You should hv done a tonne of stuff on bias
Straight out of Excel book - Bias occurs when atttiude of person collecting data affects what data is included in the sample.
eg collecting data from mainly whoite, Anglo Saxon males about a general population issue would result in biassed data.
or my example tends to be
a phone survey only conducted between 10am and 2 pm would be bias against students and people who worked, since they wouldnt normally be there to give their opinion.
The data would not be truly representative and thus not really accurate.
Surveys often ask the same question, but in different ways to minimise bias in response. Questionairres that do not have to have your name on it .. can give both good and bad responses, honest - cause you arent scared of consequences, inaccurate cause without yr name you dont give a damn.