Hi,
I try to identify the influence of managers' sustainable behaviour on subordinates's sustainable behaviour. I create 8 Likert Scale question to measure the pro-environmental behaviour of managers and their subordinates with the answer ranging from never, rarely, sometimes, frequency and always.
I am thinking to use Chi-square test to test the dependent/ independent of behaviours between managers and subordinates. Is this will aid me to get the conclusion?
Can anyone give me some suggestion?
Thank you so much,
Best Regards,
Nhat
Hi Nhat,
You are dealing with latent constructs in your research. I advise you to measure their perceived behavior using interval scales, instead of ordinal scales, that is measuring how much they agree or disagree with the item statement. By measuring data using interval scales, you can employ the parametric analysis like Structural Equation Modeling instead of non parametric analysis (chi-square) to determine the relationship between two latent constructs.
Hi, Good Morning,
to identify the influence of managers' sustainable behaviour on subordinates's sustainable behavior. You are used the 8 Likert Scale question to measure the pro-environmental behavior of managers Non-parametric tests such as chi-squared test, Mann–Whitney test, Wilcoxon signed-rank test, or Kruskal–Wallis test are often used in the analysis of Likert scale data.
So, you have to use the Chi-Square test in your research.
All the best.
If you are going to use Likert type answer format with 8 answer categories, you can think of treating it as a continuous variable... some say that Likert with 4 categories and more may be treated as continuous so I would not be so sure if chi-square is the only option here. I think Chi-square is more useful in categories such as type of employment, ethnicity, etc. Hope that helps a bit
You have ordinal variable with 4 categories , the best choice is Kruskal - Wallis and in case of significant of the test , then you can use Mann - Whitney test for each pair categories comparison .
Good Luck
Hello Nhat,
Also, even if you obtain significant differences between the groups using chi-squares you may not be sure where exactly they are. For example you have two independent samples, and you compare the results between them using chi-square in five categories (A,B,C,D, E). It is not obvious in chi-square if statistically significant differences are between A and B, C and E and so on... Hence it is not easy to interpret the statistically significant differences with chi-squares (regressions and t tests are much more obvious).
Good luck with your research
- Badges
- Science topic