Validity Flashcards
(17 cards)
What is validity?
concerned with accuracy and legitimacy of a study
Internal validity
the extent to which you are measuring what you claim to measure. E.g if in an experiment your results are due to the manipulation of the IV rather than CVs/EVs this means your results have internal validity.
External validity
The extent to which the findings can be generalised.
Ecological validity
The extent to which findings can be generalised beyond the situation the study was conducted in.
Population validity
The extent to which findings can be generalised beyond the participants used.
Common issues of internal validity in experiments
- Demand Characteristics: a participant will provide answers to conform to the study’s aim)
- Confounding Variables: if the testing environment differs across conditions
- Participant variables: in an IMD individual differences between the group may bias results
How can you solve the issues of validity in experiments?
- Deception don’t reveal the true aim of the study and debrief at the end
- Standardisation of test conditions
- Use a different design e.g Matched Pairs/ RMD
Common issues of internal validity in self reports
- Demand Characteristics (where a participant will provide answers to conform to the study’s aim)
- Social Desirability (where a participant will provide an answer to put themselves in a better light)
- Leading Questions: these can lead to participants providing a desired answer
Common issues of validity in observations
- Demand Characteristics: in an overt design (causing participants to behave in line with the study’s aim)
- Observer Bias: when the researcher interprets the participants behaviour to fit their hypothesis
- Flawed Coding Systems: where the measures for recording behaviour contain errors
How can you solve the issues of validity in self reports?
- Deception or Double Blind Technique
- Confidentiality and Anonymity
- Pilot Study: highlights designs flaws
How can you solve the issues of validity in observations?
- Deception or covert design use
- Double Blind Technique
- Pilot Study
Ways of assessing validity
- Face Validity (before the study)
- Content Validity (before and after the study)
- Concurrent validity (after the study)
- Predictive validity (after the study)
- Construct validity (after the study)
Face validity
The details of the study can be checked using a potential naive participant(s) (e.g as part of a pilot study.) If the purpose of the study is clear to them then the study can be seen as having face validity.
Content validity
The details of the study can be checked using an expert in the field of research psychology. If the expert thinks that the method of measuring the behaviour to be investigated is accurate then the study can be seen as having construct validity.
Concurrent Validity
The proposed study and an established study that investigates the same behaviour needs to be conducted on the same group of participants at the same point in time. If the results of the 2 are comparable then the proposed study can be seen as having concurrent validity.
Predictive validity
The results of the new study need to be established as an accurate predictor of future behaviour in a similar situation (e.g GCSE grades predicting AS results) If the results can be used as an accurate predictor then the new study can be seen as having predictive validity.
Construct Validity
The results of the new study need to be established as an accurate reflection of the results of other wider research into behaviour under investigation. If the results can be used as an accurate reflection of wider research then the new study can be seen as having construct validity.
