Reliability and Validity In Psychology AQA A Explained
Internal & External Reliability within Psychology A Level (AQA A)
Reliability means consistency. There are different types of reliability: External reliability, Internal reliability and Inter-rater reliability.
External Reliability Within Psychology A Level
External reliability in psychology is all about consistency. If something is measured repeatedly, for it to be considered reliable the same results should be apparent every time when recreating the experiment using the same set of procedures. So for example: If I measure my height and the tape measure states I am 175cm tall, when the experiment is conducted again using the same procedure and the same tape measure – I should still be 175cm tall. If the tape measure reads I am now 185cm tall then the tape measure lacks reliability. This is because under the same conditions similar results should be found when the test is repeated for it to have external reliability (consistency) in findings.
An example of a study that has external reliability is Bandura’s Social Learning Theory. When repeated by other researchers to confirm the findings they have found similar results themselves with children highlighting the reliability of this study. The results will never be exactly the same (due to individual differences among samples or culture) but there is an apparent trend and positive correlation between exposure to aggressive models and subsequent aggressive behaviour in children. When repeating a study finds similar results it could be argued to have external reliability. When studies find different results you could argue the study lacks external reliability.
Testing External Reliability Using The Test-Retest Method
Testing external reliability in psychology is possible using the test-retest method. This see’s the same participant(s) being tested over a period of time on two or more separate occasions using the exact same procedures as previous. If the results are the same or closely similar you can argue the study has external reliability. Repeating the study between cultures and finding similar results would mean the study also has external reliability.
Internal Reliability Within Psychology A Level AQA A
Internal reliability in psychology refers to how well a measure is consistent within itself. For example, taking the tape measure I mentioned earlier; It should measure the exact same distance between 10cm to 20cm as it does 20cm to 30cm.
Testing Internal Reliability Using The Split-Half Method (Psychology AQA)
The split-half method within psychology is one way of testing the internal reliability of a test. This is achieved by splitting the participants into two groups (or more depending on the different conditions being measured) and having them tested between the different conditions one at a time. If the tests show similar results from each test group between the different conditions it can be argued the test has internal reliability. If the results are vastly different between conditions then the study lacks internal reliability.
Inter-Rater Reliability Within Psychology (AQA A)Inter-rater reliability in psychology is when you have two or more “judges” who independently assess and score the same participant or subject. The scores are then compared between judges and if there is inter-rater reliability then the score will be similar (consistent) among the different judges. If the study lacks inter-rater reliability the judges will score differently and the test lacks inter-rater reliability.
Testing Inter-rater Reliability Using Spearmans Rho & Multiple Judges.
Inter-rater reliability in psychology is tested by using the statistical test Spearman’s Rho. This measures the strength of the correlation between two variables with r=1 meaning a perfect positive correlation while r=-1 would mean a strong negative correlation. You can also test the inter-rater reliability of a test by having two judges simply judge the same test independently and similar results would suggest inter-rater reliability while different results suggests a lack of inter-rater reliability.
Validity in Psychology (AQA A)
Validity in psychology is all about whether the study is measuring actually what it claims to measure. For example take Bandura’s social learning theory as an example of testing validity in psychology. Some people argued that the study lacked validity because the children who displayed the most aggression within the study were most aggressive normally outside the laboratory environment anyway. Therefore it could be argued that the study lacked validity because we were not necessarily observing how aggressive models influence children but more about how they influence children with pre-existing personality traits.
External Validity Within Psychology A Level (AQA A)
External validity in psychology refers to how much of the study can be generalised beyond the laboratory setting to the wider population or other groups. For example you can argue that social learning theory can be applied beyond the laboratory setting as much of the behaviours observed by the male model could be argued to occur in the real world (witnessing aggressive acts, this being reinforced somehow). But a study may also lack external validity if it involves only children as it may not apply to adults, or only if the children are all from a particular background or ethnicity or the fact that the study was a laboratory study which was artificial and unrepresentative of the real world. Therefore external validity is all about how you can apply the findings in the study to a wider demographic based on how it was conducted.
Internal Validity Within Psychology A Level (AQA A)
Internal validity in psychology is all about making sure the study or test measures what it is suppose to measure. For example a study may lack internal validity if for any reason you can argue that the results are not caused by the Independent variable. This can be argued to happen through confounding variables the researcher has not factored in that can affect the findings. These can be: personality factors, age differences between different groups, gender, socialisation, the setting, differences in group size or even extraneous variables that the researcher is unaware of (background experiences of participants). Anytime you can find a way to argue the dependent variable may be affected by anything other than the Independent variable you can state the study may lack internal validity.
Ecological Validity In Psychology A Level (AQA A)
Ecological validity in psychology refers to the extent the findings and research itself can be applied to real-world settings. This looks at how natural the setup of the experiment was and if it was truly representative of real life. For example laboratory experiments have low ecological validity due to being set in the laboratory and therefore not being indicative of studying behaviour within the natural world. For example If you wanted to measure how much people are liked dependent on whether they smile or frown, you may use pictures to represent peoples emotions. The study could be argued to have low ecological validity because its not natural and using a false environment but also false props (pictures) that do not truly represent real people. So anything that you can use to argue is not indicative of “natural” real world settings could be argued to have low ecological validity. If the study is conducted using a natural experiment within natural settings outside the laboratory you can argue it has high ecological validity.
Population Validity In Psychology A Level (AQA A)
Population validity in psychology is all about how the findings of the study can be generalised beyond the current participants and to other populations of people. For example a study involving only English people may lack or have low population validity because we cannot say for certain the findings will generalise to other groups of people (e.g. people in other countries for example). If the study involves a wide variety of people from different population groups the study can be argued to have high population validity.
Content Validity In Psychology
Content validity refers to the extent to which an assessment procedure gathers all the detailed relevant information required. For example diagnostic interviews for depression may be argued to have content validity provided they gather all the detailed information required covering all the symptoms of depression.
In any form of assessment you can argue it has criterion validity when anyone “assessed” with a disorder for example falls into behaviours or predictable patterns that would differ from someone who is not diagnosed or assessed similarly. For example someone assessed with depression must differ in predictable ways to someone not diagnosed with the disorder.
Construct validity refers to the extent in which a hypothesis is supported by evidence.
Some studies and their findings are a “child of their times” and lack temporal validity – What this means is the social attitudes and mindset from participants were limited to that time period and would unlikely generate the same findings if conducted now. An example of this is asking people about their opinions on same-sex marriage 80 years ago; Many would no doubt be against this and the results would likely reflect this. However if you conduct the same study now the findings would likely be much more positive and therefore concluding from a study conducted 80 years ago that the majority of people in the UK are against same sex marriage would lack temporal validity.
The shift from a collectivist culture to individualistic culture may also bring about different results. After the second world war people were use to working together collectively to rebuild their respective countries so studies like Milgrams shock experiment may find a high proportion of people obeying the researcher as they are use to obeying authority during this time period. If conducted again now you could argue we are more individualistic than 65 years ago and the results may be different. If you can argue this within your evaluation points you can highlight the study and findings lack temporal validity.
How to cite this article:
Saj Devshi – Loopa Psychology – http://www.loopa.co.uk/internal-external-reliability-and-validity-in-psychology-aqa-a-explained-easily/