Power analysis of trials with multilevel data / Mirjam Moerbeek and Steven Teerenstra.
Material type:
TextSeries: Chapman & Hall/CRC interdisciplinary statistics seriesPublication details: Boca Raton : CRC Press, ©2016.Description: xix, 268 p. : illustrations ; 25 cmISBN: - 9781498729895 (alk. paper)
- 000SA.01 23 M694
| Item type | Current library | Call number | Status | Date due | Barcode | Item holds | |
|---|---|---|---|---|---|---|---|
| Books | ISI Library, Kolkata | 000SA.01 M694 (Browse shelf(Opens below)) | Available | 137214 |
Browsing ISI Library, Kolkata shelves Close shelf browser (Hides shelf browser)
Includes bibliographical references and indexes.
1.Introduction --
1.1.Experimentation --
1.1.1.Problems with random assignment --
1.2.Hierarchical data structures --
1.3.Research design --
1.3.1.Cluster randomized trial --
1.3.2.Multisite trial --
1.3.3.Pseudo cluster randomized trial --
1.3.4.Individually randomized group treatment trial --
1.3.5.Longitudinal intervention study --
1.3.6.Some guidance to design choice --
1.4.Power analysis for experimental research --
1.5.Aim and contents of the book --
1.5.1.Aim --
1.5.2.Contents --
2.Multilevel statistical models --
2.1.The basic two-level model --
2.2.Estimation and hypothesis test --
2.3.Intraclass correlation coefficient --
2.4.Multilevel models for dichotomous outcomes --
2.5.More than two levels of nesting --
2.6.Software for multilevel analysis --
3.Concepts of statistical power analysis --
3.1.Background of power analysis --
3.1.1.Hypotheses testing --
3.1.2.Power calculations for continuous outcomes --
3.1.3.Power calculations for dichotomous outcomes --
3.1.3.1.Risk difference --
3.1.3.2.Odds ratio --
3.2.Types of power analysis --
3.3.Timing of power analysis --
3.4.Methods for power analysis --
3.5.Robustness of power and sample size calculations --
3.6.Procedure for a priori power analysis --
3.6.1.An example --
3.7.The optimal design of experiments --
3.7.1.An example (continued) --
3.8.Sample size and precision analysis --
3.9.Sample size and accuracy of parameter estimates --
4.Cluster randomized trials --
4.1.Introduction --
4.2.Multilevel model --
4.3.Sample size calculations for continuous outcomes --
4.3.1.Factors that influence power --
4.3.2.Design effect --
4.3.3.Sample size formulae for fixed cluster size or fixed number of clusters --
4.3.4.Including budgetary constraints --
4.4.Sample size calculations for dichotomous outcomes --
4.4.1.Risk difference --
4.4.2.Odds ratio --
4.5.An example --
5.Improving statistical power in cluster randomized trials --
5.1.Inclusion of covariates --
5.2.Minimization, matching, pre-stratification --
5.3.Taking repeated measurements --
5.4.Crossover in cluster randomized trials --
5.5.Stepped wedge designs --
6.Multisite trials --
6.1.Introduction --
6.2.Multilevel model --
6.3.Sample size calculations for continuous outcomes --
6.3.1.Factors that influence power --
6.3.2.Design effect --
6.3.3.Sample size formulae for fixed cluster size or fixed number of clusters --
6.3.4.Including budgetary constraints --
6.3.5.Constant treatment effect --
6.4.Sample size calculations for dichotomous outcomes --
6.4.1.Odds ratio --
6.5.An example --
7.Pseudo cluster randomized trials --
7.1.Introduction --
7.2.Multilevel model --
7.3.Sample size calculations for continuous outcomes --
7.3.1.Factors that influence power --
7.3.2.Design effect --
7.3.3.Sample size formulae for fixed cluster size or fixed number of clusters --
7.4.Sample size calculations for binary outcomes --
7.5.An example --
8.Individually randomized group treatment trials --
8.1.Introduction --
8.2.Multilevel model --
8.2.1.Clustering in both treatment arms --
8.2.2.Clustering in one treatment arm --
8.3.Sample size calculations for continuous outcomes --
8.3.1.Clustering in both treatment arms --
8.3.1.1.Factors that influence power --
8.3.1.2.Sample size formulae for fixed cluster sizes --
8.3.1.3.Including budgetary constraints --
8.3.2.Clustering in one treatment arm --
8.3.2.1.Factors that influence power --
8.3.2.2.Sample size formulae for fixed cluster sizes --
8.3.2.3.Including budgetary constraints --
8.4.Sample size calculations for dichotomous outcomes --
8.4.1.Clustering in both treatment arms --
8.4.2.Clustering in one treatment arm --
8.5.An example --
9.Longitudinal intervention studies --
9.1.Introduction --
9.2.Multilevel model --
9.3.Sample size calculations for continuous outcomes --
9.3.1.Factors that influence power --
9.3.2.Sample size formula for fixed number of measurements --
9.3.3.Including budgetary constraints --
9.4.Sample size calculations for dichotomous outcomes --
9.4.1.Odds ratio --
9.5.The effect of drop-out on statistical power --
9.5.1.The effects of different drop-out patterns --
9.5.2.Including budgetary constraints --
9.6.An example --
10.Extensions: three levels of nesting and factorial designs --
10.1.Introduction --
10.2.Three-level cluster randomized trials --
10.3.Multisite cluster randomized trials --
10.4.Repeated measures in cluster randomized trials and multisite trials --
10.5.Factorial designs --
10.5.1.Continuous outcome --
10.5.2.Binary outcome --
10.5.3.Sample size calculation for factorial designs --
11.The problem of unknown intraclass correlation coefficients --
11.1.Estimates from previous research --
11.2.Sample size re-estimation --
11.3.Bayesian sample size calculation --
11.4.Maximin optimal designs --
12.Computer software for power calculations --
12.1.Introduction --
12.2.Computer program SPA-ML.
The book gives a thorough overview of power analysis that details terminology and notation, outlines key concepts of statistical power and power analysis, and explains why they are necessary in trial design. It guides you in performing power calculations with hierarchical data, which enables more effective trial design.
There are no comments on this title.
