43 bytes added ,  16:31, 14 December 2010
no edit summary
Line 16: Line 16:     
==Information bias==
 
==Information bias==
Information bias results from errors in measurement or classification of exposures or outcomes of interest amongst the individuals included in the study. In the case of analytic studies, this may be classified as '''differential''' or '''non-differential''' (or ''measurement error''):<br>
+
Information bias results from errors in measurement or classification of exposures or outcomes of interest amongst the individuals included in the study. In the case of analytic studies, this may be classified as '''differential''' or '''non-differential''':<br>
 
* Differential bias occurs when the chance of bias is different for the different groups being compared. For example, in a study investigating whether Boxer dogs have a higher incidence of mast cell tumours than other breeds, Boxers may be more likely to be diagnosed by vets as having these, due to a postulated breed disposition.
 
* Differential bias occurs when the chance of bias is different for the different groups being compared. For example, in a study investigating whether Boxer dogs have a higher incidence of mast cell tumours than other breeds, Boxers may be more likely to be diagnosed by vets as having these, due to a postulated breed disposition.
* Non-differential bias occurs when the chance of bias is not affected by the group the individuals belong to.
+
* Non-differential bias occurs when the chance of bias is not affected by the group the individuals belong to. '''Measurement error''' is a common example of non-differential bias.
 
Non-differential bias will tend to reduce the strength of any association present, and will increase the probability of a [[Random variation#Hypothesis testing and study power|type II error]], whereas the effect of differential bias cannot be predicted.
 
Non-differential bias will tend to reduce the strength of any association present, and will increase the probability of a [[Random variation#Hypothesis testing and study power|type II error]], whereas the effect of differential bias cannot be predicted.
  
700

edits