Statistical Interference

Question 1

What is a distribution?

Answer 1

• Describes the frequency (or probability) of
  occurrence for a given value
• Describes the shape of the data

Question 2

What are examples of probability distributions for continuous variables?

Answer 2

e.g. -Height, Age, - Normal, skewed

Question 3

What are examples of frequency distributions for discrete variables?

Answer 3

e.g. GP visits - Poisson, Binomial

Question 4

Why is knowing the distribution useful?

Answer 4

• We can use the distribution of a sample to make inferences about a wider population
• to generate confidence intervals ( assessing
  variability of estimates)
• test hypotheses
• calculate sample size

Question 5

What is the normal distribution?

Answer 5

A probability distribution that describes data that is symmetric around a mean

The normal has two parameters:
- mean
- standard deviation (SD)

Question 6

What is skewness?

Answer 6

a measure of the asymmetry of the distribution

Question 7

What would a negative skew look like?

Answer 7

Elongated tail at the left. More data in the left tail than would be expected in a normal distribution.

Question 8

What would a positive skew look like?

Answer 8

Elongated tail at the right. More data in the right tail than would be expected in a normal distribution.

Question 9

How do you distinguish between outcome and exposure?

Answer 9

By formulating a research question using PICO

• Determining whether the intervention
  influenced the size or occurrence of the
  outcome.

Question 10

Why are statistical tests used?

Answer 10

Ensure data on a sample can be represented on the overall population.

Statistical methods are needed when outcomes are unpredictable

Question 11

What is the null hypotheses?

Answer 11

= Outcome in not associated with exposure

Question 12

What is the hypotheses?

Answer 12

= Outcome is associated with exposure

Question 13

What is a Type I Error False - Positive Alpha (α)

Answer 13

Occurs if an investigator rejects a null hypothesis that is actually true in the population.

Question 14

What is the significance level?

Answer 14

• The probability that you will find an effect
  that does NOT actually exist
• Strength of evidence needed to reject NULL
  hypothesis
• Normally set to 5%

Question 15

What is Standard Deviation and is it a summary or inferential statistic?

Answer 15

SD = a measurer of how variable individual
measures are

• how spread out the values are of this
  variable

= a summary statistic

Question 16

What is Standard Error and is it a summary or inferential statistic?

Answer 16

SE = an estimate of how variable a statistic would be if we repeated out study numerous times
- how precise the estimate of true mean is
based on the sample mean

• a statistic is a value calculated from our sample

Question 17

Is Standard Error a summary or inferential statistic?

Answer 17

= a inferential statistic

Question 18

What is Standard Error usually used for?

Answer 18

Used to create a confidence interval

e.g. We can be 95% confident that out true population mean lies in our 95% confidence interval

Question 19

What is the equation for standard error?

Answer 19

SE = SD/ √n

Question 20

What is a P-value ?

Answer 20

P-value tells us the strength of the evidence against the null hypotheses (that there is no association)

It is the probability that we observed an effect size as large as we did if the null hypotheses is true e.g. effect size is zero

Question 21

What is a confidence interval?

Answer 21

A confidence interval gives us the range of values within which we are reasonably confident the true difference lies.

Question 22

What are both p values and confidence intervals based on?

Answer 22

They are both based on standard errors.

The smaller the error, the smaller the p-value and narrower the confidence interval.

Question 23

What happens to the null hypothesis as the p-value decreases?

Answer 23

As the p-value decreases the evidence against the null hypothesis increases.

Question 24

What does a p-value of 0.1 mean?

Answer 24

Weak evidence against the null hypothesis

Question 25

What does a p-value of 0.01 mean?

Answer 25

Increasing evidence against the null hypothesis with decreasing p value.

Question 26

What does a p-value of 0.001 mean?

Answer 26

Strong evidence against the null hypothesis

Question 27

What do the confidence intervals mean?

Answer 27

Confidence intervals show the range of values in which the true effect size is likely to lie.

Question 28

What does a 95% confidence interval mean?

Answer 28

Tells us that in 95% of replicate experiments the true value will lie in the interval.