; the latent means and variances in the US sample were freely estimated. The study findings suggest that the information in the US data stabilized the overall model parameters, and the inclusion of the US sample did not influence on the norms of the SIS-C Spanish Translation.

Randomized Trial of PMTO in Foster Care: Six-Month Child Well-Being Outcomes

Objective: This study tested the effectiveness of Parent Management Training, Oregon (PMTO) model on child social–emotional well-being. Methods: Using a randomized controlled design and three measures of social–emotional well-being, the study investigated effectiveness of PMTO with families of children in foster care with serious emotional disturbance (SED). Participants included children (3–16 years) and parents who were randomly assigned to PMTO (n = 461) or services as usual (n = 457). Study condition was known to participants and assessors. Six months after baseline, analysis of covariance models examined the intervention’s overall effect and time interactions using intent-to-treat analysis. Follow-up analyses identified salient predictors of well-being. Results: PMTO demonstrated small but significant positive effects on three primary outcomes: social–emotional functioning (Cohen’s d = .31), problem behaviors (Cohen’s d = .09), and prosocial skills (Cohen’s d = .09). Conclusion: Results suggest that PMTO was effective at improving short-term social–emotional well-being in a high-risk population of children with SED.

Randomized Study of PMTO in Foster Care: Six-Month Parent Outcomes

Objective: This study examined the effects of Parent Management Training, Oregon (PMTO) model on parenting effectiveness and caregiver functioning. Method: Children in foster care with emotional and behavioral problems were randomized to PMTO (n = 461) or services as usual (n = 457) in a nonblinded study design. Using an intent-to-treat approach, analysis of covariance models tested the intervention’s overall effect and time interactions for parenting and caregiver functioning. Additional analyses were conducted to identify significant predictors of outcomes. Results: PMTO did not significantly affect parenting practices; however, positive effects were observed on caregiver functioning in mental health (odds ratio [OR] = 2.01), substance use (OR = 1.67), social supports (OR = 2.37), and readiness for reunification (OR = 1.64). While no time interactions were detected, several child, parent, and case characteristics were associated with improvements in 6-month outcomes. Conclusion: This study extends evidence on PMTO to biological families of children in foster care, including those with older youth.

The Support Needs of Children with Intellectual Disability and Autism: Implications for Supports Planning and Subgroup Classification.

The Supports Intensity Scale—Children’s version (SIS-C) was developed to provide a standardized measure of support needs of children with intellectual disability. Over half of the norming sample had a secondary diagnosis of autism. Using this subset of the sample, we engaged in exploratory analysis to examine the degree to which latent clusters were present in the data, and after identifying these clusters, the degree to which they mapped on the SIS-C standard scores. A four latent class solution provided the best fit to the data. When mapped on SIS-C standard scores, specific patterns of differences were found in life activity domain scores and overall support needs scores. Implications for future research and practice are discussed.

The Self-Determination Inventory–Student Report: Confirming the Factor Structure of a New Measure

The Self-Determination Inventory–Student Report (SDI-SR) was developed to address the need in the field for new, theoretically aligned measures of self-determination. The purpose of this study was to establish the most robust and efficient set of items to assess the self-determination of adolescents with and without disabilities on the SDI-SR. Confirmatory factor analysis (CFA), using mean and covariance structures, was used to evaluate the factor structure of the SDI-SR to inform decisions on scale reduction. The items were tested across 20 groups generated by crossing disability (i.e., no disability, learning disability, intellectual disability, autism spectrum disorders, and other health impairment) and race/ethnicity (i.e., White, Black, Hispanic, and Other) groups. A robust set of 21 items that align closely with their associated constructs were identified. These 21 items showed strong measurement properties, including measurement invariance at the item level across the 20 groups. Implications for future research and practice are discussed.

Getting beyond the Null: Statistical Modeling as an Alternative Framework for Inference in Developmental Science

We describe statistical modeling as a powerful alternative to null hypothesis significance testing (NHST). Modeling supports statistical inference in a fundamentally different way from NHST which can better serve developmental researchers. Modeling requires researchers to fully articulate their beliefs about the processes under study and to communicate that understanding through the structure of a probabilistic model before testing specific hypotheses. Research hypotheses are assessed through estimated parameters of the model and by conducting model comparisons. We conclude the paper with a series of worked examples that highlight the merits of the statistical modeling approach as a tool for scientific inference.

A Comparison of Methods for Creating Multiple Imputations of Nominal Variables

ABSTRACT Many variables that are analyzed by social scientists are nominal in nature. When missing data occur on these variables, optimal recovery of the analysis models parameters is a challenging endeavor. One of the most popular methods to deal with missing nominal data is multiple imputation (MI). This study evaluated the capabilities of five MI methods that can be used to treat incomplete nominal variables: multiple imputation with chained equations (MICE) using polytomous regression as the elementary imputation method; MICE based on classification and regression trees (CART); MICE based on nested logistic regressions; the ranking procedure described by Allison (2002); and a joint modeling approach based on the general location model. We first motivate our inquiry with an applied example and then present the results of a Monte Carlo simulation study that compared the performance of the five imputation methods under conditions of varying sample size, percentage of missing data, and number of nominal response categories. We found that MICE with polytomous regression was the strongest performer while the Allison (2002) ranking procedure and MICE with CART performed poorly in most conditions.

The Application of the VIA Classification of Strengths to Youth With and Without Disabilities

Abstract Considering strengths when planning for postschool outcomes is critically important for all youth, including youth with disabilities, as strengths should guide the identification of meaningful postschool goals. However, there are a limited number of strengths-based assessment tools that have been validated with youth with disabilities. This article reports the results of a pilot study of the application of the short form of the VIA Inventory of Strengths for Youth (VIA—Youth) to secondary students with and without disability labels. Although the VIA-Youth has been studied in youth without disabilities, it has not been applied to youth with disabilities. Similarities in the reliability of the scores were found across youth with and without disabilities. However, students with disabilities tended to score lower on character strengths than students without disabilities. We were unable to replicate, using confirmatory factor analysis, the theoretical structure used to develop the VIA-Youth, although ...