Using SPSS

There are multiple statistical analyses you can complete with SPSS. Choose from the options listed below to begin.

Importing Data from an Excel Worksheet

Running Descriptive Statistics

Running a Paired Samples T-test

Running a One-way T-test

Running a One-way ANOVA

Running a Two-way ANOVA

Running a Correlation

Running a Linear Regression

Step 1

Click on the SPSS Icon and start the program.

Step 2

A window similar to the one above will open asking you what file you wish to open.

Step 3

Select Open Another File.

Step 4

You should see the SPSS file explorer.

Step 5

Under “Files of type”, select Excel Data (*.xls ,*.xlsx, *xlsm).

Step 6

You will need to go to the directory where your Excel data file is kept. Select the File name you wish to use (in this case we are using Frequencies and Descriptives Dataset.xlsx). Click Open.

Step 7

A window will open showing you the format of the Excel file and preview of the data. Click on OK.

Step 8

SPSS will import your data into the SPSS data window and you are now ready to run your analyses.

Step 1

Start with opening the data set you want to analyze. This would be an Excel worksheet with your data input into the worksheet as outlined in the Instructions for importing data from an Excel worksheet section.

Step 2

For this example, we will use the “Frequencies and Descriptives Dataset.xlsx” Excel file. Once this file has been imported and opened in SPSS, begin by selecting Analyze.

Step 3

Select Descriptive Statistics.

Step 4

Select Descriptives.

Step 5

SPSS will automatically select all the numerical variables to include as possible variables where you may want to know descriptive statistics. Descriptive statistics is a way to describe features of a data set by generating parameters such as the mean, minimum, maximum, variance, range, etc., as a way of summarizing the data.

Step 6

To have SPSS calculate the descriptive statistics for a set of variables (in this case Age, Weight, Height, and Shoe Size), simply highlight the variable or variables and click the blue arrow that points to the empty Variable(s) box on the right and the highlighted variables will be placed into the Variable(s) box. Select "OK" to show the calculations.

Results

Above are the results shown from the Descriptive Statistics.

Step 1

Step 2

For this example, we will use the “Paired Samples T-test data set.xlsx” Excel file. Once this file has been imported and opened in SPSS, begin by selecting Analyze.

Step 3

Select Compare Means.

Step 4

Select Paired Samples T-Test. There are a number of different types of T-tests that can be used depending on the type of data you collect, how you wish to analyze the data, and how your research problem is set up. In this case, the data is coming from the same participants with two different trials, so it will be a Paired Samples T-Test. All T-tests in SPSS are selected from this menu.

Step 5

Once you select Paired Samples T-Test a window will pop up asking you to place your variables in specific places. The placement of these variables will determine how the variables will be used in your calculations, and whether or not your T-Value will be positive or negative. In this case you will want to name your Trial 2 data as Variable 1, and your Trial 1 data as Variable 2. Highlight Trial 2 Time and click on the blue arrow in the middle of the window to move the Trial 2 Time data into the Variable 1 slot. Highlight Trial 1 Time and click on the blue arrow in the middle of the window to move the Trial 1 Time data into the Variable 2 slot.

Step 6

Once Trial 2 Time shows up in the Variable 1 slot, and Trial 1 Time shows up in the Variable 2 slot, click on “OK” to run the T-Test. The SPSS window will now show output in the SPSS viewer window entitled “Output1 (Document 1) - IBM SPSS Statistics Viewer” window and will have the results of the T-Test you just processed.

Results

SPSS outputs a lot of extra information that we do not need at this point. What we are looking for is 1) the means and standard deviations, and 2) the t value with the degrees of freedom and significance.

Step 1

Step 2

For this example, we will use the “Independent Samples T-test data set.xlsx” Excel file. Once this file has been imported and opened in SPSS, begin by selecting Analyze.

Step 3

Select Compare Means.

Step 4

Select Independent Samples T-Test There are a number of different types of T-tests that can be used depending on the type of data you collect, how you wish to analyze the data, and how your research problem is set up. In this case, the data is coming from the two groups of participants with two different trials, so it will be an Independent Samples T-Test. All T-tests in SPSS are selected from this menu.

Step 5

Once you select Independent Samples T-Test a window will pop up showing you “Group” and “Score”. “Group” is your Grouping Variable and “Score” is your Test Variable. Highlight “Group” and click the blue arrow next to Grouping Variable to place the variable into the Grouping Variable slot.

Step 6

Select Define Groups.

Step 7

Another smaller window will pop up asking you to define Group 1 and Group 2. You can get complicated and use some descriptive name that describes each group, or you can simply define each group as “1” and “2”. We will simply define each group as “1” and “2” because the actual group names and reason for each group will be different for each research project, and this is just an example for use in our SPSS guide to help you learn to use the functions available in SPSS.

Place 1 in the box next to Group 1, and place 2 in the box next to Group 2. Once done click Continue.

Step 8

Highlight “Score” and click on the blue arrow next to the Test Variable box to place “Score” into the Test Variable box.

Step 9

Click OK. The next window that pops up is the Output window that contains the information from the analysis you just ran. The SPSS window will now show output in the SPSS viewer window entitled “Output1 (Document 1) - IBM SPSS Statistics Viewer” window and will have the results of the T-Test you just processed.

Results

SPSS outputs a lot of extra information that we do not need at this point. What we are looking for is 1. the means and standard deviations, and 2. the t value with the degrees of freedom and significance.

Step 1

Step 2

For this example, we will use the “One-way ANOVA data set.xlsx” Excel file. A One-way ANOVA data set simply has the data you want to analyze (the dependent variable) in one column, and the other variable is called your levels or Factors in SPSS.

Step 3

Once this file has been imported and opened in SPSS, begin by selecting Analyze.

Step 4

Select Compare Means.

Step 5

Select One-Way ANOVA. Another window will pop up that shows your variables and an empty box called “Dependent List”, and “Factor”. You will see that in this case, “Type of Car” is highlighted. This variable is known as a “Factor”, and the other variable, “Maintenance Cost” will be your “Dependent” variable.

Step 6

The variable that is already highlighted is the “Factor”, so leave this highlighted and click on the blue arrow that is to the left of the “Factor” box. This will move the variable “Type of Car” into the Factor box.

Step 7

Next, highlight the “Maintenance Cost” variable, and click the blue arrow next to “Dependent List” and move that variable into the Dependent List box.

Step 8

Since there are more than two levels of variables to compare (3 in this case), select Post Hoc and then a window will pop up with all different types of post hoc tests. Your individual situation may call for a different type of post hoc test to be performed, but for this example we will choose LSD.

Step 9

Select Continue.

Step 10

Select Options, check off Descriptive and Means plot.

Step 11

Select Continue and Select OK. The output window will not populate with all the analyses you selected.

Step 12

SPSS sometimes outputs data we do not need at this time, but this time all the output is relevant. We are interested in 1) Descriptives, 2) ANOVA output, 3) Post Hoc Tests, and 4) Means Plots.

Step 1

Step 2

For this example, we will use the Two-way ANOVA data set.xlsx Excel file. A Two-way ANOVA data set simply has the data you want to analyze (the dependent variable) in one column, and the other columns usually contain independent variables of some sort.

Step 3

Once this file has been imported and opened in SPSS, begin by selecting Analyze.

Step 4

Select General Linear Model.

Step 5

Select Univariate. Another window will pop up that shows your variables and an empty box called “Dependent List”, and “Fixed Factors”. You will see that in this case, “Type of Car” is highlighted. This variable is knows as a “Fixed Factor”, and the other variable, “Made in USA” in another “Fixed Factor”. The final variable, “Maintenance Cost” will be your “Dependent” variable.

Step 6

“Type of Car” is already highlighted, so select the blue arrow next to “Fixed Factors” and place “Type of Car” into the Fixed Factors Box. Highlight “Made in USA” and select the same blue arrow next to “Fixed Factors” and place “Made in USA” into the Fixed Factors Box. Finally, highlight “Maintenance Cost” and select the blue arrow next to “Dependent Variable” and place “Maintenance Cost” into the Dependent Variable box for analysis.

Step 7

Select the Post Hoc box and highlight “Type of Car”, select the blue arrow next to the empty “Post Hoc Tests for:” box and place “Type of Car” into that box and select LSD for your post hoc test.

Step 8

Hit Continue. Select Options and then check off Descriptive Statistics, Estimate of Effect Size, and Observed Power.

Step 9

Hit Continue. Hit OK to run your selected analysis. An SPSS output window will pop up with 1) a Between Subjects Factors count, 2) Descriptive Statistics, 3) Tests of Between Subjects Factors, and Post Hoc Tests.

Step 10

SPSS sometimes outputs data we do not need at this time, but this time all the output is relevant. We are interested in 1) Between-Subjects Factors item count, 2) Descriptive Statistics, 3) Test of Between Subjects Effects, and 4) Post Hoc Tests. These results can be seen in Step 9.

Step 1

Step 2

For this example, we will use the Correlation data set.xlsx Excel file. Once this file has been imported and opened in SPSS, begin by selecting Analyze.

Step 3

Select Correlate.

Step 4

Select Bivariate.

Step 5

Another window will pop up with your variables on the left side and an empty “Variables” box on the right side. “Years Employed” should already be highlighted, so just click on the blue arrow to the left of the empty Variables Box to move Years Employed into the Variables Box. Highlight “Salary in thousands” and click on the blue arrow to the left of the Variables Box to move Salary in thousands into the Variables Box. Pearson box should be checked.

Step 6

Select OK. Next window that opens should be the Output1 window with the results from the correlation that was just processed.

Step 1

Step 2

For this example, we will use the Correlation data set.xlsx Excel file. Correlation and Linear Regression are somewhat related. One gives you the correlation coefficient with an indication of the strength of the correlation, and the other gives you a linear regression line or “line of best fit” along with the equation for that line of best fit, to allow you to predict one variable using the values from another variable. Once this file has been imported and opened in SPSS, begin by selecting Analyze.

Step 3

Select Regression.

Step 4

Select Linear. Another window will pop up so you can place the variables in either the “Dependent” box or the “Independent” box.

Step 5

Since we are trying to predict “Salary in thousands” from “Years Employed”, in this case we put the “Years Employed” in the Independent box by highlighting “Years Employed” and clicking on the blue arrow to the left of the Independent box. Once this is done and “Years Employed” appears in the Independent box, highlight “Salary in thousands” and click on the blue arrow to the left of the Dependent box.

Step 6

Select Statistics and check off the Descriptives box, then select Continue.

Step 7

Select OK.

Step 8

The next window that pops up is the Output1 window with the results of your Linear Regression.

Step 9

SPSS sometimes outputs information that we don't need at this point. In this case, we are interested in 1) Descriptive Statistics, 2) Correlations, and 3) Model Summary. Everything else is not needed right now.

Step 10

This gives us the correlation information but no regression line and no line formula just yet. In order to get this, we must click on Graphs.

Step 11

Select Legacy Dialogs.

Step 12

Select Scatter/Dot.

Step 13

Select Simple Scatter and click on Define.

Step 14

We then have a final window pop open that asks us which variable we want to show up on the Y Axis and which variable we want to show up on the X Axis. Since we are trying to predict Salary in thousands based on Years Employed, We will place Years Employed in the Y Axis box by highlighting Years Employed and clicking on the blue arrow to the left of the Y Axis box. This will place Years Employed in the Y Axis box. We then want to place Salary in thousands in the X Axis box so we will highlight Salary in thousands and click on the blue arrow to the left of the X Axis box. This will place Salary in thousands in the X Axis box. Select OK to generate the regression line or line of best fit.

Step 15

The next output is the scatter plot. The scatter plot appears, and to get the regression line or line of best fit, and the equation to that line, we double click the graph.

Step 16

When a window appears with the scatter plot, we click on the option that says Add fit line at Total. A line of best fit will appear on the graph, and the equation for that line will also appear.

Step 17

Click on the RED “X” in the upper right corner of the graph with the line and equation listed on it and window will close and the line of best fit and the line equation will now appear on the output graph.