Validity

Internal Validity

Things that happen within the study, asks whether we actually measured what we intended to measure

External validity

Outside of a study, extent to which findings of a study can be generalised beyond the research setting

5 causes of validity issues

1. Situational variables- time of day, temp, noise levels. Something about situation of experiment could act as an EV if it has an effect on DV

2. Participant variables- difference in participants: age,gender, IQ etc. difference between each participant can cause change in DV how ppl perform

3. Investigator effects- behaviour and language of the experimenter may influence behaviour if the participants. Way an experimenter asks a question (researcher bias) leading questions may consciously or unconsciously alter how participants respond

4. Demand Characteristics- looking for cues as to how to behave, participants may guess aim of study- may cause them to change their behaviour

5. Participant effects- participants aware they’re in an experiment, may behave unnaturally

Ways to overcome issues of validity

1. Use of standardised procedures to ensure all participants tested under same conditions

2. Using repeated measures designed, matched pairs can also be used

3. Double blind technique to overcome investigator effects

4. Double blind design so participants can’t guess the aim

5. Redesigning study so participants can’t guess aims- reduce participant effect & create situations with high mundane realism

Face validity

Look at the method you are using and decide whether or not it seems to be measuring what you intended to measure; based on intuition- simplest method

Content validity

Panel of experts asses and measure for validity

Concurrent (agree) validity

Testing participants with both new and established test- if test had concurrent validity, high agreement between scores on both measures

Construct validity

Definite what it is we are aiming to measure and ensure all parts of that definition are being measured

Predictive validity

Compare results of our measure with other measures to see whether our tests predicts what we expect them to predict