Stats Flashcards Preview

PS2002 > Stats > Flashcards

Flashcards in Stats Deck (24):
1

Give 2 conditions for using factorial designs.

More than 1 IV is thought to effect a DV and/or ignoring an IV detracts from the explanatory power of our experiments.

2

Give 2 limits of between-subjects designs.

Participant variables and number of participants required.

3

Give 2 limits of within-subjects designs.

Practice effects and long testing sessions.

4

What are the assumptions in mixed factorial ANOVA?

Interval/ratio data, normal distribution, homogeneity of variance (between-subjects) and sphericity of covariance (within-subjects).

5

What should you do if the assumptions of mixed factorial ANOVA are violated?

Process with caution, report the violation, and use corrected results if possible.

6

If there are interactions in a mixed factorial ANOVA, which formulae do you use?

Within-subjects formulae.

7

What are the assumptions for Pearson’s correlation?

Linear variable relationship, interval/ratio data and normal distraction free of outliers.

8

What are the assumptions for Spearman’s correlation?

Monotomic variable relationship and ordinal/interval/ratio data.

9

Why do distribution and outliers not influence Spearman’s correlation?

They use ranks, not means and SDs.

10

How are degrees of freedom calculated in a Pearson’s correlation?

df - n*pairs - 2

11

Give 2 factors that affect the significance threshold of a correlation coefficient.

The magnitude of the correlation and the number of observations.

12

When are tests of regression used?

When causal relationships between variables are likely.

13

What type of tests are regression?

Inferential statistical tests of association.

14

What are the assumptions of regression?

Linearity, interval/ratio data, normal distribution free of outliers and homoscedasticity.

15

Describe residuals.

The distance between the actual outcome and predicted outcome.

16

What do you do if there is heteroscedasticity in a regression?

Proceed with caution and inform the reader.

17

What is multicollinearity and how do you test for it?

When predicted variables are highly correlated, tested by running simple correlations.

18

How much should variables correlate to be considered to have multicollinearity?

(+/-) 0.8

19

What do mixed factorial designs use very efficiently?

Participant numbers and participant time.

20

What is homoscedasticity?

When residuals have the same degree of variation across all predictor variable scores.

21

What is multiple regression?

Predicting one outcome variable from more than one predictor variable.

22

Describe simultaneous regression.

All predictors are entered at the same time - used for explanatory analysis.

23

Describe hierarchical regression.

Predictors are entered in a pre-defined order - used when informed by theory.

24

Describe stepwise regression.

Predictors are entered in an order driven by how well they correlate with the outcome.