Not zero, but what you learned is highly subject to bias and error from your small sample of individuals. You are likely to have drawn conclusions from your con

Not zero, but what you learned is highly subject to bias and error from your small sample of individuals. You are likely to have drawn conclusions from your conversation that are not broadly applicable or possibly even true for those individuals because in 5 informal conversations likely: * you didn't interview enough people from different backgrounds * you didn't control for the specific psychological conditions of your interviewees or what treatments/medications were in play at the same time or before / after the program * you didn't test your question set to ensure that it doesn't have leading questions or questions can otherwise influence the results * and so on. I could go on, but the point is anecdotes are not _nothing_ but they pale in comparison to a well built scientific study. If you want to really know if a type of treatment works well and who it works well for with a minimum of bias then science is by far the best option. It's normal and fine to use anecdotes to guide personal day to day decisions in life, but if you want to make broad claims that are taken seriously by professionals trying to say help folks manage mental health issues then we have a higher standard than anecdote and rightly so.