Structural Equation Modeling is a multivariate technique popular in social sciences. 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.

learn more… | top users | synonyms

1
vote
0answers
13 views

R, lavaan package, latent change model: Definition of change parameter?

I'm currently struggling with the latent change model that I try to realize in R with the lavaan package. R can't estimate all standard errors, which makes me believe that the model is not fully ...
1
vote
1answer
28 views

Ill-labeled survey repairable by analysis?

Suppose a survey has five items with a six-point Likert scale. It has been rolled out with ill-formed labels: *the (2) was labelled by "somewhat disagree" instead of "disagree" - and vice versa the ...
0
votes
1answer
27 views

Structural Equation Models - building with a large sample, and testing with a small one?

I'm planning a study using Structural Equation Modelling to test different accounts of language learning. I would like to study a group of children with language difficulties. However, they are very ...
0
votes
0answers
10 views

gsem goodness of fit

does anyone have any advise how to obtain goodness of fit information after a gsem command (for a logit path model) in Stata verion 13? Or would you advise to use BIC and AIC instead? Thank you for ...
0
votes
1answer
37 views

Identifying Instrumental Variables to use in tsls function in sem R package

I want to reproduce the Example from Introduction to Econometrics by G. S. Maddala in R using tsls function from ...
1
vote
0answers
46 views

Is it possible to fit a Graded Response Model in Stata?

I've been reading about Item Response Theory during the past few weeks and I'd like to use it to examine how my scales are functioning. The response categories are ordinal. If I understood well, the ...
1
vote
1answer
31 views

How to cure Poor Fit Indices for CFA?

I am facing problem in getting the good fit indices. I have only 163 respondents with a total of 4 variables and 70 items (A: 5 components, 3-4 items each; B: 3 components, 8 items each; C: 4 ...
0
votes
0answers
18 views

Mediation analysis with two waves of data in AMOS

Does anyone know the best way of carrying out a mediation analysis when the IV, mediator and DV have all been measured at two time points? I'm wanting to do this using SEM in AMOs.
0
votes
0answers
12 views

Can specific residuals be compared in multigroup analysis using Lavaan in R?

Is there a way to statistically compare r-squared across 2 groups using nested models in multigroup analysis? I know how to use lavaan to test various other parameters across groups (e.g. regression ...
3
votes
1answer
80 views

Structural Equation Models (SEMs) versus Bayesian Networks (BNs)

The terminology here is a mess. "Structural equation" is about as vague as "architectural bridge" and "Bayesian network" is not intrinsically Bayesian. Even better, God-of-causality Judea Pearl says ...
0
votes
1answer
15 views

In CFA model, why observed measures operate as dependent variables?

I am curious that why in CFA model, observed measures operate as dependent variables. In my understanding, observed measures served as data point that estimating other parameters, such as latent ...
1
vote
0answers
32 views

Covariates in Path model

I am building a path model which links different psychological constructs. This is the replication of a model I have previously tested in another sample. The model fits the data well. When however I ...
0
votes
0answers
30 views

Why does this autocorrelation formula hold?

I have been unable to understand the highlighted subsection of David Kenny's Correlation and Causality, which I downloaded from Kenny's website.
1
vote
1answer
41 views

Why is error variance important in CFA?

I am reading the book related to SEM (Byrne, 1998) and it is stated that regression of the observed variables on the factor, and the variances of both the errors of measurement and the factor, as well ...
0
votes
0answers
25 views

Determining characteristics of sampling sets for EFA/CFA/SEM

Dividing sample data into several sets seems to be a common approach in statistics. This is especially evident in predictive modeling, where samples are traditionally divided into two sets, usually ...
3
votes
3answers
148 views

Reporting the use of different likert-scales in survey for structural equation modelling?

Suppose you are doing a survey and from the data you want to build a structural equation model. The research model consists of 5 constructs and each of them has between 2 and 5 items. The survey uses ...
0
votes
0answers
34 views

SEM: sample covariance can not be inverted

I am trying to fit a structural equations model, which looks like this: {model<- 'ambition=~NA*imprich+ipshabt+ipsuces+iprspot+0*ipeqopt+0*iphlppl altruism = ...
0
votes
0answers
25 views

Using transformed variables in SEM (using Amos)

Typically when I'm using Amos graphics I access my SPSS datafile and use individual items from a survey (observed variables) to make up my constructs (latent variables). In my current study, the ...
1
vote
1answer
103 views

How to create item parcels in SPSS Amos?

I am trying to run a SEM to see the causal relationship among latent variables. I ran the CFA and SEM using Amos, but I realized that the model is very complicated since one factor, which I originally ...
0
votes
0answers
23 views

having probllem with fit indices in SEM

I ve developed CFA to validate my model. fit indices show good fit. since every variable contain multiple questions. I impute to make composite variable for regression and checked the final ...
0
votes
0answers
15 views

Poisson regression on partially missing covariates

I'm reading an article where the author fits a Poisson regression on some data. Some of the covariates (but not the outcome) are partially missing. The author writes that he: conducted the ...
0
votes
0answers
16 views

Market Segmentation Test for SEM

I am working with SEM and would like to know how I could conduct a market segmentation test. I am trying the same model on two different samples and would like to test whether it's better to have them ...
2
votes
2answers
68 views

What test should I perform on a structural equation model?

This model is a replication of an earlier study. After factor analysis in SPSS, I need to test my hypothesis (as you can see there are 7) all should give significant positive relation with each ...
1
vote
1answer
42 views

SEM & Identification

Consider the following demand and supply model: Demand equation: $q=a_1p+a_2y+e^d$ Supply equation: $q=b_1+p+e^s$ The supply equation is identified because $y$ can be used as an ...
3
votes
1answer
39 views

Calculating and comparing weighted means with changing number of daily values

I have daily data where the number of values changes with each day. I would like to calculate the weighted mean of this data, where days with more values are weighted higher than days with less. ...
0
votes
0answers
285 views

Calculating average variance extracted (AVE) in R for checking discriminant validity (Fornell-Larcker criterion)

We are using lavaan in R to calculate CFAs (confirmatory factor analyses) and SEMs (structural equation models). We now want to ...
0
votes
2answers
67 views

factor analysis problem in structural equation model

I am using an Extended Technology Acceptance Model and have adopted an instrument from a research paper. After performing factor analysis in SPSS, most questions related to different constructs ...
0
votes
0answers
36 views

MIMIC fail to converge

I am trying to fit a MIMIC model measuring poverty in a household sample. The model looks like this: (HUMCAP -> education job) (HOUSINGQUAL -> floor wall dwelling tv refrigerator radio electricity ...
1
vote
0answers
39 views

Differences between Multilevel-SEM and PLS-SEM

I would be grateful if you someone could explain to a layperson the differences between 1a) multilevel structural equation modeling and partial least squares structural equation modeling. 1b)What ...
1
vote
1answer
91 views

Difference in AIC and BIC values between sem and lavaan packages in R

I ran the same SEM model in sem and lavaan. I got the same parameters and - generally - very close test values, with the exception of AIC and BIC which were immensely different between the two ...
3
votes
0answers
39 views

How can IRT-Models be understood in GLM/ SEM Framework? (Predict Learning with added Paradata-Covariates)

I'll be working with data from an intelligent tutor system similar to one studied in the KDD-Cup 2010 on student performance prediction and plan to use IRT models to infer item and ability parameters. ...
2
votes
1answer
77 views

Observed variables in R's lavaan

I am running a SEM on lavaan in R. The model consists of latent variables [N1 to N3], two dummy choice variables [du1 and du2] and observed satisfaction variable [SV] (beside the observed variables). ...
0
votes
0answers
11 views

Performing CFA & Structural Models

I have developed a conceptual framework based on literature. The conceptual framework comprises antecedents and consequences. The antecedents factor 1, factor 2 and factor 3 affects factor 1 and ...
2
votes
0answers
22 views

Is it possible to combine bayesian SEM with PLS SEM?

I have already read some books about both two structural equation models. It seems both SEMs are suitable to the situation with small observations and large variables. I assume to use combine both ...
0
votes
0answers
25 views

SEM for Non-Normal Ordinal Data

In short, I conducted a satisfaction survey in which surveyees are required to answer on a satisfaction scale from 1 to 7: 285 observations, 37 satisfaction variables. Here is an example of what the ...
0
votes
0answers
22 views

Estimating the parameters of a model, which method should I use?

I am trying to estimate a system of macroeconomic (simultaneous) equations, and I've learned about the 'existence' of various methods including Structural Equation Models, Simultaneous Equations ...
0
votes
0answers
43 views

How to test measurement invariance with PLS?

I have two questions about structural equations modeling: To perform comparisons of model estimates for different groups (3 clusters), do I always have to analize measurement invariance first? If ...
0
votes
1answer
55 views

How do I convert unstandardized RMR estimates into standardized RMR estimates?

I am having difficulty turning unstandardized RMR estimates from AMOS software output in a SEM into standardized estimates to enable suitable interpretation. Can someone give me advice on where to ...
1
vote
1answer
47 views

Why did fixing the covariance value cause AMOS to mis-estimate variance?

I have 100 data points on two variables, a and b. The correlation between the two is .3 and the SD is 1. When I run the ...
0
votes
0answers
9 views

What's the minimum number of measured variables required when fitting a one latent variable model?

I am envisaging a situation in which I have designed a small questionnaire, and want to assess whether the measured variables are measuring the same thing. I'm not really interested in practical tips ...
0
votes
0answers
29 views

Differentiating strength of relationships between categorical and continuous variables in SEM?

I'm currently planning an SEM study designed to investigate attitudes toward energy technologies. My past research has suggested that the persuasive effect of positive/negative information is minimal, ...
2
votes
0answers
221 views

Optimal sampling strategy for EFA, CFA and SEM

I'm wondering what should be the optimal sampling strategy for my dissertation research. I have four data sources (two open source software projects meta-repositories and two global startup ...
0
votes
0answers
16 views

parameter estimation in bivariate autoregressive sem with individual heterogeneity - troublesome likelihood surface

I am working with models for panel data of individuals, wherein: my latent variables Yt= $\beta$Yt-1+Ai+Q Individuals level of the process A~N(l,h) Innovation of the latent Q~N(0,q) observed variables ...
3
votes
1answer
288 views

ML vs WLSMV: which is better for categorical data and why?

I was wondering which is a better estimator to use for categorical data: ML or WLSMV. I saw on a discussion on the Mplus website that they recommend WLSMV for categorical data but didn't explain why. ...
0
votes
0answers
63 views

power/sample size for a complex set of equations

I run an experiment in which several teams perform several tasks each (not all teams perform all tasks, but in general, I can make sure that enough teams perform any given task, and that each team ...
1
vote
0answers
88 views

Structural equation model: Bootstrap, Bayesian statistic or rescaling binary data to deal with small sample size?

I’m trying to solve a SEM. Mi model includes 4 latent variables and 11 observed variables. My data is binary. Question 1: I just have 47 observations and in the literature a sample size over 150-200 ...
1
vote
0answers
28 views

Any thoughts in PLS as an alternative to SEM if you have a small sample size?

I have put a question on here before re SEM and small sample sizes. I have 3 factors each with 3 manifest variables each. I only have 90 records. I want to find out the amount of covariance between ...
1
vote
1answer
237 views

Transforming a relationship from quadratic to linear

Assume that I'm running a path analysis and I have discovered that certain relationships, empirically, are quadratic rather than linear. In order to model the data as such, I want to transform the ...
0
votes
0answers
54 views

Sample size for SEM. What are the alternatives

I have 3 latent variables with 3 measures each. I understand the conventional sample size would be 15 participants per measure (i.e. 9 x 15 = 135). I have 85 participants and don't think I will have ...
4
votes
2answers
153 views

A question on SEM and path analysis

For my PhD dissertation, I am offering a conceptual model on teacher professional development. My model has one mediator. I wonder if I can test the mediation through Baron and Kenny's procedure or ...