I am looking for the history of the student t test. I was wondering about it origins. I have to write an essay on the subject for a Research Methods class, I am a student at The College of St. Scholastica in Minnesota. Thanks, Lisa Lundgren

-- Lisa Lundgren (lundgren5@charter.net), January 16, 2003

Answers

This is not specific to the t test but it may be useful. See: Stephen M. Stigler (november 1992) A historical view of statistical concepts in psycholohy and educational research. American Journal of Education,101,pp60-70.

-- Rock Faulkner (rock.faulkner@umontreal.ca), January 16, 2003.

The t-test was developed by "Student," whose actual name was Gossett, but the company he was working for, Guiness Brewery, wouldn't let him publish under his own name. There are a number of histories of statistics that tell the story. In addition to Stigler, mention above, is _The Empire of Chance_ by Gerd Gigerenzer and a bunch of other folks. Ian Hacking has also written an excellent two-volume history of probabilty and statistics more generally.

-- Christopher Green (cgreen@chass.utoronto.ca), January 16, 2003.

You might also want to look at Michael Cowles' text _Statistics in psychology : an historical perspective_. Like many statistical innovations, the t-test was developed primarily to satisfy practical needs (in this case, making inferences from small samples instead of trying to collect whole populations). Gosset is generally credited with deriving the sampling distribution of t in 1908. He was originally recruited by Guiness Brewery in 1899 by Lord Guiness. The approach by Gosset helped to pave the way for Fisher's Analysis of Variance.

In addition to the text above, the following websites contain some general info about Gosset and the t-test:

http://www-gap.dcs.st-and.ac.uk/~history/Mathematicians/Gosset.html

http://nimitz.mcs.kent.edu/~blewis/stat/tTest.html

-- Daniel J. Denis (dand@yorku.ca), January 16, 2003.