research methods part 2 Flashcards
define reliability
refers to how consistent a measurement devise is internally or externally
what are 3 ways of assessing reliability
- test retest
- inter-observer reliability
- measuring reliability
test retest
- repeating an experiment on multiple occasions
- sufficient time between
- correlate results to make sure they are significant
inter-observer reliability
- use two psychologist to observe the same behavior.
- applies to behavioral categories
- data should be correlated to test reliability
how is reliability measured
with a correlation analysis.
the result should have a co-efficient should be higher than +0.80 to be significant
how to improve reliability in a questionnaire
- deselect or rewrite questions
- use closed questions
how to improve reliability in interviews
- use the same trained interviewer
- avoid using ambiguous questions
- use structed interviews
how to improve reliability in observations
- probably operationalized behavioral categories
- train the observers
define validity
the extent to which an observed affect is true or generalizable
what are the 4 type of validity
- internal validity
- external validity
- temporal validity
- ecological validity
internal validity
relates to if the observed effects are coursed by the manipulation of the independent variable or other external factors.
(mainly affect by demand characterists)
external validity
relates to external factors outside the investigation. e.g. generalizability
a type of external validity is ecological validity
temporal validity
relates to the issue if finding or studies hold true overtime
ecological validity
is a type of external validity that relate to if the study can be applied to different settings
2 ways of assessing validity
- face validity
2. concurrent validity
face validity
- looks at if a study/research is measuring what it is suppose to.
- done by giving it to an expert or checking if it seems correct
concurrent validity
- seeing if the results seem similar to previous studies
- correlation of the to scores should excessed +0.80
how can you improve validity in experiments
- use control group (make sure the iv is affect the dv)
- standardized procedures
- single/ double blind procedures (no demand characteristics)
how to improve validity in questionnaires
- lie scale (assess the consistency of the results and control social desirability bias)
- anonymous responses
how to improve validity in observations
(tend to have high ecological validity)
- use covert observation
- make sure the behavioral categories aren’t to board
how to improve validity in qualitative research
high ecological validity = less interpretative methods - conclusion
- make sure experimenter and participants perceive the research in the same way
- triangulation
define the sign test
a statistical test used to analysis the difference in scores between related items. the data used is nominal.
define probability
look at the certain likelihood an event will occur
define a peer review
is the assessment of scientific work by others specialists in the same field, to ensure that any research intended for publication is of high quality.
