SPSS AMOS Mediation Analysis with Multiple Mediators

Mediation Analysis with Multiple Mediators

Mediation Analysis with Multiple in SPSS AMOS

The Focus of the video tutorial is guiding how to analyze a model with multiple mediators in SPSS AMOS.

Introduction

In the last session, i explained the concept of mediations and tested a simple model with a single mediator. Click here for the last session.

What if I Have Multiple Mediators

  • As stated earlier, when you request indirect effects in AMOS, this will give you all the possible indirect effects in your model. If you have a single mediator, this function is quite handy, but if you have multiple possible mediators in a model, this function can be problematic.
  • The reason for this problem is AMOS will assess the indirect effect from an independent to a dependent variable through all possible mediators.
  • The program will try to examine the total indirect effect though all possible mediators.
  • This type of analysis is not that helpful because mediation really needs to be assessed with each individual mediator instead of the collective group.
  • To find the indirect effect between two variables when there are numerous possible mediators, we have to use a different method from how we assessed mediation in the previous example.
  • To examine the individual relationships in a multiple mediator model, AMOS has a function called “estimands” that will allow you greater flexibility in the analysis of complex models. The estimands function is syntax based.
  • There are no icons or drop-down menus we can use to accomplish our task of examining a specific indirect effect when multiple mediators are present.
  • Thus, we are going to use the syntax-based estimands function to examine a specific indirect effect within a larger model.
  •  

A Simple Mediation Model

In this example, there are two mediators in the relationship between Organizational Learning and Organizational Performance

    • –Collaborative Culture
    • –Organizational Commitment

 

  •  

Step 1: Analysis Properties

  • To determine if the indirect effect of OL to OP is significant, we need to request from AMOS the indirect, direct, and total effects in the output. This will give all possible indirect effects in the model.
  • To do this, select the Analysis Properties button , and when the Analysis Properties pop-up window appears, go to the Output tab at the top. On that tab, you will see at the top right an option for “Indirect, direct, and total effects”. Select this option.

Step 2: Bootstrap Properties

  • Next, we need to request a bootstrap analysis in AMOS. To do this, go to the bootstrap tab at the top of the Analysis Properties window. On that tab will be a checkbox called “Perform bootstrap”; click that box. AMOS will initially give you a default number of 200 samples. This is way too small. Change the number of samples to 5,000.
  • You will also need to select the “Bias-corrected confidence intervals” checkbox. AMOS will default a 90% confidence interval, but significance in most research is at the .05 level, so you need to change this to a 95% confidence level.

Step 3: Name the Paths

  • When you have multiple mediators, AMOS will require you to denote what specific relationships you are concerned with in testing for the indirect effect. To do this, you need to label all the indirect parameters for the specific relationships you are concerned with in the model.
  • You will need to label the paths from IV to MVs and MVs to DV.

In this example I will label the paths as

    • a1 path = OL to CC
    • b1 path = CC to OP
    • a2 path = OL to OC
    • b2 path = OC to OP
  • Double clicking on the Single Headed Arrow on the identified paths and bring up the Object Properties for that specific arrow. (You can right click on the arrow and select Object Properties, too.) You need to label this specific parameter.
  • The label will be in the regression weight field of the Object Properties window. Note that you cannot have spaces in the name you give the parameter.

Step 3: Model with Labels

Once, Labeled, the model will look like the model as shown below

Step 4: Defining the Estimands

  • Once the Regression Weights are labelled, the next step is to define Estimand for calculating individual specific indirect effects.
  • In Order to Define the Estimand, Click “Not estimating any user-defined estimand”.
  • Just simply click on it, and Select Define New Estimands
 

Analyzing the Results - Step 1

Let’s look at the output to determine if mediation is present.

  • To do so, In the Estimates link, you want to select the “Matrices” link. This will let you see the total effects, direct effects, and indirect effects for each relationship in your model.
  • We want to select the indirect effects. AMOS will give you the option to examine the unstandardized or standardized indirect effect. With most mediation analyses, you will see the unstandardized indirect effect reported.
  • Click the Indirect Effects, and from Estimates/Bootstrap -> Bootstrap Confidence, Select Two Tailed Significance (BC)
  • In this example we have only One IV and One DV, do we only see only indirect effect from OL (In the 1st Column) to OP (In the 3rd Row).

Analyzing the Results - Indirect Effects

  • If we look at the intersection of OL and OP, the unstandardized indirect effect is .250.
  • Again, to calculate an indirect effect, it is a very simple process.

 

Next, We can also manually calculate these specific indirect effects. 

The indirect effect is the sum of the (a1*b1) and a2*b2.

  • a1: Organizational Learning to Collaborative Culture = 0.741
  • b1: Collaborative Culture to Organizational Performance = 0.054
  • a2: Organizational Learning to Organizational Commitment = 0.619
  • b2: Organizational Commitment to Organizational Performance = 0.340
  • a1*b1 = 0.741*0.054 = 0.040
  • a2*b2 = 0.619*0.340 = 0.210
  • Indirect Effect from OL to OP = a1b1+a2b2 = 0.250
 
The Same specific indirect effects are also provided using use defined estimands
 

Analyzing the Results - Step 2: Which Indirect Paths are Significant

  • a1b1 Represents Indirect Effect of OL on OP through CC
  • a2b2 Represents Indirect Effect of OL on OP through OC
  • Based on these results, we can conclude that the influence of Organizational Learning to Organizational Performance flows only through Organizational Commitment (P < 0.05).
  • The mediating role of CC is insignificant (p > 0.05).
  • If you have more relationships to test, you can label those parameters and make adjustments in the estimands function.
  • By using the estimands function, you have the ability to isolate an indirect effect even when multiple mediators are present.

Analyzing the Results - Step 3:Type of Mediation

  • Based on these results, we can conclude that the influence of OL to OP flows only through OC and mediating role of CC was insignificant.
  • Since the Direct Effect of OL to OP was significant (p<0.05).
  • We can conclude that OC partially mediates the relationship between OL and OP.
 
Assess, the Direct effect from OL to OP in presence of the mediators.

Calculating T-Statistics

  • To calculate t-statistics, we need Beta Coefficient and Standard Error.
  • To retrieve beta coefficient, Refer to Estimates -> User-defined estimands and from the Estimates/Bootstrap select Bias-corrected percentile method
  • Divide the Beta Coefficient by the Standard Error (Estimate/SE) to retrieve the t-statistics
  • For Path a1b1 = 0.040/0.060 = 0.666
  • For Path a2b2 = 0.210/0.057 = 3.684

From the aforementioned table, take Estimate and take Standard Error from the following table

Reporting Mediation Analysis

Mediation Analysis

The study assessed the mediating role of Collaborative Culture (CC) and Organizational Commitment (OC) on the relationship between Organizational Learning (OL) and Organizational Performance (OP). The results revealed an insignificant indirect effect of OL on OP through CC (b= 0.040, t = 0.666, p = .412), not supporting H1. Analyzing the mediating role of OC, the study found a significant mediating role of OC on the linkage between OL and OP (b – 0.210, t = 3.684, p = 0.004), supporting H2. Furthermore, the direct effect of OL on OP in presence of the mediators was also found significant (b = 0.373, p = 0.000). Hence, OC partially mediated the relationship between OL and OP. Mediation analysis summary is presented in Table 1.

Reference

Collier, J. E. (2020). Applied structural equation modeling using AMOS: Basic to advanced techniques. Routledge.

Video Tutorial

Additional SPSS AMOS Tutorials