Statistics

Question 1

when would you use a Chi squared test?

Answer 1

• number of individuals in two or more categories have been counted
• you want to know if there is a significant difference between observed and expected results

Question 2

when would you use Spearman’s Rank Correlation?

Answer 2

• measurements have been taken
• you want to know if there is an association/correlation between different measurements from the SAME sample

Question 3

when would you use a t-test?

Answer 3

• measurements have been taken
• you want to compare the means of two sets of data

Question 4

paired vs unpaired t-test

Answer 4

paired: two measurements are collected from the SAME individual

unpaired: two measurements are collected from DIFFERENT individuals

Question 5

six assumptions made to use an unpaired t-test

Answer 5

• groups are independent
• same variable has been measured for an unbiased sample
• variable is continuous
• continuous variable is normally distributed for each group
• each group has approximately equal variances (similar SD)
• sample sizes for each group are roughly equal

Question 6

five assumptions made to use a paired t-test

Answer 6

• two groups of with a specific relationship with each other (e.g. same individuals used in each sample)
• same variable has been measured for both unbiased samples
• variable is continuous
• continuous variable is normally distributed
• each group has similar variance/SD

Question 7

two assumptions made to use Spearman’s Rank Correlation

Answer 7

• there is a set of items with data for BOTH variables for each item
• both variables are ordinal (can be ranked)

Question 8

assumption made to use the Chi squared test

Answer 8

• sample size must be big enough so that all “expected values” are greater than 5

Question 9

what probability should be used to find the critical value (if not stated)?

Answer 9

p=0.05 (95%)

Question 10

degrees of freedom (one sample with t test or spearman’s rank)

Answer 10

DF = N - 1
(N = sample number)

Question 11

degrees of freedom (two samples with t test or spearman’s rank)

Answer 11

DF = N1 + N2 - 2

Question 12

what does the comparison of CV and t value mean? (with conclusion)

Answer 12

if t > CV, reject H0

if t < CV, accept H0

reject H0, we are 95% confident that something has caused the difference and it is not just down to chance

Question 13

what does the comparison of CV and Χ² value mean? (with conclusion)

Answer 13

if X² > CV, then reject H0, there is a significant difference between observed and expected frequencies

if X² < CV, then accept H0, there is no significant difference between observed and expected frequencies

reject H0, there is less than a 5% probability that our results are due to chance

Question 14

how do you find the CV for Spearman’s Rank Correlation?

Answer 14

use n (number of sample) to find CV in a table

(DO NOT USE DF)

Question 15

what working should be shown on the data table for Spearman’s Rank Correlation?

Answer 15

• rank for BOTH variables
• difference between the ranks (and d²)

Question 16

degrees of freedom (Chi-squared)

Answer 16

DF = N - 1

where N = no. of categories (rather than no. of samples)