Questions tagged [sem]

Structural Equation Modeling is a multivariate technique. It is based on formulating a set of linear relations between variables, some of which may be latent, and estimating the whole system, typically by analyzing the covariance matrix of the observed variables.

Filter by
Sorted by
Tagged with
1
vote
0answers
25 views

What methods can I use to quantify the impact of multiple variables on a response variable?

Suppose I have observations for both the predictors ${x_1,x_2,...,x_p}$ and the response $y$, what I am interested in is the controls of these predictors on the response $y$. There are several cases ...
0
votes
1answer
11 views

How to conduct a post hoc power analysis for a just-identified path model

I am doing secondary data analysis on an existing dataset. No power analysis was conducted a priori. My sample size is low, n = 122. Thus, I plan to do a post hoc power analysis. My model is just-...
0
votes
0answers
12 views

SEM model in Lavaan - information matrix could not be inverted

Can anyone tell me what is wrong with my model? R is saying that it may not be identified! This is what im running: ...
0
votes
0answers
14 views

Can proportional binomial regression be used in piecewise structural equation modelling (piecewiseSEM)?

I'm using generalized linear mixed models from the lme4 package to do structural equation modelling using piecewiseSEM. It works ...
1
vote
0answers
25 views

What distribution to use for left-skewed data in generalized linear mixed models (for use with structural equation modelling)?

I'm trying to run a GLMM with a response variable that is left-skewed. Eventually this model will form part of a piecewise structural equation model (using piecewiseSEM). I have data from 480 plots, ...
0
votes
0answers
21 views

SEM - How to specify an interaction of latent variable and manifest variable?

It is a known fact that if we want to test if there is an interaction effect between latent variables, we have to specify interaction terms with manifest variables those latent variables are composed ...
0
votes
0answers
7 views

How to interpret lavaan output from clustering and categorical endogenous variable?

I ran a structural equation model with serial mediation and the outcome variable is a binary observed variable (coded 0 or 1 for "no" and "yes"). The rest of the variables are ...
0
votes
0answers
19 views

How to obtain a latent curve model with autoregressive disturbances from an autoregressive latent trajectory with exogenous covariates

Hamaker (2005) explains that we can write a constrained autoregressive latent trajectory (ALT) model (Bollen & Curran 2004) as a latent curve model (LCM) with autoregressive disturbances as long ...
0
votes
0answers
5 views

CLPM with count data in lavaan

I am hoping to run a crossed-lagged model with two timepoints and two variables. One variable is continuous, and the other is symptom count (zero-inflated). What would be the best approach for dealing ...
0
votes
0answers
14 views

How to calculate convergent and discriminant validity in SEM?

Using the average variance extract (AVE), the method is explained easily by doing some calculations on the factor loadings. I only found such procedure on CFA models, where we don't have an endogenous ...
1
vote
0answers
17 views

Is it possible to fit a piecewiseSEM model with datasets of different sizes?

I'm trying to compare different piecewiseSEM models in R evaluating the factors that directly and indirectly influence the difference in median sizes of cohorts from one year to the next. My terminal ...
0
votes
0answers
28 views

Unrefined scores for confirmatory factor analysis

I have read the paper entitled "Understanding and Using Factor Scores: Considerations for the Applied Researcher" by C. Distefano, M. Zhu, Diana Mîndrilă (2009), Practical Assessment, ...
0
votes
0answers
16 views

Non positive covariance matrix of latent variables when constraining residual covariances to be equal across time

I am trying to build a bivariate latent growth curve model with structured residuals using lavaan. When I try to constrain residual covariances to be equal at each time point, however, I get a warning ...
0
votes
0answers
34 views

Latent variable interactions/moderation using OpenMx

I am currently trying to create a latent interaction regression model in OpenMx where I have four latent variables (C, H, T, S) that predicts a fifth latent variable (A). In addition, interaction ...
1
vote
0answers
14 views

How to find degree of freedom in SEM-CFA model? [Inquiring Clarification]

I found a link that explains it. http://reifman-sem.blogspot.com/2007/04/some-of-students-wanted-review-of.html It is $$Degree \;of\;Freedom (DF) = Total \;Elements - Total \;Free \;Parameters$$ [...
1
vote
0answers
24 views

How to Use Package “simsem” for Sample Size Determination in SEM: Validation Project? [Step-by-step Inquiry] [closed]

To provide context: My model is a latent model; 3 Latent Variables, 3 Indicators for each of two Latent Variables and the last Latent Variable with 4 indicators. Hence, Total of 3 Latent and 10 ...
0
votes
0answers
10 views

Regress statistical composite variable and against latent variable in an SEM using lavaan.mi

I am trying to estimate the association between a statistical composite variable, developed following the methods described here, and a latent variable using lavaan.mi (estimated from a series of ...
0
votes
1answer
59 views

How to interpret the negative variances

I had used the already published Likert scale for the survey. And the responses to the survey from 98 participants were collected. The survey likert scale was from <...
0
votes
0answers
16 views

Multigroup SEM analysis - parameters test

I'm working with SEM in social-mobility framework as I want to use different survey-questions to define the cognitive ability as a latent factor and in turn use it as an exogenous regression path. The ...
0
votes
0answers
8 views

Measurement invariance/differential item functioning - Is it okay to have 2 indicators per factor?

I'm trying to run a moderated nonlinear factor analysis (MNLFA) but my factor has only two items assessing it, so I'm wondering if it's okay to use this approach. https://psycnet.apa.org/record/2016-...
2
votes
0answers
81 views

Calculate R² from regression estimates (beta coefficients, variances of variables etc.)

I need a generalisable way to to calculate $R^2$ from given regression estimates (regression coefficients, variances of variables, covariances between variables, residual variance of the response ...
0
votes
0answers
9 views

SEM: Moderation / interaction

What are the steps to prove interaction between Latent variable A that influences Latent variable B, subject to the moderation/interaction of Age groups? is it to test for invariance assuming b (...
0
votes
0answers
11 views

Dynamic Panel Data GMM with Endogenous X

Assume the regression model is $ y_{it} \ = \ \alpha \ + \ \rho y_{i,t-1} \ + \ \beta_1 x_{it} \ + \ \beta_2 w_{it} \ + \ e_{it} $ Where $x_{it}$ is strictly exogenous and $w_{it}$ is endogenous, and ...
2
votes
1answer
29 views

SEM: can GLS estimation be used under severe nonnormality?

In Randall E. Schumacher's and Richard E. Lomax's book "Beginer guide to SEM", the writers keep saying that if non-normality is sensed, where you can't use ML, you can go for GLS/ADF/WLS. ...
4
votes
1answer
43 views

In SEM, what is the difference between ADF and WLS estimation methods?

In some books both WLS and ADF are considered different methods. In other books, they acknowledge that ADF = WLS, so they are used interchangeably throughout the book. How solve this confusion?
4
votes
1answer
33 views

Structural Equation Model no solution found

I'm trying to use the lavaan package in R for a SEM. I'm using ~30 variables. I scaled the variables with the scale() function. This is how my model looks like: ...
2
votes
1answer
20 views

Comparing alternative structural equation models x -> m -> y VS m->x->y

I'm new to SEM and I'm not sure if my way of testing mediations sounds right... ...
0
votes
1answer
27 views

Missing data in Structural Equation Modeling

I am new to the topic of SEM, so my question may seem a bit naive. I have about 60 observed variables that will be grouped in some latent variables to explain one outcome measure. One of the variables ...
1
vote
0answers
8 views

Multi-level Latent Class Growth Curve Model of a double-repeated measures data structure

I have an interesting data structure I'm trying to analyze. I've decided that the best route is most likely using some type of LCGA, but am having trouble conceptualizing the model. Unfortunately, I'm ...
0
votes
1answer
25 views

Factor loading calculated by factor_analyzer smaller than -1

I used factor_analyzer Python package for CFA, but matrix of factor loadings that I got contains value smaller than -1. Is a bug in the package, or maybe I ...
6
votes
1answer
42 views

Difference between manifest and observed variables

I try to code SEM model using the semopy package (https://arxiv.org/pdf/1905.09376v1.pdf), but there are 3 types of variables there - except latent, there are ...
3
votes
1answer
63 views

Treatment effect and Causal Structural Models (Pearl-type)

How can I check whether the condition of no confounding is satisfied in a Structural Causal Model? In particular is there a way to do this by applying the do-calculus? The condition of no confounding ...
1
vote
0answers
24 views

Positive definite matrix and categorical variables - SEM - (lavaan) [closed]

I'm trying to estimate the following SEM model with the lavaan package: ...
0
votes
0answers
34 views

moderation (latent variable * gender) analyses using lavaan

Hi I'm following https://stackoverflow.com/questions/24399353/r-lavaan-coding-latent-variable-interactions to run moderation analyses. My model is a bit different from the example in the link so I ...
0
votes
1answer
19 views

Is a reversed path analysis a nested model? (SEM)

I'm trying to compare whether a forward/direct path analysis is a better fit to the same data than a reversed model. I'm using the SEM function of the ...
0
votes
0answers
19 views

Can SEMs with latent variables be consistently predicted using a Bayesian approach with a MCMC (pymc3)?

Alright this question is more for validation and a discussion around this topic to see if im missing something... Ive been thinking about SEMs and how you can implement them in the simplest way ...
0
votes
1answer
37 views

Constraining correlations between two latent variables

I am running a multiple mediation model using SEM, where my 2 latent variables are my mediators, and my main predictor and my outcome are both exogenous variables. I have specified that my LVs are ...
0
votes
1answer
55 views

Mediation analysis (SEM): 2 mediators, insignificant direct and indirect effect and significant total effect

I ran 2 mediation models each with two mediators in R (lavaan) and am looking for help with interpretation. Model 1: Paths a1 and a2 = significant Paths b1 and b2 = not significant Path c' = ...
0
votes
0answers
15 views

SEM: Difference between standardized and non-standardized p-values?

I'm currently working on regressions of one latent factor on another latent factor with ``lavaan'`. More specifically, I want to examine if depression predicts procrastination behavior. However, I did ...
2
votes
1answer
34 views

Should allowing multiple DVs to covary in SEM influence beta coefficients?

I am running a replication study to test a model with 8 IVs and 3 DVs (all variables are continuous). In the initial study, I had a moderate sample (≈ 200), and thus relied upon multivariate multiple ...
1
vote
1answer
59 views

What is the difference between Factor Analysis and Path Analysis in Structural Equation Modelling?

I am still a beginner in Structural Equation Modelling and I don't have a solid background in Statistics, so I apologize if my question is a bit naïve. What I traditionally do when creating a ...
1
vote
0answers
37 views

How should the Bollen-Stine bootstrap $p$-value be calculated?

Writing about an original test statistic $T_{ML}$, Nevitt and Hancock (2001) write on p356 that The bootstrap-adjusted $p$ value is calculated as the proportion of bootstrap model test statistics ...
0
votes
0answers
26 views

SEM for within subject data

My questionnaire is novle,100 listeners listen to 10 sound files and then rate each file on a 12-items likert type scale. So I have 100 participants but 1000 cases. Now my theory is 3 factors can be ...
2
votes
0answers
19 views

Alternatives to SEM/Path analysis with smaller sample size

I have a study design with one between groups independent variable with five levels (different behaviours). I then have a number of latent dependent variables, measured by self-report questionnaires. ...
1
vote
1answer
21 views

In Structural Equation Modelling, what is the difference between Exploratory and Confirmatory Factor Analysis?

I am trying to understand the difference between Exploratory and Confirmatory Factory Analysis. I have usually only done CFAs because I would test existing models. Though I know EFA usually precedes ...
0
votes
0answers
18 views

Which analysis method to choose: two between club variations need to be explained using many other variables? Scheme provided

I am a master student and need help in approaching the following problem. I have a large dataset about athletes from different professional level clubs (athlethes nested in clubs), their performance (...
1
vote
1answer
24 views

Sem with groups

My purpose is to test the same SEM model on two independent groups. However, some of the variables that participate in the model have significantly different values between the two groups (e.g. age is ...
2
votes
1answer
92 views

Multilevel SEM with three or more levels in OpenMx

I am trying to figure out how to conduct analyses using CFA and SEM models with three or more levels. Pritikin et al. (2017) investigates how to conduct a three-level and up to five-level SEM using ...
0
votes
0answers
6 views

How to group and test interaction patterns of variables

IV: 1st Gender of pichtures (male/ female) 2nd Assumed posture (open/closed) DV: couple a variables (power, emapthy etc.) rated on a likert scale In a first pilot study I found evidence for ...
0
votes
0answers
56 views

Creating a composite variable with data that is missing by design

I have a dataset of 20 variables and over 2,000 observations. These variables are paired. A respondent is first asked if they endorse a goal. If they endorse that goal, a follow-up question asks how ...

1
2 3 4 5
16