You need to determine an effect size at first , a confidence level and a power. Then you could determine sample size. It will depend on also the test you are going to perform.
You can make use of GPower, a free software to calculate sample size. If your design is more complicated then you could use PASS.
You can go with power analyses, as Mehmet noted, if there is a test (or tests) which works well for your purposes, but I prefer to avoid hypothesis tests.
See, FYI, https://www.researchgate.net/publication/262971440_Practical_Interpretation_of_Hypothesis_Tests_-_letter_to_the_editor_-_TAS
I prefer to look for a sample size that may yield a desired standard error.
Regardless as to what you do, even with hypothesis testing (done properly), you will be assuming you have reduced your bias to a negligible level (not always true), and then sample until variability is under control. That should be relevant to any kind of data and study.
A pilot study may be helpful.
Cheers - Jim
Article Practical Interpretation of Hypothesis Tests - letter to the...
Dear, Dr. Alta'ee firstly samplecalculate size in medical search depending on the study design type if it is simple or complex experiment and if it is in Vivo or in Vetro , toxic or non toxic agents .......ect. And many other affecting factors.So sir, to determine the size sample you must previously assess for degree confidence level and you can applied the G powerful analysis as you can benefit from the use of PASS and freely software calculation sample and the significant to do an a pilot (small group) study to be avoidance the bias.
Determining the sample sizes involve resource and statistical issues. Usually, researchers regard 100 participants as the minimum sample size when the population is large. However, In most studies the sample size is determined effectively by two factors: (1) the nature of data analysis proposed and (2) estimated response rate.
For example, if you plan to use a linear regression a sample size of 50+ 8K is required, where K is the number of predictors. Some researchers believes it is desirable to have at least 10 respondents for each item being tested in a factor analysis, Further, up to 300 responses is not unusual for Likert scale development according to other researchers.
Another method of calculating the required sample size is using the Power and Sample size program (www.power-analysis.com).