Sunday, October 1, 2017
Statistical power and minimum sample size requirements
The WarpPLS menu option “Explore statistical power and minimum sample size requirements”, available starting in version 6.0, allows you to obtain estimates of the minimum required sample sizes for empirical studies based on the following model elements: the minimum absolute significant path coefficient in the model (e.g., 0.21), the significance level used for hypothesis testing (e.g., 0.05), and the power level required (e.g., 0.80). Two methods are used to estimate minimum required sample sizes, the inverse square root and gamma-exponential methods. These methods simulate Monte Carlo experiments, and thus produce estimates that are in line with the estimates that would be produced through the Monte Carlo method.
Related YouTube video:
Explore Statistical Power and Minimum Sample Size in WarpPLS
http://youtu.be/mGT6-NKUe3E
Article in the Information Systems Journal discussing the methods:
http://onlinelibrary.wiley.com/doi/10.1111/isj.12131/full
Subscribe to:
Post Comments (Atom)
5 comments:
Thank you for the very informative and valuable information you provide on your blog.
I am wondering in cases where the data violates the minimum sample size, what kind of considerations one should be involved with? and what other alternatives one may attempt?
Hi. The pub. linked below covers these issues, as well as other related issues.
http://cits.tamiu.edu/kock/pubs/journals/2018/Kock_Hadaya_2018_ISJ_SampleSizePLS.pdf
Full ref., just in case:
Kock, N., & Hadaya, P. (2018). Minimum sample size estimation in PLS-SEM: The inverse square root and gamma-exponential methods. Information Systems Journal, 28(1), 227–261.
Hi, i have read the paper, 'Kock, N., & Hadaya, P. (2018). Minimum sample size estimation in PLS-SEM: The inverse square root and gamma-exponential methods. Information Systems Journal, 28(1), 227–261.'. Sorry but i am not sure if i understand it correctly. If there are no relevant past empirical research that can suggest a path coefficient value that i can refer to, can i safely follow the general rule of thumb of a minimum sample size of 160 (inverse square root method)?
That is a general rule of thumb that is based on the assumption that the minimum absolute significant path coefficient in the model is greater than or equal to 0.197 (rationale discussed in the article: Kock and Hadaya, 2018). It should be used prior to data collection, as a target minimum required sample size of 160. After data collection, you should then re-estimate the minimum required sample size with WarpPLS, using the actual minimum absolute significant path coefficient in the model, to be sure you meet that required sample size. For example, if your minimum absolute significant path coefficient in the model is 0.130, you may need more data to properly assess that path (and thus the model) with at least 80 percent power. This assumes that the effect size associated with the path is above the minimum acceptable level (see WarpPLS User Manual); otherwise you may want to reject the path based on the effect size being too low (even if the path is statistically significant).
Post a Comment