Hello Appalayya,

Perhaps if you could give a little more explanation about the specific problem you're trying to solve, that would help.

I presume you are trying to determine the sample size needed in order to estimate a population parameter within some desired level of precision (with some desired level of confidence). Is that correct. If so:

First, you'll need to specify which population parameter you'd like to estimate (e.g., mean, median, mode, interquartile range, or something else?).

Second, I'm not sure how one determines that the population variance for scores from a 7-point scale = 7/6. As an example, if I have sample values of [1,1,1,7,7,7] the population variance estimate is 10.8. It is true that, if you have a set of untied ranks of scores (1..n), then the variance of the ranks may be determined by the sample size, n: variance = n * (n + 1) / 12. But I don't think that applies to this situation.

Third, we could no doubt come up with examples for which the population variance of a characteristic measured with a 7-point scale is less, more, or equal to that for another characteristic measured with a 5-point scale. It's not just the number of points on a finite scale, it's the shape of the distribution as well.

Good luck with your work.

Thanks a lot, Prof. David Morse.

Please read the following file.

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.486.8295&rep=rep1&type=pdf