Required Sample Size for Correlation Analysis

Abstract

Sample size calculation prior to data collection is still relatively rare in Russian research practice. This situation threatens validity of the conclusion of many projects due to insufficient statistical power to estimate the parameters of interest with desired precision or to detect the differences of interest. Moreover, in a substantial proportion of cases where sample size calculations are performed simplified formulas with assumption of a normal distribution of the studied variables are used in spite of the fact that this assumption does not hold for many research questions in biomedical research. Correlation analysis is still one of the most commonly used methods of statistical analysis used in Russia. Pearson’s correlation coefficient despite its well-known limitations appears in a greater proportion of publications that non-parametric coefficients. We calculated minimal sample sizes for the parametric Pearson’s coefficient as well its non-parametric alternatives — Spearman’s rho and Kendall’s tau-b correlation coefficients to assist junior researchers with the tool to be able to plan data collection and analysis for several types of data, various expected strengths of associations and research questions. The results are presented in ready-for-use tables with required sample size for the three abovementioned coefficients within the range from 0,10 through 0,90 by 0,05 for statistical power 0,8 and 0,9 and alpha-error or 5% as well as for estimation of the same correlation coefficients with the 95% confidence intervals width equal to 0,1 and 0,2.

The Adverse Environmental Impact Factors Analysis on Fly-In-Fly-Out Personnel at Industrial Enterprises

“…To determine the relationship of the combined influence of factors of adverse environmental impact with different levels of differential analysis of the professional activity of FIFO personnel, a correlation analysis was used, after which a correlation pleiad was compiled ( Figure 5 ). In the Pleiad, the correlations from 0.15 were considered at p = 0.047, due to the fact that for a sample of 359 or more, the Pearson coefficient was 0.15 [ 75 , 76 ]. A table with correlation values and significance levels is presented in Appendix A .…”

Cardiac remodeling in patients with heart failure with mildly reduced ejection fraction and metabolic disorders: association with biomarkers and autonomic nervous system parameters

“…Для корреляционного анализа использовался критерий Спирмена. Размер выборки и характер распределения признаков при уровне статистической мощности 0,8 и доверительной вероятности 95% позволяли выявлять средние и сильные корреляционные связи между изучаемыми параметрами (величина коэффициента корреляции Спирмена (r) ≥0,60) [15]. Различия считали статистически достоверными при значениях р<0,05.…”

Oral care practices and dental status among children in Arkhangelsk

“…Минимальный объем выборки был рассчитан с помощью программного обеспечения PASS-2000. При расчете минимального размера выборки на уровне доверительной вероятности 95% и статистической мощности 80% достаточно 123 участников для выявления корреляционных связей между изучаемыми признаками от 0,25 и выше [14]. В связи с тем, что распределение таких признаков, как индекс гигиены и индекс кп / кп+КПУ / КПУ, как правило, имело правостороннюю асимметрию, было принято решение использовать для анализа непараметрические критерии.…”