Threats to Internal Validity Flashcards
What is chance?
Where we have the pressence of random error which is inherent in all measurement.
More error is not good
How do we measure chance?
Using the p-value
What is a P value?
Probability representing the strength of evidence to support hte null hypothesis
What is a Large P value of >0.05 represent?
Supports the Null hypothesis hence their is no statistically significant difference between to two groups
What is a samll p value of <0.05 represent?
We reject the null hypothesis hence there is a statistical significance between the two groups’ results
What is the overall theme of what is the p value really saying?
The probability the results are due to chance rather than a real treatment effect
How do we deal with chance?
Increase sample size
Recognize the extent via statistics and interpret the results accordingly
What is important with respect to chance and statistical significance?
Does not necessarily mean clinical significance
What is confounding?
When some factors other than the intervention or exposure under study, influence the overall outcome
What are the ways we can help with confounding?
Randomization
Stratification
Matching
What is randomization?
Ensures groups are similar in all aspects, known and unknown factors
What is stratification?
Where we attempt to force an equal distribution of people from similar groups (Regina and Regina, Sask/Sask)W
What is matching
Where we try and match people in the test/placebo group based on sex, age, and similar characteristics
What is bias?
Problems with the way a study was designed conducted or analyzed that leads to incorrect results or conclusions
What is selection bias?
Problems with how the study subject were selected
What is information bias
Problems with measuing collecting or classifying information (Exposure and or outcomes)
What does selection bias generally affect?
External validity
Where do we see selection bias occur?
Self-seleciton/volunterr bias
Health workers bias
Attrition bias
What are the two ways information bias can be categorized?
Outcome error
Exposure error
What is outcome errors with respect to information bias?
RCT and observaiton studies where we have problems with measuring tools
Problems with actual measurements
What is the exposure errors with respect to information bias
Problems with how subjects are categorized and problems with measuring tools
What is an example of recall bias?
INdividuals remember things differentlu
What is an example of interviewer bias?
Interviewers asks about exposure/outcome differently
What is an example of surveillance/detection bias
One study group followed more closely than the other
How do we deal with bias?
Recognize and acknowledge it
Standardization
Collect the same information form all subjects
blinding
