Planned missing data designs with small sample sizes: How small is too small?

Fan, Junwei; Moore, E. Whitney G.; Kinai, Richard; Crowe, Kelly S.; Schoemann, Alexander M.; Little, Todd D.

doi:10.1177/0165025414531095

Cited by 39 publications

(39 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Missing data were handled using Full Information Maximum Likelihood (FIML) while fitting confirmatory factor analyses and structural equation models. FIML has been shown to perform better at smaller sample sizes than multiple imputation (Jia et al 2014). The Little MCAR test conducted using R package BaylorEdPsych showed that the data were missing completely at random (MCAR), X 2 (1913) = 1196.204, p = .09.…”

Section: Methodsmentioning

confidence: 99%

Relations between Parent Psychological Control and Parent and Adolescent Social Aggression

2018

View full text Add to dashboard Cite

Parent-child interactions and parenting behavior may be related to social aggression among adolescents, and adolescents’ social aggression may relate to parents’ social aggression. This study investigated (a) whether parent psychological control predicted future adolescent and parent social aggression in their own peer relationships, (b) whether parents’ social aggression was related to their use of psychological control within the parent-adolescent relationship (c) whether adolescents’ and parents’ social aggression was associated with changes in each other’s social aggression over time, and (d) change in psychological control. Participants were 174 racially/ethnically diverse parent-adolescent dyads assessed longitudinally for four years. Adolescents were approximately 15-years-old at the first time point. The adolescent sample was 52% girls and 56% identified as White, 22% as Black or African American, 16% as Hispanic, and 5% as mixed race/ ethnicity. Ten percent of the parent participants were fathers. Parents self-reported their psychological control and social aggression, and their adolescents’ teachers reported adolescents’ social aggression. Hypotheses were tested using longitudinal structural equation modeling and a latent growth curve analysis. The hypothesized effect of parent’s psychological control on parent’s future aggression with their own peers was supported. Psychological control positively predicted parent aggression from T2 to T3 (β = .28, p < .05) and from T3 to T4 (β = .37, p < .05). Other hypotheses were not supported. The findings suggest that the parent-child relationship may influence the parent’s functioning in their own peer relationships. Parents’ peer relations seem to have important implications for their own wellbeing and the parent-child relationship.

show abstract

Section: Methodsmentioning

confidence: 99%

Relations between Parent Psychological Control and Parent and Adolescent Social Aggression

2018

View full text Add to dashboard Cite

show abstract

“…Recent studies (e.g., Jorgensen, Schoemann, McPherson, Rhemtulla, Wu, & Little, 2013;Jia, Moore, Kinai, Crowe, GROWTH CURVE PM DESIGNS 6 Schoemann, & Little, 2013) have studied the performance of item-level missingness by imposing the 3-form design at each time point in longitudinal panel models, where the predictive relations between constructs over time are of interest (e.g., longitudinal mediation). To our knowledge no studies have examined the effect of item-level planned missingness in the context of latent growth curve models.…”

Section: Item-level Planned Missingnessmentioning

confidence: 99%

Planned missing designs to optimize the efficiency of latent growth parameter estimates

Rhemtulla

Fan

et al. 2014

International Journal of Behavioral Development

Self Cite

View full text Add to dashboard Cite

Planned missing designs to optimize the efficiency of latent growth parameter estimates Rhemtulla, M.; Jia, F.; Wu, W.; Little, T.D. General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. AbstractWe examine the performance of planned missing (PM) designs for correlated latent growth curve models. Using simulated data from a model where latent growth curves are fitted to two constructs over five time points, we apply three kinds of planned missingness. The first is item-level planned missingness using a 3-form design at each wave such that 25% of data are missing. The second is wave-missingness such that each participant is missing up to 2 waves of data. The third combines both forms of missingness. We find that 3-form missingness results in high convergence rates, little parameter estimate or standard error bias, and high efficiency relative to the complete data design for almost all parameter types. In contrast, wave missingness and the combined design result in dramatically lowered efficiency for parameters measuring individual variability in rates of change (e.g., latent slope variances and covariances), and bias in both estimates and standard errors for these same parameters. We conclude that wave missingness should not be used except with large effect sizes and very large N.Keywords: Planned Missing Designs, Latent Growth Curves, Three-Form Design, Wave Missingness, Longitudinal Planned Missingness. GROWTH CURVE PM DESIGNS 3Planned missing data designs allow researchers to reduce the testing burden on participants, leading to higher-quality data with less unplanned missingness and smaller fatigue and practice effects (Harel, Stratton, & Aseltine, 2012). Planned missingness can be applied to many complex models resulting in no added bias and minimal power loss. In the present paper, we apply both 3-form planned missingness, where participants are assigned to miss a subset of items at every time point (Hansen et al., 1988;Graham, Taylor, Olchowski, & Cumsille, 2006;Graham, Hofer, & MacKinnon, 1996;Graham, Hofer, & Piccinin, 1994) and wave missingness, where participants are assigned to miss a subset of measurement occasions (Graham, Taylor, & Cumsille, 2001) to simulated latent growth curve data. We model latent growth curves of two constructs, allowing the interce...

show abstract

“…For longitudinal studies, using a random assignment of form to participant at each measurement occasion also enhances the power of this design. For a detailed simulation studies on the use of the three-form design for longitudinal research, see Jorgensen et al (2014) and Jia et al (2014).…”

Section: Planned Missing Data Designsmentioning

confidence: 99%

Core Principles of Life Course Health Development Methodology and Analytics

Little

2017

Handbook of Life Course Health Development

Self Cite

View full text Add to dashboard Cite

The complexity of life course health development research demands rigorous methodology and analytics. Advanced techniques, improved estimation algorithms, and user-accessible software/hardware developments signal what I see as a paradigm shift in how life science research will be conducted now and in the future. Although techniques such as structural equation modeling (SEM), multilevel modeling (MLM), and mixture distribution modeling have permeated the research agenda in many fields, the capability of these tools has hardly been fully utilized. The craft of life course health development methodology and analytics requires dedication, sophistication, and a knack for molding the analysis tool to the research question at hand. For life course health development research to be at its most effective, utilizing the expertise of collaborative teams is essential. The days of one person being both expert theoretician and methodologist are waning. Partnering with dedicated experts on the various issues that I have outlined here will bring the needed sophistication and knack to execute research at its finest and most impactful levels. In this chapter, I highlight many of the issues that life course health development researchers will need to address and practices to adopt.

show abstract

Planned missing data designs with small sample sizes: How small is too small?

Cited by 39 publications

References 28 publications

Relations between Parent Psychological Control and Parent and Adolescent Social Aggression

Relations between Parent Psychological Control and Parent and Adolescent Social Aggression

Planned missing designs to optimize the efficiency of latent growth parameter estimates

Core Principles of Life Course Health Development Methodology and Analytics

Contact Info

Product

Resources

About