Entering Data

It is customary in SPSS for rows to be used to identify cases (or subjects) and columns to identify variables (or measures). (If necessary, review the discussion of ID as a variable.) You will not need to label the rows, but it is critical that variables be clearly and succintly labeled. To do this, position the cursor at the top of the column you wish to label and doubleclick. Notice that this takes you out of Data View (where raw data is entered and stored) and into Variable View (where descriptive information on your variables is input). This is a wise and valuable prompt, reminding you of the importance of entering the requested data on each variable as it is added to the data set. Toggle now to SPSS and try clicking on a column header.

In Variable View use your cursor arrow key to move from column to column. In the first column (Name), enter a meaningful abbreviation of your variable e.g., "read" for "reading score" or "tchid" for "teacher identification number". (Depending on the version of SPSS you are using, you may be limited in the number of characters available for naming variables. Prior to release of version 12.0 variable names were limited to no more than 8 characters. In general, do not use spaces, symbols, or capital letters in variable names. Variable names should not begin with numbers.) Since this is how the variable name will appear in the column header in Data View, you will want it to be easily recognizable and identifiable.

In the second column (Type) you determine important characteristics of how you wish the data to be displayed in Data View. Notice that upon entering this cell a small gray box with three dots in the center appears on the righthand cell margin. This happens every time you, the researcher, have to make a decision. Click on the gray icon to see which decisions you have to make here. Point and click to the type of data you have for the variable in question (is it numeric, calendar dates, dollar or other currency?). Another decision you can make here is the width of the column in which the data points for the variable are to be entered. You need less width to enter "1" than you do to enter "$12,000". Finally, you can decide here how many decimal places you'd like each data point to contain. Do you want your "1" to be just "1"? Or do you want it to be "1.0" or "1.00", etc.? Once you have made these decisions, click on the OK button to continue. Notice that the decisions you made in column two automatically default into columns three (Width) and four (Decimals).

Column five (Label) gives you an opportunity to succinctly describe (maximum of 40 characters) what the variable is. This is especially important when analyzing a data set with many variables. It may help you remember that "read" is a score on an achievement measure rather than, say, the rate or level at which the subject reads. The description you enter here will appear when you position the cursor over the variable name in Data View.

The Value column (column 6) is critical when working with categorical (nominal) data. It functions essentially as the key or legend to the meaning of values assigned to a variable by you, the researcher. For example, a common categorical variable is gender. After you have decided how you wish to code female (perhaps "1") and male (perhaps "2" ), you enter that information in the Value column. To do that, click on the small gray box at the right cell margin. In the "Value" field, enter "1" . Tab to the "Value Label" field and enter "female." Tab once more to the "Add" button and hit enter. Now repeat the process for males, with "2" identifying this group. Hit the OK button and you will see these values, as assigned by you, appear in the Values column. You can imagine how valuable this information is when you are working with data that is not your own!

The next two columns (Missing and Column) are not critical at this stage and are not discussed here.

The ninth column (Align) is self-explanatory. Placing the cursor in the cell prompts a pull down menu to appear, and allows you to determine where in the cell you wish each data point to appear (left, right, or center).

The final column in Variable View, Measure, asks you to identify the type of data captured by the variable. That is, here you classify the data as nominal (an artificial identifier indicating membership in a group or category), ordinal (rank order data which gives information on the relative order within a data set), or scale (interval data, where the units between numbers are equal, or ratio data, where the scale has equal units and a meaningful zero point). Use the pull down menu to make the appropriate selection.

At this stage you have identified all the major elements of your first variable. If you have not followed in SPSS through to this stage, it is suggested you do so now. Click on the Data View tab at the bottom of the file to see your results. You should immediately notice the new name of the variable as determined by you appears in the column header.

Now you are ready to enter the data on this variable. Position the cursor in the first cell of the column and enter a series of make believe numbers or, if you have them, your actual data points. You will see they appear exactly as you determined they should. Practice naming variables and entering data for them. As a way to test your skills and to fully explore the possibilities, be creative when making your decisions. You may wish to purposefully make mistakes and see if you can correct them.