4 scales of measurement stats

Nominal Scale

Categories without numeric meaning (gender, eye color)

Ordinal Scale

Ordered categories without equal intervals (educational level, pain scale)

Interval Scale

Numeric scale with equal intervals but no true zero point (Time of day, Temperature)

Ratio Scale

Numeric scale with equal intervals and a true zero point (weight, distance, age)

Independent Variable

What you change (the cause, ex. Hours of sleep)

Dependent Variable

What you measure (the effect, how much you lift)

Continuous Variables

Can take any value within a range, measured (decimals, fractions)

Discrete variables

Can only be specific, separate values, counted (number of people, number of reps)

Test-retest reliability

Same test, different times. Checks consistency over time

Inter-rater reliability

Different people scoring. Checks if they agree with each other

Parallel forms reliability

Two different versions of the same test. Checks if both versions give similar results

Internal Consistency reliability

Questions within the same test. Checks if they all measure the same thing

Internal validity

Correctly identifying causal relationships. Did the IV actually cause the DV? (cause & effect accuracy)

External validity

Generalizability of findings. Can results apply to other people/settings? (generalize it)

Construct validity

Measuring intended theoretical constructs. Are you truly measuring what you say you are?

Face validity

Measure appears effective superficially. Does it look like it measures what it should? (on the surface)

Ecological validity

Real-world relevance. Does it reflect real-life situations?