Research and Statistical Support

Module 1. Familiarization with SPSS.

First, we offer a review of some commonly used terms and definitions.

What is statistics? There is no generally accepted answer.

"Statistics is considered by some to be a mathematical science pertaining to the collection, analysis, interpretation or explanation, and presentation of data, while others consider it to be a branch of mathematics concerned with collecting and interpreting data. Because of its empirical roots and its focus on applications, statistics is usually considered to be a distinct mathematical science rather than a branch of mathematics" (Wiki).

Generally speaking there are two accepted types of statistics. Descriptive statistics are used to summarize groups of numbers and make them understandable (describing the data). Inferential statistics are used to draw conclusions based on the numbers actually collected during a research study, but going beyond these numbers (making inferences about the data and potential data or populations).

Operational definitions: Operational definitions allow us define variables with measurement. Think quantitatively. What is the quantity of this characteristic, phenomena, feature, behavior, emotion, etc.? Defining a variable operationally means defining it in such a way that description and observation are not the only benefits, but measurement as well. How do you define success in college? How do you define drunkenness? How do you define sadness?

What is PASW / SPSS and why would we want to use it? Originally, SPSS was an acronym for Statistical Package for the Social Sciences. The PASW name was applied when recently IBM bought SPSS. From this point forward, we will use SPSS to refer to PASW / SPSS. Regardless of the name or version you use, SPSS is a statistical software package that allows us to organize, assess, manipulate, and analyze data. The simple answer for "why would we want to use SPSS" is that is allows us to do statistical calculations much quicker than by hand or with other statistical software. This is the only real strength of SPSS over other packages; its ease of use. SPSS has garnered market share because the majority of its functions are available as point-and-click operations, while other software packages require the user to input syntax, code, or script to perform functions. However, other software packages have the benefit of newer, more sophisticated functions available than what is offered in the base SPSS installation.

1.) Creating a data file.

Open SPSS: --> Start, Programs, SPSS. The initial window (center of the screen) will be asking you if you want to open an existing file; close that for now by clicking the "Cancel" button.

What you will be looking at is the Data window; one of three windows generally used when working with SPSS. The other two are the Output window and the Syntax window; both of which will be discussed below. For now, notice that within the Data window, each row corresponds to a case or observation and each column represents a variable. There are two displays of concern within the Data window; Data View and Variable View, accessed with tabs in the lower left corner of the Data window.

Data View is used to input and access data. The Variable View is used to specify the details of each variable in the data file. Click on the Variable View tab. You'll notice the following details can be specified for each variable. In Variable View, each row corresponds to a variable and each column corresponds to some detail or characteristic which can be specified for each variable.

Name is used to type a short or abbreviated name of the variable; this will appear as the column name when in Data View. Type allows you to specify the type of variable this is (e.g. numeric, string, date, etc.). Width refers to the column width this variable will have in the Data View. Decimals refers to how many places to the right of the decimal you would like displayed in Data View. Label is used to type a description of this variable (i.e. non-abbreviated). The Label will appear in Data View if one holds his or her cursor over the Name at the top of the column. Values are used to assign names to each value of the variable (i.e. what will each number refer to). Missing allows the user to specify how missing values are coded for recognition by SPSS. Columns allows the user to specify more than one column (in Data View) for this variable. Alignment allows the user to specify the left, center, or right alignment of data within the column of this variable. Measurement allows the user to specify the type of variable; here SPSS uses Nominal, Ordinal, and Scale (which refers to both Interval and Ratio). Role can also be used to specify the type of variable (input, target, both, none, partition, split).

An example for creating and setting up a data file.

1. Click on the Variable View tab at the bottom of the spreadsheet.
2. Click on the first row under Name.
3. Type the word “ID” (this will stand for the Identification number of each participant).
4. Press <enter>
5. Click on the cell under the Decimals column and type a zero (0).
6. Click on the cell under the Label column.
7. Type “Participant Identification”
8. Click on cell below the Measure column and select Nominal.
9. Click on the Name cell of the next variable.
10. Type “IV” (this will stand for Independent Variable [or condition]).
11. Press <enter>
12. Click on the cell under the Decimals column and type a zero (0).
13. Click on the cell under the Label column
14. Type “Condition”
15. Click on the Values cell.
16. You will have to click the definition button (…) in the cell. A new window will open.
17. Type 1 in the Value box, and then click on the Value Label box.
18. Type “Control” and click Add.
19. Repeat steps 17 – 18 using the value “2” and the value label “Experimental”.
20. Click okay.
21. Click on the cell under Measure, then select Nominal.
22. Click on the Name cell of the next variable.
23. Type “DV” (this will stand for Dependent Variable).
24. Click on the cell under the Decimals column and type a zero (0).
25. Click on the cell under the label column.
26. Type “Number Correct”.

Now, three variables are defined: the participant number (ID), the levels of the IV (IV), the number correct on the memory test (DV).

Using the Data View tab will open the data spreadsheet. It is time to enter the data. The variable names that were typed under the Name column in the Variable View should be at the top of the first three columns. In the Data View, each row represents data for one participant. Data should be entered under each variable for each participant. To enter data simply position the cursor in the appropriate cell and type the number. Pressing the “enter” key will move the highlighted position down one row. Pressing the “tab” key after entering a value will move the position over one column to the right. So, the user can either enter all the values for one variable at a time by using “enter” or all the variables for one participant can be entered by using “tab.” Now enter the following data for 12 participants with the first 6 in the control condition and the second 6 in the experimental condition. Their number correct (from the top): 10, 8, 14, 12, 11, 13, 22, 23, 22, 19, 20, 24.

Notice that when you hold the cursor over the column headings, the Label for that column is displayed.

Also notice that when you click on the Value Labels button (shown below), the Value Labels (names) are displayed instead of the Values (numbers).

2.) Open an existing data file.

One of the benefits to newer versions of SPSS is the ability to have multiple data files open at once.

In the SPSS tool bar at the top of the Data window, go to File, Open, Data..., C drive, Program Files.

Find and open the SPSS directory, then open the folder "Samples" then "English" and notice all the example data sets. Move the slider to the right and find the "carpet.sav" data file; and open it.

Now, in the SPSS toolbar at the top of the Data window, go to Analyze, Descriptive Statistics, Frequencies.

Select "Preference [pref]" and move it into the variable box; then click the OK button.

The output will be displayed in the Output window. The left side of the Output window shows all the output in outline form, which is often handy for navigating between many different sections of output. The right side of the Output window actually displays the tables and figures of the output and syntax associated with the task performed.

Notice that in the output, there is a 'Log' section above the primary output that displays the SPSS syntax. You can create a dedicated syntax file for each function or analysis you run in SPSS by clicking "Paste" instead of "OK" in the dialog box for the function or analysis you specify.

Returning to the Data window, click on Analyze, Descriptive Statistics, Frequencies...  Notice the last run is still specified. Also notice that we could have clicked paste--do that now to open the syntax window.

You'll notice the Syntax window is similar to the Output window in displaying an outline of tasks on the left and the actual syntax on the right.

Saving SPSS files is similar to most other programs. Saving data* is done from the Data window and files carry the .sav extension (e.g. dataname.sav). Saving output is done from the Output window and files carry the .spv extension (older versions used the .spo extension). Syntax files are saved from the Syntax window and carry the .sps file extension.

*As of PASW Statistics 18, you can now save data in SAS data file format.