Variability

Question 1

Variability

Answer 1

Data collected from humans is varied

Value used to represent natural variation between ppt scores and sample means is standard deviation

Total variance = systematic and unsystematic variance

Error bars provide indication of spread of data (variation)

Wider error bars indicate more variation in scores

Question 2

Standard deviation

Answer 2

Average amount of deviation between ppt and sample mean

Unsystematic variance

Question 3

Mean difference between experimental conditions

Answer 3

Systematic variance

Question 4

Standard error

Answer 4

Estimate of the deviation between our sample and population means

Large value = a lot of variation

Sample mean will unlikely be exactly the same as the population mean

Question 5

Increasing unsystematic variation increases uncertainty

Answer 5

95% Cl error bars estimate he effects we should find in our population

Uses standard error to take into account amount of deviation we would have

Top bar= upper 95% confidence interval

Middle circle= mean

Bottom bar= lower 95% confidence interval

Question 6

Effect

Answer 6

Mean difference

Eg., mean - mean =

Question 7

Error

Answer 7

Standard error

SD of mean differences / standard error =

Question 8

Inferences

Answer 8

T test compares estimate of error against effect

Effect / error =

T(…)=…., p= ….

Question 9

P value

Answer 9

Small p value means probability of finding effect if no effect in population is very small

Probability of null hypothesis is true

If the probability of finding an effect that doesn’t exist in the population is

Question 10

Inferential statistics

Answer 10

We need to generalise

Allows us to generalise from a small sample population

Do not know population mean and standard deviation

Small samples will not be representative of population

Results in flawed estimates of error

Question 11

Effect size

Answer 11

Standardised measure of how important the difference / effect is that you found in the sample

(Magnitude of effect)

Question 12

Magnitude of effect for tests with 2 comparisons

Eg t tests

Answer 12

Cohen’s d

Provides measure of distance between 2 means in standard deviations

Amount of overlap in %

High overlap = small d

Low overlap = High d

Question 13

Magnitude of effect

Answer 13

Partial eta squared

np2

How much of the total variance is accounted for by the treatment effect (systematic variance)

Value provided by SPSS converter into proportion eg 0.375 = 37.5%