Pilot Study, Reliability (Assessing), Validity (Assessing and Improving) Flashcards
Pilot Study
Small-scale prototype of a research investigation.
Carried out on a small number of participants.
Aim of piloting - To find out whether there are any problems with the design.
Reliability
Consistency in terms of the findings of research investigations or measuring devices.
Ways of Assessing Reliability
Test-Retest Method
Inter-observer Reliability
Test-Retest Method
Giving the same participants the same test the next day.
If the results are similar the test is reliable.
Inter-Observer Reliability
Getting two observers to observe the same phenomena and see if the rating they gave were similar.
Split-Half Method
Tests for Internal Reliability.
Splitting the test in half and comparing the results.
If results are similar test is reliable.
Interview/Questionnaire
Administer a questionnaire with valid questions as well as an interview and then comparing the results. If the interview score was reliable, there was a strong positive correlation between the scores.
Validity
Internal and External Validity.
Internal Validity - A test or measurement to measure what its supposed to measure.
External Validity - Comprised of Population, Ecological and Temporal Validity.
Ecological Validity
Findings can be generalised to other situations or settings.
Temporal Validity
Findings can be generalised to other historical times and eras.
Assessing Validity
Face Validity - Scrutinising a measure to determine whether it appears to measure what it intends to measure.
Concurrent Validity - Correlating the scores from the new procedure with scores from an alternative procedure.
Improving Validity
Single Blind - Participants do not know the condition they are in.
Double Blind - Both participants and experimenter do not know which condition they are in.
Randomisation
Use of chance when designing materials to control the effect of bias.
Standardisation
Giving participants the exact same instructions and procedures in a research study.
