I am trying to work out how many respondents I need for my questionnaire. I have come across different sample size strategies which have confused me. I intend to run SEM analysis and at the same time looking at justifying my sample representativeness. I would appreciate if someone could clarify the right sample size strategy and that how to work out a target sample size from target population in order to be able to reach a certain number of responses.
Many thanks.
Saying you're going to use SEM does not help much, given that structural equations models can range very simple to extremely complex. (E.g., a simple linear regression model can be estimated via SEM.) How large the sample size needs to be will depend on the complexity of your model. If you describe your model more fully, someone who knows more about SEM than I do might be able to offer some advice about sample size.
HTH.
Bruce Weaver thanks for your comment. I am looking at the effect of a process oriented approach and modularization on the level of internal supply chain integration. I have 6 observed variables for the process approach, measured based on 7 point likert scale and each has multiple scales. Modulatization is measured based on 5 items on 7 point likert scale.
You may find answers via google searches such as the link below as a starter. Consider the purpose of your testing and be able to justify the sample size you chose. Experience suggests a larger sample size, if you can afford to do it in cost and time terms, is better as it lessens the risk of sampling error.
https://en.wikipedia.org/wiki/Sample_size_determination
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size