Biostatistics Flashcards
Confounding
a 3rd variable that hasn’t been considered but muddies the exposure & outcome bc is is related to both
i.e. (EtOH & bladder cancer seem associated until you realized smoking & bladder cancer are assoc. and smokers are more likely to drink)
Lead-time Bias
Artificial increase in pt survival time who actually have an unchanged prognosis, when in reality they were just diagnosed earlier
T-test
compares the mean of 2 groups of subjects
Chi-square test
compares the mean of >/= 3 groups; categorical data
Correlation Coefficient
0= no correlation, 1= perfect correlation, -1= perfectly inverse correlation
Positively Skewed Distribution
Mean > Median > Mode
Negatively Skewed Distribution
Mean
Hawthorne Effect
subjects change behavior when they find out they are being studied
Pygmalian Effect
Bias based on investigator bias and the inability to see the results in an unbiased manner