Stratified sampling is the first step by identifying those homogenous groups (strata). I suppose you have already got these strata. The random sample to be selected from each depends on the variability in each subgroup.

The next stage is to determine which data is required from each strata to address the research objectives. It can be primary and/or secondary data taking into account availability of the data.

Dear Faisal

1. The secondary data is research data that has previously been gathered and can be accessed by researchers. The term contrasts with primary data, which is data collected directly from its source.

2. One of the most important benefits of secondary data is the cost of sampling, i.e. it is almost zero.

3. Then, you can increase the sample size to get better estimators in order to decrease the sampling error.

4. 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”.

5. There several types of random samples, Some of which are: Simple Random Sampling, Stratified Random Sampling, Double-stage Random Sampling...

6.Each sampling design has it is principles and conditions and need to be satisfied.

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

Stratified random sampling can be applied on data which have strata. We can take sample from each of strata with certain rule ex. proportional sample.