Validity Flashcards
What is Validity?
- Validity is the extent to which a psychological test, tool, experiment or observation ACCURATELY MEASURES WHAT IT CLAIMS TO
What is Internal Validity?
- Validity within the experiment
- To what extent a researcher can confidently claim that the results are due to the experimental conditions
What are the factors that can affect internal validity?
Investigator Effects
- Behaviours that the investigator displays that may have an effect on a participants performance other than the IV
Demand Characteristics
- When the participants guess or understand the aims of the study and act accordingly either to please the researcher or to screw the researcher
Extraneous/confounding Variables
- Any variables other than the IV than could affect behaviour
- Confounding is the same as Extraneous just found after the experiment has concluded
Social Desirability bias
- Refers to the tendency participants provide answers to show themselves in a good light
Poor Operationalisation
- Behaviour needs to be clearly defined and measurable so they are interpreted the same by everyone
Problems with Measuring tool
- Tool should accurately measure what it claims to be and not prone to bias
What is External Validity?
- Whether the results are valid beyond the confines of the experimental setting
What are the types of external validity?
Population Validity
- Extent to which the results can be generalised to groups of people other than the participants used
Ecological Validity
- Extent to which the task used in a research study is representative of real life.
Temporal/Historical Validity
- Whether or not the findings from a particular study or concept holds true over time.
How do we asses Validity?
- Face Validity
- Concurrent Validity
What is Face Validity?
- Simplest technique
- Independent experts asses a measuring tool and see if it appears appropriate
- Subjective so not very strong
- Only takes place in the early stages of constructing a measuring tool
What is Concurrent validity?
- Comparison between 2 tests to measure if a new one is valid
- Obtaining 2 sets of scores from an already valid test and the new test
- When scores are obtained they will then be correlated to asses the new tool
- If correlation is 0.8+ the it can be classed as valid
How do we improve validity of experiments?
Use Control Groups
- Using a control group to assess if a new technique or method works on the DV
Use Standardised Procedures
- Ensure all participants experience identical environment conditions including the instructions, time given and task unless this differs as the IV
Use single or double blind procedures
- SB = Participants aren’t aware of the aims until they’ve taken part
- DB = Neither the Participant or Investigator knows who has done what to increase validity
How do we improve validity of Questionnaires?
Use a lie scale
- Assesses consistency o a participants responses to control the effects of social desirability
Anonymity
- Removes participants biographical details and encourages them to answer honestly
How do we improve Validity of Observations?
Use a Covert observation
- Participants are unaware they are being observed so less likely to display demand characteristics
Independent observations
- When there’s more than one observer they should work independently to reduce influence over each other
