Home / Summaries / Class notes - scientific and statistical reasoning and psychological testing / item-curve-factors

Pca and utility

19 important questions on Pca and utility

What do you essentially do with a factor analysis on a test?

You cluster different items into a few clusters/factors/dimensions to make the results more interpretable.

How do you perform a factor analysis?

Go to analyse, dimension reduction, factor
put all the questions into variables
extraction; select principle axis factoring for method, check scree plot (after looking at the scree plot, select fixed number of factors; amount of factors)
rotation; select direct oblimin,
options; press sorted by size, (optional;) suppress small coefficents .3, do this just know that some values will not be shown.

What does the table of total variance explained tell us?

The amount of factors spss extracted based on the eigenvalue of the factor being bigger than 1.

In a scree plot, where is the point of inflection?

The dot from where the line goes to be horizontal

How can you tell the amount of factors from a scree plot?

The amount of dots left from the point of inflection.

What do do after you've determined the amount of factors you have?

Rerun the analysis, this time indicate that you want x amount of factors.

What can you see in the pattern matrix?

You can see which questions form a cluster together
which questions belong to which factor
what the overarching theme of a factor is.

What are the advantages and disadvantages of the classical test theory?

Advantages:

Intuitive and easy to apply

It is in SPSS and it is easy to do in excel
No large sample sizes/many items needed

Disadvantages:

– Focus on the test, not on the items
– Test properties depend on the population

E.g.,: Reliability and difficulty of a test

Person properties depend on the test

I.e., sum score is higher if the test is easy and lower if the test is difficult

What do you do when using the item response theory?

You specify a measurement model or a formula, which mathematically links the item scores to the construct as a latent variable.

What are the assumptions of the item response theory?

Uni-dimensionality; you only measure one construct
local independence; the latent trait explains all item correlations
monotonicity; the higher the trait the higher the expected score.

How can you tell the item difficulty from an item characteristic curve?

The number of the x axis that the middle of the curve is on.

How can you tell the item discrimination from the item characteristic curve?

The steepness of the curve, if the curve is very steep then a small increase in laten trait will result to a big difference in chances correct.

From the results of an item response analysis, how do you know which items are weak?

When the discrimination is negative

At which point of the item characteristic curve is the most information?

If the slope of the curve is maximum

What is a scale information function?

An information function of the whole test instead of just one item.

How can you tell that items are unfair towards certain groups?

If you suspect a group having a disadvantage on an item check the item curves for both groups, if they're different there is an disadvantage

What is computerized adaptive testing

Instead of giving everyone the same items and the whole range of difficulty
computerized adaptive testing will adjust difficulty of the next item based on the answer of the last item.

How does the computer pick new questions in computerized adaptive testing

Based on the information of the last question, and estimate is calculated of the extend to which the trait is present
the next question is picked to have the highest information on that degree of trait presence

What are the advantages and disadvantages of item response theory

Advantages

Population and test independent
Focus on items

Disadvantages

Statistical complex
Needs large samples

The question on the page originate from the summary of the following study material:

scientific and statistical reasoning and psychological testing

View summary

A unique study and practice tool
Never study anything twice again
Get the grades you hope for
100% sure, 100% understanding

Remember faster, study better. Scientifically proven.

Pca and utility

19 important questions on Pca and utility

What do you essentially do with a factor analysis on a test?

How do you perform a factor analysis?

What does the table of total variance explained tell us?

In a scree plot, where is the point of inflection?

How can you tell the amount of factors from a scree plot?

What do do after you've determined the amount of factors you have?

What can you see in the pattern matrix?

What are the advantages and disadvantages of the classical test theory?

What do you do when using the item response theory?

What are the assumptions of the item response theory?

How can you tell the item difficulty from an item characteristic curve?

How can you tell the item discrimination from the item characteristic curve?

From the results of an item response analysis, how do you know which items are weak?

At which point of the item characteristic curve is the most information?

What is a scale information function?

How can you tell that items are unfair towards certain groups?

What is computerized adaptive testing

How does the computer pick new questions in computerized adaptive testing

What are the advantages and disadvantages of item response theory

Summaries related to Class notes two way anova

Class notes - scientific and statistical rea…

Cognitive Psychology

Research Methods in Psychology Evaluating a …

Psychology A Concise Introduction

An Introduction to Developmental Psychology

Abnormal Psychology

Statistics The Art and Science of Learning f…

Organizational Behavior

A conceptual introduction to psychometrics

Electrical Engineering: Concepts and Applica…

Class notes - Psychological assessment

Class notes - sexology