Flashcards
Reliability and validity
Cards in this deck (14)
- 1Reliability and validity-psychologists who are committed to using scientific methods in their research investigations try to design studies that can be replicated or repeated.
-to achieve this, it is vital that the measurement tools used in an investigation are reliable and that the data they produce are valid. - 2ReliabilityThe extent to which a method of measurement or test produces consistent findings.
-replicate study to check for consistent results - 3Validity-in addition to our measurements being consistent, they also need to be valid.
-validity means that we are actually measuring what we claim to be measuring.
-whether it has tested what was set out to be measured.
-there are 2 broad categories of validity
1. internal validity
2. external validity - 4Internal validity-the validity of an experiment in terms of the context in which it is carried out.
-concerns events within the exp as distinct form external validity. - 5External validity-the validity of an exp outside the research situation itself; the extent to which the findings of a research study are applicable to other situations, especially 'everyday' situations
-External validity is affected by internal validity- you cannot generalise the results of a study that was low in internal validity because the results have no real meaning for the behaviour in question. - 6Experimental validity2 types
1. internal validity
2. external validity
1. internal- refers to inside the study
e.g. the measurements tools, procedure
2. external- concerns what goes on outside the exp
e.g. it is about representativeness and generalising. - 7Types of Internal validity Assessing internal validity1. Face
2. Concurrent
3. Predictive
1. face validity- assess validity by asking yourself if the test looks as if it could/is measuring what it is suppose to (e.g. a tape measure isn't going to measure IQ, only the size of your head).
2. concurrent validity- involves comparing the results found in a new test with those from an existing, already well established test that has good validity.
(e.g. score 148 on an established test, but only 110 on the new one- this raises Q's on the validity of the new test.
3. predictive validity- the ability that a test can predict performance on future tests
e.g. performance on your GCSE tests should help to predict the type of grades you are capable of achieving at A level. - 8Type of External validity Assessing external validity1. Population
2. Ecological
1. population validity- whether the findings can be generalised to the people outside of the sample population.
2. ecological validity- how true to real life were the findings, could they be generalised to other settings. - 9Internal valditiy being affected- extraneous variables can affect internal validity if the extraneous variables aren't controlled properly, then they could directly affect the DV's, but we only want the IV affecting the DV.
-if a study doesn't have face value, concurrent (concur- agree) and predictive validity the validity is very low. - 10More on external validity-the sampling method can affect population validity because the findings may not be a generalisation to other people outside of the sample population.
-choice of method- this can affect the ecological validity because the method chosen may be quite artificial, so isn't very true to real life findings.
-so the method needs to be able to be generalised to other settings. - 11external validity-the more you can replicate a study, in different settings, with different people, in varying time periods, the higher the external validity becomes.
- 12Examples of external validity1. population- Milgram (1963) lacked population validity
2. Ecological- Asch (1940) wouldn't be generalised as being a normal everyday life activity. - artificial - lacks ecological validity.
- a lot of exp's that are lab exp are low in ecological validity.
3. historical- Dement and Kleitman (1957) and Milgram (1963) - 13Types of ReliabilityInternal
-refers to what is going on inside the study
-e.g. do all of the Q's on a questionnaire measure the same thing?
External
-refers to outside of the study
-e.g. observations are repeated.
-similar/consistent results each time in a test outcome. - 14ways of assessing reliabilitiyInternal
split half method
-this is to compare a person's performance on 2 halves of a questionnaire or test.
-if the test is assessing the same thing in all its Q's then there should be a close correlation in the scores derived from both halves of the test.
External
Test-retest
-where a person is given a questionnaire/interview/test on 1 occasion and then this is repeated again after a reasonable interval (e.g. week/month)
-if the measure is reliable, the outcome should be the same every time.
observer (inter rater) reliability
-observers use the same behavioural category check list and observe the same event separately
-you can then compare their findings to see how similar they are
Information
- Description
- as level Psychology Flashcards on Reliability and validity , created by tara4444 on 10/05/2014.
- Created by
- tara4444
- Views
- 46
- Rating