Define internal reliability and how it is reduced
The extent to which a measure is consistent within itself
- usually issues in procedure: operationalisation of variables, order effects (repeated measures design), lack of standardisation of procedures, accuracy of measuring tool
How is counterbalancing used to reduce effect of order effects?
- population is split to 2 halves where one half do step 2 then 1, and the other do step 1 then 2
Explain standardisation
- process of ensuring the experience of the research is identical for all Ps
- provide a standard set of instructions using same environment to control as many variables as possible
- allows data collection to establish a cause and effect
—> utilise scientific measuring tools where possible so data is easy to compare
Explain external reliability and what can reduce it
- the extent to which a test produces consistent results over several occasions with the same Ps
- issues commonly relate to the consistency of results over time and across different populations
—> location can be a problem as acts as a barrier to external reliability. Only solution is to conduct research in controlled conditions (ie lab) but this reduces validity
Experimental way to show internal reliability
Split half: split Ps into 2 or more groups and test conditions one at a time
—> if tests show similar results it can be argued it has internal reliability
Experimental way to show external reliability
Test re-test: same Ps tested over a period of time or separate occasions using same procedures
—> if results similar then external reliability
Explain inter-rater reliability
Where two or more psychologists produce consistent results by using a standardised procedure, an agreed coding system or a correlation of their data
—> prevents variation of data collected by multiple psychologists / researchers
What is validity?
- reasonably certain that findings of study reflect real life behaviour
External validity
- whether the study paints a true picture of real life behaviours and whether findings would be applicable to other places, times and people
- population validity, ecological validity, historical validity
Internal validity
- the study measures what it intends to measure as confounding variables have been controlled and will not affect results
- accurate and effects on DV caused by IV
What can cause validity issues?
Extraneous variables
How to prevent participant variables influencing validity
- use a matched pairs design
- assign Ps randomly so chance of individual differences is lower
How to prevent situational variables affecting validity
- keep all aspects the same for all Ps
- counterbalancing to reduce order effects
How to prevent experimenter variables impacting validity
- experimenter act in neutral manner
- standardised instruction set for Ps
- double blind study
- pilot study
How to prevent participant effects impacting validity
- field experiment
- anonymous Ps
- single blind study
How to prevent stimuli variables influencing validity
- independent measures
- repeated measures designed to
