Thank you for your insight. I am in material engineering, and I emphasize having representative data, but I have heard at conferences that the results shown are sometimes the top outliers, outside of the average.

I completely agree about the publication of negative results. Many times I have wondered how many people have tried the same idea, only to find out it didn't work and did not or could not publish it. And thus another team will spend effort and money because nothing was ever reported.