Allometry in Social Insects


Graduate students: Kenneth McKenna and Richard Gawne

Faculity instructor: Fred Nijhout

Class: BIO 546L Entomology


Undergraduate education in the life sciences often emphasizes the memorization of important concepts, with examinations functioning as a means of gauging the extent to which a student has mastered these ideas, and can use their knowledge to solve basic problems. However, as anyone who has actively participated in scientific research will tell you, "real" biology bears little resemblance to the scripted assignments most students are familiar with. Biological research almost always involves the analysis of raw data, which tends to be an extraordinarily messy affair.

The aim of our Data Expeditions course was to give Duke undergraduates a window into how science is actually done. Over the course of two class periods, students in BIO-547L (Entomology) were given the opportunity to explore an unpublished dataset on social insect morphology, using the statistical analysis program JMP. The dataset contained four different species of social insects, and the students were asked to compare morphological relationships between and within species.

Students were introduced to the concept of morphological allometry and its significance in social insects. They were then given a basic introduction to data analysis in JMP. The students were put into three groups and asked to develop their own hypotheses, and to test those hypotheses by manipulating the data provided to them. This gave them hands-on experience with exploratory data analysis and hypothesis testing. Each group utilized different aspects of the JMP software to elucidate significant differences within and between species. Among other things, students analyzed correlation matrices and log transformed regressions, revealing that caste systems evolve drastically different morphological relationships. In all, the students came away from this Data Expeditions course with a much better understanding of how biology really works. 




Data Expeditions