Peter Ji1, David L DuBois, Brian R Flay, Vanessa Brechling. 1. Department of Public Health, Institute for Health Research and Policy, University of Illinois at Chicago, 1747 W Roosevelt Rd, Room 558, MC 275, Chicago, IL 60608, USA. petji@uic.edu
Abstract
BACKGROUND: Recruiting schools into a matched-pair randomized control trial (MP-RCT) to evaluate the efficacy of a school-level prevention program presents challenges for researchers. We considered which of 2 procedures would be most effective for recruiting schools into the study and assigning them to conditions. In 1 procedure (recruit and match/randomize), we would recruit schools and match them prior to randomization, and in the other (match/randomize and recruitment), we would match schools and randomize them prior to recruitment. METHOD: We considered how each procedure impacted the randomization process and our ability to recruit schools into the study. After implementing the selected procedure, the equivalence of both treatment and control group schools and the participating and nonparticipating schools on school demographic variables was evaluated. RESULTS: We decided on the recruit and match/randomize procedure because we thought it would provide the opportunity to build rapport with the schools and prepare them for the randomization process, thereby increasing the likelihood that they would accept their randomly assigned conditions. Neither the treatment and control group schools nor the participating and nonparticipating schools exhibited statistically significant differences from each other on any of the school demographic variables. CONCLUSIONS: Recruitment of schools prior to matching and randomization in an MP-RCT may facilitate the recruitment of schools and thus enhance both the statistical power and the representativeness of study findings. Future research would benefit from the consideration of a broader range of variables (eg, readiness to implement a comprehensive prevention program) both in matching schools and in evaluating their representativeness to nonparticipating schools.
RCT Entities:
BACKGROUND: Recruiting schools into a matched-pair randomized control trial (MP-RCT) to evaluate the efficacy of a school-level prevention program presents challenges for researchers. We considered which of 2 procedures would be most effective for recruiting schools into the study and assigning them to conditions. In 1 procedure (recruit and match/randomize), we would recruit schools and match them prior to randomization, and in the other (match/randomize and recruitment), we would match schools and randomize them prior to recruitment. METHOD: We considered how each procedure impacted the randomization process and our ability to recruit schools into the study. After implementing the selected procedure, the equivalence of both treatment and control group schools and the participating and nonparticipating schools on school demographic variables was evaluated. RESULTS: We decided on the recruit and match/randomize procedure because we thought it would provide the opportunity to build rapport with the schools and prepare them for the randomization process, thereby increasing the likelihood that they would accept their randomly assigned conditions. Neither the treatment and control group schools nor the participating and nonparticipating schools exhibited statistically significant differences from each other on any of the school demographic variables. CONCLUSIONS: Recruitment of schools prior to matching and randomization in an MP-RCT may facilitate the recruitment of schools and thus enhance both the statistical power and the representativeness of study findings. Future research would benefit from the consideration of a broader range of variables (eg, readiness to implement a comprehensive prevention program) both in matching schools and in evaluating their representativeness to nonparticipating schools.
Authors: Isaac J Washburn; Alan Acock; Sam Vuchinich; Frank Snyder; Kin-Kit Li; Peter Ji; Joseph Day; David DuBois; Brian R Flay Journal: Prev Sci Date: 2011-09
Authors: Niloofar Bavarian; Kendra M Lewis; Alan Acock; David L DuBois; Zi Yan; Samuel Vuchinich; Naida Silverthorn; Joseph Day; Brian R Flay Journal: J Prim Prev Date: 2016-02
Authors: Kendra M Lewis; Marc B Schure; Niloofar Bavarian; David L DuBois; Joseph Day; Peter Ji; Naida Silverthorn; Alan Acock; Samuel Vuchinich; Brian R Flay Journal: Am J Prev Med Date: 2013-06 Impact factor: 5.043
Authors: Kendra M Lewis; Samuel Vuchinich; Peter Ji; David L DuBois; Alan Acock; Niloofar Bavarian; Joseph Day; Naida Silverthorn; Brian R Flay Journal: Appl Dev Sci Date: 2015-05-29
Authors: Kendra M Lewis; Niloofar Bavarian; Frank J Snyder; Alan Acock; Joseph Day; David L Dubois; Peter Ji; Marc B Schure; Naida Silverthorn; Samuel Vuchinich; Brian R Flay Journal: Int J Emot Educ Date: 2012-04
Authors: Niloofar Bavarian; Kendra M Lewis; David L Dubois; Alan Acock; Samuel Vuchinich; Naida Silverthorn; Frank J Snyder; Joseph Day; Peter Ji; Brian R Flay Journal: J Sch Health Date: 2013-11 Impact factor: 2.118