My questionnaire data is a combination of categorical data and continuous data. According to Cochran, the formula of measuring sample size varies with the data type. So, which formula should I use?
Sample size calculations depend upon the design of your study, the effect size you are expecting and the alpha and beta levels you choose to use (normally 0.05 and 0.2). The best software to use is G*Power.
Cochran is good. I'd stay away from more nebulous hypothesis tests and stick with just standard errors as in Cochran and be careful of bias and nonsampling error.
I think you may be referring to continuous data versus yes/no questions. Both covered in Cochran for different probability designs.
You will need to check both to be sure your sample is adequate for every important y variable. Of those that are important, you'd have to have the biggest sample required, if they are all on the same survey.
Hope I understood your question correctly.
Best wishes.
PS -
I'm referring to Cochran, W.G(1977), Sampling Techniques, 3rd ed., John Wiley & Sons.
There are other good texts as well.
I have a number of texts, but Cochran is a very good one.
Sampling, 3rd ed, 2012, by Steven Thompson, John Wiley & Sons, is a recent addition to my library, and it seems quite good also. A used copy of that may be less expensive. The Pennsylvania State University online course info, freely available on the internet, seems to reference that book.
Thanks, Knaub. Actually, Cochran provided two formulas to calculate sample size. separately. One is for categorical data and another is for continuous data. However, What will be the formula for calculating sample size if I use both kinds of data in my questionnaire?
You need to check each question, using whichever method is appropriate for that particular question - individually. Among the important questions, the maximum sample size needed for any such question is what is needed.
In order to answer this question/problem, several remarks have to be studied.
General remarks:
Research studies are usually carried out on sample of subjects rather than whole populations. The most challenging aspect of fieldwork is drawing a random sample from the target population to which the results of the study would be generalized. The key to a good sample is that it has to be typical of the population from which it is drawn. When the information from a sample is not typical of that in the population in a systematic way, we say that error has occurred. In actual practice, the task is so difficult that several types of errors, i.e. sampling error, non-sampling error, Response error, Processing error,…
In addition, the most important error is the Sampling error, which is statistically defined as the error caused by observing a sample instead of the whole population. The underlying principle that must be followed if we are to have any hope of making inferences from a sample to a population is that the sample be representative of that population. A key way of achieving this is through the use of “randomization”. There several types of random samples, Some of which are: Simple Random Sampling, Stratified Random Sampling, Double-stage Random Sampling... Moreover, the most important sample is the simple random sample which is a sample selected in such a way that every possible sample of the same size is equally likely to be chosen. In order to reduce the sampling error, the simple random sample technique and a large sample size have to be developed.
Specific remarks:
The following factors are highly affected the sample size and need to be identified:
• Population Size,
• Margin of Error,
• Confidence Level (level of significance) and
• Standard of Deviation.
Then, the sample size can be estimated by,
Necessary Sample Size = (z-score or t-value)2 * StdDev*(1-StdDev) / (margin of error)2 .