Brainscape's Knowledge GenomeTM

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


See full index

MST > Theory > Flashcards

Theory Flashcards

1
Q

What is a statistical generalisation technique?

A

It is an instrument that can be used when making statements about population characteristics on basis of data from a sample.

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

Factorial design

A

A design with more than one independent variable

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

What 4 assumptions do you need to satisfy when you want to use an ANOVA test?

A

(1) Data at interval level of measurement, (2) all observations are independent of each other, (3) scores are normally distributed within each group and (4) the variance has to be approx. equal between groups

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

What is analysis of variance?

A

A tool you can use to compare variances in two or more different groups while including multiple independent variables.

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

What does analysis of variance measure?

A

Whether the values of groups with different conditions differ more from each other than you would expect from chance fluctuations.

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

What is the F-ratio?

A

The ratio between the two components of variance in ANOVA. F = variance between groups / variance within groups.

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

What does the value of F mean?

A

Reject H0 if F > 1 and P(F|H0) = Alpha. H0 is true if F = 1.

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

What is the total variance?

A

The mean squared (positive) deviation between each measurement and the group mean.

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

What is the formula for the total variance over all N obersvations from all groups and conditions?

A

st^2 = ( (sum of j=1, j=k) nj (mean of xj - mean of x)^2 ) / (N-1)
With j=condition, x=measurement, st=standard deviation.

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

What three variances are interesting for ANOVA?

A

(1) The total variance of all observations from all groups and conditions (=t), (2) the variance between the groups and conditions (=b) and (3) the variance within the groups or conditions (=w).

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

How do you calculate the degrees of freedom for t, b and w?

A

Total variance: (N-1)
Variance between groups: (k-1)
Variance within groups: (N-k)

With N= total number of observations, k= number of groups.

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

Do the degrees of freedom add up to each other?

A

Yes!
df(t) = df(b) + df(w)

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

Is MSt equal to the normal variance sx^2?

A

Yes
MS= mean squares

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

How do we define F?

A

F = (sb^2) / (sw^2)
The ratio of the two variance components, using two degrees of freedom (k-1) and (/) (N-k)

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

How can you summarize ANOVA results in one sentence?

A

The mean scores are (not) equal for group a, b and c [ F (a, b) = c, p=.d, omega^2 =

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

What does the effect size tell us?

A

It tells us how much of the variance in the sample can be assigned to the independent variable.

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

What is eata squared?

A

An effect size measure that tells us how much of the variance in the sample can be assigned to the independent variable

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

What is omega squared?

A

An effect size measure that estimates the proportion of the variance in the population which can be attributed to the independent variable, based on the sample.

19
Q

What are planned comparisons?

A

When you want to look at a specific difference and ignore all others.

20
Q

How do you check if a contrast is orthogonal?

A

Assign a weight to each condition for that contrast. If the sum of the product of the population mean and the weight is zero, the contrast is orthogonal.

21
Q

What is a post hoc comparison?

A

When you decide to look more closely which conditions are different only after you found a significant effect.

22
Q

What is conservatism in statistics?

A

A tendency not to reject H0, to keep H0.

23
Q

What is liberalism in statistics?

A

A tendency to reject H0.

24
Q

What is Tukey’s Honestly Significant Difference?

A

Tukey’s HSD, a test for post hoc comparisons between pairs of conditions that has a 95% confidence interval.

25
Q

What do you do if you want to know the descriptive statistics?

A

Frequencies or mean and standard deviation + APA table

26
Q

What do you do if you want to know the relation between two variables?

A

Use a Pearson (interval) or Spearman (ordinal) correlation.

27
Q

What do you do if you want to know the reliability of raters?

A

Use Cronbach’s Alpha. It should be positive. OK if higher/equal to 0.7

28
Q

What do you do when you want to investigate an effect / wheter X has effect on Y?

A

Compare means. Use t-test or ANOVA.

29
Q

What is the difference between parametric and non-parametric tests?

A

The non-parametric test requires a larger effect and/or a larger sample, and generally has less power than a parametric test when seeking out an effect.

30
Q

Name two parametric tests:

A

T-test and analysis of variance

31
Q

Name a non-parametric test:

A

the chi-squared-test

32
Q

What is the X^2 test based on?

A

On differences between the expected and observed frequencies. Data on nominal level of measurement.

33
Q

What does the X^2 test indicate?

A

how high the probability is of finding this (uneven) distribution of outcomes if H0 is true

34
Q

What is the X^2 test also referred to?

A

as a test of ‘goodness of fit’

35
Q

How do you use a X^2-test for a research design with 1 nominal variable observed in multiple samples?

A

The question is whether the distribution of the observations over the categories is equal for the different samples. Comparable with t-est for 2 different samples.

36
Q

How do you use a X^2-test for a research design with two nominal variables in a single sample?

A

The question is whether the distribution of observations of the second variable’s categories is equal for the different categories of the first variable.

37
Q

What 3 assumptions need to be satisfied in order to use a X^2-test?

A

1) Data needs to be on nominal level of measurement
2) All observations are independent of each other and based on random observation/assignment.
3) Sample has to be large enough so that expexted frequency E per caegory is at least 5.

38
Q

When can you use non-parametric tests?

A

Data is not measured on an interval level, or if the probability distribution deviates from normal distribution.

39
Q

How does a sign test work?

A

Look only at the sign (positive or negative) of the difference D between two paired observations.

40
Q

How does theWilcoxon signed-ranks test work?

A

A nonparametric ordinal counterpart of t-test for paired observations. Investigates wheter the medians differ.

41
Q

What is the relation between the Wilcoxon signed-ranks test and the t-test?

A

If the Wilcoxon is significant, the t-test is as well.

42
Q

What is H0 in the median test?

A

The distributions of the two samples do not differ from each other, and that approximately half of the observations in both samples lie above the joint median and the other half lies below it.

43
Q

When can you use a X^2-test?

A

If you want to compare numbers withiin groups and they are at nominal level of measurement

MST (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions