Introduction

Question 1

Definition of falsifiable

Answer 1

Can be proved to be false

Question 2

Parsimonius

Answer 2

simple model to explain

Question 3

Definition of theory

Answer 3

Set of principles that explain a topic on which a new hypothesis can be made

Question 4

Descriptive statistics

Answer 4

Summarise a collection without inferences made

Question 5

Inferential stats

Answer 5

Draws inferences about a population from estimation or hypothesis testing

Question 6

Quantitative

Answer 6

Measured on interval/ratio scale or ordinal data

Question 7

Qualitative

Answer 7

Assign objects into labelled groups without natural ordering

Question 8

Interval variables

Answer 8

Equal intervals e.g. age

Question 9

Ratio variables

Answer 9

Equal intervals with a clear 0

Question 10

Binary

Answer 10

2 categories

Question 11

Nominal

Answer 11

More than 2 categories

Question 12

Average used for nominal data

Answer 12

Mode

Question 13

Ordinal data

Answer 13

More than 2 categories with an order (e.g. 1st, 2:1, 2:2

Question 14

Average used for ordinal data

Answer 14

Median

Question 15

why does a variable type matter?

Answer 15

Alters tests that can be used

Question 16

Measurement error

Answer 16

Discrepancy between actual number and one recorded

Question 17

Systematic variance

Answer 17

Due to dependent variable

Question 18

Random error

Answer 18

Random variance

Question 19

What is validity

Answer 19

Measure of how well it measures what it’s supposed to measure

Question 20

Problem with hypothesis testing

Answer 20

encourages all or nothing thinking, just because null hypothesis is rejected doesn’t mean it’s true

Question 21

One-tailed when to reject H0

Answer 21

Reject null hypothesis if in extreme 5%

Question 22

Two tailed when to reject H0

Answer 22

Reject is in either 2.5%

Question 23

Type 1 error

Answer 23

Rejection of true null hypothesis, incorrectly preduct that variance is accounted for by the model, accepted p

Question 24

Type 2 error

Answer 24

Fails to reject null hypothesis, incorrectly predict that too much variance is unaccounted for by the model, acceptable p=.2 at beta level

Question 25

Benefit of one tailed (error)

Answer 25

Lower chance of type 2 error and more power BUT only in one direction have to be sure of result

Question 26

Meaning of effect size

Answer 26

degree to whihc the mean of H1 differs from mean od H0 in terms of SD OR how much variance is explained

Question 27

Ways to reduce type 1 error

Answer 27

Look at effect size (standardises results and not reliant on sample size)

Question 28

Calculate cohen's d

Answer 28

Mean1 - Mean2 / pooledSD

Question 29

Calculate pearson's r

Answer 29

Cov(xy) / SxSy

Question 30

Pooled SD

Answer 30

Squarert ((SD12 + SD22)/2)

Question 31

What does more power mean in terms of error

Answer 31

Reduces type 2 error, better chance of correctly rejecting null hypothesis with bigger sample

Question 32

Ideal power

Answer 32

.8

Question 33

Big r

Answer 33

.5

Question 34

Big d

Answer 34

.8

Question 35

Linearity

Answer 35

Variables are linearly related

Question 36

Addivity

Answer 36

Several predictors combined effects best described with addition

Question 37

Normality

Answer 37

Sampling distribution should be normal check with Kolmogorous-Smirnov (non-sign = normal) or Kurtosis (0=normal)

Question 38

Homogenity of variance

Answer 38

Samples should have similar variances and outcome stable across predictor check (also referred to as heteroscedasticity) Levene's test (non sig = variance assumed) or variance ratio (2 = assumed)

Question 39

Independence

Answer 39

Errors should not be related

Question 40

Checking for heteroscedasticity

Answer 40

Check scatterplots, want random pattern of standardised values

Question 41

How to reduce bias

Answer 41

trimming data (or standardised), windowinising (replace outliers with highest value), bootstrapping or transforming (e.g. logs)