Search

A common challenge in developmental research is the amount of incomplete and missing data that occurs from respondents failing to complete tasks or questionnaires, as well as from disengaging from the study (i.e., attrition). This missingness can lead to biases in parameter estimates and, hence, in the interpretation of findings. These biases can be addressed through statistical techniques that adjust for missing data, such as multiple imputation. Although this technique is highly effective, it has not been widely adopted by developmental scientists given barriers such as lack of training or misconceptions about imputation methods and instead utilizing default methods within software like listwise deletion. This manuscript is intended to provide practical guidelines for developmental researchers to follow when examining their data for missingness, making decisions about how to handle that missingness, and reporting the extent of missing data biases and specific multiple imputation procedures in publications.

Latino sexual minority men (LSMM) are affected by HIV and behavioral health disparities. Evidence-based HIV-prevention and behavioral health (BH) services are not sufficiently scaled up to LSMM. The current study identified multilevel barriers and facilitators to LSMM's use of HIV-prevention and BH services. LSMM (N = 290) in South Florida, a US HIV epicenter, completed a battery of measures potentially associated with pre-exposure prophylaxis (PrEP) and BH treatment use. Stochastic search variable selection (SSVS) followed by multiple linear regression analyses identified variables associated with engagement in PrEP and BH treatment. Multilevel determinants of PrEP and BH treatment engagement were identified, with most identified determinants being at the relational level (e.g., stigma, discrimination based on income and immigration status, personal recommendation for treatment). Individual (e.g., knowledge, self-efficacy) and structural (e.g., financial stress) determinants were also identified. Accordingly, modifiable leverage points to enhance the reach of PrEP and BH treatment to LSMM include educating and enhancing the perceived relevance of services, de-stigmatizing and normalizing via peer examples, bolstering self-efficacy, and building trust.

Adequately addressing missing data is a pervasive issue in the social sciences. Failure to correctly address missing data can lead to biased or inefficient estimation of parameters, confidence intervals, and significance tests. Multiple imputation is a statistical technique for handling missing data that involves using existing data to generate multiple datasets of plausible values for missing data that each incorporate random components to reflect their uncertainty. Each dataset is analyzed individually and identically, and parameter estimates are pooled into one set of estimates, variances, and confidence intervals. Although this technique is widely used, there is little consensus on what constitutes best practices in multiple imputation, including with regard to assessing the extent of missing data bias and reporting multiple imputation procedures in publications. This decision tree was crowdsourced at the 2021 annual meeting of the Society for the Improvement of Psychological Science (SIPS) and revised thereafter. This document is intended to provide practical guidelines for researchers to follow when examining their data for missingness and making decisions about how to handle that missingness. We primarily offer recommendations for multiple imputation, but also indicate where the same decisional guidelines are appropriate for other types of missing data procedures such as full imputation maximum likelihood (FIML).

Background: Depression and anxiety disorders are a significant public health concern for youth and their co-occurrence places youth at a higher risk for poor psychosocial outcomes. This comorbidity is often examined through the lens of the common cause model that emphasizes shared latent variables; however, this approach minimizes the role of individual symptoms as meaningful agents contributing to comorbidity in youth. In the present study, we use network analysis to investigate the role of and interactions among individual depression and anxiety symptoms in a treatment-seeking clinical sample. Methods: We estimate regularized partial correlation networks for youth- and parent reported symptoms in a transdiagnostic sample of youth (N= 417, ages 8-18). We examined features of the symptom-level networks such as network stability, centrality, bridge symptoms, and communities. Moreover, we compare the youth- and parent-reported networks to examine informant discrepancies in such features. Results: Results indicate stable networks with disorder-specific clustering, such that symptoms were more interconnected within compared to between disorders. Symptoms related to self-comparison to peers and negative views of the future were most central in both networks. In the youth-reported network, worries for the future and hopelessness bridged the two symptom domains, whereas self-comparison to peers and low self-efficacy bridged depression and anxiety sub-networks in the parent network. Distinct symptom clusters emerged in the parent- and youth-reported networks. Conclusions: Our findings indicate that negative self-evaluations and negative views of the future are influential symptoms in the presentation and co-occurrence of depression and anxiety and as such may be promising targets in the treatment and prevention of depression and anxiety in youth. These findings contribute to a growing body of literature suggesting network analysis is an effective tool for elucidating transdiagnostic symptoms of psychiatric disorders.

Given the importance of education and the role educators play on the outcomes of their students, the current research investigates implicit attitudes among a sample of mathematics educators. Implicit attitude measures are ones which investigate automatic associations between a target social group and a positive or negative feeling. These attitudes are important because they predict behaviors toward members of the target social group (e.g., Cameron, Brown- Iannuzzi, & Payne, 2012; Greenwald, Poehlman, Uhlmann, & Banaji, 2009). The current research also extends beyond previous research by investigating the association between educators’ implicit racial attitudes toward black and white people and students’ trust and respect for their educator. If educators’ implicit attitudes are associated with how they interact with their students, it may be that we can uncover this relationship by asking students about their attitudes toward their educators. To analyze these questions, this study leverages data from the National Study of Learning Mindsets (NSLM), the largest-ever randomized controlled trial of a growth mindset program in the U.S. in K-12 settings, in which a brief online growth mindset program was administered to 9th grade students during the 2015-2016 academic year. The results suggest that educators have a more implicit positive reaction to women than men, and slightly more implicit positivity toward black men than white men. These implicit associations, however, are unrelated to students’ trust and respect for their educator. The findings and implications are discussed.

In this paper we address the problem of selecting important predictors from some larger set of candidate predictors. Standard techniques are limited by lack of power and high false positive rates. A Bayesian variable selection approach used widely in biostatistics, stochastic search variable selection, can be used instead to combat these issues by accounting for uncertainty in the other predictors of the model. In this paper we present Bayesian variable selection to aid researchers facing this common scenario, along with an online application (https://ssvsforpsych.shinyapps.io/ssvsforpsych/) to perform the analysis and visualize the results. Using an application to predict pain ratings, we demonstrate how this approach quickly identifies reliable predictors, even when the set of possible predictors is larger than the sample size. This technique is widely applicable to research questions that may be relatively data-rich, but with limited information or theory to guide variable selection.