|MacOS 8.5 Incompatibility:
Note that P-Stat is not compatible with MacOS 8.5. I hope to solve this problem, but currently am unable to do so. As soon as an 8.5-friendly version of P-Stat is available, it will be posted on this web site. Meanwhile, you can still use P-Stat on MacOS 8.1.
Progress Report (1 March 1999): It looks grim. The bug is clearly in the commercial interface modules rather than in my own code. These interface modules are no longer published or updated. Therefore, in order to fix P-Stat I'll have to rebuild it from the ground up using more up-to-date development tools. And that means I'll have to learn a new programming language (C++ instead of Pascal), a new IDE (Metrowerks instead of THINK) and a new set of interface tools (such as PowerPlant). Actually, I would like to learn all these things, but don't have a timetable for it at this time.
Frequently Asked Questions
|What is P-Stat?||Macintosh software for permutation statistics. Permutation tests, sometimes called "randomization tests," are the most basic and powerful kinds of statistical methods. In fact, some would say that they form the logical basis for all other procedures.|
|How can I obtain these programs?||They can be downloaded from this web site.|
|How much do they cost?||They are freeware, which means you don't have to pay, but I retain the copyright.|
|Does P-Stat have any Y2K problems?||No. The only date or time handling by P-Stat is done via the MacOS toolbox. According to Apple, this should present no problems.|
|What is the latest version of P-Stat||Version 0016 (beta) is currently available. It fixes a bug in the contingency table function of previous versions in which the probability of the observed table was sometimes added twice.|
|What if I have trouble getting these programs to work?||Well, keep in mind that the main reason I don't charge money for these programs is so that I don't have to support them. So you are pretty much on your own. If you have some familiarity with the Macintosh and with statistics you should have no trouble. Also, there is some (incomplete) documentation that comes with the software.|
|But what if I really have to contact you about this software for some reason?||You can send me email.|
|What is the "categorical statistics" package that is also available for downloading from this web site?||These are programs for doing Fisher's Exact Test, Monte Carlo contingency table tests and binomial tests, that I wrote before P-Stat. The routines in P-Stat supersede some but not all of the older programs, so you might want to download both. For example, you can use P-Stat to do the two-tail version of Fisher's Exact Test, but you need the older Fish6 for a one-tail test. Unfortunately, some of the older programs are only partly compatible with recent (system 7 and beyond) versions of MacOS. They work, but it isn't pretty.|
|Are these programs PowerPC-native?||No, I'm sorry to say. They were done on an old 68K compiler from Symantec, and I have not converted them to work with Metrowerks. However, with Speed Doubler you can get them to run blazingly fast.|
|Did discussions with Jeff Rohl, Frank Ruskey and Dominique Roelants van Baronaigien contribute to this software?||Yes! And I thank them for their ideas and suggestions.|
|What sort of legal mumbo-jumbo does your employer ask that you include?||
This software is provided "as is", without express or implied warranty, and with no support nor obligation to assist in its use, correction, modification or enhancement. We assume no liability with respect to the correctness of information provided, infringement of copyrights, trade secrets, or any patents, and are not responsible for consequential damages. Proper use of this software is entirely the responsibility of the user. This software may not be sold or used for profit.
|Will there ever be a DOS or Windows version?||Not by me.|
|Do you have any other free software available?||Yes, check out Amplify if you are interested in PCR.|
Download P-Stat or Categorical Statistics
Go to Bill Engels's home page