Determine the types of stratified random sampling

There are two kinds of stratified random sampling; (a) proportionate stratified sampling (b) disproportionate stratified sampling. In proportionate stratified sampling, sample size from every stratum is dependent on that stratum's size relative to defined target population. Thus, larger strata are sampled more heavily using this technique as they make up a larger percentage of target population. Conversely, in disproportionate stratified sampling, sample selected from each stratum is independent of that stratum's proportion of total defined target population. There are numerous benefits of stratified sampling including assurance of representativeness, comparison between strata and understanding of every stratum and its unique characteristics. One of the main difficulty however, is to identify correct stratifying variable.