[EMAIL PROTECTED] (Scheltema, Karen) wrote: >A matched case-control design is one in which an investigator has identified >individuals (cases) with Disease Y. There is also a group (controls) who has not >contracted Disease Y. The investigator wishes to determine if Exposure X is related >to Disease Y. However, the cases and controls differ on more than just the exposure >variable. Therefore, the investigator selects characteristics that may be >confounders, such as age, gender, socioeconomic status, etc. and for each case >identifies a control who has the same values on the confounding variables that are >being matched on. Sometimes, investigators will select more than 2 or more controls >for each case, but it's a matched design. Each case has its own control. Hope this >helps.
OK, now I got it. Thanks. Peter . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
