This Article Full Text (OnlineFirst PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Tao, G.-L. Articles by Tang, N.-S. Search for Related Content PubMed PubMed Citation Articles by Tao, G.-L. Articles by Tang, N.-S. Social Bookmarking                         What's this?

Sample size determination for the non-randomised triangular model for sensitive questions in a survey

¹ Department of Statistics and Actuarial Science, The University of Hong Kong, Pokfulam Road, Hong Kong, P. R. China and Division of Biostatistics, University of Maryland Greenebaum Cancer Center, MSTF Suite 261, 10 South Pine Street, Baltimore, Maryland 21201, USA
² Department of Mathematics, Hong Kong Baptist University, Kowloon Tong, Hong Kong, P. R. China
³ Division of Biostatistics, University of Maryland Greenebaum Cancer Center, MSTF Suite 261, 10 South Pine Street, Baltimore, Maryland 21201, USA
⁴ Division of Biostatistics, University of Maryland Greenebaum Cancer Center and Department of Epidemiology and Preventive Medicine, MSTF Suite 261, 10 South Pine Street, Baltimore, Maryland 21201, USA
⁵ Department of Statistics, Yunnan University, Kunming 650091, P. R. China

^* To whom correspondence should be addressed.

	Abstract

Sample size determination is an essential component in public health survey designs on sensitive topics (e.g. drug abuse, homosexuality, induced abortions and pre or extramarital sex). Recently, non-randomised models have been shown to be an efficient and cost effective design when comparing with randomized response models. However, sample size formulae for such non-randomised designs are not yet available. In this article, we derive sample size formulae for the non-randomised triangular design based on the power analysis approach. We first consider the one-sample problem. Power functions and their corresponding sample size formulae for the one- and two-sided tests based on the large-sample normal approximation are derived. The performance of the sample size formulae is evaluated in terms of (i) the accuracy of the power values based on the estimated sample sizes and (ii) the sample size ratio of the non-randomised triangular design and the design of direct questioning (DDQ). We also numerically compare the sample sizes required for the randomised Warner design with those required for the DDQ and the non-randomised triangular design. Theoretical justification is provided. Furthermore, we extend the one-sample problem to the two-sample problem. An example based on an induced abortion study in Taiwan is presented to illustrate the proposed methods.

First published on February 16, 2009
Statistical Methods in Medical Research 2009, doi:10.1177/0962280208099444


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?