IBM SPSS Statistics 23

Part 3: Regression Analysis

Winter 2016, Version 1

Table of Contents
Introduction ......................................................................................................................................2
Downloading the Data Files.................................................................................................................... 2
Simple Regression ..............................................................................................................................2
Scatter Plot ............................................................................................................................................. 2
Predicting Values of Dependent Variables ............................................................................................. 5
Predicting This Year’s Sales with the Simple Regression Model ............................................................ 6
Multiple Regression ...........................................................................................................................8
Predicting Values of Dependent Variables ............................................................................................. 9
Predicting This Year’s Sales with the Multiple Regression Model........................................................ 10
Data Transformation ........................................................................................................................ 11
Computing ............................................................................................................................................ 12
Polynomial Regression ..................................................................................................................... 13
Regression Analysis .............................................................................................................................. 14
Analyzing the Results ..................................................................................................................... 14
Chart Editing .................................................................................................................................... 15
Adding a Line to the Scatter Plot .......................................................................................................... 15
Manipulating the Scales on the X and Y Axes....................................................................................... 17
Adding a Title to the Chart ................................................................................................................... 19
Adding Color to the Chart..................................................................................................................... 19
Applying a Background Color ............................................................................................................... 20

For additional training resources, visit www.calstatela.edu/training.

Introduction
SPSS stands for Statistical Package for the Social Sciences. This program can be used to analyze data
collected from surveys, tests, observations, etc. It can perform a variety of data analyses and
presentation functions, including statistical analysis and graphical presentation of data. Among its
features are modules for statistical data analysis. These include (1) descriptive statistics such as
frequencies, central tendency, plots, charts, and lists; and (2) sophisticated inferential and multivariate
statistical procedures such as analysis of variance (ANOVA), factor analysis, cluster analysis, and
categorical data analysis. IBM SPSS Statistics 23 is well-suited for survey research, though by no means is
it limited to just this topic of exploration.

This handout provides basic instructions on how to answer research questions and test hypotheses
using linear regression (a technique which examines the relationship between a dependent variable and
a set of independent variables). The value of the dependent variable (e.g., salesperson’s total annual
sales) can be predicted based on its relationship to the independent variables used in the analysis (e.g.,
age, education, and years of experience). The two research questions proposed in this handout are as
follows:
• How much money will each salesperson make this year?
• Who will qualify for a $1,000 bonus?

Downloading the Data Files

This handout includes sample data files that can be used to follow along the steps. If you plan to use the
data files, download the following ZIP file to your computer and extract the files. It is recommended to
save the data files on your desktop for easy access.
• IBM SPSS Statistics 23 Part 3 Data Files

Simple Regression
Simple regression estimates how the value of one dependent variable (Y) can be predicted based on
the value of one independent variable (X). The linear equation for simple regression is as follows:
Y = aX + b

Simple regression can answer the following research question:

Research Question # 1
Based on last year’s sales, how much money will each salesperson make this year?

Scatter Plot
A scatter plot displays the nature of the relationship between two variables. Before performing a
regression analysis, it is recommended to run a scatter plot to determine if there is a linear relationship
between the variables. If there is no linear relationship (i.e., points on a graph are not clustered in a
straight line), then a simple regression would not be the appropriate analysis to use for this data set.

To run a scatter plot:

1. Start IBM SPSS Statistics 23, and then open the Regression.sav file.
2. Click the Graphs menu, point to Legacy Dialogs, and then click Scatter/Dot (see Figure 1).

2 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 Figure 1 – Scatter/Dot Selected on the Graphs Menu

3. In the Scatter/Dot dialog box, make sure that the Simple Scatter option is selected, and then
click the Define button (see Figure 2).
NOTE: The Simple Scatter plot is used to estimate the relationship between two variables.

Figure 2 – Scatter/Dot Dialog Box

4. In the Simple Scatterplot dialog box, select the Last year sales variable in the left box, and then
click the transfer arrow button to move it to the Y Axis box (see Figure 3).
5. Select the Years of experience variable in the left box, and then click the transfer arrow button
to move it to the X Axis box.

3 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 Figure 3 – Simple Scatterplot Dialog Box

6. Click the OK button. The Output Viewer window opens and displays a scatter plot of the
variables (see Figure 4).
NOTE: The scatter plot in Figure 4 indicates that a linear relationship exists between the
variables Last year sales and Years of experience. The next step is to find a line that best
accommodates the pattern of points in this scatter plot. The steps for enhancing the graph
appearance are covered in the last section of this handout.

Figure 4 – Scatter Plot

4 | IBM SPSS Statistics 23 Part 3: Regression Analysis

Predicting Values of Dependent Variables
Judging from the scatter plot above, a linear relationship seems to exist between the two variables.
Therefore, a simple regression analysis can be used to calculate an equation that will help predict this
year’s sales.

To run a simple regression analysis:

1. Switch to the Data Editor window.
2. Click the Analyze menu, point to Regression, and then click Linear.
3. In the Linear Regression dialog box, select the Last year sales variable in the left box, and then
click the transfer arrow button to move it to the Dependent box (see Figure 5).
4. Select the Years of experience variable in the left box, and then click the transfer arrow button
to move it to the Independent(s) box.
5. Click the OK button.

Figure 5 – Linear Regression Dialog Box

The following tables in the Output Viewer window present the results of a simple regression. R Square
(.918) indicates that this model accounts for almost 92% of the total variation in the data (see Figure 6).

Figure 6 – Model Summary Output

5 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 Figure 7 – Coefficients Output

The slope and the y-intercept as seen in Figure 7 should be substituted in the following linear equation
to predict this year’s sales: Y = aX + b. In this case, the values of a, b, X, and Y will be as follows:
a = 1954.658
b = 440.987
X = Years of experience (values of independent variable)
Y = Last year sales (values of dependent variable)

Predicting This Year’s Sales with the Simple Regression Model

To predict this year’s sales for each salesperson, substitute the values of a and b in the following linear
equation:
Y = aX + b
Last year sales = (a * yearexpe) + b
This year sales = (1954.658 * yearexp2) + 440.987
a = 1954.658
b = 440.987
X = Years of experience [yearexp2]
Y = This year sales
NOTE: The new independent variable yearexp2 is used instead of yearexpe in order to predict
this year’s sales.

To predict this year’s sales using the computing function:

1. Switch to the Data Editor window.
2. Click the Transform menu, and then click Compute Variable.
3. In the Compute Variable dialog box, type Simple in the Target Variable box (see Figure 8).
4. In the Numeric Expression box, enter the following equation by typing or selecting from the
dialog box keypad:
1954.658 * yearexp2 + 440.987
NOTE: It is recommended to select the yearexp2 variable directly from the Variable box on the
left side of the Compute Variable dialog box to prevent typing mistakes.

6 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 Figure 8 – Compute Variable Dialog Box

5. Click the OK button. The results are displayed in the Simple column in Data View (see Figure 9).

Figure 9 – Simple Regression Results

To change the data type for the Simple variable:

1. Click the Variable View tab in the lower-left corner of the Data Editor window (see Figure 10).

Figure 10 – Variable View Tab

7 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 2. Locate the Simple variable in row 6, click in the next cell under the Type column, and then click
the Ellipses button that appears.
3. In the Variable Type dialog box, select the Dollar option button, select the $###,###,### format
(12 digit width with 0 decimal places), and then click the OK button (see Figure 11).

Figure 11 – Variable Type Dialog Box

4. Click the Data View tab in the lower-left corner of the Data Editor window.
NOTE: The prediction of this year’s sales for each salesperson is computed under the new
variable named Simple (see Figure 12).

Figure 12 – Simple Regression Prediction

Multiple Regression
Multiple regression estimates the coefficients of the linear equation when there is more than one
independent variable that best predicts the value of the dependent variable. For example, a
salesperson’s total annual sales (the dependent variable) can be predicted based on independent
variables such as age, education, and years of experience. The linear equation for multiple regression is
as follows:
Z = aX + bY + c

8 | IBM SPSS Statistics 23 Part 3: Regression Analysis

Predicting Values of Dependent Variables
The previous section demonstrated how to predict this year’s sales (the dependent variable) based on
one independent variable (number of years of experience) by using simple regression analysis. Similarly,
this year’s sales (the dependent variable) can be predicted from more than one independent variable
(such as Years of experience and Years of education) by using multiple regression analysis.

To run a multiple regression analysis:

1. Click the Analyze menu, point to Regression, and then click Linear. The Linear Regression dialog
box opens (see Figure 13).
NOTE: If there are variables in the Dependent or Independent(s) boxes, click the Reset button
before performing steps 2 and 3 below.
2. Select the Last year sales variable in the left box, and then click the transfer arrow button to
move it to the Dependent box.
3. Select the Years of experience and Years of education variables in the left box, and then click
the transfer arrow button to move them to the Independent(s) box.
NOTE: You can select multiple variables by clicking the first variable, holding down the Ctrl key,
and then clicking each of the other variables.
4. Click the OK button. The Output Viewer window opens.

Figure 13 – Linear Regression Dialog Box

NOTE: The table should look similar to Figure 14. R Square = .976 indicates that this model can
predict this year’s sales almost 98% correctly.

9 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 Figure 14 – Model Summary Output for Multiple Regression

Figure 15 – Multiple Regression Output

To predict this year’s sales, substitute the values for the slopes and y-intercept displayed in the Output
Viewer window (see Figure 15) in the following linear equation: Z = aX+ bY + c.

In this case, the values of a, b, X, and Y will be as follows:

a = 1874.5
b = 609.391
c = (-8510.838)
X = Years of experience (independent variable)
Y = Years of education (independent variable)
Z = This year sales (dependent variable)

As indicated in the output table, the coefficient for Years of experience is 1874.5 and the coefficient for
Years of education is 609.391.

Predicting This Year’s Sales with the Multiple Regression Model

To predict this year’s sales for each salesperson, substitute the values of a, b, and c in the following
linear equation: Z = aX + bY + c.
This year sales = 1874.5 * Years of experience + 609.391 * Years of education + (-8510.838)

To predict this year’s sales by multiple regression analysis:

1. Switch to the Data Editor window.
2. Click the Transform menu, and then click Compute Variable.
3. In the Compute Variable dialog box, click the Reset button (see Figure 16).
4. In the Target Variable box, type Multiple.
5. In the Numeric Expression box, enter the following equation by typing or selecting from the
dialog box keypad:
1874.5 * yearexp2 + 609.391 * educatio - 8510.838

10 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 Figure 16 – Compute Variable Dialog Box

6. Click the OK button. The Multiple column in Data View displays the results (see Figure 17).
NOTE: The sales prediction for each salesperson using two independent variables is listed under
the new variable named Multiple.

Figure 17 – Multiple Regression Results

Data Transformation
Situations may arise when data transformation is useful. Most data transformations can be performed
with the Compute command. Using this command, the data file can be manipulated to fit various
statistical performances.
Research Question # 2
Who will earn a $1,000 bonus?

11 | IBM SPSS Statistics 23 Part 3: Regression Analysis

Computing
Each salesperson’s yearly sales were predicted using multiple regression analysis. The salespeople who
made $2,000 or more than their predicted values will receive a $1,000 bonus. Use the Compute
command to compare the values of this year’s actual sales with the predictions from multiple regression
analysis computed in the previous lesson to find eligible salespeople. The first step in predicting who will
receive a bonus is to calculate the difference between this year’s actual sales and the prediction of this
year’s sales from the multiple regression analysis.

To predict who qualifies for the bonus:

1. Open the Bonus.sav file.
2. Click the Transform menu, and then click Compute Variable.
3. In the Compute Variable dialog box, type bonus in the Target Variable box, type 1000 in the
Numeric Expression box, and then click the If button (see Figure 18).

Figure 18 – Compute Variable Dialog Box

4. In the Compute Variable: If Cases dialog box, select the Include if case satisfies condition
option button (see Figure 19).
5. Enter the following expression by typing or selecting from the dialog box keypad:
thissale - multiple >= 2000

NOTE: It is recommended to select the variables and the >= sign directly from the Variable
box and keypad provided in the dialog box to prevent mistakes.

12 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 Figure 19 – Compute Variable: If Cases Dialog Box

6. Click the Continue button, and then click the OK button.

NOTE: Salespersons Ivett (#44) and Jason (#49) are two of the sales personnel who qualify for
the $1,000 bonus because they made $2,000 over their predicted sales from the last lesson (see
Figure 20).

Figure 20 – Bonus Results

Polynomial Regression
This type of regression involves fitting a dependent variable (Yi) to a polynomial function of a single
independent variable (Xi). The regression model is as follows (see Table 1 for the meaning of the
variables):
Yi = a + b1Xi + b2Xi2 + b3Xi3 + … + bkXik + ei

Table 1 – Breakdown of the Variables

Variable Meaning
a Constant
bk The coefficient for the independent variable to the k’th power
ei Random error term

13 | IBM SPSS Statistics 23 Part 3: Regression Analysis

Regression Analysis
To look at the growth relationship between weight and age:
1. Open the Growth.sav file.
2. Click the Analyze menu, point to Regression, and then click Curve Estimation.
3. In the Curve Estimation dialog box, transfer the wght variable to the Dependent(s) box and the
age variable to the Independent Variable box (see Figure 21).
NOTE: The dependent variable wght is predicted using the independent variable age.
4. Deselect the Plot models check box.
5. Select the Display ANOVA table check box.
6. Under Models, deselect the Linear check box and select the Cubic check box.
7. Click the OK button.

Figure 21 – Curve Estimation Dialog Box

Analyzing the Results

This cubic model has an R Square value of 99.567% (see Figure 22). The F-ratio indicates a highly
significant fit. The best fitting cubic polynomial is given by the following equation:
Yi = 0.052 – 0.017 Xi + 0.010 Xi2 – 0.001 Xi3 + ei
(where Yi is weight and Xi is age)

Polynomial regression can find the line of best fit for polynomials consisting of two or more variables. If
X is the dependent variable, use the Transform and Compute options of the Data Editor (as discussed
earlier in this lesson) to create new variables X2 = X*X, X3 = X*X2, X4 = X*X3, etc., then use these new
variables (X, X2, X3, X4, etc.) as a set of independent variables for the regression analysis.

14 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 Figure 22 – Polynomial Regression Summary Results

Chart Editing
During the final stage of research, enhancing the appearance of charts and figures can help viewers
understand what may seem to be confusing statistics. The following steps explain some useful methods
for enhancing a chart’s appearance.

Adding a Line to the Scatter Plot

Adding a straight line to fit the scattered pattern of a data chart can help emphasize the linear
relationship between the data.

To add a line to the scatter plot:

1. Click the Graphs menu, point to Legacy Dialogs, and then click Scatter/Dot.
2. In the Scatter/Dot dialog box, select the Simple Scatter option, and then click the Define
button.
3. In the Simple Scatterplot dialog box, transfer the age variable to the X Axis box and the wght
variable to the Y Axis box, and then click the OK button. A chart appears in the Output Viewer
window.
4. Double-click the chart to modify it. The Chart Editor window and the Properties dialog box
open.
5. In the Chart Editor window, right-click a chart marker, and then click Add Fit Line at Total on the
shortcut menu (see Figure 23).
6. In the Properties dialog box, on the Fit Line tab, select the Cubic option button under Fit
Method, deselect the Attach label to line check box, and then click the Apply button (see Figure
24).
7. Close the Chart Editor window.

15 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 NOTE: Notice that the Add Fit Line at Total does not automatically capture the way the data
curves, but the Cubic method is almost a perfect fit (see Figure 25). Make sure to select the best
fit line for the data. Once a best fit line is applied, it will stay on the graph. Selecting Add Fit Line
at Total again will add a new fit line to the graph. To edit the fit line that has already been
applied, select the fit line that is already on the graph and change the properties.

Figure 23 – Chart Editor Window

Figure 25 – Fit Line Added to the Scatter Plot

Figure 24 – Fit Line Tab of the Properties

Dialog Box

16 | IBM SPSS Statistics 23 Part 3: Regression Analysis

Manipulating the Scales on the X and Y Axes
Adjust the X axis and Y axis to enhance the overall appearance and readability of a chart. Various
elements of the axes can be manipulated (such as scale, ticks and grids, number format, and axis label).

To manipulate the scales on the X and Y axes:

1. If necessary, open the Regression.sav file, and then run a scatter plot with Last year sales
assigned to the Y Axis and Years of experience assigned to the X Axis.
2. In the Output Viewer window, double-click the chart.
3. In the Chart Editor window, right-click a chart marker, and then click Add Fit Line at Total on the
shortcut menu.
4. In the Properties dialog box, on the Fit Line tab, deselect the Attach label to line check box, and
then click the Apply button.
5. To select and manipulate the X axis, click the Select the X axis button on the Standard
toolbar.
6. In the Properties dialog box, on the Scale tab, change the value in the Lower margin (%) box to
0, and then click the Apply button (see Figure 26).
7. In the Properties dialog, select the Labels & Ticks tab (see Figure 27).
8. In the Major Ticks section, select the Display ticks check box, select Inside from the Style list,
and then click the Apply button.

Figure 26 – Scale Tab of the Properties Dialog Figure 27 – Labels & Ticks Tab of the Properties
Box (X Axis) Dialog Box (X Axis)

9. Click the Show Grid Lines button on the Standard toolbar. The Properties dialog box
displays the Grid Lines tab (see Figure 28).
10. Select the Major ticks only option button, click the Apply button, and then click the Close
button.

17 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 11. To select and manipulate the Y axis, click the Select the Y axis button on the Standard
toolbar.
12. In the Properties dialog box, on the Scale tab, change the value in the Lower margin (%) box to
0, click the Apply button, and then click the Close button (see Figure 29).

Figure 28 – Grid Lines Tab of the Properties Figure 29 – Scale Tab of the Properties Dialog
Dialog Box (X Axis) Box (Y Axis)

NOTE: Below is the chart before and after manipulating the X and Y axes (see Figure 30 and
Figure 31).

Figure 30 – Chart Before Manipulating the X Figure 31 – Chart After Manipulating the X and
and Y Axes Y Axes

18 | IBM SPSS Statistics 23 Part 3: Regression Analysis

Adding a Title to the Chart
Adding a title to a chart is a simple process that enhances the chart’s appearance.

To add a title to the chart:

1. In the Chart Editor window, click the Insert a title button on the Standard toolbar. A text
box with the word Title is inserted above the chart and the Properties dialog box opens.
2. Delete the placeholder text in the text box, and then type Relationship Between Last Year Sales
and Years of Experience.
3. To enter a line break, click where you want to break the line, and then press Shift+Enter.
4. To format the title, click the border of the text box to select it.
5. In the Properties dialog box, select the Text Style tab, select the desired font size and style in
the Font section, select the desired color from the color palette, click the Apply button, and
then click the Close button (see Figure 32).
6. If necessary, move or resize the text box. The changes are applied to the chart (see Figure 33).

Figure 33 – Title Added at the Top of the Chart

Figure 32 – Text Style Tab of the

Properties Dialog Box

Adding Color to the Chart

All chart elements can be colored differently to add emphasis or to distinguish between elements.

To add color to the chart:

1. In the Chart Editor window, select the chart element to change or add color to (such as the
chart markers).
2. Click the Show Properties Window button on the Standard toolbar.
3. In the Properties dialog box, select the Marker tab (see Figure 34).

19 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 4. To change the marker color, select the desired color from the color palette.
5. To change the marker type, click the Type arrow in the Marker section and select the desired
symbol from the list.
6. View the changes in the Preview section, click the Apply button, and then click the Close button.
The changes are applied to the chart (see Figure 35).

Figure 35 – Chart After Changing the Marker Type

and Color

Figure 34 – Marker Tab of the Properties

Dialog Box

Applying a Background Color

You can change the chart’s background color to make it stand out from other chart elements.

To apply a background color:

1. Click in the background area of the chart to select it.
2. Click the Show Properties Window button on the Standard toolbar.
3. In the Properties dialog box, on the Fill & Border tab, select the Fill swatch , and then select
the desired color from the color palette (see Figure 36).
NOTE: You can also apply a background pattern by clicking the Pattern arrow and selecting the
desired pattern from the list.
4. Click the Apply button, and then click the Close button. The changes are applied to the chart
(see Figure 37).

20 | IBM SPSS Statistics 23 Part 3: Regression Analysis

 Figure 37 – Chart After Applying a Background Color

Figure 36 – Fill & Border Tab of the

Properties Dialog Box

21 | IBM SPSS Statistics 23 Part 3: Regression Analysis