i wish i can help:

you can select:

multistage cluster sampling method

it is an statistical operation by which you can select aggregates of study units from a list of aggregates presented as geographical areas rather than individual in a sampling frame. i.e. your sample frame here is a  list of provinces each province here is called cluster because it is refers to a group of individuals.   you select 1/3 or say one. another stage of clustering can be done till you reach manageable cluster that can be studied.

Continuing the previous answer, once you reach the mangaeable cluster, make a sampling frame from voter's list or some similar list and then sample (simple/systematic) from that.

You will have to add a design effect in your sample size when there are stages in sampling.

Attached is an old document that helps understanding the design effect. I found it useful. I hope it is useful for you.

I think that you need to go on as follows:

You should apply one of the following methods which helps you to achieve a conceptualized model through a proper Explanation :

• Delphi
• ELECTRE

If you choose the ELECTRE method which is more proper than Delphi, you should go as follows:

Section 1: Explanation: Finding the model through Expert's opinion

You should firstly arrange the variables which are in your study 5x5 in which you can produce your question as follows:

• 5x5: 25-5=20 questions
• using the Likert scale

Set the questionnaire by making any communication between the variables.

After all you can set up the Opinion Matrix for any Expert dominated in your research.All opinion should be changed into unique answer through G-AHP. The group matrix resulted from G-AHP can be processed through ELECTRE method and Shannon's Entropy.

Finally, the matrix achieved by ELECTRE will show you how is the relations and their impacts between the variables and this is your legalized model of the study.

Section 2: Verification: Proving the model in practice

After achievment of the conceptual model of your study you need to extract the second type questionnaire for the statistical society which is difference with Experts' society mentioned in the section (1) You can choose any industry or nation for your purpose as the new society. So you can start the study questionnaire involving the explained conceptual model the study.

---------------------------------------------------------------------------------------------------------

You should also apply the Reliability and the Validity of Measurement Instruments involving relevant scale and model:

A. Reliability of measuring instruments through one of the following methods:

• Re-execution (retest method)
• Parallel method or using peer tests
• Split method (two halves)
• Coter-Richardson Method
• Cronbach Alpha Method
• Also you should consider the factors affecting the reliability coefficient

B. Validity of measuring instrument through one the following methods:

• Content validation
• Criterion validation
• Construct validation
• Agent validation

For more information you can read the article as attached to find respective methods for using in your study.

Finally I recommend you to read second attached article to consider SAM model in your study for evaluation.

Article Explanation of the " Strategic Marketing Management " model ...

Article Explanation of the“Strategic Assessment Model” and verificat...