Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
26 Cards in this Set
- Front
- Back
Prevalence of B
|
Proportion of a population that has B at a point in time
|
|
Incidence of B
|
Rate of NEW cases of B that appear in a population
|
|
Confounding
|
A not really the cause of B. C is instead.
|
|
Temporality
|
Cause precedes effect
|
|
Case Report
|
-Observational study of an individual
-Will be at least one case where A is associated with B. |
|
Case Series
|
-Observational study of a group
|
|
Cross-sectional
|
-Observational study at one point in time
-Studies "prevalence" -Pros: easy to conduct -Cons: No temporality |
|
Case Control
|
-Observational & Retrospective
-Start with people who have the disease and go back and look at what they have been previously exposed to. |
|
Cohort
|
-Observational & Longitudinal
-Look at the "incidence" -Follow a group of "healthy" people forward in time, waiting for axe to fall. -Temorality -Non-interventional |
|
Randomized, control trial
|
-Interventional & longitudinal
-Look at the "incidence" -NOT susceptible to confounding -temporality established |
|
The only study that is not susceptible to confounding?
|
Randomized, controlled trial
|
|
Studies that establishes temporality?
|
Cohort & randomized control
|
|
Study that may have ethical issues?
|
Randomized controlled trial
|
|
Null hypothesis (Ho)
|
The predicted difference does not exist
|
|
Alternative hypothesis (H1)
|
The predicted difference does exist.
|
|
Statistics
|
Give us an estimate of the probability that a particular observed difference occurred by chance alone.
|
|
What factors affect probability (p)
|
1. Size of observed difference (bigger size, lower p)
2. # of subjects in the sample population (more subjects, lower p) 3. Variability of outcome (more consistent difference, lower p) 4. lower p means less probability of getting a result by chance; so better chance that our hypothesis is right. |
|
If p=0.001
|
.1% probability of finding this difference in the sample alone
Reject the null hypothesis |
|
If p=0.31
|
Accept the null hypothesis
|
|
alpha = tolerance for error
|
alpha=0.05
|
|
if p less than alpha
|
-Difference is statistically significant
-We accept H1 |
|
If p is greater than alpha
|
-Difference is not statistically significant
-We accept Ho |
|
Alpha
Type I, False positive |
-There is a statistically significant difference when there really is no difference
-Type I, False Positive |
|
1-alpha
No error, true negative |
-Concluding that there is no statistically significant difference when there really is none
-No error -True negative |
|
Beta
Type II, False negative |
-Concluding there is no statistically significant difference when there really is a difference
-Type II error -False negative |
|
1-beta
No error, true positive |
Concluding there is a statistically significant difference when there is a difference
|