Statistics

Question 1

Define sampling

Answer 1

The selection of a subset of individuals from within a statistical population

Question 2

What is sampling bias?

Answer 2

Where some individuals are more likely than others to be included in the study

Question 3

What is recall bias?

Answer 3

When individuals cannot remember specifics

Question 4

What is social-desirability bias?

Answer 4

When individuals tell us incorrect information because they feel a societal pressure

Question 5

What is a confounding factor?

Answer 5

Something that is related to the outcome and the characteristics of interest

Question 6

What is a case control study?

Answer 6

1. Take a sample of individuals with outcome, and similar group without
2. Look back retrospectively to see who had exposure

Question 7

What are the advantages and disadvantages of a case control study?

Answer 7

+ Good for investigating rare outcomes
+ Relatively cheap and quick

• Subject to recall bias
• Can only investigate a single disease

Question 8

What is a cross sectional study?

Answer 8

Look at what is happening now (snapshot of time)

Question 9

What are the advantages and disadvantages of a cross sectional study?

Answer 9

+ Very cheap, quick and easy

• No time scale
• Not suitable for rare diseases

Question 10

What is a cohort study?

Answer 10

1. Collect information on a sample without the outcome

2. Follow up over time, looking at exposure, to see who gets the outcome

Question 11

What are the advantages and disadvantages of a cohort study?

Answer 11

+ Can look at a variety of outcomes

• Time consuming and expensive
• Not great for rare outcomes or outcomes that take a lot of time to develop

Question 12

What is a randomised control trial?

Answer 12

1. Have multiple (at least two groups ) referred to as arms
2. Give different exposures to each arm
3. Compare outcomes

Question 13

.What are the advantages and disadvantages of RCTs?

Answer 13

+ Minimises bias and confounding factors and has statistical reliability
+ Comparative study design

• Not always suitable, there can be ethical issues
• Expensive

Question 14

What is the equation for proportion / probability?

Answer 14

Event / Total

Question 15

What is the equation for odds?

Answer 15

Event / Non event

Question 16

What is the equation for absolute risk difference?

Answer 16

Probability - Probability

Question 17

What is the equation for risk ratio?

Answer 17

Probability / Probability

Focus group goes on top

Question 18

What is the equation for odds ratio?

Answer 18

Odds / Odds

Focus group goes on top

Question 19

What is negative skew?

Answer 19

Where the median is greater than the mean

Question 20

What is positive skew?

Answer 20

When the median is less than the mean

Question 21

What data would be best to publish for a normal distribution?

Answer 21

Mean and standard deviation

Question 22

What data would be best to publish for a non symmetric, skewed distribution?

Answer 22

Median and IQR

Question 23

In normally distributed data, what percent of data lies within 1 SD of the mean?

Answer 23

Approx 68%

Question 24

In normally distributed data, what percent of data lies within 1.96 SD of the mean?

Answer 24

95%

Question 25

What would a Pearson’s correlation of 1, 0 and -1 mean?

Answer 25

1 = Perfect positive linear association
0 = No linear relation
-1 = Perfect negative linear association

Question 26

What is the equation for standard error?

Answer 26

SE = Standard Deviation / √n

Question 27

What is the equation for 95% confidence interval?

Answer 27

95% CI = mean ± 1.96*SE

Question 28

What is the 95% confidence interval

Answer 28

The range of numbers you can be 95% confident that contain the true population mean

Question 29

What is a p value?

Answer 29

The probability of obtaining your results if the null hypothesis is true

Question 30

For the p value to be significant, what two things have to be true?

Answer 30

1. It has to be less than 0.05

2. The null value cannot be in the 95% confidence interval

Question 31

What is regression?

Answer 31

A mathematical process to explore the association of multiple factors on an outcome

Question 32

What is regression?

Answer 32

A mathematical process to explore the association of multiple factors on an outcome