I have a research article wherein the authors have conducted a study. They have selected a particular industry and surveyed people using a questionnaire and the response rate is only 32%. However, the result has been generalized to the whole of the country! I'm wondering if that is good enough? is it fair to generalize?
The response rate of survey questionairs more than 30% is sufficiant for nanlysis purpose.
It is good enough to generalize the results but it also depends upon the sample design parameters . Say population and sample size and confidence level.
U may use cochrans formulae for actual calculations.
At the same time you may refer to different types of research validity such as construct, content .conclusion/internal etc. to validate your results. For further details, u may refer to
1. Gubrium, J., & Holstein, J. (2002). Handbook of interview research: Context & method.Sage Publications
2. Runeson, P., & Höst, M. (2009). Guidelines for conducting and reporting case study research in software engineering.Empirical Software Engineering, 14(2), 131–164
Depends on the size of the sample. If 32% refers to more than a thousands of answers, the reserach error margin would stay bellow 3%. If the number of answers sits bellow one hundred answers, the margin of error will go up much higher, above 10%. To calculate the sample Margin of error at 95% confidence is approx 0.98/sqrt {n}, being n the sample size
- Badges
- Science topic
- Similar topics
- Econ
- Manufacturing
- Industry