*Information written and mounted by
Chris Evans
(Email to: http://www.psyctc.org/cgi-bin/mailto.pl?webmaster)
on 17.vi.95, last updated 31.vi.14. All information mounted under a
Creative Commons License. Please respect the licence but feel free to reuse anything here within the terms that licence allows.
*

- This link takes you to an HTML file that, I think, shows that the issues really are serious and not hard to understand starting from the fairly well known bivariate case of correlations or regressions where there is group mean structure as well as a simple within group regression relationshhip.
- This is the PDF version of the file
- and this is the Rmd file that was used to generate those in Rstudio, see <www.rstudio.com>. It can also be used to generate ODF and M$ Word format files now too if you have OpenOffice or LibreOffice on any system or M$ Word on a Windoze system.

- Program to compute Jacobson et al. "Reliable Change (RC)" criterion for a measure updated to give more explanation of the results (with thanks to Mary Jinks). See http://www.psyctc.org/stats/rcsc1.htm
- Program to compute the confidence interval around an observed (Pearson or Spearman) correlation coefficient

and

Program to compute the meta-analytic confidence interval around observed (Pearson or Spearman) correlation coefficients - Program to compute the confidence interval around an observed sample Cronbach alpha value
- Program to generate all nC2 pairs from n objects for n < 33
- Program to generate various extensions of kappa for multiple raters in inter-rater reliability studies

- Probably the most important information here is that about
the CORE (Clinical Outcomes in Routine Evaluation) system
That's not very statistical as yet but some of the information in there, particularly the powerpoint presentations,
do have statistical aspects.
*Mounted 17.viii.03* - Similarly important on theoretical grounds (I think!) is
my work on probabilities for matching scores "derangements"
This is a very simple method of demonstrating the statistical significance of information transfer through
idiographic techniques.
*Mounted 25.v.01, updated 21.xii.03* - Then there's information about determining reliable and clinically significant change
*Mounted 31.v.2001, tweaked several times in 2003, latest 21.xii.03* - Finally in the top rank of my statistical utilties are my interactive calculators for confidence intervals
of any required spread (95% or other):
- confidence interval of a proportion
*Mounted 12.viii.03* - confidence interval of a difference between two (independent) proportions
*Mounted 14.viii.03* - confidence interval of a mean from mean, s.d. (or s.e.) and
*n**Mounted 16.viii.03, last updated 22.xii.03* - confidence interval of a Cronbach alpha

- confidence interval of a proportion
- Then there's the issue of the confidence intervals for, and inferential testing of, "internal reliability" or "internal consistency" which I think is very much
avoided to the huge detriment of much psychological and psychiatric research. I've mounted some programs:
- feldt1.sas {1kb} for a single value (somewhat less necessary than it was now that you can get this from SPSS RELIABILITY as the CI of the "two-way mixed, consistency" ICC)
- feldt1.ssc similar program for S+ (and probably fine in R too)
- feldt1-table.ssc silly little extension of the above to deal with tables of alphas easily
- feldt2.sas {1kb} which compares two values for independent samples and the same number of items (number of participants can be different)
- feldt2.ssc similar program for S+/R
- feldt3b.sas {2kb} which will accept a listing of pairs of values to be compared

- Then we're down to something I really must come back to develop more some time: repertory grids:
- my SAS/IML program to do an INGRID style analysis of a repertory grid along the lines of the famous program written by the late Patrick Slater
- similar S+ program (I'll try to fix up an interactive version some time)

- Then we're down to a rather random collection of programs I wrote because I couldn't find them elsewhere:
- interactive random permutation generator (1). This calls a perl script random1.prl which takes the lowest and highest number of sequential numbers you want permuted, and a seed, and returns those numbers permuted. E.g. if you ask for numbers between 10 and 18 to be permuted, with a seed of 89 you will get: 13 12 10 16 11 18 15 17 14. Another seed would give you another sequence. Uses the perl rand function. I'd guess it's the same as the typical C function but don't know more! As ever, NO liability accepted!!
- spreadsheet templates for plotting receiver operating characteristics (ROCs or R.O.C.s!; very simple I'm afraid)
- trite SAS program to generate confidence intervals of means from output from SPSS DESCRIPTIVES!
- SAS/IML program gives the probability and cumulative probabilities of r significant
results from n tests under the (generally rather implausible) general null hypothesis.
Crude but can handle large n which simpler implementations in MathCAD 6 which
I was using at the time wouldn't where they produced overflow errors, leastways, they did back then.
*mounted 16.iii.98*