Argument-Driven Inquiry as a Way to Help Undergraduate Students Write to Learn by Learning to Write in Chemistry

Abstract

This exploratory study examined how undergraduate students’ ability to write in science changed over time as they completed a series of laboratory activities designed using a new instructional model called argument-driven inquiry. The study was conducted in a single section of an undergraduate general chemistry lab course offered at a large two-year community college located in the southeast USA. The intervention took place over a 15-week semester and consisted of six laboratory activities. During each laboratory activity, the undergraduates wrote investigation reports, participated in a double-blind group peer review of the reports, and revised their reports based on the reviews. The reports written during each laboratory activity were used to examine changes in the students’ writing skills over time and to identify aspects of scientific writing that were the most difficult for the undergraduates in this context. The reviews produced by the students during each report were used to evaluate how well undergraduates engage in the peer-review process. The results of a quantitative and qualitative analysis of the reports and reviews indicate that the participants made significant improvements in their ability to write in science and were able to evaluate the quality of their peers’ writing with a relatively high degree of accuracy, but they also struggled with several aspects of scientific writing. The conclusions and implications of the study include recommendations for helping undergraduate students learn to write by writing to learn in science and new directions for future research.

Keywords:

• Chemistry education
• Laboratory
• Learning activities
• Writing in science
• Argument

Acknowledgments

We would like to thank Carol Zimmerman for helping us to develop the lab activities used in the study and for allowing us to conduct this study inside her classroom. We also would like to thank Katrina Kurtek and Jon Swanson for scoring the investigation reports and peer-review guides.

Notes

The authors found statistically significant improvements in the undergraduates’ use of references on the second paper the students submitted.

The maximum value of κ is 1.0, indicating perfect agreement; a value of 0.0 indicates that the observed agreement is the same as that expected by chance, and the minimum value of κ falls between −1.0 and 0.0. Landis and Koch (1977) suggest that values of κ above 0.60 show good to excellent agreement between the two raters’ scores, and values of ≤0.40 show \stop fair to poor agreement.

A hypothesis is a tentative explanation (or a claim). A prediction is an expected outcome that is generated from a hypothesis and a proposed test of that hypothesis (see Lawson, 2002, 2003).

The students in the non-ADI sections of the course complete twice as many lab activities but only write short answer responses to a series of ‘analysis questions’ at the end of each lab activity.