PASS Statistical Concepts Flashcards
Population meaning in stats
All possible observations of an experiment/study variable
Sample meaning in stats
A selection of observation taken from the population - a small group of the population
Types of errors
Random error due to chance
Systematic error due to bias
What is random error due to?
Due to chance + sampling variation
How to reduce random error
Increasing sample size reduces random error
What is bias quantified by?
The difference between the true value and the expected value
Relationship between random error and sample size
As sample size increases, random error decreases
Relationship between systematic error and sample size
Changing sample size has no effect no systemic error - does not reduce
What is a hypothesis?
A statement that an underlying truth of scientific interest takes a particular quantitative value
What is a p-value?
The probability of getting an observation as extreme as or more extreme than the one observed in the hypothesis is true
When do you reject the null hypothesis at the 5% level?
When p-value < 0.05
What do you do with the null hypothesis at the 5% level when the p-value < 0.05?
Reject null hypothesis
Limitation of hypothesis testing
Rejecting a hypothesis is not always useful:
-statistically significant ≠ clinically important
- statistical significance depends on sample size
What is a 95% confidence interval?
The range within which we can be 95% certain that the true value of the underlying truth really lies
Relationship between confidence interval andsample size increases
As sample size increases, random variation + uncertainty decreases
Null value of ratio
1
Null value for a difference
0
What is a null value?
An estimate that shows no difference between two groups
Relationship between confidence interval and p-value
Will always agree on statically significance
Null hypothesis inside 95% CI - p >0.05
Null hypothesis outside 95% CI - p<0.05
When is the 95% confidence interval wider?
Greater the variation in the population values
Smaller the size of thesample used to calculate it
What does PICO stand for?
Population
Intervation
Comparison
Outcome
What is the population to be studied?
What is the intervention of interested?
What is the comparision/control?
What is the outcome of interest?
Descriptive epidemiology study designs
Ecological study
Cross sectional surgery
Unit of analysis in ecological study
Groups
Unit of analysis of cross sectional survey
Individuals
Analytical epidemiology study designs
Case-control study
Cohort study
What ratio is related to case-control study?
Odds ratio
What ratio is related to cohort study?
Rate or odds ratio
Issues for ecological studies
Definition of characteristics
Measurement variation
Confounding
Random error
