European Journal of Scientific Research - EuroJournals
European Journal of Scientific Research - EuroJournals
European Journal of Scientific Research - EuroJournals
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
Determination <strong>of</strong> Sample Size 320<br />
minimize the variance. (Greenland, 1988; Samuels, 1992; Buderer, 1996; Satten & Kupper,<br />
1990; Streiner, 1994 and Beal, 1989).<br />
iii) Issues and Prior Information: If the process has been studied before, then the prior<br />
information can be used to reduce sample sizes. This can be done by using prior mean and<br />
variance estimates and by stratifying the population to reduce variation within groups. Increase<br />
in sample size may reduce the sampling error but may cause increase in non-sampling error.<br />
iv) Clinical Trials: At the planning stage <strong>of</strong> a clinical trial a key question is "How many patients<br />
do we need?” This requires a sample size that ensures sufficient statistical power to detect a<br />
clinically relevant improvement. Sample size must be planned carefully to ensure that the<br />
research time, patient effort & support and costs invested in any clinical trial are not wasted.<br />
v) Power Analysis: power analysis and sample size estimation is an important aspect <strong>of</strong><br />
experimental design, because without these calculations, sample size may be too high or too<br />
low. If sample size is too low, the experiment will lack the precision to provide reliable answers<br />
to the questions it is investigating. If sample size is too large, time and resources will be wasted,<br />
<strong>of</strong>ten for minimal gain.<br />
vi) Cost: The cost <strong>of</strong> sampling issue helps to determine how precise the estimate should be. When<br />
choosing sample sizes, it is required to select risk values (affordable error). If the decisions<br />
made from the sampling activity are very valuable, then these will have low risk values and<br />
hence larger sample sizes. So some cost functions in different situations have been discussed.<br />
Choosing the sample size is a problem faced by anyone doing a survey <strong>of</strong> any type. ‘What<br />
sample size do I need?’ is one <strong>of</strong> the most frequently asked questions to the statisticians. The response<br />
always starts “It depends on...” The sample size must depend on what you want to know about and<br />
how well you want to know about it. In order to make rational sample size choices, both the quantities<br />
to be estimated and the precision required must be specified.<br />
Sample-size planning is very important and almost always difficult. It requires care in eliciting<br />
scientific objectives and in obtaining suitable quantitative information prior to the study. Successful<br />
resolution <strong>of</strong> the sample-size problem requires the close and honest collaboration <strong>of</strong> statisticians and<br />
subject-matter experts (Russell, 2001).<br />
Sample-size problems are context-dependent. For example, how important it is to increase the<br />
sample size to account for such uncertainty depends on practical and ethical criteria. Moreover, sample<br />
size is not always the main issue; it is only one aspect <strong>of</strong> the quality <strong>of</strong> a study design (Russell, 2001).<br />
The objective <strong>of</strong> the paper is to mention available literature relevant to the determination <strong>of</strong><br />
sample size under various situations and to develop new formulae with reference to cost and affordable<br />
error.<br />
2. Sample size Estimation<br />
In this Section a review <strong>of</strong> above mentioned points is given with reference to estimating parameters,<br />
right variance, issues <strong>of</strong> interest, clinical trials, effect size, and power analysis.<br />
2.1. Parameters<br />
Chochran (1977) had given the simplest case for determination <strong>of</strong> sample size is concerned with<br />
infinite normal population with known variance. Desu and Raghavarao (1990) had provided brief<br />
review on sample size methodology. Further more the paired sample approach is employed by Dupont,<br />
(1988); Parker & Bregman, (1986); Nam, (1992); Lu & Bean, (1995); Lachenbruch, (1992); Lachin,<br />
(1992); Royston, (1993); Nam, (1997). A brief review has been given in literature to determine the<br />
sample size for the tests <strong>of</strong> proportions by Chochran, 1977; Casagrande, Pike & Smith, 1978; Feigl,<br />
1978; Haseman, 1978; Fleiss, 1981; Lemeshow, Hosmer & Klar, 1988; O’Neill, 1984; Thomas, 1992;<br />
Whitehead, 1993 and Gordon & Watson, 1996 for infinite population.