terms Flashcards
reliability
Reliability refers to how consistent a measuring device is (repeatable). A measurement is said to be reliable or consistent if the measurement can produce similar results if used again in similar circumstances.
measures of reliability
This involves splitting a test into two and having the same participant doing both halves of the test. If the two halves of the test provide similar results this would suggest that the test has internal reliability -s.r
standardised procedure where each of the p experiences the experiment the same way ex
pilot study mini study before the main allows you to iron out problems with the procedure check of reliability for your measures s.r
test-retest method this involves doing a test once giving a q and then doing a again later then results can be compared to see if consistent s.r
validity
Validity refers to whether a study measures or examines what it claims to measure or examine (truthful). Self Report studies are said to often lack validity eg participants may lie; give answers that are desired etc. It is argued that qualitative data is more valid than quantitative data
concurrent validty
Used to assess the validity of self report measures. It compares the results of the self report with another self report on the same topic.
face validity
the degree to which a test measures what it appears to look as though it probably measures what it supposed to
cost benefit analysis
This is a method of balancing all the ethical costs of a project (eg likelihood of harm, loss of self esteem)… against the benefits of the research (new knowledge, business opportunities)
confounding v extraneous v
• Confounding Variables: An extraneous variable not controlled because it would confound (spoil) the results of the study. One variable would obscure the effects of another variable.
extraneous/Controlled Variable: A variable that remains unchanged or held constant to prevent its effects on the outcome and therefore may verify the behaviour of and the relationship between independent and dependent variables. Control variables are important in scientific experiments to test the validity of the results.
internal and external validity
Internal: concerned with the extent to which we can be sure that the results are due to the IV not EV. Study will lack internal validity if it is poorly designed, to improve; choose appropriate design, control EV, reduce likelihood of demand characteristics.
External: concerned with extent to which result findings can be generalised (population and ecological validity). Study will lack external validity if findings cannot be generalised, to improve: use representative sample (i.e. age, gender, class, ethnicity), do investigation in realistic setting.
inter rated reliability
Inter-rater reliability: Involves comparing or correlating the observations of two or more observers and checking for consistency
DIRECTIONAL/ONE TAILED:
Predicts the direction of differences in results (or correlation in a correlational study) e.g. Men are slower at…. Women will make fewer errors… Disabled people will find more…
NON-DIRECTIONAL/TWO TAILED:
Does not predict the direction of any differences in results (or correlation in a correlational study) e.g. there will be a difference between men and women’s scores at….. Attractive people will receive different length sentences than less attractive people.
NULL HYPOTHESIS:
A prediction that there will be no differences between the results from the different conditions (or no correlation) e.g. there will be no differences between men and women in….
Operationalise:
the process of strictly defining variables into measurable factors. The process defines concepts and allows them to be measured, empirically and quantitatively. This allows other researchers to repeat the same procedures
