23. choosing a statistical test

what are statistical tests

also called inferential tests

• used in psychology to determine whether a significant difference or correlation exists - an thus whether the null hypothesis should be rejected or retained

factors affecting the choice of a statistical test

1. test of difference or association

2. experimental design

3. level of measurement

test of difference or association

should be clear from the wording - whether an experiment or correlation

types of experimental design

1. related = repeated measures, matched pairs

2. unrelated = independent groups

what are the levels of measurement

• nominal data

• ordinal data

• interval data

nominal data

• represented in the form of categories

• discrete as one item can only appear in one of the categories

• also known as categorical data

ordinal data

• data is ordered in some way eg a scale

limitations of ordinal data

• does not have equal intervals between each unit (unlike interval)

• lacks precision as is based on subjective opinion rather than objective measures

for these reasons its sometimes referred to as unsafe data

interval data

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

• more preserved than ordinal data

• eg public scales of measurement that produce data based on accepted units of measurement - stopwatch, thermometer etc

table of tests

mneumonic for the table across in rows

cows sing country

music while sipping

unusually rare punch