Validity and Reliability

Question 1

external validity

Answer 1

extent to which results of a study can be generalised and applied to real life settings

Question 2

internal validity

Answer 2

measuring the right thing

Question 3

types of validity

Answer 3

temporal, ecological, population

Question 4

factors affecting internal validity

Answer 4

• demand characteristics
• confounding variables e.g. extraneous
• social desirability bias
• investigator effects
• poorly operationalised variables/behavioural categories

Question 5

dealing with validity issues

Answer 5

• experimental realism - ppts forget it’s an experiment if it’s realistic and engaging
• pilot study - test study with a small group of ppts
• single blind study- prevents DC
• double blind study - prevents DC and researcher bias

Question 6

Face validity

Answer 6

• extent to which a measure looks like it’s measuring the right thing
• improved by changing/rewording questions

Question 7

how is face validity assessed?

Answer 7

• researcher checks their tools are measuring the right thing
• also ask a peer to check for us
• doesn’t require expert knowledge of the topic

Question 8

concurrent validity

Answer 8

• extent to which your measure is similar to another measure of the same thing
• improved by amending tasks/questions and checking concurrent validity again

Question 9

how is concurrent validity assessed?

Answer 9

• compare yours to another previously validated measure on the same topic
• ppts given both method to complete and results are compared
• if similar then you have concurrent validity

Question 10

reliablity

Answer 10

• extent to which research is consistent
• need to show an understanding of how you might test this and improve it

Question 11

ways of assessing reliability

Answer 11

• test-retest at spaced intervals
• inter-rater reliability
• experiment: check how dv is measured
• observations: use operationalised behavioural categories
• self-reports: test-retest for questionnaires, inter-rater for interviews