FACULTY NOTES

The "step one-step two" approach is explained in more details in the "Interpreting Figures and Tables" essay. As the essay explains, students often have difficulty interpreting figures because they do not realize that understanding figures takes time. This approach slows them down and requires them to pay attention to axes and other aspects of a figure.

Students should read the background material and then work in pairs or small groups to interpret the figure. They should be able to figure out that the graph shows that the number of aggregations (feeding rate on large prey) was (significantly) higher where birds were excluded than where they were present. Thus, the birds steal food the ants would otherwise get, making the relationship parasitic.

The "paired difference" in the figure shows essentially the same thing as comparing the means from the control and exclusion treatments, but is more representative of what the statistical analysis (a paired t-test) was testing. The researchers also found that the cost of the birds to E. burchellii increased proportionally to the number of birds in a flock (see Fig. 2 in the third section of the main Figure Set table). An alternate conclusion from this figure is that ant-following birds are eating the ants themselves, reducing the number of aggregations. According to Willis and Oniki (1978), ant-following birds do not deliberately eat army ant workers, although they may occasionally incidentally consume a few that are attached to something else they are eating.

Students who have been taught that parasites, such as tape worms, are invasive organisms may not recognize resource competition as a type of parasitism. Definitions of parasites differ; for example, in his ecology text Molles (Ecology: Concepts and Applications. 3^rd edition, 2005. McGraw Hill Higher Education, New York, NY) defines parasites as organisms that live on the tissues of their host, reducing fitness. A discussion about resource competition and parasitism as defined by Molles will help students work through their confusion.

Students may not be familiar with box and whisker plots. If they are not, explain that box plots are designed to visually represent the dispersion of data and is therefore a graphical summary of the data. Quartiles are used to divide the data into four groups, each containing 25% of the values. The box contains the "middle" 50% of the data, and the vertical line within the box indicates the median point. Each “whisker” shows 25% of the data and therefore the extremities of the whiskers are the minimum and maximum values.

For assessment, you can have students turn in a written explanation of the figure or have them (individually) answer a short essay question, such as:

In this experiment the researchers always ran the control first, followed by the exclusion. What is the significance of the control treatment always being first in their paired trials? Is this the best design of an experiment? How does it affect your interpretation of their results?

In my experience, students generally have no difficulty determining the nature of the relationship from the figure. Questions about the experimental design, such as the one suggested for assessment, pose more difficulty. Some students may frame their answers in terms of "realism" of the experiment, rather than the statistical or logical appropriateness of the design. For example, in nature, "birds don't suddenly depart from an ant swarm," or, "birds aren't always present first, then absent." Class responses to this question can be used as an opportunity to discuss the significance of randomization and independence in experimental design.