Lecture One - Psychometrics Flashcards
Introduction to Course
In general, what does individual differences psychology measure?
Personality, Intelligence
Dispositional attributes: abilities, personality, non-intellective traits, interests etc.
How does Darwinian Theory relate to Individual Differences?
The theory is completely reliant on individual differences. Darwin’s theory proposes that we have random genetic mutations that either increase or decrease our chances of survival as a species. Those traits within a species that help it fit its environment are selected and passed on heritably through DNA. Therefore the differences between individuals are essential.
What are the two different approaches in scientific psychology?
Experimental = Looking at psychology through experiment
Correlational = Based on the relationship between different variables and figuring out how they’re related. (Individual differences is focused on this)
What is a trait? What are the key assumptions we make about them?
A trait is a distinguishing quality or characteristic that belongs to an individual.
We typically assume they are;
- Stable across time
- Stable across situations
- They are ‘inside us’
- They influence our behaviour
What is the difference between Nomothetic and Idiographic?
Who introduced them to Psychology?
Nomothetic approach believes there are a finite set of traits that can be used to describe personality broadly (Individual Differences uses this). They try to find the traits that occur across individuals.
The idiographic approach focuses on the individual - they all have an individual structure and that is what we should focus on. They study one individual at a time.
Introduced to Psychology by Allport (in 1937)
- What is psychometrics?
- How does psychometrics work? What is involved?
- Is the measurement of individual differences.
- Using psychometric tests that attempt to measure traits.
This involves test construction, validation, assessment of measurement error and bias.
What is the difference between measuring in physical sciences and in psychometrics?
Measurement in science requires precise measurement
Scientific measurement assumes that any attribute has a quantitative structure (time-seconds, length-CM etc.)
Scientific measurement requires units of measurement, quantitative structure and ratio scales to be precise. It is about finding out how phenomena relate to each other in terms of ratios.
Psychological measures do not have a true 0 and therefore don’t have ratio scales and therefore isn’t like measurement in the physical sciences. Traits are defined by the tests that measure them and therefore the quantitative structure is ignored or assumed.
Psychometrics is still very useful and a powerful tool but we must be cognisant of its weaknesses.
On which scale do psychometrics measure?
Nominal, Ordinal or Interval scales - cannot be ratio as there is no true zero
What does a psychometric test need for it to be a good psychometric test?
- It should measure one trait - unidimensionality
- It should be accurate and can be repeated with the same result, little influence of random errors - reliability
- Inferences about the meaning of a score should be correct - validity
- Should have discriminatory power, not everyone should get the same score - a large scatter of scores
- Tests should be representative of the population that the test is designed for - standardisation
What is the mean?
What is it sensitive to?
How do you calculate mean?
The mean summarises all of the observations in a data set.
It is the arithmetic average of all the observations. It is therefore sensitive to extreme observations.
Add up all the scores and then divide by the number of people. (R code in lecture slides)
What is a Deviation score?
What is the sum of a deviation score?
Deviation Score:
Once you calculate a groups mean you can then calculate how far away a subject is away from that mean.
When you calculate all these peoples deviation scores you can summarise them
Problem: If you add up everyone’s deviations scores the answer (sum) will be 0.
This is because you are subtracting the mean from every score and the mean is right in the middle.
What is the solution to the issue of the sum of Deviation Scores?
‘Sum of Squared Deviation Scores’
If you add up everyone’s deviations scores the answer (sum) will be 0.
This is because you are subtracting the mean from every score and the mean is right in the middle.
Therefore you need to subtract the individual scores from the mean, then square the individual scores and then add the squared scores so you get the ‘Sum on Squared Deviation Scores’
This summarises how dispersed the scores are.
What is Variance?
If you average the ‘Sum of Squared Deviations’ you get a measure of dispersion and that is called a variance.
What is Standard Deviation?
The Standard Deviation of a set of observations is the square root of the Variance.
What is the process you must go through to get a Standard Deviation?
- Work out the Mean of a group (average the numbers)
- Then for each number, subtract the Mean and square the result
- Then work out the mean of those squared differences
- Take the square root of that (the variance) and you have your Standard Deviation
