Measurement, Effect Size & Meta-Analysis

Measurement

the process of assigning numbers to represent attributes of people or things

Levels of Measurement

nominal, ordinal, interval, ratio

Nominal

categories without order

Ordinal

categories with order

Interval

equal intervals without true zero

Ratio

equal intervals with true zero

Effect Size Letter

d

effect size

standardized number showing how strong a relationship or difference is allowing comparison across studies

effect size significance

shows whether the change is important in real life

random error

small, unpredictable distortions (not caused by bias)

systematic error

consistent mistakes cause by bias

meta-analysis

a set of statistical methods for combining results of previous studies

weighted average

method that gives more importance to a studies with larger sample size

-1 effect size

low or negative relationship

0 effect size

no relationship

0 < effect size

positive or strong relationship