Robert,

You (or the user) should be able to investigate that your/herself by running a PROC FREQ on the concerning variables. I don't need to know it, you don't need to know it, but I think the user should know it, as she is the one working with the data and she should know her data. I would anyway.

Besides why should a missing 2-digit year be coded as 99? 99 is a very valid value. It might better be coded as something like -9. Maybe it already was coded useful (see the output of PROC FREQ). Are you sure there are no legal values of 99? If I were responsible for that I would forbid the use of 99 that way; besides I would make years 4 digits and leave or make missings really missing (a period or some other distinctive missing value).

Regards - Jim. -- Y. Groeneveld, MSc IMRO TRAMARKO tel. +31 412 407 070 senior statistician P.O. Box 1 fax. +31 412 407 080 5350 AA BERGHEM IMRO TRAMARKO: a CRO J.Groeneveld@ITGroups.com the Netherlands in clinical research

"My job is to keep my computer working." - Jim Groeneveld

> -----Original Message----- > From: Robert Andrew Bowell Jr. [SMTP:bowell@nbcs.rutgers.edu] > Sent: Friday, October 29, 1999 2:51 PM > To: Jim Groeneveld > Subject: RE: If/then > > On Fri, 29 Oct 1999, Jim Groeneveld wrote: > > > Are you sure the missing data actually were coded as . (a period)? > > Maybe they were coded differently, e.g. one of ._, or .A to .Z or even > > just some negative number. That would all fit your successful recode > > you indicate below. > > She might have coded them differently, yeah, but I doubt it > because then I think she'd've known that, right? In any case, I asked her > for the complete program, but since she feels the problem has been solved, > she doesn't want to go any further and send me the code. > > Y'obt., > Rob Bowell > ------------------------------------------------- > Programmer/Analyst, RUCS User Services Group > Hill Center, Room 209 > Phone: 732-445-0481/Fax: 732-445-5539 > -------------------------------------------------