Brief critical numbers Flashcards
Sampling bias
Sampling bias – sample does not represent the population of interest (also called selection bias?)
(e.g. sampling only from one hospital is not a representative of all hospitals)
Recall bias
Recall bias – inaccurate recall of past events/exposures/behaviours
Information bias
Information bias – incorrect measurement e.g. miscalibrated machine
The hawthorne effect
The ‘Hawthorne’ effect – participants change their behaviour when they know they are being observed
Attrition bias
Attrition bias – differential dropout of participants from the study e.g. sicker participants drop out so our outcome is only measured on healthier participants
Confounding variable
A variable that influences both the dependent and independent variable, leading them to have a spurious association.
Randomised control trials
Definition, pros and cons
Research studies in which participants are randomly assigned to different interventions or control groups.
Pros
- Gold standard
- Randomisation reduces potential for confounding
- Can reduce bias via control and blinding
- Can determine causal effects
Cons
- Could potentially be unethical (e.g. withholding treatment from control groups)
- Expensive
- It requires expert management
Cohort study
Definition, pros and cons
An observational study that follows a group of individuals over a period of time to investigate the relationship between exposure to certain factors and the development of specific outcomes
Pros
- Useful when random allocations are not possible
- Can work for rare exposures (as you can select participants)
- Can examine multiple outcomes
Cons
- May require long to follow up
- Can be expensive
- Not ideal for rare outcomes
Case control studies
Definition, pros and cons
Retrospective observational studies that compare a group of individuals with a specific outcome to a group without the outcome. (identifies association between risk factors and occurrence of outcome)
Pros
Faster (using past data)
Useful for rare outcomes
Cheaper
Cons
- Prone to bias
- Hard to show causal relationship
- Not ideal for rare exposures
Cross sectional study
Definition, pros and cons
Observational studies that collect data from participants at a single point in time.
Pros
- Relatively quick
- Cheap
- Can access multiple exposures
Cons
- Susceptible to bias
- Cannot prove causal relationship
- Not ideal for rare exposures
Ecological studies
Definition, pros and cons
Observational studies that analyse group level data to explore the relationship between exposures and outcomes, focusing on populations (instead of individuals)
Pros
- Large scale comparison
- Can quantify geographical trends
Cons
- Cannot make inference at individual level
Describe the categorical and numerical variables
Categorical
Binary - 2 categories (yes, no- positive, negative)
Ordinal - categories with an order (stage of cancer, moderate - severe pain)
Nominal - categories with no order (blood group, ethnicity)
Numerical
Discrete- Observations can take numerical values (number of children, number of GP visits)
Continuous- Observations can take any value within a range (age, height, temperature)
Difference between proportion and odds
Proportion- Number of people with disease / total number of people
Odds- Number of people with disease / number of people without the disease (probablity / 1 - probability)
(Therefore odds ratio is odds in one group divided by odds in another)
How to calculate standard deviation?
1) Calculate the mean
2) Square the (differences of each value from the mean)
3) Add all the squares
4) Divide by the sample size minus 1
5) Square root it
What does a right skewed distribution mean?
What does a left skewed distribution mean?
Median is lower than the mean
Median is higher than the mean
