Is SEM Causal Modeling?

You will often hear SEM referred to as a causal modeling approach. SEM does not determine causation between two variables. This is a misnomer that is used quite often with SEM. As stated earlier, SEM uses a covariance matrix as its input, so you are essentially looking at cor- relations between variables to determine

A Confirmatory Approach to SEM

Testing a SEM model is said to take a confirmatory approach. Put another way, a conceptual model is determined a priori and then data is collected to test how well the model fits the data, thus trying to “confirm” the researcher’s hypotheses about how constructs influence one another. Joreskog (1993) outlines how SEM testing

Theory Should Lead Conceptualization for SEM Model

A problem you will see with many SEM studies is when a model is conceptualized apart from any theory. A theory should be a guide to understanding a phenomenon of interest and ulti- mately help to explain why two variables are influencing each other. Sadly, it seems like many SEM models are conceptualized first

Assumptions of SEM

With any statistical technique, assumptions are made. Here are a few of the assumptions with SEM that you need to be aware of going forward: Multivariate Normal Distribution of the Indicators—there is an assumption that the data has a normal distribution. Dependent variables need to be continuous in SEM—while the independent variables do not

Understanding Diagram Symbols in SEM Model

SEM uses diagrams to denote relationships to be tested. It is important that you understand what these diagram symbols mean because AMOS is going to make you draw out your concep- tual model. One of the frustrating aspects of SEM is that there are often multiple terms that mean the exact same thing.You can

Independent vs. Dependent Latent Variables in SEM Model

Independent variables (also called exogenous variables) are the constructs that influence another variable. Dependent variables (called endogenous variables) are constructs influenced by independent variables. Figure 1.6 Source: Thakkar, J.J. (2020). “Procedural Steps in Structural Equation Modelling”. In: Structural Equation Modelling. Studies in Systems, Decision and Control, vol 285. Springer,

How to Measure an Unobserved Construct in SEM Model

With an unobservable construct, we are trying to use indicators to measure the concept.With any unobservable analysis, you will rarely be able to say you are capturing the “true” or actual score without some degree of error. Thus, an indicator is a function of the “true” score plus error: Construct Indicator =True or Actual

Greek Notation and SEM

Many SEM programs produce their results using Greek notations. They use Greek letters to represent different constructs and relationships. AMOS does not produce output using Greek notations, but you need to be familiar with them so you can understand the results from other SEM programs. What Do All These Greek Symbols Mean? Latent Constructs

Data Screening for SEM Model

The first step before analyzing your SEM model is to examine your data to make sure there are no errors, outliers, or respondent misconduct. We also need to assess if you have any missing data. Once your data has been keyed into a data software program like Excel, SAS, or SPSS, the first thing

Screening for Impermissible Values in the Data

There are times when respondents simply key in a value wrong or list an invalid response to an inquiry.To test if an answer is outside of an acceptable range, you need to go to your SPSS file, select the “Analyze” option at the top, and then select “Descriptive Statistics”. Next, you will select the

How Do I Assess If I Have Missing Data?

We have already addressed how to find respondent abandonment, but finding missing data that takes place in a random manner can be more challenging. To initially see if any data is miss- ing, let’s start in the SPSS data file. In SPSS, go to the “Analyze” option at the top, then select “Descriptive Statistics”,

How Do I Address Missing Data?

Before we address what to do with missing data, we need to understand why data goes miss- ing. Missing data is typically classified in three ways: (1) missing completely at random, (2) missing at random, and (3) missing not at random. The first category, missing completely at random, is where the missing data is

Assessing Reliability for SEM Model

After you have screened your data on both the respondent and variable levels, the next step is to assess the reliability of your indicators to predict the construct of interest.While having a single indicator for a construct might be easy, it does not provide us a lot of confidence in the validity of the

Identification With SEM Models

Identification in regards to a SEM model deals with whether there is enough information to identify a solution (or in this instance, estimate a parameter). A model that is “under- identified” means that it contains more parameters to be estimated than there are ele- ments in the covariance matrix. For instance, let’s say we

How Do I Calculate the Degrees of Freedom of SEM Model?

As stated earlier, AMOS will calculate your degrees of freedom, but if you have a problem or want to verify another researcher’s work, you need to know how to calculate degrees of freedom.To determine your degrees of freedom, you can use this simple formula outlined by Rigdon (1994) for your measurement model: df =

What Do I Do if My SEM Model Is Under-Identified?

If your model is under-identified, you have two primary solutions to fix this problem. First, you can reduce the number of proposed parameter estimates. This means that you can delete a covariance or structural relationship. Second, you can add more exogenous (independent) variables. By adding more exogenous variables, you increase the number of observations

Sample Size: How Much Is Enough?

With covariance-based SEM, one of the major assumptions is that this technique requires a larger sample size than other statistical techniques. SEM relies on tests which are sensitive to sample size as well as to the magnitude of differences in covariance matrices. There are a litany of suggestions in regards to necessary sample size

Understanding the Validity of Measures for SEM Model

After screening your data and assessing if the measures are reliable, you need to examine the validity of your constructs and indicators. There are numerous validity tests that a researcher needs to be aware of to support the legitimacy of their findings. Before moving on, I want to initially introduce what validity means, and

Overview of the AMOS Graphics Window

When you open the AMOS graphics program, the software will display a window that looks like it has a white page in the middle of the screen. This is your working area in AMOS.You need to try to keep your model within the confines of the white page because the software program can have

AMOS Functions Listed as Icons in Pinned Ribbon

This function allows you to draw an observable variable.You can drag the square in the graph- ics window to the size of the box you want. Example 3.1: This function allows you to draw an unobservable variable.You can drag the circle in the graphics window to the size of the circle you want. Note: