Cut-and-paste then run the SPSS syntax at the end of this message. this should help you check your results.
this is a .pdf document on how to calculate the significance of a t . http://www.spss.com/tech/stat/algorithms/11.0/app02_sig_t.pdf this is another .pdf document that shows how get cdf and idf (p. 11) http://www.spss.com/tech/stat/algorithms/11.0/app12_cdf_pdf_rand.pdf Hope this helps. Art [EMAIL PROTECTED] Social Research Consultants University Park, MD USA (301) 864-5570 * this will read t and df and output p. new file. data list free/ t, df. begin data -7.22 21.5 3.99 14 5.2 20 1.96 300 1.645 300 -1.96 50 -.33 22.5 -2. 18 -1.66 66 end data. compute p= cdf.t(t,df). format t (f7.2) df (f7.1) p (f7.3) . list. *this will read p and output t. new file. data list free/ p, df . begin data .06 10 .051 12.5 .050 15 .0501 17.5 .049 22 .0499 33 .01 27.5 .001 27.5 end data. compute t= idf.t(p,df). format t (f7.2) p (f7.3) df (f7.1). list. Martin Russell wrote: > I've been trying to find out for a few days now how to calculate the value > of t that's exceeded given the degrees of freedom and the probability the > value will be exceeded. I've found some ways of calculating the probability, > given the t value and df, but it seems that they all fail for at least one > value for df. So, could anyone help me with this? > > The reason I'm asking is I'm trying to program my calculator to work it out, > and it doesn't have many stat functions. Also, I could take an answer in > some computer language and convert it to work on my calculator, provided it > isn't too complex. Thanks in advance :) > > . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
