A colleague of mine sent me the following reference which could also be useful to you:

doi:10.1016/j.foodqual.2013.08.012

Actually, it depends on the statistical method which you will be using for data analysis.

A sample size of 100 is enough for Exploratory Factor Analysis, if the model parameters are clear and reliably drawn from strong theoretical bases, but for complicated models, sample should be preferably above 200.

However, for Confirmatory factor analysis, a general rule of minimum of 5 to 10 cases per measure are recommended. 

Thanks

http://goo.gl/jr2y0n

Muhammad -

If I am reading your question correctly, you do not have a population size of N = 150.  You have at least 150 different populations, for each question you ask, and the other caregories, like respondent experience, might be used for stratification, but if you need to estimate and publish for each of those categories, you need to mutiply your number of populations by the number of further categories, to see how many populations you are considering.  You need to consider the population size and sample size from within each of those categories, by question. 

Your sample size depends upon variability of the data for each question of importance to you, and you must consider the type of data.

For continuous and yes/no data, here are two, among many, books you might consult: 

Cochran, W.G(1977), Sampling Techniques, 3rd ed., John Wiley & Sons. 

Blair, E. and Blair, J(2015), Applied Survey Sampling, Sage Publications.

The following book is a clearly written alternative as well: 

Lohr, S.L(2010), Sampling: Design and Analysis, 2nd ed., Brooks/Cole.

I don't know how big your actual population sizes are, but when deciding how many to sample in each case, you should consider not only data variability, but also nonsampling error such as measurement error.  Also you may need to consider finite population correction (fpc) factors. 

For quantitative data, there are two major ways to do inference. Primarily available to you would be randomized design (probability of selection) methods.  (The books above concentrate on that for continuous data, I think you could say.)  There are also model-based methods, and combinations of these two methods.  The sampling methodology and the estimation methodology need to be compatible.

If you are doing a convenience sample, your data will be far less meaningful, and possibly misleading.

Regardless of the kind of data, stratification can help improve accuracy.

You might want to seriously consider a pilot study to help work out your procedure, details, and obtain an idea of variability to help estimate sample size needs in your full-blown study.  

Cheers - Jim

PS - Something else might apply to you.  For example, network sampling/Monroe Sirken.  It depends upon your collection method, data types, needs, etc.  

 Dear Mr. Jim,

Many thanks for your suggestions. Actually, I will conduct survey among the experts who are working in power plant construction projects in my country. So, it is in a specific domain. So far I know, construction of 30 mega projects are going on. The targeted experts are 7-categories, for example, contractors, sub-contractors, vendors, project director (PD), project manager (PM), site engineer, and consulting engineer (or consultant). The other variables are project size in terms of power generation capacity, budget, project location, experts' experience (year), and academic qualification. Could you please suggest me in a precise way to save my time? I am now in a critical moment. I have a presentation just after couple of weeks. Thank you for your patience and time. 

Muhammad -

You have not noted if you are collecting continuous data, some other quantitative data, or qualitative data.  That is very important.  It relates to your goals, what you want to know and how you expect to learn it.  It is also hard to tell when you mean a category of the population, or a question on your survey.  Big difference.  It does not sound like you have stratification, because for anything you categorize, it sounds like you want information on each specific category. We also still do not know what your population size is for each category.

For an example of the confusion here, you note that "The other variables are project size in terms of power generation capacity, budget, project location, experts' experience (year), and academic qualification."  Are those categories (subpopulations),  or do they refer to questions (variables) asked on each survey?

There appears to be a great deal for you to figure out before you talk about sample sizes.  Once you do, you need to have some knowledge of the variability of your data within each group.  I don't think you will be ready to discuss final details for quite some time. Perhaps your presentation should be on the need for a pilot study for some exploratory learning.

Best wishes - Jim

Dear Mr. Jim,

Thank you for your valuable advise.

Actually, the data type is qualitative like the level of a particular risks would be extremely high, very high, high, moderate, low, very low, and extremely low. The 63 identified risks are the observed variables for power plant project. I have attached the questionnaire herewith for your kind attention. The population size in each category of the respondent is unknown to me as well. 

Yes, my presentation is the preliminary presentation but they may ask about number of samples so far I would require to collect for my research.

Thank you again.

Dear Mr. Jim,

Thank you for your valuable advise.

Actually, the data type is qualitative like the level of a particular risks would be extremely high, very high, high, moderate, low, very low, and extremely low. The 63 identified risks are the observed variables for power plant project. I have attached the questionnaire herewith for your kind attention. The population size in each category of the respondent is unknown to me as well. 

Yes, my presentation is the preliminary presentation but they may ask about number of samples so far I would require to collect for my research.

Thank you again.

Sounds like likert-scale data, with which I did not work.  I think it might technically be quantitative, but I'm not certain. Something like generation capacity could be continuous data, but it does not sound like you are using it that way.  Best if you find a book on likert scales. I don't know any. 

Variability of your results is still important for determining sample size needs, which still means preliminary information is needed on that. So a pilot study may be helpful for that reason, and to test your questionnaire and work out logistics problems.

Results for some questions will be more variable than others. 

Suggest you research "likert scales" and "pilot studies." 

Thank you for all of the experts who cordially response to my questions facilitating my research.

Dear Mr. Chalamalla Srinivas,

Many thanks for your response. I hope it will be helpful for me. 

Sekaran (2003) wrote:

Roscoe (1975) proposes the following rules of thumb for determining sample size:

1. Sample sizes larger than 30 and less than 500 are appropriate for most research.

2. Where samples are to be broken into subsamples; (male/females, juniors/seniors, etc.), a minimum sample size of 30 for each category is necessary.

3. In multivariate research (including multiple regression analyses), the sample size should be several times (preferably 10 times or more) as large as the number of variables in the study.

4. For simple experimental research with tight experimental controls (matched pairs, etc.), successful research is possible with samples as small as 10 to 20 in size.

Reference

Sekaran, U., 2003. Research methods for business: A skill building approach. John Wiley & Sons.