This chapter discusses methods for presenting your data and findings in your reports.  Most of this chapter is devoted to introducing you to methods for creating and editing charts.  Then, we review ways to edit the tabular output from the various statistical procedures so that you convey just the information you need.  Finally, we show you how to copy your work from the SPSS output screen into a word-processing document (i.e. Microsoft Word).

Charts

Deciding on a Chart Type
Charting a Single Variable: DEGREE

First, let’s decide how we might display education as measured by DEGREE.  Click on “Graphs,” then “Gallery.”  The Chart Gallery is a guide that SPSS 11 provides to help you to decide which type of chart fits your needs, and to show you the steps for obtaining the chart you choose.  The Chart Gallery box is shown in Figure 9-1.

It looks like a Bar Chart would be a good way to display DEGREE.  Click on “Bar” in the Chart Gallery to see some examples of bar charts.  You will see a box like the one in Figure 9-2.

You see three types listed:  Simple, Clustered, and Stacked.  Since we are displaying only one variable, click on “Simple.”  A box will appear that shows three types of “simple” bar charts (see Figure 9-3).

It looks like Chart A, “Summaries for Groups of Cases,” is the best choice.  It shows the number of cases which fall into each category (i.e. DEGREE value).  If you click the “How To” button next to the chart, you will see the steps to follow to create the chart (see Figure 9-4).

Creating Your Chart

Lets go through the steps in Figure 9-4.  On the SPSS menu bar, click on “Graphs,” then “Bar.”  You will see the box shown in Figure 9-5.

You can see that “Simple” is already selected, because it has a black box around it.  “Summaries for groups of cases” is also already selected.  So, click on “Define” so that you can tell SPSS which variable you want to show in your chart.  Find “Degree” in the list of variables on the left, then click the arrow below “Category Axis” to enter degree into the box.  See Figure 9-6 to see how it should look.  (Note:  In this exercise, you are going to create a chart that shows the number of respondents in each category.  If you preferred to show the percentage of respondents, you would click on “% of cases” in the “Bars Represent” section.)

Click “OK” to create the chart.  Your chart should look like the one in Figure 9-7.  Notice that the chart includes a bar representing the cases having “missing” data (i.e. there is no degree level reported for those cases).

You may not be interested in showing the “missing” category.  There are two ways to exclude this category.  The first way is as you are creating the chart.  Go through the steps of creating the chart again (“Graph” “Bar,” “Define”), so that you get to the box shown in Figure 9-6.  Notice that there is an “Options” button in the lower right corner.  Click on it, and in the box that appears you will notice that “Display groups defined by missing values” is checked.  Click on it to remove the check (thereby “turning off” the option).  Click “Continue,” which will close the box, then click “OK.”  You will now see a chart that looks like the one in Figure 9-8, with no “Missing” category displayed.

What if you don’t feel like re-creating the chart?  There is a second way to exclude this bar.  You can edit the chart in the Output Viewer to remove the missing category.  Do this by first double clicking on the chart (the one that looks like Figure 9-7).  The chart will open in its own window, called the “SPSS Chart Editor.”  Then, working within that window, double-click on any of the bars in the chart.  You will see a box appear, as shown in Figure 9-9.

In the bottom right corner of that box, you will see the categories of your variable.  Click on “Missing,” and move it into the “Omit” space on the left.  Then click “OK.”  Your chart now looks like the one shown in Figure 9-8.

While your chart is still open in its own window, you can make a variety of other changes.  Try this:  Double-click on “COUNT,”  the title on the y-axis.  A box will open (see Figure 9-10) which gives several editing choices.

Delete “Count” in the “Axis Title” section, and type in “Number of Respondents.”  You can also center the axis title by clicking on the arrow next to the “Title Justification” area, and choosing “Center” from the drop down menu.  Click “OK,” and you will see the new axis title, and it will be centered.  You can make similar edits to the x-axis title by double clicking on “RS HIGHEST DEGREE.”  You will see a dialog box such as the one in Figure 9-11.

Change the title to Highest Degree, and center it.  Then, click “OK.”  You can also show the number of respondents represented by each bar.  To do this, click on “Format,” then “Bar Label style.”  You will see a box like the one in Figure 9-12.  Click on “Standard,” then “Apply All.”

Once you are finished with your edits, close the “SPSS Chart Editor.”  The chart shown in the SPSS Viewer (the output view) now reflects all the changes you made.  Now, working within this view, lets make one more change.  Double click on the chart title (where it says GRAPH).  When you do that, the title will be surrounded by a hatched line, indicating that it is ready to edit (see Figure 9-13).  Delete  the word “Graph,” and type in “Figure 1: Respondent’s Highest Degree.”  Your final product should look like the one shown in Figure 9-14.

We could have chosen other chart types to display the data in the DEGREE variable.  You can explore some of the other types by clicking on “Graphs,” then “Gallery.”  When the “Main Gallery” box opens (refer to Figure 9-1), click on the “See Also” button located near the top.  A dialog box will open (see Figure 9-15).

Highlight “Chart Galleries: By Data Structure” by clicking on it, then click “Display.”  You will see a box like the one in Figure 9-16.

You will want to chart a single variable, which is measured in categories (for the purposes of charts, ordinal variables are treated very much like nominal variables).  So, click on the square to the left of “Single categorical variable.”  You will see a new box like the one in Figure 9-17.

Clicking on any of the chart types will bring up a new box which describes the chart.  You can then click on the “How To” button to find out the steps for creating the chart you are interested in.

Charting Several Variables: Abortion Attitudes

When you are working with several related variables, you may want to show them in one chart, rather than having separate charts for each.  Let's show how responses to the questions about abortion attitudes vary with the condition under which abortion would be allowed.  Lets also make it more interesting by showing the similarities and differences in men’s and women’s attitudes.

To obtain this chart, click on “Graphs,” then “Bar.” When the dialog box opens, click on “Clustered.”  Doing this will “cluster” the data by respondents’ gender.  In the lower section, click on “Summaries of Separate Variables.”  This will allow you to show the results for all of the ‘abortion attitude’ variables.  The dialog box should look like the one in Figure 9-18.  Click “Define.”

A dialog box for entering your variables opens once you click “Define.”  Move each of the ‘abortion attitude’ variables into the larger box (the one labeled “Bars Represent.”  The order you add them into the box will determine the order they will print out in the chart.  You may want to enter them in the following order:

ABHLTH, ABRAPE, ABDEFECT, ABNOMORE, ABPOOR, ABSINGLE, ABANY.  Then, click on SEX, and move it into the space labeled “Category Axis.”  At this point, your dialog box should look like the one shown in Figure 9-19.

Notice that in the dialog box, the attitude variables are preceded with MEAN.  This tells you that the default is for the bars to represent the mean of each variable.  If we were charting interval or ratio variables, such as education or income, this would make sense.  However, the possible meaningful responses to the abortion questions are “yes” and “no.”  It seems that reporting the percentage who answered “yes” would convey the information much more meaningfully. Do this by using the “Change Summary” button.  First, click on all of the abortion variables to highlight them (you will need to hold down the Shift key while you do this).  Then, click the “change Summary” button.

The dialog box that opens gives you several choices about how you want your data to be summarized and displayed.  Recall that for the abortion variables, a “Yes” response was coded as 1, and “No” was coded as 2.  (If you aren’t sure how variable categories were coded, just look at the labels in the “Variable View.”)  We can tell SPSS to display the percent who said “Yes” by clicking on “Percent below,” and entering the number 2 in the “Value” box.  This way, only those responses coded with a value lower than 2 will be charted, which in this case, are the “Yes” responses.  When you finish, your dialog box should look like the one in Figure 9-20.  Click “Continue,” then “OK.”

Figure 9-20

Figure 9-21

The resulting chart is shown in Figure 9-21.  Clearly, this chart needs some editing!  Recall that to edit a chart, you must double click on it so that it opens in its own window.  Do this now.  The most obvious area needing attention is the legend, which is the key that designates the meaning of each bar.  It is located to the right of the chart.  Notice that some of the legend is cut off.  If we reduce the font, the entire legend will fit in its space.  Click once on the legend, then on the Chart Gallery menu bar, click on “Format,” then “Text.” Change the font size to 6 (just backspace over the number that is there, and enter 6).  Click “Apply,” then “Close.”  You should now see the entire legend.

You can also edit the way the legend is labeled.  This time, double click on the legend.  A dialog box will open, that looks like Figure 9-22.  You can see that SPSS breaks the text between Line 1 and Line 2.  Click on the variable name you want to edit, then, in the Line 1 and Line 2 spaces, edit the text until it looks the way you want it to appear.  Do this for each variable.

Figure 9-23 shows the bar chart with the legend edited, as well as other changes (review the discussion on how to edit the DEGREE chart to find out how the other edits were done).  See if you can make your chart look like the one in Figure 9-23.

Tables

Let's create a crosstabulation of SEX and FEAR.  Click on "Analyze,"  then "Descriptive Statistics," then "Crosstabs". Put FEAR in the row box and SEX in the column box (recall that in crosstabulations, the independent variable always goes in the column position).   Now click on "Cells" and select column percents, and then click on "Continue."  Finally, click on "Statistics" and select “Chi-square,” click "Continue," and click  "OK". The Output Window will appear, and your screen should look like Figure 9-24.
                  Crosstabs

Figure 9-24

The left hand frame will contain an outline of the tables in the right hand frame. Click on " FEAR  AFRAID TO WALK AT NIGHT IN NEIGHBORHOOD * SEX  RESPONDENTS SEX Crosstabulation.”  This will select that table in the right hand frame.  The red arrow to the left of the table indicates that it has been selected. Double click anywhere inside the selected table.  The border of the table will change and it will have a series of cross hatches on it. You can now edit the table. Let's say you want to change the title. Point your mouse at the title and click once. This highlights the title (see Figure 9-25).

Figure 9-25

You can now edit the title.  Delete the old title, and type in: “Table 1: Crosstabulation of Fear of Walking at Night, by Gender,” then click anywhere in the table (outside of the title).  You can click in any area of the table, and make changes.  By clicking in the various cells and deleting and/or editing contents, you can simplify the table so that it looks like the one shown in Figure 9-26.

Figure 9-26

Copying and Pasting Charts and Tables to a Document

Since you will probably be using a word processing program to prepare the report of your results, it will be useful to copy your charts and tables from SPSS into your word-processing document.  Lets start with the table we just created.  There are two ways to do this.  The simplest way is to click on the table using the right mouse button.  A small menu will appear; click on “Copy object.” Then, go to your word-processing document, and right-click where you want the table to appear.  The small menu will appear again; click “Paste.”

The second way to copy the table is by using the menu commands.  Make sure the table you want is selected (you will see the red arrow pointing to it, and the table will have an outline around it).  Click on “Edit” on the menu bar, then click on “Copy Objects.  Switch over to your word-processing document.  Click the mouse where you want to paste your table.  Then, click on “Paste Special”.  Either way you accomplish this task, your pasted table should look like the one in Figure 9-27.  Once you have pasted the table into your document, it would be a good time to add a footnote conveying the Chi-square result (i.e. p < .05).

Figure 9-27

The method for copying and pasting charts is exactly the same as the method as for copying and pasting tables.  Note that in both cases, what you are pasting into your word processing document is an object: a graphic (or picture).  This means that you can resize it to make it bigger or smaller, but you cannot edit it.

1. Make a bar chart of TRUST.  Do not show the “missing values” category on your chart.  Using Figure 9-14 as a guide, make a proper title, and edit the y-axis label.  Show the number of respondents represented by each pair (see Figure 9-12).  Copy and paste the chart into a word processing file.  Write a few sentences that describe the pattern shown in the chart.

2. Are happily married people more trusting?  Create a “clustered” bar chart which shows levels of TRUST grouped by HAPMAR (hint: the HAPMAR categories should appear along the x-axis, and the TRUST categories should appear in the legend).  This time, have the bars show the percentages of respondents in each category, rather than the number (so, click on “% of cases” in the “Bars Represent” section of the dialog box).  Be sure to exclude the “missing values” category for both variables.  Show the % represented by each bar, and make sure the chart is properly titled and labeled.  Copy and paste the chart into a word processing file.  Write a few sentences that describe the patterns shown in the chart.

3. Do a cross-tabulation of HAPMAR and TRUST.  Since HAPMAR is the independent variable, place it in the column location, and show column percentages (see Chapter 5 for a review).  Edit the table, using Figure 9-28 as a guide.  Be sure that your table is properly titled.  Copy and paste the table into a word processing file.  Write a few sentences that discusses the relationship of the  information shown in the table to the information shown in the chart you created for Question 2.