Statistical Software and the Year 2000
April 6, 1998
We hear enough about the catastrophic computer problems that will arise after the change to the new millennium, but how will this affect data used in statistical software?
At this point, most statistical software programs have been upgraded to be "year 2000 compliant." SAS, SPSS and Minitab, amongst other statistical software, have specific formats and functions to enter and manipulate dates. To be "year 2000 compliant" means that the software will be able to interpret the dates correctly, and give the right result after data manipulations with dates (for example, when calculating elapsed time between two events). An important prerequisite for this, though, is that you must have entered years with four digits.
If you have entered years with only two digits, SAS, SPSS and Minitab will assume that the years are from the twentieth century and thus will add 19 as the first two digits. SAS and SPSS have options to overwrite this default.
In case you entered only two digits for year, version 12 of Minitab (the newest version) will use as the default first digits 19 if years are >=20 and 20 if years are <20. Note that even when you enter four digits for the years in Minitab's date format, only two digits will appear in the worksheet. A quick test indicates, though, that Minitab has a correct interpretation for this date compared to just entering a two-digit year.
One good resolution, if you have not yet done so, would be to start entering years in four digit form from now on.
If you need additional information or if you need some help assessing if statistical programs involving dates are still adequate for the next century, feel free to contact the Office of Statistical Consulting.
Author: Francoise Vermeylen
(This newsletter was distributed to faculty and graduate students in the Division of Nutritional Sciences and the College of Human Ecology, and faculty in the College of Agriculture and Life Sciences, by the Office of Statistical Consulting. Please forward it to any interested colleagues and research staff. Anyone not receiving this newsletter who would like to be added to the mailing list for future newsletters should contact statcons@cornell.edu. Information about the Office of Statistical Consulting can be obtained at World Wide Web address http://www.cscu.cornell.edu
.)