Biostats 5 Flashcards
What type of graph is used to represent a chi squared test
bar graph; no whisker
no variance or distribution
What is the use of correlations
can establish predictions, do not imply causation
What is the Bradford-Hill criteria for causality
strength consistency specificity temporality biological gradient plausibility coherence experiment analogy
What is strength relate to causality
larger association is more likely causal effect
What is consistency related to causality
stronger association when same findings observed by different people and different samples
How does specificity relate to causality
specific population with specific disease and no other likely explanation
Larger specificy means what with causation
bigger probability of causal effect
How does temporality relate to causation
effect happens after a cause
How is biological gradient related to causation
more exposure= greater risk
How is the degree of correlaiton measured
Pearson product-moment correlation coefficient
What does the pearson product moment correlation coefficient measure
linear relationship between two variables
strength and direction of association
What are the values of Pearson Product coefficient
+1 to -1
What does a Pearson Product of 0 mean
no correlation
What is R2
coefficient of determination; determines how well data fits a statistical model
What is the alternative test for pearson test if data is skewed
Speraman’s rank
What caution must be taken when using R2 values
do not know if the correlation is positive or negative, must have a graph to report that
How is the degree of correlation measured in the Spearman’s rank test
p (rho)
What does rho measure
extent to which as one variable increases, the other variable tends to increase
What cannot a confidence level include
0
What will be indicated if the confidence level includes 0
not considered significant
What is the use of survival analysis
used to analyze data in which time until event is of interest
What is selection bias
sample is not representative of the population
What is Berkson bias
disease prevalance is estimated from hospital records
What is measurement bias
leading questions
What is Hawthorne effect
participant behavior altered by knowing they are being studied
What is experimenter bias (pygmalion effect)
expected results are accidently communicated to participants
What is lead-time bias
false estimate of survival rates; patients appear to survive longer after new screening test
What is recall bias
Participants do not accurately recall events
Issue in retrospective studies
What is late-look bias
participants with severe disease less likely to be identified because they die first
What is confounding bias
a factor in a study is related to co-factor being examined
WHat is design
choosing inappropriate study design to answer questions
