Power Analyses for Moderator Effects in Three-Level Cluster Randomized Trials

Dong, Nianbo; Kelcey, Benjamin; Spybrook, Jessaca

doi:10.1080/00220973.2017.1315714

Cited by 43 publications

(52 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The above two steps are no different from the common practice used in designing CRTs. To calculate the sample size for detecting HTE, one could further inflate (or deflate) N OTE by a second design effect, Θ(m), defined in Equation (24). This requires knowledge of the relative effect sizes (RDES), the covariate ICC, the adjusted outcome ICC, and the adjusted total variance.…”

Section: Discussionmentioning

confidence: 99%

“…Of note, model (3) is a direct extension of those studied in Raudenbush, 29 Li et al, 30 and Yang et al, 31 where only main effects of W i and X ij are considered. Further extensions of model (3) to allow for random coefficients for X ij can be found in Jaciw et al 32 and Dong et al 24 To proceed, we let W = E(W i ) denote the proportion of clusters that are randomized to the intervention group. When half of clusters are randomized to the intervention arm, W = 1∕2, but our results allow W to be any value in (0,1).…”

Section: Statistical Modelmentioning

confidence: 99%

See 1 more Smart Citation

Sample size requirements for detecting treatment effect heterogeneity in cluster randomized trials

Yang

Li²,

Starks

et al. 2020

Statistics in Medicine

View full text Add to dashboard Cite

Cluster randomized trials (CRTs) refer to experiments with randomization carried out at the cluster or the group level. While numerous statistical methods have been developed for the design and analysis of CRTs, most of the existing methods focused on testing the overall treatment effect across the population characteristics, with few discussions on the differential treatment effect among subpopulations. In addition, the sample size and power requirements for detecting differential treatment effect in CRTs remain unclear, but are helpful for studies planned with such an objective. In this article, we develop a new sample size formula for detecting treatment effect heterogeneity in two-level CRTs for continuous outcomes, continuous or binary covariates measured at cluster or individual level. We also investigate the roles of two intraclass correlation coefficients (ICCs): the adjusted ICC for the outcome of interest and the marginal ICC for the covariate of interest. We further derive a closed-form design effect formula to facilitate the application of the proposed method, and provide extensions to accommodate multiple covariates. Extensive simulations are carried out to validate the proposed formula in finite samples. We find that the empirical power agrees well with the prediction across a range of parameter constellations, when data are analyzed by a linear mixed effects model with a treatment-by-covariate interaction. Finally, we use data from the HF-ACTION study to illustrate the proposed sample size procedure for detecting heterogeneous treatment effects.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Statistical Modelmentioning

confidence: 99%

Sample size requirements for detecting treatment effect heterogeneity in cluster randomized trials

Yang

Li²,

Starks

et al. 2020

Statistics in Medicine

View full text Add to dashboard Cite

show abstract

“…The purpose of grand-mean centering is to conduct a natural standardization on the coefficients of main effects so that the coefficient estimates for main effects could also serve as the "effect size" to measure the relative strengths of the magnitude of the main and moderation effects (Dong, Kelcey, & Spybrook, 2018). The purpose of grand-mean centering is to conduct a natural standardization on the coefficients of main effects so that the coefficient estimates for main effects could also serve as the "effect size" to measure the relative strengths of the magnitude of the main and moderation effects (Dong, Kelcey, & Spybrook, 2018).…”

Section: Statistical Analysesmentioning

confidence: 99%

“…Grand mean centering was also applied to the dependent variable of teacher victimization. The purpose of grand-mean centering is to conduct a natural standardization on the coefficients of main effects so that the coefficient estimates for main effects could also serve as the "effect size" to measure the relative strengths of the magnitude of the main and moderation effects (Dong, Kelcey, & Spybrook, 2018).…”

Section: Statistical Analysesmentioning

confidence: 99%

Teacher victimization by students in China: A multilevel analysis

Yang

Jenkins

Fredrick

et al. 2018

Aggressive Behavior

View full text Add to dashboard Cite

The goal of this study is to examine individual-level and school-level predictors of teacher victimization (TV) by students in China based upon the multilevel socialecological framework. A sample of 1711 teachers (7-12th grade) from 58 schools from eight provinces in mainland China completed measures of teacher victimization (i.e., physical TV, verbal TV, social TV, cyber TV, sexual harassment, and personal property offenses) by students, school-wide bullying and disciplinary practices, and demographics. In the present sample, 25.1% of teachers reported that they experienced at least one of the six forms of victimization from students in the past school year.Prevalence of teacher victimization ranged from 4.0% (physical victimization) to 16.8% (social victimization). Male teachers were more likely to experience all forms of TV and homeroom teachers were more likely than non-homeroom teachers to experience social TV. Schools with fewer students, but higher number of teachers, also had higher levels of TV. Student bullying and punitive disciplinary practices at the teacher level were associated with higher levels of most forms of TV. Implications of these findings are discussed. K E Y W O R D SChinese schools, disciplinary practice, multilevel, teacher victimization

show abstract

“…We utilized PowerUp! 5 software to determine the required sample size for detecting treatment-by-moderator effects in this trial (Dong et al, 2018;Dong & Maynard, 2013;Spybrook et al, 2016). Accordingly, with a minimum detectable effect size of 0.37 and an alpha level of 0.007, we would still ensure 80 percent power with a sample size of 1500 participants.…”

Section: Sample Sizementioning

confidence: 99%

The impact of a “carry-around” soft commitment device on savings and temptation spending: Evidence from a field experiment in Indian slums

Steinert¹

2018

AEA Randomized Controlled Trials

View full text Add to dashboard Cite

We study the impact of a portable "soft" commitment device on the financial behavior of low-income slum dwellers in urban Pune, India. The portable device may add value to existing commitment designs by activating its binding appeal precisely at the point in time when spending decisions are made. 1650 individuals will be randomly allocated to receive either a zip purse and a lockbox (treatment arm) or a lockbox only (control arm). Both groups are asked to formulate a savings goal and commit to a step-by-step individualized savings plan. We will estimate the causal impact of receiving the portable device on total savings amounts and temptation spending. Further, we will compare study arms with regards to their borrowing activity, expenditure levels, resilience to economic shocks, financial self-efficacy, and female empowerment. Findings from this study can enhance the current understanding of how commitment devices may help alleviate behavioral biases and temptations.

show abstract

Power Analyses for Moderator Effects in Three-Level Cluster Randomized Trials

Cited by 43 publications

References 32 publications

Sample size requirements for detecting treatment effect heterogeneity in cluster randomized trials

Sample size requirements for detecting treatment effect heterogeneity in cluster randomized trials

Teacher victimization by students in China: A multilevel analysis

The impact of a “carry-around” soft commitment device on savings and temptation spending: Evidence from a field experiment in Indian slums

Contact Info

Product

Resources

About