issues of reliability, validity and sampling

Question 1

What is reliability

Answer 1

How much we can depend on any particular measurement for example the measurement of a table, a psychological characteristic such as iq or the findings of a research study

We want to know exeter if we repeat the exact same measurement/test/study we can be sure that we would get the same result. If not out measurement is unreliable

Question 2

What is validity

Answer 2

Related ti reliability

If a measurement is not reliable then a study can’t be valid

E.g. researcher might measure intelligence using an intelligent test. If the same person is tested on several occasions using the same test and the results change each time the study lack reliability hence validity cz scores are meaningless

But a measurement can be reliable but invalid

E.g. iq test taken over several months. Their score may be the same so it’s reliable but items on the may test may be assessing what a person learned at school rather than intelligence which lacks validity

Question 3

Talk about experimental research in terms of issue of reliability

Answer 3

For experiments
Reliability refers to the ability to repeat a study and obtain the same result
Esencial that all conditions are the sane otherwise any changes in the result may be due to changed conditions

Question 4

Observational techniques in terms of reliability

Answer 4

Two ir more observers should produce the same record to show reliability

Inter rater reliability is the extent to which the observers agree

Reliability if observations can be improved by training observers in the use of behaviour checklists

Question 5

Selten report techniques for reliability

Answer 5

2 different types of reliability when it comes to this. Internal and external

Question 6

What us internal reliability

External reliability

Answer 6

A measure of the extent to which sonething is consistent within itself

A measure of consistency over several different occasions

Question 7

What is inter interviewer reliability

Answer 7

Whether 2 interviewer produce the same outcome

Question 8

2 ways of a assessing reliability in self report techniques

Answer 8

Split half method
Compare a persons performance on 2 halves of a questionnaire or test. If the test is assessing the sane thing in all its questions then there should be a close correlation

Test retest method
Person given a questionnaire test or interview on one occasions and then this is repeated again after a reasonable interval. If the measure is reliable the outcome should be the same every time

Question 9

2 types of validity

Answer 9

Internal
Whether the researcher did test what they intended to test

External
Extent to which the results of the study cab be be generalised to other situations and people

Aka ecological validity

Question 10

Experimental research on validity

Answer 10

Internal validity is affected by extraneous variables which may act as an alternative iv

Lab low in external validity. Natural Hugh in external validity but this isn’t always the case

Whether the Pps were aware they were being studied and whether the task itself was artificial and thus low in mundane realism and reduce Generalisability

Question 11

Observational techniques on validity

Answer 11

Observations won’t be internally valid if the behaviour checklist is flawed

Cz some observations may be long in more than one category or some behaviours may not be code able which reduced internal validity

Internal validity also affected by observers bias. Observations are influenced by expectations

Observations likely to have high ecological validity cz they involve more natural behaviours

Question 12

Self report techniques on validity

Answer 12

2 ways of assessing
Face validity
Concurrent validity

Face validity
Are the questions related to the topic in question

Concurrent validity
Established by comparing performance on a new questionnaire or test with a previously established test on the same topic
High concurrent validity if Pps performance on both tests show high correlation

External validity affected by the sampling Strategies

Question 13

5 sampling techniques

Answer 13

Opportunity sample
Volunteer sample
Random sample
Stratified and quota sample
Snowball sampling

Question 14

Opportunity sample

Answer 14

Using people who are most easily available
Easiest method to use
Biased cz sample is drawn from a small part of the target pop
E.h. sample from ppl walking in city centre on Monday is biased cz doesn’t involve professional people who are working at the time

Question 15

Volunteer sample

Answer 15

Asking for volunteers

Example an advert in college

Can access a variety of Pps so more representative

But biased cz Pps are probe more motivated or with extra time on their hand os volunteer bias

Question 16

Random sample

Answer 16

Pps are selected using a random number technique

All members of the target pop are identified

In divs selected by random number generator or lottery method

Unbiased cz all have equal chance of being selected

But may still end up biased cz some may not want to participate

Question 17

Stratified and quota samples

Answer 17

Subgroups within a populations are identified

Predetermined number of Pps is taken from each sub group in proportion to their representation in the target pop

Stratified sampling is done using random techniques

Quota sampling is done using opportunity sampling

More representative cz there is a proportional representation of sub groups

But selections within each subgroup may be biased eg.g. through use of opportunity sampling

Question 18

Snowball sampling

Answer 18

Start with one or 2 people with eating disorder and ask them to direct you to other with Ed

Useful when conducting research with Pps who are not easy to identify

Prone to bias cz researchers may only contact ppl within a limited section of the pop

Question 19

Systematic sampling

Answer 19

Taking every tenth person from a register