Hi,

Cluster sampling is a good way to go.

Now the financial issue, I can suggest the following:

after the schools are selected, you may try to identify research assistants from the same area,, through Internet. you have to weigh the cost of your travelling vs. offering the RA some money for their work.

or, you can offer them to be as coauthor the publicaiton.

regarding the point you raised, that some universities will not allow you to do the study,,, then you repeat the random cluster selection to replace the refusing University.

hope this help

good luck

Alysson  -

The idea of cluster sampling is to reduce travel cost.  If the universities are of different sizes, this changes the probabilities though. If you were collecting quantitative data like continuous or yes/no data, then this is made more clear in textbooks such as

Cochran, W.G(1977), Sampling Techniques, 3rd ed., John Wiley & Sons, and may be helpful, regardless. 

With regard to the second stage, you said "...if the university will allow me to apply the survey with all students."  However, you could just pick a random selection of students (or classes).  A sequential sample may be close enough to a random sample, if your "professors' network can just take a class list and select every 3rd or 4th person, whatever you decide.

The total number of students you need depends upon both stages of selection, and the variability of your answers (standard deviations for quantitative data), for each of the questions, so you may need to look at which ones are most important to you. 

It may be best to start with a small pilot study to work out logistical problems, test your question, estimate sample size needs for the full study, and work out any other problems.

Cheers - Jim

A sua ideia de cluster é muito boa. De fato ela é consistente com uma amostragem estratificada da população. 

Cluster sampling is a good way to go. Regarding the financial issue, I can suggest the following:

after the schools are selected, you may try to identify research assistants from the same area,, through Internet. you have to weigh the cost of your travelling vs. offering the RA some money for their work.

or, you can offer them to be as coauthor the publicaiton.

regarding the point you raised, that some universities will not allow you to do the study,,, then you repeat the random cluster selection to replace the refusing University.

Refusals generally would mean 'not missing at random,' which creates bias for any inference based on probability of selection, if you just replace them with more cooperative responders.  You could consider response propensity groups if you have a nonresponse problem. 

Dear

In order to answer this question/problem, several remarks have to be studied.

1. 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.

2. 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 .

In addition, there are several types of Probability Sampling which can be used in your research. Some of which are:

· Simple random sampling,

· Stratified Random Sampling,

· Systematic Sampling,

· Cluster Random Sampling and

· Multi-Stage Random sampling .

Also, there several types of Non-Probability Sampling Some of which are:

· Convenience Sampling,

· Haphazard Sampling,

· Purposive Sampling,

· Expert Sampling:

· Heterogeneity Sampling / Diversity Sampling,

· Modal Instance Sampling,

· Quota Sampling and

· Snowball Sampling.

For more details, please refer to

Cochran,W.G. (1977). Sampling techniques, Wiley.

Snedecor, W.G. and Cochran,W.G. (1989). Statistical Methods, Blackwell.

Regards,

Zuhair

If the population is heterogeneous, you should use the cluster sampling. Also, you can utilize the stratified sampling. But, cluster sampling is the best.