Vaidity

Question 1

Internal validity

Answer 1

The extent to which changes in a DV are legitimately caused by a change in an /V or is it by some other extraneous variable.
Likewise in a correlation study is
the strength of relationship
legitimate or caused by extraneous variables

Question 2

Internal issues of internal validity

Answer 2

Demand Characteristics - Ppts guess the aim of the study and change their natural course, of behaviour in order to help the researcher obtain favourable results.
Researcher bias - a researcher’s
interpretations of behaviour or the way they set up the study may be biased to support their hypothesis.
Social Desirability - Ppts will change answers or natural course of behaviour in to present themselves in more socially acceptable ways to show themselves in a better light.
Test is inaccurate- the measure used in the study does not measure the desired behaviour, e.g. an IQ test that measures solely mathematical ability.

Question 3

How to overcome internal validity issues

Answer 3

Single blind study (to deal with demand characteristics )- do not inform Ppts the aim of the study or take measures to ensure they do not guess the aim of the study, such as adding bogus activities.
- Double blind study ( to deal with researcher bias) - Ensure the researcher collecting the data and the participants do not know the aim or hypothesis therefore restricting bias in interpretation (and demand characteristics).
Ensure confidentiality (to deal with social desirability )- if participants believe that their behaviours or responses cannot be attribute to them they will more likely be truthful.
Pilot-study ( to deal with test is inaccurate)- ensure test is accurate by trialling it with a focus group then consult with participants and make changes as needed.

Question 4

How to assess validity

Answer 4

Use CCCFP
Content validity - use an independent expert to assess the content (measures) of the study
Concurrent validity - compare test in study to one that is already seen as valid - same
Ppts do both tests (NB compare tests not studies). If similar results then it’s valid
Construct validity - use a construct (theory or idea) to establish if a test is legitimate
Face validity - the researcher looks at their test to see if it is measuring what it claims to measure
Predictive - researcher makes predictions of the outcome of test then compares to actual findings eg predicts their personality test will identify different personality types

Question 5

External validity

Answer 5

The extent to which the findings of a study can be legitimately generalised beyond the setting of a study

Question 6

External validity issues

Answer 6

Ecological validity - the extent to which findings of a study can be generalised to real life settings, often a problem for laboratory studies.
Population validity -the extent to which the findings from a study can be generalised to the target population or to alternative groups, usually an issue when there is small or biased sample groups.

Question 7

How to overcome external validity issues

Answer 7

Design the study to be carried out in real life settings( to deal with ecological validity) - field studies or natural experiments (form of quasi-experiments)
Carry out the study on different groups of people or increase the size of the sample group (250+
Ppts).(to deal with population validity )

Question 8

What is validity

Answer 8

The extent to which a test are research study accurately measures what it intends to measure