I conducted a research study on banking sector. I have floated 1000 questionnaire through simple random sampling strategy. And received 459 filled questionnaire. My calculated required sample size as per Krejice and Morgan illustration of finite population was 365 respondents. Although the received 493 responses fullfills the required minimum sample size (365). But the response rate is too low (45% only).
This is also highlighted by reviewer of my paper and I have to provide justification for the generalization of my results.
At this stage, I cannot go back for data collection as my research paper has cleared initial screening and other recommendations of reviewer.
Please anybody help in light of existing body of knowledge, what can be done and what justification can be provided?
Thanks in advance
You could have distributed the questionnaire electronically, via email or any social media.
Hi, Malkah,
What information is publicly available that might let you compare some of the demographics between the respondents and those who did not respond? If you have access to that kind of data, you can compare the two groups and, perhaps, make the argument that there are few or no apparent differences between those who responded and those who did not. It isn't the strongest argument possible, but it would support generalizability within reasonable limits.
Hi Malkah, go with what Prof Peter said.
Alternatively, you can evaluate the pattern of the non-response. If it’s at random — i.e the non-response is not from a particular section — you can conveniently make the case of non-response being at random. This means the non-respondents and the respondents were not significantly different in distribution.
Another qualitiative way to go about it is to further search the literature to find studies which support such low response rates (e.g. 46% in your case) as valid in cross-sectional studies. There are such studies which have used real sample and simulations to demonstrate such.
Good luck.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size