Fundamentals of estimating sample size.

Abstract

Estimating sample size is an integral requirement in the planning stages of quantitative studies. However, although abundant literature is available that describes techniques for calculating sample size, many are in-depth and have varying degrees of complexity. To provide an overview of four basic parameters that underpin the determination of sample size and to explain sample-size estimation for three study designs common in nursing research. Researchers can estimate basic sample size if they have a comprehension of four parameters, such as significance level, power, effect size, and standard deviation (for continuous data) or event rate (for dichotomous data). In this paper, these parameters are applied to determine sample size for the following well-established study designs: a comparison of two independent means, the paired mean study design and a comparison of two proportions. An informed choice of parameter values to input into estimates of sample size enables the researcher to derive the minimum sample size required with sufficient power to detect a meaningful effect. An understanding of the parameters provides the foundation from which to generalise to more complex size estimates. It also enables more informed entry of required parameters into sample size software. Underpinning the concept of evidence-based practice in nursing and midwifery is the application of findings that are statistically sound. Researchers with a good understanding of parameters, such as significance level, power, effect size, standard deviation and event rate, are enabled to calculate an informed sample size estimation and to report more clearly the rationale for applying any particular parameter value in sample size determination.