 |
 |
Dear all I agree on stats - get an expert. Stacie Conchie in our Fieldwork is good study was a psychology PhD student drafted in to look after statistical analysis of the data we collected. She was excellent and is now on staff in our psycho dept. Regards Alan ------------------------------------------------------------- Dr. A.P. Boyle School of Environmental Sciences, Herdman Building, 4 Brownlow Street University of Liverpool, Liverpool L69 3GP, U.K. Tel: (0151) 794 5154; FAX:(0151) 794 5196; Departmental home page: http://www.liv.ac.uk/earth "The great tragedy of Science - the slaying of a beautiful hypothesis by an ugly fact." (Huxley, 1825 - 1895) ------------------------------------------------------------- -----Original Message----- From: Cesar Delgado [mailto:[log in to unmask]] Sent: 17 June 2010 14:48 To: [log in to unmask] Subject: Re: research in education Jeff Dodick's comment is very true - you can't always just use a t- test. The nature of the variables play a role in determining what test is appropriate. This site can provide some insight: http://www.ats.ucla.edu/stat/spss/whatstat/ but it would be best to consult a stats expert. Cesar Delgado * Assistant Professor * Science and Mathematics Education * 1 University Station D5700 * University of Texas at Austin * Austin, TX 78712 * Phone: (512) 232 9682 On Jun 17, 2010, at 4:09 AM, Britta Bookhagen wrote: > Greetings everyone > > I was wondering whether someone could recommend one or two educational > papers, that use pre-and posts tests to assess students achievements > from a program taught (I actually have two post-tests, a long term and > a short term). > > I am about to write things up for a scientific paper and I am not sure > whether I should use repeated measurements or two series of t-tests to > analyze the quantitative questions. Does anyone have experience with > this topic? There are just too many papers out there and I am not sure > about their reliability. > > I would very much appreciate any suggestions. > > Thank you very much > > Britta