Tutorial SEM PLS dengan Variabel Moderasi Menggunakan SmartPLS 4 FULL

Tabrani Education

11 May 202413:50

Summary

TLDRThis tutorial video from the Tabrani Education channel offers a comprehensive guide on using Smart PLS 4 for Structural Equation Modeling (SEM) with moderation variables. It covers the process of importing data, setting up a PLS model with creativity as an independent variable, employee performance as dependent, and compensation as a moderating variable. The video explains hypothesis testing, model evaluation, and interpretation of results, including the significance of creativity and compensation in enhancing employee performance. It concludes with a demonstration of Goodness of Fit calculation, emphasizing the model's strong explanatory power.

Takeaways

📚 The tutorial is about using Smart PLS 4 for Structural Equation Modeling (SEM) with moderation variables.
🔍 It involves three variables: an independent variable (employee creativity), a dependent variable (employee performance), and a moderating variable (compensation).
✍️ The tutorial covers the hypothesis testing, with H1 suggesting a positive impact of creativity on performance and H2 that compensation strengthens this relationship.
📈 The data used in the tutorial includes three indicators for creativity, four for performance, and five for compensation, all saved in an Excel file.
💻 The Smart PLS 4 software is used for the analysis, and the tutorial guides through creating a new project and importing data.
📊 The outer model evaluation is performed to check for convergent validity and discriminant validity using the Fornell-Lacker criterion.
📉 Composite reliability and validity are assessed, with a cutoff of 0.7 for reliability and loading factors above 0.5 indicating convergent validity.
📐 The R Square value of 0.775 indicates that the model explains 77.5% of the variance in employee performance, suggesting a strong model.
🔢 The significance of the path coefficients is tested, with creativity having a positive and strong effect on performance (p-value of 0.029).
🔄 The interaction effect of compensation on the creativity-performance relationship is also significant, with a p-value of 0.019, indicating that compensation strengthens the positive impact of creativity on performance.
📉 The Goodness of Fit (GoF) value of 0.805 suggests that the overall model performance is in the 'large' category, indicating a good fit.

Q & A

What is the main topic of the tutorial in the provided script?
-The main topic of the tutorial is the use of Smart PLS 4 for Structural Equation Modeling (SEM) with moderation variables.
What are the three variables involved in the tutorial's model?
-The three variables are one independent variable (employee creativity), one dependent variable (employee performance), and one moderation variable (compensation).
What are the hypotheses being tested in the tutorial?
-There are two hypotheses: H1 states that creativity positively affects employee performance, and H2 suggests that compensation strengthens the impact of creativity on employee performance.
What is the significance of the R Square value in the context of this tutorial?
-The R Square value of 0.775 indicates that employee creativity and compensation together explain 77.5% of the variance in employee performance, suggesting a strong model.
How is the interaction effect between creativity and compensation represented in the model?
-The interaction effect is represented by multiplying the compensation variable with the creativity variable in the model, and it is automatically generated when the model is set up in Smart PLS 4.
What is the significance of the coefficient value and P-value in hypothesis testing?
-The coefficient value indicates the strength and direction of the relationship, while the P-value determines the statistical significance of the effect, with a value less than 0.05 typically considered significant.
What is the role of the Fornell-Larcker criterion in the tutorial?
-The Fornell-Larcker criterion is used to assess discriminant validity, ensuring that the square root of the average variance extracted (AVE) for each construct is greater than its correlations with other constructs.
What is the importance of composite reliability in the context of this tutorial?
-Composite reliability, measured by values like Cronbach's Alpha, indicates the internal consistency and reliability of the constructs in the model, with a value greater than 0.7 considered acceptable.
How is the Goodness of Fit (GoF) calculated in the tutorial?
-The GoF is calculated by multiplying the average variance extracted (AVE) by the R Square value and then taking the square root of the result, providing an overall measure of model performance.
What does the acceptance of H2 imply about the role of compensation in the model?
-The acceptance of H2 implies that compensation significantly strengthens the positive effect of creativity on employee performance, indicating a moderating role.
How does the script guide users in setting up the model in Smart PLS 4?
-The script provides a step-by-step guide, from creating a new project and importing data to setting up the model, adjusting the paths, and running calculations for model evaluation and hypothesis testing.

Outlines

00:00

📚 Introduction to PLS Moderation Tutorial

The video script starts with a greeting and introduces a tutorial on Structural Equation Modeling (SEM) using Smart PLS 4 with moderation variables. The presenter outlines the research variables: an independent variable 'employee creativity', a dependent variable 'employee performance', and a moderating variable 'compensation'. Two hypotheses are presented: H1 suggests a positive impact of creativity on performance, and H2 posits that compensation strengthens the influence of creativity on performance. The script mentions the data used for analysis, with three indicators for creativity, four for performance, and five for compensation, and notes that the data is stored in an Excel file. The presenter guides viewers through the initial steps of setting up a new project in Smart PLS 4, checking for missing data, and importing the dataset.

05:01

🔍 Data Analysis and Model Evaluation

The second paragraph delves into the data analysis process, starting with the evaluation of the outer model in Smart PLS 4. The presenter explains the importance of checking for convergent validity and discriminant validity, using the Fornell-Lacker criterion to ensure the model's validity. The reliability of the constructs is assessed using the composite reliability index, and the R-Square value is discussed as an indicator of the model's explanatory power. The presenter also explains the significance of the F-statistic in testing the hypotheses and the importance of the interaction effect between creativity and compensation on employee performance. The results show that both creativity and the interaction effect of creativity and compensation significantly influence employee performance, supporting the hypotheses.

10:02

📈 Goodness of Fit and Conclusion

The final paragraph focuses on the Goodness of Fit measure, which assesses the overall performance of the model. The presenter uses the average of AVE (Average Variance Extracted) multiplied by the R-Square value to calculate the GoF (Goodness of Fit) index. The calculated GoF value of 0.805 indicates a strong model fit, suggesting that the research model is well-supported by the data. The presenter wraps up the tutorial by summarizing the findings and stating that compensation significantly strengthens the positive impact of creativity on employee performance. The video concludes with a sign-off greeting, and the presenter promises to cover more topics in subsequent videos.

Mindmap

Keywords

💡SM PLS

Second Moment Partial Least Squares, a statistical technique used for data analysis, particularly in the context of this video, for examining the relationship between variables with a focus on moderation analysis.

💡Variable Moderation

A statistical concept where the effect of an independent variable on a dependent variable is influenced by a third variable, known as the moderator. In the video, compensation is the moderating variable affecting the relationship between employee creativity and performance.

💡Independent Variable

In the context of the video, employee creativity is the independent variable, meaning it is presumed to influence another variable, in this case, employee performance.

💡Dependent Variable

Employee performance is the dependent variable in the video's narrative, as it is the outcome that is potentially affected by the independent variable, creativity, and moderated by compensation.

💡Hypothesis

The video discusses two hypotheses (H1 and H2), which are proposed statements to be tested. H1 suggests a positive impact of creativity on performance, while H2 posits that compensation strengthens this impact.

💡Indicators

Indicators are specific measures used to represent a variable in the analysis. The script mentions three indicators for creativity, four for performance, and EMP indicators for compensation.

💡Convergent Validity

A measure of how well two or more measures represent a single concept, which is crucial for ensuring the reliability of the constructs used in the analysis. The video explains that factor loadings greater than 0.5 indicate convergent validity.

💡Discriminant Validity

This refers to the extent to which a measure differentiates between two concepts. The video uses the Fornell-Larcker criterion to assess discriminant validity, where the square root of the average variance extracted should be greater than the correlations with other constructs.

💡Composite Reliability

A measure of the internal consistency and stability of a set of items within a construct. The video states that a composite reliability value above 0.7 indicates that the variables are reliable.

💡R Square

A statistical measure that represents the proportion of the variance for a dependent variable that's explained by an independent variable in a regression model. The video mentions an R Square value of 0.775, indicating that creativity and compensation explain 77.5% of the variance in employee performance.

💡Goodness of Fit

A measure used to evaluate how well a model fits the data. In the video, the goodness of fit is calculated using the average of the square root of the average variance extracted multiplied by the R Square, resulting in a value of 0.805, which suggests a good fit.

Highlights

Introduction to a tutorial on Structural Equation Modeling (SEM) with moderation using Smart PLS 4.

The tutorial covers three variables: an independent variable (employee creativity), a dependent variable (employee performance), and a moderating variable (compensation).

Two hypotheses are presented: H1 suggests a positive influence of creativity on employee performance, and H2 proposes that compensation strengthens the impact of creativity on performance.

Data used in the tutorial includes three indicators for creativity, four for performance, and five for compensation.

Instructions on how to save and import data into Smart PLS 4 are provided, with a preference for the latest version of Excel.

A step-by-step guide on creating a new project and importing data into Smart PLS 4 is demonstrated.

The importance of checking for missing data before importing is emphasized to ensure data completeness.

Explanation of how to set up the model in Smart PLS 4, including naming the model and assigning variable roles.

The tutorial highlights the automatic line feature in Smart PLS 4 for aligning the model's structure.

Instructions on manually creating a moderation variable in the model, as it cannot be done automatically.

Evaluation of the outer model's measurement, including calculating PLS algorithm and checking for convergent validity and auto-loading.

Criteria for acceptable factor loadings and the use of the Fornell-Lacker criterion for discriminant validity are discussed.

Reliability assessment using composite reliability and the acceptable threshold of over 0.7.

Calculation and interpretation of R Square to understand the variance explained by the model.

Explanation of the F statistic and its significance in testing the model's hypotheses.

Hypothesis testing using the bootstrap method in Smart PLS 4 and the interpretation of P values.

Acceptance of H1 and H2 based on the positive coefficients and P values less than 0.05.

Goodness of Fit assessment using the average of AVE and R Square, and its interpretation.

A step-by-step guide on calculating the Goodness of Fit in Excel for practical application.

Conclusion of the tutorial with a summary of the findings and a preview of upcoming topics.