Note that a "small sample" does not mean the same thing as a "represntative sample". You really care about the latter, and the former will only sometimes skew t

Note that a "small sample" does not mean the same thing as a "represntative sample". You really care about the latter, and the former will only sometimes skew the latter. "Sample size is too small, so I don't want to consider X paper" always drives me nuts. A better argument is that since its voluntary, people who are lazy will cancel and never explain and perhaps they are a majority of the user population. Please use that kind of reasoning instead of the knee jerk "300 sounds like a small number". (Sorry, serious pet peeve)