statistical testing

Question 1

define statistical testing

Answer 1

• provides a way of determining whether hypotheses should be accepted or rejected
• they tell us whether differences or relationships between variables are statically significant OR have occurred by chance

Question 2

why do we need to complete a statistical test ?

Answer 2

to see whether the difference found within studies haven’t occurred by chance/coincidence

Question 3

define sign test

Answer 3

a statistical test used to analyse the difference in scores between related items

Question 4

what is an example of ‘related items’ ?

Answer 4

the same participant tested twice

Question 5

what are the 2 types of hypotheses ?

Answer 5

• the original/alternate hypothesis
• the null hypothesis

Question 6

what is the difference between original and null hypothesis ?

Answer 6

O - states there will be an effect
N - states there will not be an effect

Question 7

give an example of original hypothesis

Answer 7

there will be a difference in participants depression score on the BDI before and after a 6-week course of CBT

Question 8

give an example of null hypothesis

Answer 8

there will be no difference in participants depression score on the BDI before and after a 6-week course of CBT

Question 9

what are the abbreviated form of original and null hypothesis ?

Answer 9

H1 - O
H0 - N

Question 10

what does the statistical test allow to decide between original and null hypothesis ?

Answer 10

to determine whether to accept or reject the null hypothesis

Question 11

what is the sign test used for ?

Answer 11

used to find out whether the difference established is significant

Question 12

what conditions need to be met in order to carry out the sign test ?

Answer 12

1. looking for a difference rather than an association
2. have used a repeated measures design
3. nominal data.

Question 13

what are significance levels used for ?

Answer 13

to check for significance differences or relationships

Question 14

what is the accepted level of probability in psychology ?

Answer 14

0.05 / 5%

Question 15

define significance levels

Answer 15

represents the threshold at which researchers decide whether to reject the null hypothesis

Question 16

what does it mean if the experimental hypothesis is accepted ?

Answer 16

this means there is less than 5%
probability that the results occurred by chance

Question 17

what do some researchers do to the significance level and why ?

Answer 17

-make a stringent (small) signifance level (0.01/1%)
- means that researchers can be more confident in their results
- to make sure that results weren’t due to a chance

Question 18

give examples of why one may want to employ 1% level of significance

Answer 18

• may involve a human cost (new drugs are being on trial )
• when a particular investigation is a one-off , and there is no possibility that it can be repeated in the future

Question 19

what can choosing the wrong significance levels result in ?

Answer 19

type 1 error + type 2 error

Question 20

define type 1 error

Answer 20

if too lenient significance level is used
(e.g. 10% )
–> results rejecting the null hypothesis that is fact true
null hyp = true
alt hyp = wrong

Question 21

define type 2 error

Answer 21

if a too stringent significance level is used
(e.g. 1%)
–> results in accepting a false null hypothesis
null hyp = wrong
alt hyp = true

Question 22

what does one -tailed and two-tailed mean ?

Answer 22

one-tailed means directional hypothesis
two-tailed means non-directional hypothesis

Question 23

what are 5 pieces of information one needs in order to reading a table of critical values

Answer 23

1. the significance level desired (always 0.05 or 5% except in the cases described
   above).
2. the number of participants in the investigation (the N value),
3. whether the hypothesis is directional (one-tailed) or non-directional (two-tailed)
4. the calculated value
5. the critical value

Question 24

what does test of difference/correlation mean ?

Answer 24

association or relationship = correlation

Question 25

what are the test designs ?

Answer 25

unrelated - independent groups
related - matched pairs or repeated measures

Question 26

define matched pairs

Answer 26

pairing up participants who are similar in important ways

Question 27

define repeated measures

Answer 27

all participants take part in all
conditions of the experiment

Question 28

define independent groups

Answer 28

• participants allocated to different groups where each group represents one experimental condition

Question 29

what does levels of measurements mean ?

Answer 29

quantitative data that can be classified into different levels of measurement

Question 30

what are the types of levels of measurement ?

Answer 30

• nominal
• ordinal
• interval

Question 31

define what nominal data means ?

Answer 31

• categorical data
  e.g - favourite dessert , smoker/non-smoker and sex
• discrete data –> only appears in 1 category

Question 32

define what ordinal data means ?

Answer 32

• data that can be ordered in some way
  e.g. rating scales
• subjective + no equal intervals between values

Question 33

define what interval data means ?

Answer 33

• data based on numerical scales
  e.g. weight, size , scores , speed
• objective
  -equal intervals between values
• most sophisticated form

Question 34

draw out the choosing test table

Answer 34

find in notes :)

Question 35

what is the way of remembering the 8 tests we need to use ?

Answer 35

Carrots - Chi 2
Should - Sign test
Come - Chi 2
Mashed - Mann-Whitney
With - Wilcoxon
Swede - Spearman’s rho
Under - Unrelated t-test
Roast - Related t-test
Potatoes - Pearson’s r