4) STATISTICAL TESTS

choosing a statistical test (3)

1) test of difference or correlation/association

2) is the experimental design related (RMD/MPD) or unrelated (IMD)

3) level of measurement (nominal, ordinal, interval)

used to determine whether a difference or correlation is statistically significant so if the results could have occurred by chance or real effect

nominal data

• categorical data - discrete (can only belong to one category)

• simplest form

• i.e no. of times something occurs

ordinal

• measurements that can be put in an order or rank (numerical ordered scale) but intervals are variable, unknown

• lacks precision as it’s based on subjective measures

• i.e asking someone how much they like dogs on a scale of 1 to 10

interval

• based on numerical scales that include units of equal, precisely defined size

• most precise

• public units of measurement

choosing a statistical test (table)