WINKS Manual -- Chapter 4, Part 1

WINKS Online Manual

Chapter 4 Part 2

Graphs and Charts

This section describes many of the plots you can produce in WINKS. Some plots are produced as a stand alone procedure, and some plots are produced as a result of a specific analysis. The plots described in this section are stand alone plots. Plots specific to a procedure are described in the section for that feature.

Histogram/Stats

The Histogram/Stats option in the Graphs menu produces the same output as the Detailed Statistic option in the Descriptive Statistics option. Once the descriptive statistics output is displayed, select the GRAPH button to display a histogram of the data. The following plot is displayed:

Figure 4.1

This plot and options were previously described in Chapter 1.

Normal Probability Plot

Another plot available in the Detailed Statistics option is the Normal Probability Plot. To display this plot first display the Graph from the Detailed Statistics screen, then click on the “Distribution button” (second from left, next to Options button). This button causes the display to cycle through several views of the data, including a probability plot, as pictured in Figure 4.2

Figure 4.2 Normal Probability plot.

XY Plot (Scatterplot)

An XY plot (scatterplot) displays the relationship between two variables. Such a plot is helpful in determining if two variables are related, and if the relationship is linear (a straight line), curvilinear, or something else.

To display an example X-Y scatterplot, follow these steps:

Step 1: From the Analyze pull-down menu, choose the Graphs option (or click on the XY-plot icon)

Step 2: Select X-Y Scatterplot. If a database is not open, choose to open the EXAMPLE database. (Or whatever database you are using.)

Step 3: Select two fields to plot. Optionally, choose a grouping field. For the data values select Time1 and Time2 and select Group for the Group field. The following plot is displayed:

Figure 4.3 Scatterplot with Groups

When you display a scatterplot, you can choose Options from the menu to select items to customize your plot. One of the options is to display a regression line. When you choose this option, a line based on a linear fit of the data will be displayed on the plot.

If you do not already have your data in a file, choose New Database from the FILE menu. Then, select the database type you want to create (usually "X-Y Chart/Scatterplot"), enter and save the data, and perform the analysis as in the above example.

For example, suppose you have height and weight data for two groups, male and female. You would create a database with Sex as the GROUP field and Height and Weight fields as the VALUE fields. An outline of this database is shown below:

SEX    HEIGHT WEIGHT
----   ------ ------
M       70    202
M       65    145
M       72    188
:       :     :
F       62    122
F       59    112

Using this database, you would choose HEIGHT and WEIGHT as the fields to plot, and SEX as the grouping variable.

Correlation Matrix Graphs

Correlation Matrix graphs allow you to examine the relationship between several pairs of variables at once. To create this graph, select Analyze/Graph Charts/Correlation Matrix Graphs. Select the variables you want to compare and a matrix of graphs will be created, as shown below: (Comparing TIME variables in the EXAMPLE database.)

Figure 4.4 Matrix of Correlation Graphs

By Group Plot

By Group Plots allow you to compare distributions of data by groups. Select Alayze, Graphs/Charts, and By Group Plots. The following example compares AGE by GROUP in the EXAMPLE database.

Figure 4.5 Group comparison plot

First Impression X-Y Scatterplot

The database structure for this plot is the same as for the analytic XY-Scatterplot. However, this plot does not allow a grouping variable. To display this version of the scatterplot, select the XY plot option from the First Impression Gallery. Once a First Impression chart is displayed, press the right mouse button to display menu choices, including printing, cpaturing, and modifying display options for the chart.

First Impression Bar Chart

A bar chart visually displays the number of counts by group. Thus, a database for this graph consists of a grouping variable (label) and counts. To display a bar chart, follow these steps:

Step 1: Choose Open Database from the FILE menu. Select the BARCHART database.

Step 2: From the Analyze menu, choose the "First Impression Gallery" option from the Graphs/Charts Statistics sub-menu. (or click on the Charts icon)

Step 3: Choose Bar Chart from the First Impression Gallery.

Step 4: Select one or more data values and a label field. For this example, select VAR1 and VAR2 as data fields, and Label as the label field. Optionally, you can right click and change the chart features such as Plot/3D. Figure 4.6 illustrates a bar chart:

Figure 4.6 First Impression Bar Chart

If you do not already have your data in a file, choose New Database from the FILE menu. Then, select the database type you want to create (usually "Simple Bar Chart" or one of the other Bar Chart options ), enter and save the data, and perform the analysis as in the above example.

Creating a First Impression Pie Chart

A pie chart is created from a list of counts. A database for a pie chart should contain a Label field and a Value field. The database is similar to a barchart database. To display an example pie chart, follow these steps:

Step 1: Choose Open Database from the FILE menu. Select the BARCHART database.

Step 2: From the Analyze menu, choose the "First Impression Gallery" option from the Graphs/Charts Statistics sub-menu. (or click on the Charts icon)

Step 3: Choose Pie Chart from the First Impression Gallery.

Step 4: Select one data value and a label field. For this example, select VAR1 as the data field, and Label as the label field. Figure 4.7 show a 3D Pie Chart.

Figure 4.7 First Impression 3D Pie Chart

Creating a First Impression Line/Area Chart

A line, area or time series chart is useful in examining data that are time related, such as profit by month, etc. The X (bottom) axis is assumed to be "time". This plot can also be used as a sequence plot, a plot of the response values in the order in which they are taken. This can be useful to check for trends, perhaps undesirable, in data gathering. A sequence plot showing a trend may call into question the randomness of the sample of observations. On this plot, the points are connected for visual clarity.

To display a time series plot, you should have a database containing one or more VALUE fields and an optional LABEL field. For example, follow these steps to display a chart:

Step 1: Choose Open Database from the FILE menu. Select the EXAMPLE database.

Step 2: From the Analyze menu, choose the "First Impression Gallery" option from the Graphs/Charts Statistics sub-menu. (or click on the Charts icon)

Step 3: Choose Line/Area Chart from the First Impression Gallery.

Step 4: Select one or more data values. For this example, select TIME1 and TIME2 as the data fields.

If you do not already have your data in a file, choose New Database from the FILE menu. Then, select the database type you want to create (usually "Line/Area Chart"), enter and save the data, and perform the analysis as in the above example.

Continue to Chapter 4. Part 3. (t-tests and ANOVA.)