Thanks for your post. Stratified sampling is a probability sampling method that is usually paired with quantitative research so that the results can be generalized. If you use this strategy, you must ensure your sample is representative of your population. I am not sure you have a clear understanding of who your population is because you won't necessarily be receiving one by asking different agencies for participants. Also, if you are conducting a quantitative study, you will need hundreds of participants to ensure you have enough power to analyze your independent and dependent variables. After saying all that, why are you opposed to convenient sampling (a non-probability sampling method), which social workers use all the time to conduct our research? I don't understand why this method is not a better option for you. Thoughts