You should print out this tutorial so you will have a printed copy in your hand while you work.
From previous assignments, you have probably realized that SPSS for WINDOWS statistical operations involve two separate windows:
1. SPSS calls the window for entering and editing data the DATA EDITOR. Within the data editor, you can be in one of two views:
When you are in the DATA EDITOR, you can toggle back and forth between DATA VIEW and VARIABLE VIEW by clicking on their respective tabs IN THE LOWER LEFT CORNER OF THE DATA EDITOR SCREENS.
2. The other window is the VIEWER, or OUTPUT FILE, in which you will find all statistical results, tables, and charts you produce.
You create your data files in the DATA EDITOR. Then, if you wish, you can save the data file by clicking on File and then on Save As, as you do for any WINDOWS file.
VIEWER files (sometimes called OUTPUT FILES) are saved in the same way but under a different name.
VIEWER (OR OUTPUT FILES) HAVE A .SPO EXTENSION.
There are several ways of getting your raw data into the DATA EDITOR:
1. One way is to type the data in using the keyboard in the same way you would type data into an electronic spreadsheet. This is probably the simplest way.
2. You can cut and paste data directly from an EXCEL spreadsheet into SPSS for WINDOWS.
3. You can also use any word processor to create the file, save it as an ASCII or text file, then load it into SPSS for WINDOWS. This works well, but is a little more involved.
You have already completed several assignments in which you entered data directly from the keyboard . In this assignment, you will learn how to enter experimental/control group data into SPSS for WINDOWS, save it, and print out the data file.
For the sake of this tutorial, let us say you want to use SPSS to run DESCRIPTIVE statistics on a test you have given. Let us say you gave the test to ten people, five in an experimental group and five in a control group. Here are the scores:
Experimental group: 85 90 82 75 99 Control group: 70 66 52 71 50
Therefore, you know right away that your data file must contain at least three variables on each person:
1. a variable that contains an identification number for each person.
2. a variable that defines whether or not the person was in the experimental or control group, and
3. a variable that has that person's score in it.
Most of the time when you use a computer to do statistics, each person or CASE occupies a row (rows are in a horizontal line).
Each VARIABLE occupies a column (columns are in a vertical line).
Each variable needs to be named. The name that appears at the top of each column has a default lmit of eight characters in length, but you can also designate a longer, more descriptive variable name if you want. The longer name is called an SPSS LABEL. The LABEL can be of practically any length.
By tradition, we usually use the first column of cells for the identification number. Also, these identification numbers should be written on the raw data sheets, test protocols, surveys, etc., depending on the study.
The variable that defines the person's group is called a "dummy variable" or "break variable." The only purpose it serves is to identify which group the person is in. That is why it is called a "dummy" variable. The term "break variable" is also sometimes used, because it is this variable the computer uses to BREAK the scores into different groups.
Here is information about each of the three variables we will use for this study:
1. We might choose an SPSS name of "id" and an SPSS LABEL of "identification number" for our variable (column) that will contain the subject identification numbers we assign.
2. We might decide to name the dummy variable with the SPSS name of "group" and an SPSS LABEL of "group designation". This variable is a CATEGORICAL variable. That is, the values it contains do not designate amounts of something. Rather, they are arbitrarily chosen simply to assign which CATEGORY the subject is in. Whenever you set up a CATEGORICAL variable in your data, you need to tell SPSS what the VALUES found in that variable mean. (In this case, a "1" designates the experimental group, "2" for control group). You will see how to specify VALUES for a categorical variable in a moment.
3. We might decide to give the variable that will contain the test scores themselves the SPSS name of "test1, and an SPSS LABEL of "midterm math test."
You decide your data file should look like this:
01 1 85 02 1 90 03 1 82 04 1 75 05 1 99 06 2 70 07 2 66 08 2 52 09 2 71 10 2 50
Look at the above data.
The first column is the "id" variable.
The second column is the "group" variable.
The third column is the "test1" variable.
In a previous lesson, you learned to do a CODEBOOK
The CODEBOOK is important when you have many scores in a study. Say you had 200 subjects. That means you would have 200 test papers, plus you might have a separate demographics page for each subject. That page might contain name, age, grade, gender, etc. It would be VERY difficult, and not a good idea at all to sit down at the computer and try to enter all the variables directly from all these papers to SPSS. This invariably results in many errors.
So, what we generally do is an intermediate step. We usually first enter the scores on coding sheets. I like to use regular IBM paper (similar to graph paper). We enter all the data on the graph paper. THEN, we type the data into SPSS. The CODEBOOK refers to the columns on the graph paper, and is essential to help you find and correct errors once you have entered the scores into the computer.
Here is a codebook for our present study:
Full SPSS Variable Variable Column Name Name Key -------------------------------------------------------------- 1-2 Identification number id continuous 3 BLANK 4 group designation group 1=experimental 2=control 5 BLANK 6-7 midterm math test test1 continuous --------------------------------------------------------------
Of course if you know you have more than 99 subjects, or if you think you may add subjects later so that the total exceeds 99, you would leave 3 columns on the graph paper for the id variable. Also, if it is possible that someone could earn a score higher than 99 on the test1 variable, you would leave three columns for the test1 variable.
The first 2 columns on the graph paper contain identification numbers, and will appear in the first SPSS column of cells.
The 3rd column on the graph paper will be blank.
The 4th column on the graph paper will contain the group variable (either a 1 or a 2) and will appear in the second SPSS column of cells.
The 5th column on the graph paper will be blank.
The 6th and 7th column on the graph paper will contain the test1 scores, and will appear in the third SPSS column of cells.
Once again, here are our scores:
01 1 85 02 1 90 03 1 82 04 1 75 05 1 99 06 2 70 07 2 66 08 2 52 09 2 71 10 2 50
Here is how to enter these scores into the SPSS computer program. The following instructions are for the computer lab in the College of Education, but of course you can use your own computer if you have SPSS installed on it.:
_____1. Turn on the main power switch on the computer if it is not already on (mounted underneath table you are sitting at).
_____2. Wait for Windows to load.
_____3. Double-click the SPSS for WINDOWS icon. The program will open. If you see the screen that looks like a spreadsheet, you are ready to type in data. If you get a menu, choose TYPE IN DATA.
_____4. Enter the data. Note that you do not need to enter zeroes in front of any numbers. When you are finished, your data should look like this:
That is fine, but you do not have any variable names. It is time to name the variables.
_____5. Look at the lower left corner of your DATA EDITOR screen. You will see two tabs there. Click the one that says Variable View. When you do, you will see this:
_____6. Click on the first cell in the Name column and type in id to replace var00001. Do the same thing for group, and test1.
_____7. Now, it is time to specify SPSS VALUES for any CATEGORICAL VARIABLES. The only categorical variable in our data set is the group variable. We need to let SPSS know that a 1 in this column means experimental group, while a 2 there means control group.
To do that click once in the second cell under the Values column (this is in the group row). A small gray square with three dots in it will appear in this cell. It should now look like this:
_____8. Click once on the gray square.
_____9. The VALUE LABELS box will appear:
_____10. Type a 1 in the Value: field, type experimental in the Value Label: field, and hit the Add button.
_____11. Type a 2 in the Value: field, type control in the Value Label: field, and hit the Add button.
_____12. Click the OK button. You will be returned to the Variable View screen. Now, click the Data View tab in the lower left of your screen to return to the Data View screen and see your data. It should look like this:
Now, you should SAVE your data file on a diskette, so that if you need to use it in the future, you will not have to enter the data over again. To save an SPSS data file:
_____1. Click on File and Save As:
_____2. Select the A drive and name the file spssdatalesson.sav. (Remember that all SPSS data files must end with the .sav extension.)
_____3. Now, print out the file. At the top of the printout, put your name and the following heading:
ASSIGNMENT #3 - assign3.exp.con.groups.html
STATISTICS COURSES TAUGHT BY CLEB MADDUX PAGE.
CEP640 Page - Educational Measurements and Statistics
CEP740 Page - Advanced Educational Measurements and Statistics
Click here to go to the page for CEP741 - Applied Research Design and Analysis in Education I