RT Book, Section
A1 Greenberg, Raymond S.
A1 Daniels, Stephen R.
A1 Flanders, W. Dana
A1 Eley, John William
A1 Boring, John R.
SR Print(0)
ID 546168
T1 Chapter 10. Variability & Bias
T2 Medical Epidemiology, 4e
YR 2005
FD 2005
PB The McGraw-Hill Companies
PP New York, NY
SN 9780071416375
LK accessmedicine.mhmedical.com/content.aspx?aid=546168
RD 2024/04/19
AB Validity  is  the  extent  to  which  a  measurement  or  study  reaches  the  correct  conclusion.  Bias  is  a  systematic  error  that  occurs  in  a  study  and  leads  to  a  distorted  result.  Selection  bias  is  a  distortion  in  the  results  of  a  study  that  arises  because  of  the  manner  in  which  the  subjects  are  sampled.  Information  (or  misclassification)  bias  is  a  distortion  in  the  results  of  a  study  that  arises  because  of  the  manner  in  which  data  are  collected  or  classified.  When  errors  in  classification  of  exposure  or  disease  status  do  not  depend  on  the  other  variable,  any  bias  is  referred  to  as  nondifferential,  and  tends  to  underestimate  the  association.  When  errors  in  classification  of  exposure  or  disease  status  depend  upon  the  other  variable,  any  bias  is  referred  to  as  differential,  and  may  lead  to  over-  or  underestimating  the  association.  Confounding  is  the  mixing  of  the  influence  of  an  extraneous  factor  with  the  effect  of  the  exposure  of  interest.