Lecture Notes P POST Midterm - Factor Analytic Research on...

Lecture Notes P POST Midterm
Download Document
Showing pages : 1 - 3 of 66
This preview has blurred sections. Sign up to view the full version! View Full Document
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Factor Analytic Research on Traits • Why factor analyze? Allport lacked a content theory (what we should be studying). However, he did produce a catalogue of trait-like words in the English language (~18,000) and then eliminates similes and words not used. There were ~3000 trait words. This gives rise to 2 problems • There are too many words. Plus, our knowledge accumulates over years, so how will we accumulate if there are still so many? • How do we know which are important and will allow us a lot of prediction? • What is factor analysis? It’s a statistical technique with a intercorrelation with a set of measures. • (Below) There are clusters of intelligence measures that are highly correlated and there are some traits that are highly correlated. • There is something redundant about this because really there are only two things happening here. • It would combine the first 4 and combine into 1 then combine the last 3 and combine into 1. It finds the most efficient way to combine the (highly correlated) variants (see statement above). This makes a factor score which is a weighted sum of items (take the average). In practice, we give an approximate factor score. Its an unweighted sum (just add them up) of “highly loading” items (the items that are most strongly related to the underlying factor). • Start with correlation matrix • Typical features of FA o Uncorrelated “independent” “orthogonal” items This is what we ask of the computer. o Emerge in a sequence They emerge in an order from big (related to the largest number of input variables) to small. So it tells you which are strongly correlated. o Name/meaning/conceptual definition unclear It just says these things go together and you have to figure out why the things are correlated together (e.g. healthy altruism). The names given are very much up for debate so its hard to assign objective names to the groups. • Reasons to be wary of FA o Its seductive because it appears to replace the usual subjective bickering between psychologists with a precise, definitive, mathematical answer. o “Carving nature at its joints”? It makes you think that the computer carves nature at its joints for you. This is not so! It is not to say that what you generate is what trying underlies personality. How many factors? • Its arbitrary. The computer does not say “4’ to you. It requires human judgment so you are balancing parsimony (just take a very) vs. comprehensiveness. How to name the factors? • This is a subjective/creative act when the psychologist names the factors. GIGO • Garbage-In, Garbage-Out • The results are only as good as the quality of data you put in. ...
View Full Document

Create a FREE account now to get started. Log In

The email address you entered is not valid. The email address you provided is already in use.
Your username must be at least 5 characters. Your username must consist of only alphanumeric characters. Your username must contain at least one letter. Your username contains inappropriate language. Another user has already claimed this username.
Your password must be at least 6 characters in length.
{[ $select.selected.label ]} Please select a valid school.
By creating an account you agree to our Privacy Policy, Terms of Use, and Honor Code.
Create my FREE account Processing...
Sign Up with Facebook

We will never post anything without your permission.

Already on Course Hero? Log In