Data Analysis

Question 1

what is effect size

Answer 1

the size of the outcomes from a determinant

Question 2

what are the measures of average

Answer 2

mean - average measurement
median - midpoint of measurements
mode - the most common value

Question 3

what are the measures of spread

Answer 3

range - extremes of all data
standard deviation - average spread of values around the mean
IQR - spread of values around the median

Question 4

what are the directions of skew

Answer 4

positive skew is to the right

negative skew is to the left

Question 5

what does normally distributed data measure

Answer 5

report mean and standard deviation

Question 6

what does skewed data measure (discrete)

Answer 6

report median and inter quartile range

Question 7

what is the relationship between mean and median in normally distributed data

Answer 7

roughly the same

Question 8

what values are dependant on distribution

Answer 8

mode and range

Question 9

what are the two ways of displaying data

Answer 9

skewed data - box plot - median, IQR and range

two continuous variables - scatterplot

Question 10

how do you measure associations between categorical variables

Answer 10

use risk, risk ratio, odds ratio

Question 11

what test would you use for a continuous outcome, or categorical exposure

Answer 11

t test or non-parametric equivalent

Question 12

what measure of association would you use between continuous variables

Answer 12

correlation or lier regression

Question 13

what is the difference between correlation and linear regression

Answer 13

c - association between two variables

lr - effect of one on the other

Question 14

what is the definition of correlation and the two types

Answer 14

measure of linear association between two continuous variables (r)
persons - both variables normally distributed
spearman’s - either or both variables skewed

-1 = perfect negative linear relationship 
0 = no linear relationship 
\+1 = perfect  positive linear relationship

Question 15

what are the pros and cons of correlation

Answer 15

simple method of association
order doesn’t matter
calculated between two variables only
assessment of straight line association only
cannot describe an exposure/outcome relationship or make predications

Question 16

what is anscombes quartet

Answer 16

4 completely different relationships that give all the same r value - con against correlation tests as can’t describe the relationship

Question 17

describe the uses of linear regression

Answer 17

models the relationship between two or more variables

can describe an exposure/outcome relationship

Question 18

what does the r2 value of linear regression mean

Answer 18

the variability in the health outcome that is explained by the given causative variable - described between 0 and 1 = higher the better

Question 19

what are the axis of linear regression

Answer 19

outcome on y

exposure on x

Question 20

in data interpretation what is the difference between correlation and regression

Answer 20

c - tells us whether both variables do or don’t change together between -1 and 1

r - quantifies how they change together - a gradient

Question 21

what would 0.32 mean in regression

Answer 21

means that for every 1% increase in exposure there is a 0.32% increase in outcome

Question 22

how do you interpret the r2 value

Answer 22

the proportion of the outcome variance that the model explains
the larger the better between 0 and 1

Question 23

what are point estimates in data interpretation

Answer 23

estimated values for particular points eg one patient comes in and diagnosed with A - other similar patient is diagnosed with B

Question 24

what is the difference between confidence level and interval width

Answer 24

c l - how often is this true
i w - the boundaries in which the truth lies at this level

confidence level is proportional to confidence width ie with an increased interval you can be more certain that the trust value lies within it

Question 25

what is the p value definition

Answer 25

the probability of a coefficient at least as big as yours, assuming the coefficient is actually zero

Question 26

describe a small vs large p value

Answer 26

small = zero-assumption is probably wrong - an effect is likely

large - zero assumption is probably right - an effect is unlikely

Question 27

what would a p value below 0.05 mean

Answer 27

statistically significant