Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Experimental Economics > Statistics > Flashcards

Statistics Flashcards

1
Q

What assumptions are required for parametric tests to be valid?

A

Observations are independent.
Observations are drawn from a normally distributed population.
Different groups must have equal variance.
Data must be at an interval scale at least.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the two most common issues with experimental data that prevent the use of parametric tests?

A

Not knowing the underlying distribution of the data.

Data not being in the interval scale.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What assumptions are required to carry out non-parametric tests?

A

Observations are independent.

Sometimes: data are drawn from a continuous underlying distribution.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Main benefits of non-parametric tests?

A

Smaller sample size required.
Can use more forms of data.
Less impact from outliers.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the main downside of non-parametric tests?

A

Tests have less power

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What does it mean when a test has less power?

A

Harder to reject the null hypothesis when it is false

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What are the four scales of data?

A

Nominal
Ordinal
Interval
Ratio

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Characterise nominal data

A

Data is categorised, no order

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Characterise ordinal data

A

Data is categorised and ordered

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Characterise interval data

A

Numerical differences between numbers has some meaning

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Characterise ratio data

A

Scale has a natural zero point at origin

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are the two one-sample parametric tests?

A

Binomial and chi-squared

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

When can you use the binomial test?

A

When population consists of only two classes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What scale of data is required at least for the binomial test?

A

Nominal scale

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

When to use the chi-squared test?

A

When population consists of at least two classes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is the minimum scale of data required for the chi-squared test?

A

Nominal scale

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What are the three two-sample non-parametric tests?

A

Fisher exact test
Mann-Whitney U test
Wilcoxon test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

When to use the Fisher exact test?

A

Two independent samples that consist of two classes

19
Q

What is the minimum scale of data required for the Fisher exact test?

A

Data can be nominal or ordinal

20
Q

How to calculate the Fisher exact test statistic?

A

Calculate the probability of a more extreme outcome than the most extreme observed, keeping marginal totals fixed.

21
Q

What experimental design is the Mann-Whitney U test used for?

A

Between-subject design

22
Q

What element of samples data does the Mann-Whitney U test examine?

A

The distribution and differences in the distribution

23
Q

What scale of data is required at the minimum for the Mann-Whitney U test?

A

Ordinal scale

24
Q

What is the general intuition of the Mann-Whitney U method?

A

Pool all data and rank each observation. If distributed similarly would expect to see similar sum of ranks for each sample.

25
Q

What experimental design is the Wilcoxon test used for?

A

Within-subject design

26
Q

When you can carry out the Mann-Whitney U test or the Wilcoxon test, which should you prefer?

A

The Wilcoxon test is stronger

27
Q

What is the minimum scale of data required for the Wilcoxon test?

A

Ordinal scale

28
Q

Describe the method of the Wilcoxon test

A

Look at differences in paired observations.
Pool all differences and rank them.
Add a minus sign to any negative observations.
Compare positive and negative sum of ranks.
If treatment had no effect, sums of ranks should be similar.

29
Q

Name the two k-sample non-parametric tests

A

Kruskaw-Wallis test

Jonckheere test

30
Q

What are the differences in expected group order that separate between the Kruskaw-Wallis and Jonckheere tests?

A

Group order is random in Kruskaw-Wallis test, while groups are ordered a priori in Jonckheere test.

31
Q

What is the minimum scale of data required for the Kruskaw-Wallis test?

A

Ordinal scale

32
Q

What element of the data does the Kruskal-Wallis test consider?

A

The median of each group

33
Q

Describe the general method of the Kruskaw-Wallis test

A

All observations are converted in to a single series.
Observations are ranked.
Sum of ranks should be similar for each group if medians are the same.

34
Q

What is the minimum scale of data required for the Jonckheere test?

A

Ordinal scale

35
Q

What do you expect of the groups to use the Jonckheere test?

A

Groups have an expected rank order a priori

36
Q

Describe the general method of the Jonckheere test

A

Count the number of times an observation in group i is preceded by an observation in group j

37
Q

In what case are the MWU, W, KW and J tests poor choices?

A

When there are lots of ties when ranking data

38
Q

What is type 1 error

A

Falsely rejecting the null hypothesis when it’s true

39
Q

What is type 2 error

A

Failing to reject the null hypothesis when it is false

40
Q

What is the power of the test?

A

1 - the type 2 error rate

41
Q

How to improve the power of the test?

A

Increase sample size
Reduce error variance
Increase treatment level variance

42
Q

How to reduce the error variance

A

Randomisation and blocking

43
Q

How to increase treatment level variance

A

Reduce number of treatments and increase spread

44
Q

When can you not use only two treatments in an experiment?

A

When a non-linear relationship is expected

Experimental Economics (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions