Please enable JavaScript.
Coggle requires JavaScript to display documents.
PS2010 (Introduction (Golden rules (Each row is one participant, Each…
PS2010
Introduction
Golden rules
Each row is one participant
Each column is one variable
missing data must be defined as impossible values (e.g.-999)
Continuous variables = scale Categorical variables = nominal
Descriptve statistics
Central tendency
Mode: Most frequent score Median: middle score Mean (1 d.p.)
Dispersion
range : highest - lowest score Standard deviation (1 d.p.) Variance
Experimental Variance
-Due to experimental manipulation -
Variability between conditions
Random Variance
-Due to measurement error -Due to individual differences -Due to unmeasured variables -
Variability within conditions
Graphs and Tables
Include units and legends Label axes with = scales (min/max)
Parametric assumptions
Interval or ratio data Independence of observations Normal distribution
Homogeneity of variance
Levene's Test
t-tests
Error
Type 1 error: FInd effect when no
Incorrectly finding a significant effect
Always possible to type 1 error unless entire population tested
Accpetable risk (alpha level) is
p<0.050
Type 2 error: Find no effect when yes
Experimental designs
Independent designs(Unrelated designs)
Independent measures t-tests
t (df) = t statistic, p value (exact)
Interpretation
There was a significant difference (stats). "A" had significant score (M,S.D.) than "B" (M,S.D.)
Graph
Mean ratings (+- 1SD) for "A" and "B" seperately
Levene's test
want not-significant, p>0.050
Assumption has been met Equal variances assumed
Independent variable: manipulation by researcher Dependent variable: outcome measured
Repeated design (Related or paired design)
Repeated measures t-tests
No independence of observation
no individual differences (random variance reduced)
Sphericity
.
t (df) = t statistic, p value (exact)
Interpretation
There was a significant increase (stats) with condition. With "A" mean (S.D.) with "B" mean (S.D.)
Graph
Mean ratings (+- 1SD) for "A" and "B" seperately
one sample t-test
compare data from a single sample to a reference value (test value)
t (df) = t statistic, p value (exact)
