Research Methods - Reliability and Validity

Question 1

Give the definition of internal reliability

Answer 1

The extent to which the procedure can be replicated

Question 2

Give the definition of external reliability

Answer 2

The extent to which the results can be replicated and checked for consistency

Question 3

Name the aspects of internal reliability and external reliability

Answer 3

internal aspect = procedure

external aspect = results

Question 4

How can we increase internal reliability

Answer 4

Through using a standardised procedure - where all aspects of the procedure are kept the same for all participants

Question 5

How can we check internal reliability

Answer 5

Use split half method ( usually for questionnaires)

use test re test method

Question 6

How can we increase external reliability

Answer 6

Through collecting quantitative data

Question 7

How can we check external reliability

Answer 7

Use split half method ( usually for questionnaires)

use test re test method

Question 8

Define inter rater reliability

Answer 8

The level of agreement between two or more researchers when they’re observing the same behaviour in the same way - aiming for an 80 percent agreement level

Question 9

Describe the ‘split half method’

Answer 9

When a questionnaire involves repetition of questions from the first half in the second half of the questionnaire to check the consistency of results

Question 10

Describe the test-retest method

Answer 10

This is when participants are tested more than once in exactly the same way.

Question 11

What is validity

Answer 11

validity refers to the extent to which a study has tested or measured what it set out to measure

Question 12

Define internal validity

Answer 12

This refers to how well a study measures what it sets out to measure at the start of research

Question 13

How can we increase internal validity

Answer 13

through implementing high levels of control in the procedure to reduce the effect of extraneous variables

Question 14

What are extraneous variables

Answer 14

variables that haven’t been con trolled that can cofound the results of a study. If these aren’t controlled then the validity of a study can be questioned

Question 15

Name the five different types of extraneous variables

Answer 15

Situational variables
Individual differences
Researcher effects
Demand characteristics
Social desirability

Question 16

What are situational variables and give an example

Answer 16

Any factors within the environment that can affect participants behaviour or the results eg - background noise, time of day

Question 17

What are individual differences and give an example

Answer 17

Any differences between participants that aren’t accounted for eg age, gender, IQ, amount of sleep

Question 18

What are researcher effects and give examples

Answer 18

If a researcher acts differently to different participants or conditions which can influence certain behaviours eg - changing their disposition or attitude in a different condition

Question 19

What are demand characteristics

Answer 19

If participants know the aim of the study, they may change their behaviour

Question 20

What is social desireability

Answer 20

When a participant behaves in a a way that makes them seem more socially desirable and acceptable which may not be truthful

Question 21

Define external validity and its two types

Answer 21

the extent to which the findings of a study can be generalised to factors outside of research

• ecological validity - if the task and or setting is representitve of real life
• population validity - whether sample is representive of the wider target population

Question 22

name the three ways we can check the validity and define them

Answer 22

face validity - extent to which something looks as if it will measure what it is supposed to

construct validity - extent to which a test measures all aspects of behaviour in question

criterion validity - extent to which the scores on one measure are able to predict the outcome on another related measure

Question 23

name and define the two types of criterion validity

Answer 23

concurrent - extent to which a test correlates well with the measure of the same thing that has been previously validated . You would compare results of your new measure with a previously validated one

predictive - extent that a measure can predict future behaviour or attitude