Population Science - Rates, Ratios And Risk

Question 1

What is probability?

Answer 1

Given information in a big population, what can you infer about a smaller group

Question 2

What is statistics?

Answer 2

Given information from a small sample, what can you infer about the wider population?

Question 3

What 3 things must a sample population be?

Answer 3

1. Representative
2. Unbiased
3. Precise

Question 4

What are the 2 types of error that can occur?

Answer 4

1. Chance (random error)
   Occurs due to sampling variation, will reduce with increased sample size
2. Bias (systematic error)
   The difference between the true value and the expected value
   Will NOT reduce as sample size increases

Question 5

Name 3 sources of selection bias

Answer 5

1. Study sample (external validity)
   Sample not representative of entire population of interest
2. Group selection within the study (internal validity)
   Groups may not be comparable e.g comparing young smokers with old smokers
3. Healthy worker effect
   Workers usually exhibit lower overall mortality than the general population

Question 6

Identify 4 possible sources of information bias

Answer 6

1. Recall error
2. Observer / interviewer error (preconceived expectations may influence result)
3. Measurement error e.g using different techniques
4. Missclassification e.g. classing participants into the wrong group

Question 7

What’s the difference between precision and bias?

Answer 7

Bias is being far away from the target/true value

Precision is all results in a similar range

High precision does not = less bias

Question 8

What is prevalence?

Answer 8

The proportion of people who have disease at any given point in time (not a rate)

Prevalence = number of people with disease / total population

Question 9

What is incidence?

Answer 9

The number of NEW cases within a given time frame

Incidence= number of new cases/ total patient time at risk = events per person, per year