How is it possible for two organizations having 25000 employees. and i need to cause and effect of business strategies with innovation? Please help with references, in real tension.
You can use SmartPLS, it has no sample restriction. But it seems you have only two organizations and for only two organizations, SmartPLS can not run it.
As a matter of fact it is not possible to mention an exact ideal number of the subjects for a good sample from a certain population. The sample size obtained for a study should be appropriate according to the rules of thumb proposed by Roscoe (1975) (as in Sekaran, 1992) such as that sample sizes larger than 30 and less than 500 are appropriate and the sample size should be several times (preferably 10 times or more) as large as the number of variables in multivariate study (including multiple regression analyses).
Best wishes!
Variables or predictor? If u don't mind, can Ii have the reference? Thanks.
Thanks Mr. Burhan, means we can run smart PLS on 30 organizations data. please give any reference.
George Marcoulides from UC Santa Barbara is an expert in the SEM field. Check the web for some of the books/articles he has written on the subject.
Some scholars propose a sample size of 200 as sufficient for SEM methods (e.g. see Garver and Mentzer, 1999; Hoelter, 1983), while others consider a sample size of above 100 observations as sufficient for latent variables models (e.g. see Harris and Schaubroek, 1990; Kelloway, 1998). Some contend that some fit indices like NNFI/TLI, CFI, IFI SRMR are independent of sample size, i.e. they are relatively unaffected by sample size (e.g. see Gerbing and Anderson, 1992; Bollen, 1990; Hu and Bentler; 1995). Generally the debate regarding the appropriate sample size for SEM analysis techniques is not conclusive and there is little consensus on the recommended sample size. While SEM is considered a large sample technique, the definition of large is still an issue of debate. Of course the chi-square index is not appropriate where the sample size falls short, since the chi-square statistic is sensitive to sample size. I would personally not go for SEM where sample sizes fall below 100 – but rather PLS.
Dear Shafiq, if your unit of analysis is the employee, then I imagine your sample s big enough. That is if you are looking at innovation at the level of employee as opposed to innovation at organisation level. Otherwise if your unit of analysis is the organisation and you have a sample of only two units, then in-depth interviews, case study methods and qualitative approaches may be helpful.
You may use Kaiser-Meyer-Olkin Measure of Sampling Adequacy. This measure varies between 0 and 1, and values closer to 1 are better. A value of .6 is a suggested minimum.
Kaiser-Meyer-Olkin Measure (KMO) measure can be utilized through factor analysis in SPSS.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publications