Planned Missingness: How to and How Much?

Planned missingness (PM) can be implemented for survey studies to reduce study length and respondent fatigue. Based on a large sample of Big Five personality data, the present study simulates how factors including PM design (three-form and random percentage [RP]), amount of missingness, and sample size affect the ability of full-information maximum likelihood (FIML) estimation to treat missing data. Results show that although the effectiveness of FIML for treating missing data decreases as sample size decreases and amount of missing data increases, estimates only deviate substantially from truth in extreme conditions. Furthermore, the specific PM design, whether it be a three-form or RP design, makes little difference although the RP design should be easier to implement for computer-based surveys. The examination of specific boundary conditions for the application of PM as paired with FIML techniques has important implications for both the research methods literature and practitioners regularly conducting survey research

Multi-Level Multi-Growth Models: New opportunities for addressing developmental theory using longitudinal designs

Longitudinal models have become increasingly popular in recent years because of their power to test theoretically derived hypotheses by modeling within-person processes with repeated measures. Growth models constitute a flexible framework for modeling a range of complex trajectories across time in outcomes of interest, including non-linearities and time-varying covariates. However, these models have not thus far been expanded to include the effects of multiple growth processes at once on a single outcome. Here, I outline such an extension, showing how multiple growth processes can be modeled as a specific case of the general ability to include time-varying covariates in growth models. I show that this extension of growth models cannot be accomplished by statistical models alone, and that study design plays a crucial role in allowing for proper parameter recovery. I demonstrate these principles through simulations to mimic important theoretical conditions where modeling the effects of multiple growth processes can address developmental theory including, disaggregating the effects of age and practice or treatment in repeated assessments and modeling age- and puberty-related effects during adolescence. I compare how these models behave in two common longitudinal designs, cohort-sequential and accelerated, and how planned missingness in observations is key to parameter recovery. I conclude with directions for future substantive research using the method outlined here.

Consortium on Vulnerability to Externalizing Disorders and Addictions (cVEDA): A developmental cohort study protocol

Background Low and middle-income countries like India with a large youth population experience a different environment from that of high-income countries. The Consortium on Vulnerability to Externalizing Disorders and Addictions (cVEDA), based in India, aims to examine environmental influences on genomic variations, neurodevelopmental trajectories and vulnerability to psychopathology, with a focus on externalizing disorders. Methods cVEDA is a longitudinal cohort study, with planned missingness design for yearly follow-up. Participants have been recruited from multi-site tertiary care mental health settings, local communities, schools and colleges. 10,000 individuals between 6 and 23 years of age, of all genders, representing five geographically, ethnically, and socio-culturally distinct regions in India, and exposures to variations in early life adversity (psychosocial, nutritional, toxic exposures, slum-habitats, socio-political conflicts, urban/rural living, mental illness in the family) have been assessed using age-appropriate instruments to capture socio-demographic information, temperament, environmental exposures, parenting, psychiatric morbidity, and neuropsychological functioning. Blood/saliva and urine samples have been collected for genetic, epigenetic and toxicological (heavy metals, volatile organic compounds) studies. Structural (T1, T2, DTI) and functional (resting state fMRI) MRI brain scans have been performed on approximately 15% of the individuals. All data and biological samples are maintained in a databank and biobank, respectively. Discussion The cVEDA has established the largest neurodevelopmental database in India, comparable to global datasets, with detailed environmental characterization. This should permit identification of environmental and genetic vulnerabilities to psychopathology within a developmental framework. Neuroimaging and neuropsychological data from this study are already yielding insights on brain growth and maturation patterns.

Treatment Effects in Longitudinal Two-Method Planned Missingness Designs: An Application and Tutorial

Planned missing data designs allow researchers to have highly-powered studies by testing only a fraction of the traditional sample size. In two-method planned missingness designs, researchers only assess part of the sample on a high-quality expensive measure, while the entire sample is given a more biased, inexpensive measure. The present study focuses on a longitudinal application of the two-method planned missingness design. We provide evidence of the effectiveness of this design for fitting developmental data. Methodologically, we extend the framework for use within causal modeling. Finally, we provide code and step-by-step instructions for how to analyze data within these frameworks.

Routinely Randomize Potential Sources of Measurement Reactivity to Estimate and Adjust for Biases in Subjective Reports

With the advent of online and app-based studies, researchers in psychology are making increasing use of repeated subjective reports. The new methods open up opportunities to study behavior in the field and to map causal processes, but they also pose new challenges. Recent work has added initial elevation bias to the list of common pitfalls; here, higher negative states (i.e., thoughts and feelings) are reported on the first day of assessment than on later days. This article showcases a new approach to addressing this and other measurement reactivity biases. Specifically, we employed a planned missingness design in a daily diary study of more than 1,300 individuals who were assessed over a period of up to 70 days to estimate and adjust for measurement reactivity biases. We found that day of first item presentation, item order, and item number were associated with only negligible bias: items were not answered differently depending on when and where they were shown. Initial elevation bias may thus be more limited than has previously been reported or it may act only at the level of the survey, not at the item level. We encourage researchers to make design choices that will allow them to routinely assess measurement reactivity biases in their studies. Specifically, we advocate the routine randomization of item display and order, as well as of the timing and frequency of measurement. Randomized planned missingness makes it possible to empirically gauge how fatigue, familiarity, and learning interact to bias responses.