China
Africa
Explore chapters and articles related to this topic
Preparing to Become Credentialed
Published in Nicole M. Augustine, Prevention Specialist Exam Study Guide, 2023
A process evaluationIs done at the completion of the program.Is done throughout the delivery of program services.Involves random assignment of participants.Involves the collection of participant information after they leave the program.
Basic Research Design:
Published in Lynne M. Bianchi, Research during Medical Residency, 2022
Lynne M. Bianchi, Luke J. Rosielle, Justin Puller, Kristin Juhasz
Random assignment is a key feature of most experimental studies and is preferred for clinical research whenever possible and ethical. With randomization, every participant has an equal chance of being placed into any group. Random assignment presumably distributes the inherent differences between group members among the groups, thereby increasing the likelihood an observed effect is due to the intervention. Randomization does not eliminate all selection biases, however, and investigators need to carefully consider how they will identify and recruit participants (see also Chapter 5).
Study Design
Published in Marcello Pagano, Kimberlee Gauvreau, Heather Mattie, Principles of Biostatistics, 2022
Marcello Pagano, Kimberlee Gauvreau, Heather Mattie
Stratified randomization ensures that an important confounder such as age or sex is more evenly distributed between treatment groups than might happen by chance alone. Using this technique, randomization is performed separately within strata, for subjects with and without the confounder. (Recall the discussion of stratified sampling in Chapter 21.) As an example, in the study of patients undergoing liver transplantation, randomization was stratified by whether or not the subject had been diagnosed with hepatitis C. In small studies, stratified randomization can improve statistical power. It has less benefit in large studies, where random assignment is more likely to ensure an even distribution of baseline characteristics across treatment groups.
Evaluation of a brief harm reduction intervention to reduce celebratory drinking among college students
Published in Journal of American College Health, 2023
Christine Arazan, Michael T. Costelloe, Mark T. Willingham
The use of an experimental design with random assignment is a strong component of this study and considered the gold standard in evaluation research. There are important limitations to note, however, in regards to the posttest. Contrary to best practice, this study was unable to obtain outcome data for a high proportion of the sample members originally randomized; thus this study had high sample attrition. It is important to note that the follow-up rate (the proportion of students who logged into the posttest survey) is approximately the same for the experimental (11%) and the control group (12%). Given that the two groups were similar on key demographic characteristics prior to the intervention and the posttest access rates were comparable, it is anticipated that attrition did not undermine the equivalence of the experimental and control group.
Addressing A Mental Health Intervention Gap in Juvenile Detention: A Pilot Study
Published in Evidence-Based Practice in Child and Adolescent Mental Health, 2023
Jennifer E. Duchschere, Samantha J. Reznik, Caroline E. Shanholtz, Karey L. O’Hara, Nadav Gerson, Connie J. Beck, Erika Lawrence
Random assignment of participants was not feasible in this study, and we had to balance the needs of the study site in maintaining procedures and safety with requirements to meet scientific evaluation of the intervention. We were able to develop a protocol that balanced the rigor of our scientific evaluation with real-world implementation challenges in a clinical setting with a vulnerable, underserved population. We were successful in implementing the intervention in the juvenile detention setting, as evidenced by completing 58 sessions over the course of 35 weeks. Despite limitations (e.g., upholding detention rules, safety requirements of a corrections officer being present for each group), participants still found the intervention to be engaging and acceptable; the dropout was remarkably low compared to similar studies (De Haan et al., 2013).
A spotlight on cross-sectional research: Addressing the issues of confounding and adjustment
Published in International Journal of Healthcare Management, 2021
Nestor Asiamah, Edwin Mends-Brew, Benjamin Kojo Teye Boison
The above illustrations (see Figures 1–5) would suggest that adjustment is necessary for reaching a valid estimate of the association between two variables in a cross-sectional research. Two main ways have been described in the literature for controlling for potential confounding variables. The first of these is the study design approach, which is concerned with setting up the study to avoid or minimize the blurring up of the ultimate effect with the influences of potential confounders [14]. Four methods have been provided under this approach, namely randomization, restriction, matching, and stratification [5,14]. In randomization, the researcher reduces confounding by randomly assigning study participants to control or/and experimental groups [4,7]. Random assignment is often done by using a computer-based system, making it possible for researcher bias to be avoided. Randomization is ideally applied to experimental studies such as randomized controlled, clinical and randomized interventional trials that make use of experimental and control groups. Perhaps, it is the best adjustment technique because it can curb or reduce confounding by more than one variable. Cross-sectional researchers can apply this approach to select participants for two groups (e.g. smokers and non-smokers) and the result of these groups compared to see which has a higher risk of a disease (i.e. lung cancer).