Simple Computer Graphing Assignment Becomes a Lesson in Critical Thinking1

NACTA Journal, Jun 2004 by Leonard, Jonathan G, Patterson, Thomas F

We have learned that to analyze, interpret and communicate information through a graph is both a "science and an art" in critical thinking. Faculty in the College of Agriculture and Life Sciences at the University of Vermont require all first year students to take a semester-long course in information technology, which includes a major section on data interpretation and graph construction. Through pretesting and a comprehensive analysis of graphing assignments, it was determined that first year students have difficulty constructing proper graphs. Common graph types and their usage are discussed. A frequency ranking of student graphing errors indicates that graphing problems are related to a combination of factors involving the software, the student, and the instruction. Suggestions for improving instruction include having students understand the data they intend to graph and the message they wish to communicate, and visualize their graph before using software to create the graph.

Introduction 4

Some could argue that the agricultural revolution was made possible by talented and detailed record keepersthose who kept meticulous track of each seed planted, each animal domesticated, the yield of each variety tried. The Bible, Near Eastern monuments, and Chinese, Greek and Roman writings all serve as historical testaments to the development of agriculture (Heiser, 1990). Modern agriculture could not exist without record keeping; indeed the computer is essential in every aspect of today's agriculture, from research to production to distribution. Mendel's hand-kept notes have been replaced by massive data sets that can only be handled by today's sophisticated information technology. While the challenge of storing raw information has largely been met, understanding, interpreting and communicating the meaning of these data is a different story.

Faculty in the College of Agriculture and Life Sciences (CALS) at the University of Vermont (UVM) have adopted a core curriculum (Patterson et al., 2001) that includes a semester-long course requirement in information technology for all first year students, so they may apply and build on these skills during their subsequent years at UVM and beyond. One major objective of the information technology course (AGRI 195 Foundations: Information Technology) is to teach students to accurately analyze, interpret and communicate data through graphs.

UVM and our Students

UVM is part of the land-grant college system (Justin Morrill was Vermont's Senator). The College of Agriculture and Life Sciences (CALS) has about 750 undergraduate students, 70 graduate students, and 70 faculty in six academic departments (Animal Sciences, Botany and Agricultural Biochemistry, Community Development and Applied Economics, Nutrition and Food Sciences, Plant and Soil Science, and Microbiology and Molecular Genetics [http://www.uvm.edu/cals/]). Every fall for the past five years, CALS has enrolled between 150 and 200 first year students.

Most of our undergraduate students enter CALS with considerable experience using personal computers, especially Microsoft Office programs including Word� and Excel�. Results from a "knowledge quiz" that is administered during the first weeks of the semester show that out of 148 first year students, 87 (59%) said they were familiar with Excel�, and 22 others (15%) said they were familiar with other spreadsheet and graphics software packages. Only 39 (26%) claimed they were not familiar with spreadsheet and graphics software. These summary statistics of first year students are similar to those reported by the University of Arkansas (Johnson et al., 2002). The majority of our students rated themselves between beginner and intermediate when asked how "proficient" they were at spreadsheets and graphics. This self-rating was supported when students were tested on specific questions about proper graphing techniques. Upon review of AGRI 195 graphing assignments using Excel�, a very high percentage of our students made significant graphing errors.

These errors included choosing the wrong type of graph for their data, having a bar (in a bar graph) or slice (in a pie graph) for a total when it was inappropriate, and trying to include too much data on a single graph. Out of the 134 graphing assignments we analyzed, only three were without significant problems!

We were interested in finding out what were the most common graphing errors that first year undergraduates make in CALS and how we could modify our teaching to address these problems.

Common Graph Types

Graphs give life to static spreadsheet tables and make data more interesting and useful by bringing to light patterns in the data that are hidden or obtuse in the spreadsheet format. Graphs show trends and relationships among data that are often lost in the obscurity of a table of numbers (Figure 1)

The type and amount of data, and the story the graph maker wants to tell, will determine the kind of graph to be chosen. What follows is a brief summary of common graph types and their uses. Details of graph construction and usage are found in texts by Cleveland (1985), Henry (1995), Schmid & Schmid (1979), and Wallgren et al. (1996).