Summary Statistics

The first part of any statistical analysis involves summarizing & describing the data. This may be achieved through the use of tables & graphs. The data file alcohol.sav will be used with SPSS to show some examples of descriptive statistics. All summary statistics in SPSS may be accessed through the Statistics menu option.

Frequencies

This procedure shows each distinct value of a variable, together with counts of the number of times each value occurs. These counts are known as frequencies.

Figure 6

In SPSS frequencies are obtained following the Statistics option, by choosing Summarize and then Frequencies (Figure 6). Variables are selected from the left box by clicking on the name, then the right arrow in the centre of the Frequencies window, the variable name will then appear in the right box. In mage (mother's age) has been selected and birthwt (birthweight) is about to placed in the right box (Figure 7).

Having clicked on OK, the results from the frequencies analysis will now be displayed in the !Output window (Figure 8).

Note: mage has been assigned value labels via the variable definition procedure (see entering data).

Figure 7

Figure 8

SPSS also gives a Percent column, giving the percentage of the total data having each value; a Valid Percent column, giving the percentage of the valid data having each value, and a (Cumulative) Cum Percent column, giving the percentage of the total data having a value less than or equal to each value. Since there are no missing values, the Percent and Valid Percent columns are identical.

Cross - Tabulations

It is often useful to know how data varies over certain groups. In the current dataset we can look at how alcohol consumption (alcd) during pregnancy varies over Maternal age (mage).

In SPSS, Statistics, Summarize are again selected then the Crosstabs option. This will give a Crosstabs window, where variables can be selected (similarly to Frequencies) to go into either the Row(s) or Column(s) dimension of the table.

A more informative table is given by including Row percentages. These may be included by clicking on Cells... and then on the Row option in the Percentages box (Figure 9).

Figure 9

Figure 10

The table in suggests that more women say they drink whilst pregnant, with increasing age (Figure 10).

Also included in the !Output window is the SPSS command syntax for producing the output you have shortcut by using the menu features.

If you click on either of the keys , the cursor will be placed at either the start of the previous or the next command syntax.

If you click on either of the keys , the cursor will move up or down by 1 SPSS defined page.

Graphs

One of the simplest ways of showing data is by use of an appropriate graph. For example, the long list of values given in a Birthweight (Bwght) frequency table are not easily envisaged. However, by choosing the Charts... option in the Frequencies window, Histogram may be selected (Figure 11).

Figure 11

The !Output window will again give the frequency table. To Examine the histogram, click on Window, and then on the new Chart Carousel option.

To edit and to save the histogram, click on the button in the Chart Carousel window. You should then have something similar to Figure 12.

Figure 12

You may wish to change the axis scaling, titles, color, and shading of the graph. The axis scaling may be changed by double clicking on the relevant axis, and following the information given in the subsequent windows. Similarly all the text may be changed by double clicking on the chosen text and then making the necessary amendments.

By clicking on one of the bars you may then change the color and shading:

This will enable you to change the color, by clicking on a new color and then clicking on Apply.

This will enable you to change the shading, by clicking on a new style of shading and then clicking on Apply.