Do Larger Samples Really Lead to More Precise Estimates? A Simulation Study

Nestor Asiamah^1, , Henry Kofi Mensah² and Eric Fosu Oteng-Abayie²

¹Africa Centre for Epidemiology, Accra, Ghana

²Kwame Nkrumah University of Science and Technology, Kumasi Ghana

Cite this paper:
Nestor Asiamah, Henry Kofi Mensah and Eric Fosu Oteng-Abayie. Do Larger Samples Really Lead to More Precise Estimates? A Simulation Study. American Journal of Educational Research. 2017; 5(1):9-17. doi: 10.12691/education-5-1-2

Abstract

In this paper, we use simulated data to find out if larger samples support estimation of population parameters by examining whether or not higher samples give rise to more precise estimates of population parameters. We simulated a normally distributed dataset and randomly drew 73 samples from it. Some basic statistics, namely the mean, standard deviation, standard error of the mean, confidence interval and the one-sample t-test significance were computed under some conditions for all samples. The correlation between sample size and each of these statistics was computed, among other statistical treatments. Our analysis suggests that larger samples produce estimates that better approximate the population parameters. The correlation between sample size and standard error of the mean is even stronger. We therefore conclude that larger samples lead to more precise estimates.

Keywords:
sampling sample size population parameters sample statistics statistical estimation

This work is licensed under a Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

References:

[1]	Bartlett, J. E.; Kotrlik, J. W.; Higgins, C. C. Organisational Research: Determining Appropriate Sample Size in Survey Research, Information Technology, Learning, and Performance Journal. 2001, 19, 43-50.
[2]	Brown, J.D. Sample size and statistical precision, JALT Testing & Evaluation SIG Newsletter. 2007, 11, 21-24
[3]	Hanley, J.A.; Moodie, E.E.M. Sample Size, Precision and Power Calculations: A Unified Approach, Journal of Biometrics & Biostatistics. 2011; 5, 1-9.
[4]	Marshal, M.N. Sampling for Qualitative Research, Family Practice. 1996, 13, 522-525.
[5]	Springate, S.D. The effect of sample size and bias on the reliability of estimates of error: a comparative study of Dahlberg’s formula, European Journal of Orthodontics. 2011, 34, 158-163.
[6]	Eng, J. Sample Size Estimation: How Many Individuals Should Be Studied? Radiology. 2003, 227, 309-313.
[7]	Krejcie, R. V.; Morgan, D. W. Determining sample size for research activities, Educational and Psychological Measurement. 1970, 30, 607-610.
[8]	Fiske, I.J.; Bruna1, E.M.; Bolker, B.M. Effects of Sample Size on Estimates of Population Growth Rates Calculated with Matrix Models, Plos One. 2009, 3, 1-6.
[9]	Jackman, S. (2000) Estimation and Inference via Bayesian Simulation: An Introduction to Markov Chain Monte Carlo, American Journal of Political Science. 2000, 44, 375-404.
[10]	Sotos, AEC.; Vanhoof, S.; Van den Noortgate, W.; Onghena, P. Students’ misconceptions of statistical inference: A review of the empirical evidence from research on statistics education, Educational Research Review. 2007, 2 98-113.
[11]	Trosset, M.W. An Introduction to Statistical Inference and Its Applications. Retrieved from file:///C:/Users/hp/Dropbox/Articles%202014/Personal/Literature%20(sample)/statistical_inference.pdf, 2006, at 12/11/2015 at 13:32 PM.
[12]	Biau, D.J.; Jolles, B.M.; Porcher, R. P Value and the Theory of Hypothesis Testing: An Explanation for New Researchers, Clinical Orthopaedic Relation Research. 2010, 468, 885-892.