Dimensions and types
19 important questions on Dimensions and types
Which types of latent variables are there?
- Continuous latent variables; like intelligence
- discrete latent variables; disorders
For both discrete- and continuous latent variables you can measure discrete and continuous observable variables. Which models do you use for certain combinations?
- Discrete OV, cont. LV; item response models
- discrete OV, discrete. LV; latent class analysis
- cont. OV, LV; factor models
- cont. OV, discrete LV; mixture models
How does the factor analysis work?
- Certain (scores on) observed variables are correlated with each other.
- these correlations are explained by certain latent variables (the factors)
- Higher grades + faster learning
- Never study anything twice
- 100% sure, 100% understanding
Describe the formula of a factor analysis.
X1=a11F1+a12F2+E1- X1 is an observed variable
- F1 and F2 are the latent variables
- a11 and a12 are numbers that determine the weight of the input of a latent variable, so it could be that F1 contributes more than F2.
- E is error
How does factor analysis influence theories?
- Our theories are adapted to our statistical toolset
- why come up with theories that can't be tested.
What is a major con of the factor analysis, and the way it influences theories?
- If you only think in terms of factor analysis you can only think in terms of gradual dimensional continuous factors
- while a lot of times in psychology traits are portrayed as a discrete, types, variable
What is often a problem with defining a construct in terms of discrete or continuity?
- Example of dyslexia
- Is dyslexia a separated category or the 5% worst readers in the normal distribution?
How can this problem be portrayed?
What are reasons to portray individual differences as types?
- Discrete causes have discrete effects;
- brain damage, genetic differences, environmental differences (type of schooling, bullying)
- these cause people to be different.
- or lack of discrete causes can be compensated by more complex system causing discrete effects.
What do we mean by phase transitions?
a phase transition is when a system changes from one stable state to another.
Describe phase transitions in terms of states of water.
- Water can be gas, liquid, or solid.
- you can change temperature very slowly, but at some point a minor change can have major effects.
What are some criteria for phase transitions?
- Bimodality
- hysteresis
Which criteria pf phase transitions is described; under the same circumstances, two or more stable states are possible (business card example).
How do we call the dependence of the state of a system on it's history? So ice melts at a different temp than it freezes.
How do we know humans are complex systems?
What is the principle of mixture models?
- The distribution of data is not described by one distribution
- the data is described by multiple distributions, a weighted sum f distributions (average)
How in this picture is the observable and latent variable shown?
- The weighted distribution (black) fits the data best.
- the data is measured continuous (OV)
- the different colored lines indicate the categories of the latent variable.
What is the principle of a latent class model?
- Propose the amount of expected classes of your data to the computer
- program returns the proportion of the classes and what the conditional probability of a correct item is given an individual is in a class.
- using this information compute expected values for your data and perform a chi squared test.
- if the test is not significant, the class model is accepted.
How can you combine distinctive and continuous views?
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding