Week 3: ANOVA assumptions, power and mean comparison Flashcards

1
Q

what is the underlying ANOVA equation?

A

Score = Grand mean + treatment effect + residual error (measurement and individual differences)

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

Essentially, how does ANOVA break down variance? and what does the comparison of this breakdown produce?

A

Into 2 parts

  • Variance due to treatment
  • Error variance

Comparison of these two variances produces the F STATISTIC

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

What are the assumptions of a between ANOVA?

A

Homogeneity of variance: SD for all groups is about the same
Normality: error is normally distributed
Independence of observation: truly between design

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

How do you test the assumption of homogeneity of variance?

A

Levenes test
- if statistically significant the null is rejected. this means that there is a difference somewhere do homogeneity is violated

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

What do you do if Levenes test is significant?

A

You can look up density plots to find the culprit

can be violated without grave consequences as long as sample sizes are equal

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

What can you use to test the assumption of normality?

A

Shapiro WIlk
If significant, assumption of normality is violated

can also look at histograms, distribution plots, skewness and kurtosis

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

What happens if normality is violated?

A

Typically its not a big issue

ANOVA still tends to be robust in terms of normality violations

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

How do you report that you have used a non-parametric version of an ANOVA?

A

‘Similar results were found using non-parametric….’

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

What are quantile-quantile plots?

A

They look at normality assumption.

Chop data up into how many scores are in each quantile and compare it against what we would expect in that quantile for a normal distribution.

Straight line = perfect
Deviations = bad

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

What happens to the ANOVA if assumptions are violated?

A

If you arent confident, can use non-parametric versions of the test
This can then increase confidence in results

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

What is skewness of data?

A

Deviations from typical bell curve of data distribution

- asymmetrical

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

Explain positive and negative data skewness

A

Negative skew: Have a long tail towards low values in the data (more low scores than expected)

Positive skew: long tail towards high values (more high scores than expected)

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

No skew value?

A

0 - perfectly symmetrical

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

What values are considered moderately skewed?

A

Between -1 and -0.5
or
Between 0.5 and 1

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

What values are considered highly skewed?

A

(-/+) 1+

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

Skewness in Quantile-Quantile plots?

A

If data drops below the line, it is skewed to the left or negatively skewed

If data rises above the line, it is skewed to the right or is positively skewed

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

What is kurtosis?

A

It is a measure of tailedness in data distribution curve

how light or heavy-tailed the data is

18
Q

What is leptokurtic distribution (kurtosis)?

A

People scoring closer to the average (light tailed)

19
Q

What is platykurtic distribution (kurtosis)?

A

Wider spread of scores so maybe more heavy tailed

20
Q

What is the optimal kurtosis value?

A

It is reported in terms of how much excess skew there is - so the optimal would be 0!!!

21
Q

How do you know when to reject normality in terms of kurtosis?

A

You are given a standard error as well as a skewness value - these can be used to determine a z-score

z score = Value of kurtosis/standard error

If z score is less than -2 or more than +2, reject normality

22
Q

What is power?

A

The probability of correctly rejecting the null

finding a difference between means if it is there

23
Q

What is power associated with?

A

Type 2 errors

24
Q

when do type 2 errors typically occur?

A

when alpha level is too strict
or
outlier increases error variance so hard to see treatment effect clearly

25
Q

What is the power equation?

A

power = 1-B

The probability of finding a real difference = 1 - the probability of not finding a real difference

26
Q

What increases power?

A
  • Increase in magnitude of difference between means (effect size): makes it more obvious
  • Alpha level (more generous): increases power but also increases type 1 errors or false positives
  • variance in scores: less variance increases power as if too much variance, makes it harder to see effects clearly
  • larger sample size: gives more accurate means - more power as can see more clearly
27
Q

A priori power estimate?

A

Estimate before the experiment is run

- helps to ensure have adequate numbers of participants to detect any real differences between treatments

28
Q

Post hoc power estimate?

A

Estimate after data has been gathered
Estimates the likelihood of being able to replicate a significant difference if repeated
- value between 0 and 1
eg 0.82 - 82%

29
Q

How can you obtain a post hoc power estimate?

A

G*power - REFER TO SLIDES
Can provide you with a power estimate if you feed it means, average variance (MSE) and sample size
- Tell it what kind of test you’ve done
- gives an effect size that can then be used to find power

30
Q

How do we find a priori power estimates?

A

Estimation of effect size (cohens d or f) - or you can use the effect size from other studies
With this effect size you can compute a non-centrality parameter (used to find the power of an experiment for any effect size)

Can then plan sample size etc (can use g*power again)

31
Q

What happens when a statistically significant comparison has more than 2 means?

A

Need further tests to determine which of the means actually differ from each other
- typically multiple comparisons

32
Q

What are the two approaches to follow up tests?

A

A priori: chosen before data is collected

Post hoc: planned after data is examined

33
Q

Errors in multiple comparison tests?

A

Major type 1 error rate risks

34
Q

Per comparison error rates vs. family-wise error rates?

A

Per comparison:
the probability of making a T1 error on any comparison - this is alpha

Family-wise:
the probability that a family of comparisons will contain at least 1 T1 error
- per comparison rate stacks up
- error rate per comparison X number of comparison

35
Q

A priori comparisons?

A

Usually when have very specific hypothesis - creates a smaller number of comparisons so T1 probability is reduced

  • contrasts
36
Q

How to do a priori comparisons in jamovi?

A

These are ‘constrasts’ and this is a tab within jamovi under the analysis set up

37
Q

What are the different types of contrasts?

A

Deviation - each level to grand mean
Simple - mean of one to each other one
Difference - mean of level to previous
Helmert - mean of level to all subsequent combined
Repeated - mean of one to each subsequent individually

38
Q

T1 error rates in contrast tests?

A

Need to do bonferonni adjustments

alpha/number of comparisons

39
Q

Critics to bonferonni?

A

Claim it may be too conservative and we may make T2 errors instead and miss actual effects

40
Q

What can we do here if we are worried that bonferonni is too conservative?

A

Linear step up!

41
Q

What is linear step up?

A

Focuses more on false discovery rate (FDR)

Rank comparisons according to their p values - divide this rank by the number of comparisons
Then multiply this by FDR which is alpha

Gives you a critical p value to determine significance against

42
Q

What other post hoc tests may be appropriate when comparing means?

A

Tukey or Holm