Design Sensitivity: Statistical Power for Experimental ResearchWhether in the laboratory or while doing fieldwork, all researchers face an important challenge - designing research that will have sufficient sensitivity to detect those effects it purports to investigate. Sample size, validity, and sensitivity, experimental error, subject variability and the type of statistical analysis all influence the sensitivity of a research design. In this volume Lipsey examines the concept of design sensitivity and explains statistical power and the elements that determine it. Through careful explanations and selection of examples he explores a variety of topics: factors that degrade design sensitivity, effect size parameters and approaches to assessing it, how to estimate statistical power for various statistical tests, and the special problems statistical power poses for treatment effectiveness research. This book is a vital resource for evaluators, methodologists, statisticians, psychologists, public health professionals and educators. |
Contents
Acknowledgments | 7 |
Useful Approaches and Techniques | 97 |
Role of Theory | 146 |
Copyright | |
1 other sections not shown
Other editions - View all
Common terms and phrases
alpha level analysis of variance approach attained Chapter characteristic circumstances Cohen compared comparison context control condition control group correlation covariate criterion contrast criterion group contrast delinquency denominator dependent measure dependent variable detect determine difference distribution Doctoral dissertation dose effect size estimates effect size parameter Evaluation example experimental conditions experimental control experimental design experimental groups group mean increase intervention issues Journal juvenile Kulik larger Lipsey measure of interest measurement error ment meta-analysis null hypothesis number of subjects one-tailed one-way ANOVA outcome paired subjects population power analysis power charts procedures programs proportion psychotherapy range recidivism relationship relatively reliability represented response sample size sampling error scores statistical power statistical test subject heterogeneity success rate t-test Table therapy tion treat treatment and control treatment effectiveness research treatment effectiveness study treatment group treatment versus control two-tailed Type II error validity values variance component variation within-groups variance σ²