Baker, Stokes S. , Roytek, Margaret A. .
Inquiry-based laboratory investigations with transgenic Arabidopsis thaliana containing a low-temperature induced reporter gene are effective in teaching the scientific method in general biology.
Transgenic Arabidopsis thaliana containing a low temperature induced chimeric reporter gene were used in inquiry-based laboratory activities to teach the scientific method. In the fall of 2003, one course section of general biology laboratory was taught using a confirmatory laboratory curriculum (control group). Three course sections (experimental group) performed three different skill-building activities. First, they learned how to use confidence intervals to evaluate stochastic data by comparing the sizes of fossil casts taken from two different rock strata. Second, students were taught histochemical methods to detect low temperature induced expression of the chimeric cor15a-gus gene in Arabidopsis. Third, they used confidence intervals to compare the growth of cold acclimated and non-acclimated Arabidopsis populations after freezing stress.
Following these activities, each team investigated the effects of different abiotic stresses on cor15a-gus gene expression (typically induced by
salt and drought) and related this expression to any resulting gain of freezing resistance (e.g., drought stressed plants are freezing resistant).
Prior to performing experiments, each student team wrote a short experimental plan and described possible outcomes, and possible interpretation of the outcomes. Following the experiment, each student wrote a formal laboratory report that included graphs with confidence intervals.
To assess the effectiveness of inquiry-based methodologies, students were given an unannounced test at termís end; whereby, they evaluated stochastic data and designed an experiment. Student responses to the evaluation were assessed using a randomized blind experimental format. χ2 analysis showed that those students given a statistics-based inquiry curriculum were more likely to analyze data involving repeated measurements by calculating a mean (p = 0.0119), were more likely to try some method to quantify variability (p = 9.4 x 10-6), and were more likely to correctly deduce that differences in means coming from samples containing large overlap are not significant (p = 0.0122).
1 - University of Detroit Mercy, Biology Department, P.O. Box 19900, Detroit, Michigan, 48219-0900, USA
2 - University of Detroit Mercy, Education Department, P.O. Box 19900, Detroit, Michigan, 48219-0900, USA
Presentation Type: Paper
Location: Superior A (Cliff Lodge)
Date: Monday, August 2nd, 2004
Time: 9:30 AM