Lab 2 Introduction to Figures[1]
Last class we discussed how to communicate useful descriptive statistics to your reader. This lab we will play with a couple of types of graphs. We will learn how to display graphs that are appropriate for various analysis when we discuss each one. For today, I just want to discuss graphs that represent the distributions of single variables.
Circle (pie) graphsare used to represent percentages and proportions of categorical (generally nominal) variables. They should not be used when the information can more easily be understood by simply given the percentages in the text. Pie charts should not be used when there are only one or two categories within the variable. For example, do not give a pie chart to represent the percentage of males and females in your sample.
For the sake of readability, no more than five variables should be compared in a single pie graph. The segments should be ordered very strictly: beginning at twelve o’clock, order them from the largest to the smallest, and shade the segments from dark to light (i.e., the largest segment should be the darkest). Lines and dots can be used for shading in black and white documents. (Purdue OWL)
Simple Bar graph or Histogram
Histograms are used for categorical variables (nominal or ordinal). Unlike pie charts, the data set does not need to add up to 100%. Histograms are ideal for data that has a naturally occurring order. The space left between the bars indicates that the variable is categorical and not continuous.
Line graphs
Line graphs are used to represent continuous variables. In general, when we graph continuous variables we convert them to categories so that each data point represents a range of scores. The line graph below graphs each age (in years) such that the percentage represented includes all individuals having arrived at the age indicated through to their next birthday.
Lab 2
Your assignment is to produce three figures; a pie chart, a histogram and a line graph. You can use the same data set you used for Lab 1.
General rules regarding Figures
Text
The font in figures, their titles and captions should be consistent with the font in the rest of your paper. The font in a figure should be in a san serif font (such as Calibri, Helvetica, Arial, or Futura). The font size must be between eight and fourteen point.
Each figure is included in the manuscript on a separate page at the end of the manuscript (after Figures but before Appendixes). The margins on these pages must be at least one inch on all edges. Be sure to refer to your figure in the text. When the publisher/editor places the figure in the final printed version, they will place it as close as possible to the first reference to the figure.
Each figure is numbered. When referring to a figure in the text of the manuscript, refer to it by its figure number (i.e., see Figure 1). The F in figure is capitalized. I think of this the same way I think about capitalizing my own name.
Pie Chart
Select a nominal variable from last weeks lab. Be sure that it has between three to six categories. If it has more than 6, use the RECODE (into same variable) option (see lab 1) to combine smaller categories.
From the main menu in SPSS (at the top), select Graphs and then “Chart builder”.
From the Gallery menu select Pie/Polar and drag in onto the preview area. Then movethe variable name onto the margin labeled “slice by”.
From the statistics menu select “percentage
(default is “count”) and for set parameters select Grand Total. Apply
“The segments should be ordered very strictly: beginning at twelve o’clock, order them from the largest to the smallest, and shade the segments from dark to light” (i.e., the largest segment should be the darkest; Purdue OWL). To do this first, click on the “edit properties menu, select Angle-Axis 1 (polar interval 1) and verify that the clock position is set at 12:00 and clockwise is selected. Next go to Group Color (polar interval 1) and put your category levels in order from largest to smallest.
Select Ok and your pie chart will be “served up” in an output file. Double click on the pie chart and it your chart will open in a separate editing window.
From the “Elements” menu, select data label mode, and then click on each slice of your pie. This will display the percentages in each slice.
By clicking on the legend, you can edit the order (select categories), text style and colors of the slices (select text style menu).
Click anywhere on white space and a menu will open giving you the option of copying your chart. Move it over to a Word Document. This color version is appropriate for Poster presentations and PowerPoints. To make a black and white, APA version, return to your editor screen, select “File”, “Apply Template” and choose APA. This will produce a black and white version. Cut and paste this version into a Word document.
Each figure is given a figure caption. These appear BELOW the figure. Captions serve as a brief, but complete, explanation and as a title. For example.
Figure 1. Percentage of respondents’ favorite color. Green was the most frequently selected color.
Now that you have some practice with graphics, explore while making a histogram (bar chart) and a (line graph).
What to hand in.
Three (Pie, histogram and line) APA formatted figures (each on their own page) with appropriate figure captions.
[1]Any non-text (i.e., graphic) material included in the manuscript are referred to as figures. This could include graphs, charts, maps, drawings or photographs.