WEEK 8

Question 1

Parametric Tests

Answer 1

• Parametric statistics are based on the population parameters
  -Assumptions about the underlying population our data is from
  -Assumption that our samples are similar to underlying probability distributions

Question 2

Non-Parametric Tests

Answer 2

• Do not make any strict assumptions about the data distribution
  -Implies that they can be used both when the assumptions are met, and when they are not met

Question 3

P VS NP

Answer 3

p:
-More assumptions
-Less Universal
-Larger Power
-should use whenever possible

np:
- fewer assumptions
-more universal
-lower power

Question 4

General Parametric Assumptions

Answer 4

1. The scale which we measure the dependent variable on should be interval or ratio level
2. the populations the sample are drawn from should be normally distributed
3. the variances of the populations should be approx equal
4. no outliers or extreme scores

Question 5

T-Test William Gossett (1908)

Answer 5

Used when:
- we want to compare the differences in means:
-two separate groups
-One group measured on two occasions

Question 6

Assumption checking, T tests

Answer 6

-Checking 1: The scale which we measure the dependent variable on should be interval or ratio level
-Checking 2: The populations the samples are drawn from should be normally distributed (t test can be used when normally distributed)
-Checking 3: The variances of the populations should be approx equal if comparing more than one group, Levene’s test for equality of variances
-Checking 4: no outliers, must be checked before running analysis

Question 7

Hypothesis for T tests

Answer 7

Null: The population mean from the two groups/ condition are equal
Alternative: The population means from the two groups/ conditions are not equal

Question 8

Repeated Measures

Answer 8

• Advantages:
  -same people: the natural differences between people are controlled for
  -The power of this design is higher than the between subject design

Question 9

Effect Size:

Answer 9

• An objective and standardised measure of the magnitude of an effect
• For t-test the effect size we use is called Cohen’s d
  -tells us the size of the difference between groups
  -tells us how many standard deviations the means differ by

Question 10

Cohen’s d (1998)

Answer 10

-small: d/0.2
-medium: d/0.5
-large: d/0.8

Question 11

formally reporting statistical results:

Answer 11

1.state what type of test has been performed and on what
2. report the test statistic =, df, stat significance
3. report the mean difference and associated confidence intervals
4. report the effect size
5. comment on the means