Sample size is determined on the basis of statistical laws. For example, if one wants to predict voting analysis and asks 1000 people even in a representative sample, the error of measurement is still few %.

If there are many questions with many possible answers, and then one wants to use some regression analysis, a sample should be definitely more than 30-50 people.The more - the better. This is because a cloud of uncertainly can hide potential correlation between variables for too small samples.

Dear Darshan,

Please check out the following links. I found them relevant and useful.

Regards,

Seyed Ehsan Afsahi

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3775042/

https://www.checkmarket.com/blog/how-to-estimate-your-population-and-survey-sample-size/

I am certain that the following is of good use:

Lui, Kung-Jong

Crossover designs : testing, estimation, and sample size

Wiley, Statistics in practice

2016

Chul Ahn; Moonseoung Heo; Song Zhang

Sample size calculations for clustered and longitudinal outcomes in clinical research

9781466556270, 1466556277

CRC Press LLC, Chapman & Hall/CRC biostatistics series

2015

Patrick Dattalo

Determining Sample Size: Balancing Power, Precision, and Practicality (Pocket Guides to Social Work Research Methods)

0195315499, 9780195315493, 9780198043089

2008

David Machin, Michael J. Campbell, Say Beng Tan, Sze Huey Tan(auth.)

Sample Size Tables for Clinical Studies, Third Edition

9781405146500, 9781444300710

Wiley-Blackwell

2008