Hello Paul, The conventional classification is due to Stevens (1946): nominal, ordinal, interval, ratio.
Current practice in statistics is ordered logistic regression, which treats the response variable as nominal (binomial). The most readable text is Hoffmann, 2004 Generalized Linear Models. There are many other GzLM texts. Hope that helps, David Schneider Quoting "Klawinski, Paul" <[email protected]>: > Hi all, > > I am trying to educate my colleagues about some things and would like to ask > you some questions as you are members of a talented and learned society. > I am aware of the controversy so am interested in what I would consider > common knowledge so, if you respond, please do so without consulting a > stats text. > > Feel free to answer off list. > > What are the different types of data (nominal, ordinal, etc.)? > > For each data type, list what types of statistical measures you can > compute for each. > > Which data type do Likert style data fall into (responses using phrases > like Strongly Disagree, Disagree, Neither Agree nor Disagree, Agree, > Strongly Agree even if you identify these responses with a number)? > > Comments or questions? > > Thanks, > > Paul > -- > Paul Klawinski, Ph.D. > > Monte Harmon Professor of Biology | Department of Biology > 816-415-7628 (office) | 816-781-7700 ext. 5565 (lab) > > William Jewell College: Live What You Learn | www.jewell.edu > <http://<http:/>www.jewell.edu/> > Please join Jewell in conservation efforts. Print this email only if > necessary. > > Luck favors the prepared. > Edna E Mode > > This electronic communication is governed by the terms and conditions at http://www.mun.ca/cc/policies/electronic_communications_disclaimer_2012.php
