📊 Predicting Employee Salaries with Linear Regression

This project demonstrates how to build a Linear Regression Model to predict employee salaries using IBM SPSS Modeler (or similar tools). The dataset contains details of 474 employees.

---

📁 1. Import and Examine the Data

🔹 Step 1: Load the Data

• From Sources palette, drag a Var. File node to a blank canvas.
• Edit the node, locate and select employee_data.txt.
• Click OK to confirm.

🔹 Step 2: Preview the Table

• From Output palette, connect a Table node to the Var. File node.
• Run the node to preview data.
• ➤ Dataset contains 474 employees.

🔹 Step 3: Run Data Audit

• Connect a Data Audit node to the same source.
• Run it to inspect value distributions, types, and missing data.

---

🧪 2. Set Field Measurement Levels and Roles

🔸 Step 1: Add and Configure Type Node

• Add a Type node from Field Ops, connected to the Var. File node.
• Click Read Values.

🔸 Step 2: Configure Fields

• Set educational_level → Ordinal
• Set fields from gender to months_previous_experience → Input
• Set current_salary → Target

---

📈 3. Create and Train Linear Regression Model

🔹 Step 1: Add Linear Model

• Add a Linear node from the Modeling palette connected to the Type node.

🔹 Step 2: Edit Build Options

• Go to Build Options tab:
  • Under Basics: Uncheck Automatically prepare data.
  • Under Model Selection: Choose Include all predictors.

🔹 Step 3: Train the Model

• Click Run to build the model.

---

📊 4. Evaluate the Model

🔸 View Model Summary

• Edit the model nugget and click Model Summary.

🔸 View Predictor Importance

• Click Predictor Importance
  • job_category → Most important predictor
  • gender → Second most important
  • region and age → Least important

🔸 Visualize Predictions

• Click Predicted by Observed
  • ➤ Predicted values show two major clusters rather than a smooth trend.

🔸 View Coefficients

• Click Coefficients by Observed
• From the Style dropdown, select Table.

---

✅ Summary

You now have a working linear regression model that:

• Uses job role, education, experience, and other features
• Identifies job category and gender as major salary predictors
• Can be used to further explore wage inequality or HR planning