01 Statistics

Independent: predictor

Dependent: outcome variable

1. Discrete

   • Nominal variables (unordered data: e.g. species: human, dog, bird…)

   • Ordinal variables (10-point scale —> ratings depend on subjective feelings)

2. Continuous

   • Interval (must be certain that equal intervals on the scale represent equal differences in the property being measured)

     • Helpfulness (5-point Likert scale)

     • Must be clear that difference between 1 and 2 is same difference as 3 and 4

   • Ratio variables (scale needs to have a true and meaningful zero point) e.g. reaction time

=> Discrepancy between numbers we use to represent the thing we´re measuring and the actual value of the thing we are measuring

• whether an instrument actually measures what it sets out to measure

  • Criterion validity = whether an instrument measures whether it claims to measure

  • Content validity = degree to which individual items represent the construct being measured

= whether an instrument can be interpreted consistenly across different situations

• Test-retest reliability: reliable instrument will produce similar scores at both points at time

=> Does x cause y

• manipulation of an independent variable

• control of confounding variables

• Experimental and control group

• randomization

=> Different groups of people take part in each experimental condition

=> between subjects, independent design

=> Manipulate the independent variable using the same participants

=> repeated measure design

1. Unsystematic variation => random factors that exist between the experimental condition (natural differences in ability, time of the day)

   => keep it to a minimum

2. Systematic variation => variation is due to the experimenter doing something to all the participants in the experimental group but not to the control group

   => careful: practice effects, boredom effects

   => counterbalancing the order in which a person participates in a condition

=> eliminates most sources of systematic variation (any systematic variation should be due to experimental manipulation)

skewness

• leptokurvic, left

• playtkurtic, right

• mode (score that occurs most frequently in the data)

• median (middle score when scores are ranked in order of magnitude)

• mean (artithmetic middle score)

• interquartile range (Q1 - Q3)

• Lower quartile

• Median = second quartile

• Upper quartile

H0 = null hypothesis

H1 = alternative / experimental hypothesis

—> WE CANNOT PROOF H1 but reject H0 (which is the opposite of H1)

—> supporting H1 by daa but NOT PROOFING (Falsification)