Choose the most effective graphic - Presenting evidence in tables and figures - Research and writing

A manual for writers of research papers, theses, and dissertations, Ninth edition - Kate L. Turabian 2018

Choose the most effective graphic
Presenting evidence in tables and figures
Research and writing

When you graphically present data as complex as in that paragraph, you have many choices. The simplest and most common are tables, bar charts, and line graphs, each of which has a distinctive rhetorical effect.

✵ ▪ To emphasize specific values, use a table like table 8.2.

✵ ▪ To emphasize comparisons that can be seen at a glance, use a bar chart like figure 8.1.

✵ ▪ To emphasize trends, use a line graph like figure 8.2.

Figure 8.1. Changes in family structure, 1970—2010

Figure 8.2. Changes in family structure, 1970—2010

While each of these forms communicates the same data, readers respond to them in different ways:

✵ ▪ A table seems precise and objective. It emphasizes individual numbers and forces readers to infer relationships or trends (unless you state them).

✵ ▪ Both charts and graphs emphasize an image that communicates values less precisely but more quickly than do the exact numbers of a table. But they differ:

o ▪ A bar chart emphasizes comparisons among discrete items.

o ▪ A line graph emphasizes trends, usually over time.

Choose the graphic form that best achieves the effect you intend.

Your choices also depend on your experience. If you’re new to quantitative research, limit your choices to basic tables, bar charts, and line graphs. Your computer software may offer more choices, but ignore those that you aren’t familiar with.

If you’re doing advanced research, readers will expect you to draw from a larger range of graphics favored in your field. In that case, consult table 8.7, which describes the rhetorical uses of other common forms. You may have to consider more creative ways of representing data if you are writing a dissertation or article in a field that routinely display complex relationships in large data sets.