Lesson 10- factors affecting choice of statistical tests including levels of measurements and design

Question 1

levels of measurements

Answer 1

used to try to categorise our data into one of four types so we can select the most appropriate statistical test

Question 2

nominal data

Answer 2

data consists of the numbers of Ps that fall into diff categories. you can only be in one category

Question 3

ordinal data

Answer 3

data can be placed in rank order from the lowest to highest

Question 4

interval data

Answer 4

data is fixed in even intervals eg height and weight

Question 5

non parametric tests

Answer 5

chi squared
spearman’s rho
mann whitney
wilcoxon

Question 6

parametric tests

Answer 6

pearsons r
related t test
unrelated t test

Question 7

parametric vs non parametric test

Answer 7

parametric test are more robust and powerful , they rely on actual data rather than examining the rank order of the data

Question 8

3 factors deciding if a parametric test can be conducted

Answer 8

1. interval level of measurement - can’t be ordinal
2. normal distribution
   3.variance of data

Question 9

what test should you use if its a test of correlation

Answer 9

spearman’s rho or Pearsons r

Question 10

what test should u use if its a test of difference

Answer 10

mann whitney , chi squared, wilcoxon , unrelated or related t test

Question 11

what test should u use if its a test of association

Answer 11

chi squared

Question 12

independant measures

Answer 12

diff Ps do diff conditions

Question 13

repeated measures

Answer 13

all Ps undergo both conditions

Question 14

matched Ps design

Answer 14

Ps matched based on characteristics to do the same test

Question 15

table of statistical tests

Answer 15

look on WS