Sample Variation, Bias, Confounding

Question 1

What does inference mean?

Answer 1

draw conc about pop from sample, never be perfect

Question 2

Define probability

Answer 2

the extent to which an event is likely to occur

measured by the ratio of the favourable cases to the whole number of cases possible

Question 3

What is the central limit theorem?

Answer 3

distribution of means of the samples have similar shape regardless of population

symmetrical bell shape

Question 4

What is a confidence interval?

Answer 4

range that is likely to contain mean pop values 95% of time: 1/20 miss.

Broader pop = wider confidence interval.

Width depends on: variation in pop and size of sample

Question 5

What does precision mean?

Answer 5

accuracy

Question 6

What is selection bias?

Answer 6

Study participants are drawn from a sample frame that is not representative of the general pop.

The groups being compared are not from same pop.

One of the groups being studies is not representative of sampling frame from which it was drawn

Question 7

What is information bias?

Answer 7

Differential recall error = remembering the wrong thing.

Differential observer/interviewer error = wanting it to be a certain way, prior knowledge.

Differential measurement error = measuring differently.

Differential mis-classification = classify things in diff ways, non-consistency

Question 8

What is confounding?

Answer 8

3rd variable in causal net is associated with both an exposure and outcome being studied.

This 3rd variable is not taken into account in a study = conc about relationship between 2 variables may be mis-interpreted.