I would like to find the first published instance of the erroneous combination of Fisher's p-value based significance testing and Neyman-Pearson error-rate based Hypothesis testing. This happened sometime between 1928 and 1940.
This paper explains the phenomenon I am talking about.
http://ftp.stat.duke.edu/WorkingPapers/03-26.pdfThe paper/book must be published before this earliest reference I have found:
Lindquist, E.F., 1940. Statistical Analysis in Educational Research. Houghton Mifflin, Boston.2 btc reward and you will have made a great contribution to modern science. Extra reward for images of the cover and page in which this occurs.
PM me if interested.