a form of probabilistic sampling in which the region or site is divided into natural zones or strata such as cultivated land and forest; units ate then chosen by a random number procedure so as to give each zone a number of squares proportional to its area, thus overcoming the inherent bias in simple random sampling.
Stratified random sampling (STRS) occurs when we divide our population into separate groups of units called strata. We then sample independetly from each stratum.
The sample is organised into different strata. A random sample is selected within each strata. Providing the sampling frame is accurate this will mean that the final sample chosen will be representative in terms of the chosen strata. This method is often used to ensure that a sample is representative in terms of geographical areas.