For applying structural equation modelling, is there any specific criteria for sample size?
for applying structural equation modeling in AMOS, these are some significance level:
1-C.R: the level of C.R shouldn't be between (+1/96 ) to (-1/96),
and (P.value) sig < 0.5 - should be bigger than (0.5).
2- CMIN OR x2.
3- x2/df 0.5 - should be bigger than(0.5).
6- GFI > 0.8 - should be bigger than(0.8).
7- AGFI > 0.8 - should be bigger than(0.8).
9- NFI >0.9 - should be bigger than (0.9).
10- NNFI >0.9- should be bigger than (0.9).
11-RFI >0.9- should be bigger than (0.9).
12- CFI >0.9 - should be bigger than (0.9).
13- TLI >0.9 - should be bigger than (0.9).
Sekaran (2003) wrote:
"Roscoe (1975) proposes the following rules of thumb for determining sample size:
1. Sample sizes larger than 30 and less than 500 are appropriate for most research.
2. Where samples are to be broken into subsamples; (male/females, juniors/
seniors, etc.), a minimum sample size of 30 for each category is necessary.
3. In multivariate research (including multiple regression analyses), the sample size should be several times (preferably 10 times or more) as large as the number of variables in the study.
4. For simple experimental research with tight experimental controls (matched pairs, etc.), successful research is possible with samples as small as 10 to 20 in size."
References
Sekaran, U., 2003. Research methods for business: A skill building approach. John Wiley & Sons.
SEM indices are sample sensitive especially the absolute fit indices like GFI, AGFI etc. Where as incremental fit indices (CFI) are not affected by sample size . Ideally 10 respondents per variable is the recommended limit by researchers (Hair et al., 2009)
Structural Equation Modelling(SEM) deals with big size sample to be more effective and to reduce measurement errors according to some references sample size shouldn't less than 100 , with small size sample a partial least square(PLS) is preferred.
What is the minimum sample size for structural equation modelling?
It depends on types of SEM (e.g. Variance-based SEM vs Covariance-based SEM), multivariate normality of the data, estimation technique, model complexity etc. Generally for Variance-based SEM / PLS-SEM, you can use the following:
For the above reference, you can refer to this book pg 18-20:
Hair, J. F., Hult, G. T. M., Ringle, C., & Sarstedt, M. (2013). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM). Sage Publications, Inc.
For Covariance-based SEM, generally you can use minimal 200 (but different scholars do argue minimal should be 100, 150, 300 etc. depending on certain conditions). For the above reference, you can refer to this book pg. 661-662:
Hair, J. F., Black, W. C., & Babin, B. J. (2010). Multivariate Data Analysis – A Global Perspective, 7th Edition. Pearson Education.
If your data comes from interval or ratio scale and you are using AMOS then sample shouldn't be less than 100. However with nominal and ordinal scale data and SMART PLS you are allowed to go less.
Sample size determination is a function of Population Size. Try the Yemane Formula
As mentioned above it is 200 but you need to specify the sample design. If you need assistance in SEM contact me [email protected]
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size