PubMed Health⌕ Search

Biomedical subjects

Devan V Mehrotra

Publications and source records attributed to Devan V Mehrotra.

3 recordsLinked to original sources
PubMed2003

Stratified experiments reexamined with emphasis on multicenter trials.

In many stratified experiments the researcher fixes the total sample size but either cannot or does not exert control over the sample size per stratum. A classic example is a randomized, two-treatment, multicenter clinical trial-the total sample sizes per treatment group are fixed by design but the sample sizes per center are allowed to vary. Standard analyses of continuous data from such trials fail to recognize the random nature of the stratum sizes. We show that this can lead to biased inference and estimation for both so-called "type II" (unequal weighting of strata) as well as "type III" (equal weighting of strata) analyses. We propose an alternative method of analysis that explicitly accounts for the randomness of the stratum sizes and illustrate its validity using simulations. A reanalysis of published data from a 29-center clinical trial serves to reinforce the key points.

Data Interpretation, Statistical↗
PubMed2003

A cautionary note on exact unconditional inference for a difference between two independent binomial proportions.

Fisher's exact test for comparing response proportions in a randomized experiment can be overly conservative when the group sizes are small or when the response proportions are close to zero or one. This is primarily because the null distribution of the test statistic becomes too discrete, a partial consequence of the inference being conditional on the total number of responders. Accordingly, exact unconditional procedures have gained in popularity, on the premise that power will increase because the null distribution of the test statistic will presumably be less discrete. However, we caution researchers that a poor choice of test statistic for exact unconditional inference can actually result in a substantially less powerful analysis than Fisher's conditional test. To illustrate, we study a real example and provide exact test size and power results for several competing tests, for both balanced and unbalanced designs. Our results reveal that Fisher's test generally outperforms exact unconditional tests based on using as the test statistic either the observed difference in proportions, or the observed difference divided by its estimated standard error under the alternative hypothesis, the latter for unbalanced designs only. On the other hand, the exact unconditional test based on the observed difference divided by its estimated standard error under the null hypothesis (score statistic) outperforms Fisher's test, and is recommended. Boschloo's test, in which the p-value from Fisher's test is used as the test statistic in an exact unconditional test, is uniformly more powerful than Fisher's test, and is also recommended.

Humans↗