Questions tagged [psychometrics]
Psychometrics has evolved as a subfield of psychology to become the science of measurement of unobservable individual characteristics.
358 questions
3 votes
3 answers
290 views
class imbalance in ML based IRT models
I’m testing an IRT model on large datasets, but running into class imbalance. most responses fall into the “correct” category. It is quite normal in IRT models I suppose as assessment data will be ...
- 47
0 votes
0 answers
44 views
Optimal composite score weights using results from a Delphi process
I've read a few questions and answers (specifically this one and this one) as well as a paper on "subjective weighting" and I can't for the life of me figure out what to do here. We've ...
- 3,979
1 vote
1 answer
67 views
POMP score across different versions of a questionnaire
I am planning to look at differences in empathy and systemising scores between cases and controls. The sample is aged 6-18 yrs where they completed questionnaires on empathy and systemizing traits, ...
- 11
1 vote
0 answers
42 views
Understanding terms of a composite reliability equation
I'm trying to compute composite reliability using the equation below (based on Wang & Stanley, 1970, and described here). $$ r_c = \frac{\sum_{i=1}^{k} r_i + \sum_{i=1}^{k} \sum_{j (\ne i)=1}^{k} ...
1 vote
0 answers
40 views
Is centrality stability assessment different for mlVAR models compared to standard VAR models?
My question is: Is there a meaningful statistical difference in how we should assess centrality stability and edge robustness in mlVAR models compared to standard VAR models? Specifically: Is the CS-...
- 11
11 votes
4 answers
585 views
Modelling 5-point Likert scale (very negative to very positive) in a ordinal logistic regression model? Are there other approaches?
I'm trying to model the relationship between an independent continuous variable (age) and a dependent categorical ordinal one (perception of an event) with the categories: very negative, negative, ...
- 121
0 votes
0 answers
22 views
How can you compare the result of a 2 AFC experiment and a 3 AFC?
(knowing that you have the same stimuli in both experiments) I searched in the net but could not find a satisfying response
- 1
0 votes
1 answer
89 views
Understanding misfit within correlated factors model
I'm working on a project in which I am evaluating scales. I'm running into an issue where despite fitting two CFA models separately and achieving good fit, when put together in a correlated factors ...
- 11
1 vote
2 answers
126 views
Question about Cross-Scale Exploratory Structural Equation Modeling (ESEM)
I'm working on a project in which I'm using Exploratory Structural Equation Modeling (ESEM). I usually see articles in which ESEM is used to look within a scale. Can you do ESEM in which there are not ...
- 11
1 vote
0 answers
58 views
Is there a general formula for M-up N-down psychophysical staircase convergence points?
Setting aside more modern adaptive methods for the moment, I'm trying to find a formula that takes M (the number of successive misses required to increase stimulus salience) and N (the number of ...
- 41
1 vote
2 answers
682 views
Exploratory factor analysis, non-normal data [closed]
The data I am analyzing are non-normal for nearly each observed variable. I would like to do an exploratory factor analysis (EFA) in R. KMO-test and Bartlett's test indicate that the data are eligible ...
- 185
2 votes
0 answers
71 views
Tucker's congruence coefficient on entire factor loadings
To my knowledge, Tucker's Congruence Coefficient is used to evaluate the congruence between factors across samples. For example, after performing EFA for a 2-factor model in R, using ...
- 21
6 votes
1 answer
482 views
Rationale to use McDonald's Omega as a reliability measure for an unweighted sum score
In recent literature reporting McDonald's omega as a reliability estimate is clearly favoured over reporting coefficient alpha (e.g. McNeish, 2017). One argument often given is that coefficient alpha (...
- 345
3 votes
1 answer
93 views
Why are two CFA models not equivalent if I drop some variables from my dataframe or restrict factor loadings to 0, keeping all variables? lavaan
Exploring nested vs. non-nested models, I was surprised with a strange result. The first model uses 21 variables from my dataframe and runs a CFA. The second model uses 28 variables from my dataframe, ...
- 244
0 votes
0 answers
34 views
How should I balance an experiment with two sets of conditions and not enough trials for every combination of those conditions?
I have a within-subject experiment that tests participants' memory. There are 80 trials, and in each trial, participants read 1 of 20 sentences. Then, they have to manipulate the sentence in 1 of 4 ...
- 1