A stratified sample of participants means the proportion of one group of participants in relation to the others in the real world is reflected in your sample. This can be useful in a study about where a minority group needs to be highlighted and a random sample would not provide enough people of that minority group. For example, investigating how much time the average 18 year old spends on social media per day.
This allows for a representative sample with people from rural towns with no internet to those maybe working in the social media industry. Therefore, one can say that the results of their study can be generalised to the entire population, making its external validity high. The disadvantage to this type of sampling is it’s time consuming nature.