Advan. Physiol. Edu. 27: 26-33, 2003;
doi:10.1152/advan.00028.2002
1043-4046/03 $5.00
ADV PHYSIOL EDUC 27:26-33, 2003
© 2003 American Physiological Society
TEACHING IN THE LABORATORY
INQUIRY-BASED LABORATORY COURSE IMPROVES STUDENTS’ ABILITY TO DESIGN EXPERIMENTS AND INTERPRET DATA
Marcella J. Myers1 and
Ann B. Burgess2
1 Department of Biology, College of St. Catherine, St. Paul, Minnesota 55105
2 Biology Core Curriculum, University of Wisconsin, Madison, Wisconsin 53706
 |
Abstract
|
|---|
We redesigned our intermediate-level organismal physiology laboratory course to center on student-designed experiments in plant and human physiology. Our primary goals were to improve the ability of students to design experiments and analyze data. We assessed these abilities at the beginning and end of the semester by giving students an evaluation tool consisting of an experimental scenario, data, and four questions of increasing complexity. To control for nontreatment influences, the improvement scores (final minus initial score for each question) of students taking both the laboratory and the companion lecture course were compared with those of students taking the lecture course only. The laboratory + lecture group improved more than the lecture-only group for the most challenging question. This evidence suggests that our inquiry-based curriculum is achieving its primary goals. The evaluation tool that we developed may be useful to others interested in measuring experimental analysis abilities in their students.
Key words: college science instruction; assessment of learning; student-designed experiments; evaluation tool
|
Introduction
|
|---|
There now is a great deal of evidence that student achievement, persistence in science courses, and attitudes toward science are enhanced by having students work in small groups on appropriate intellectual tasks (5). Furthermore, a large number of reports summarizing the state of science education emphasize that inquiry-based approaches are essential (1, 3, 4). The National Science Education Standards (3) states, "Inquiry is central to science learning. When engaging in inquiry, students describe objects and events, ask questions, construct explanations, test those explanations against current scientific knowledge, and communicate their ideas to others."
Even though many of us are convinced that inquiry-based laboratories improve students’ critical thinking skills and their understanding of the process of science, it is difficult to obtain data that demonstrate this (6). We describe here our laboratory course and the tool that we used to evaluate students’ ability to analyze data and experimental design. Our course structure is similar in many ways to the physiology course based on student-designed experiments described by Kolkhorst et al. (2), although we arrived at it independently. However, our assessment method is different and measures the development of thinking and analysis skills.
|
DESCRIPTION OF THE LABORATORY COURSE
|
|---|
Organismal Biology constitutes the third semester of the four-semester Biocore honors sequence at the University of Wisconsin-Madison. It follows courses that deal with evolution, ecology, and genetics and with cell and molecular biology. Students can elect to take the lecture class by itself (3 credits) or the laboratory-lecture combination (5 credits). Those who enroll in the laboratory course tend to be either students who need another laboratory class to meet degree requirements or students who are especially interested in organismal physiology.
For the laboratory course, each student attends one 50-minute discussion period per week and one three-hour laboratory. In addition to the learning of physiological principles, the primary goals of the course are to improve the ability of students to design experiments and analyze data. To promote these objectives, students work all semester in research teams of three or four on some stage of designing or carrying out their own experiments. Early in the semester, they focus on plant physiology, later on aspects of their own physiology. As the semester progresses, we stress increasingly complex elements of experimental design (hypothesis formation, literature review, randomization, blinding, controls, sample size estimation) and analysis (data manipulation, graphing, statistical tests, comparison with previous findings). As Table 1 illustrates, the course is designed to provide students with the raw materials to ultimately develop a complete understanding of the experimental process. By the end of the semester, most research teams are functioning quite independently, consulting with instructors only when necessary.
Many of the experiments developed by students make use of a computerized data acquisition system that transduces, records, and displays physiological responses in real time (Fig. 1). The system allows students to immediately analyze their data, to work together to solve problems and decide what to do next, and finally to draw conclusions from their results. Several times during the semester, the research teams report their findings through oral presentations or by writing, peer reviewing, and revising papers following the format of scientific journal articles. The last five weeks of the semester are devoted to the "Proposal+" project. During this period, each research team develops a detailed proposal on a novel research question, including literature review, testable hypotheses, detailed (and pilot-tested) methods, and preliminary results. If research teams definitively answer their experimental question by the end of the semester, they write up and present this final project as a scientific paper. If, on the other hand, there is no clear answer (as frequently happens), but they have developed a better method to test the original idea, the research team writes up and presents its efforts as a pilot-tested proposal for future work (hence the name Proposal+). Some of the Proposal+ projects conducted by student teams are shown in Table 2.
View larger version (113K):
[in this window]
[in a new window]
|
FIG. 1 Research teams use sensors that interface with a computer data acquisition system to test their ideas.
|
|
|
ASSESSMENT OF COURSE GOALS
|
|---|
Administration of Evaluation Tool
To assess how well students in the course were achieving our primary learning objectives, we developed a customized evaluation tool. The tool consisted of an experimental scenario and data, followed by a series of four questions of increasing complexity (described below). This tool was administered at both the beginning and the end of the semester to students taking the lecture course only (n = 65) and to students taking the laboratory + lecture combination (n = 43). Students were given the evaluation during a discussion session for the lecture course and had no idea of the purpose of the exercise or that they would see the same evaluation again at the end of the semester. Students were given 20 minutes to work on the evaluation and were simply asked to "put a good effort" into the assignment and to show as much of their work and thinking as possible.
The evaluation tool was scored (using criteria described below) by a biostatistician who did not know whether a particular test was from the beginning or end of the semester or whether it was from a student who took the lecture course only or the laboratory + lecture combination. For each student, an improvement score was calculated for each question on the evaluation as the difference between the student’s final and initial scores. We expected some pre-to-post improvement in scores because students were seeing the evaluation for a second time at the end of the semester. Furthermore, they had the opportunity to learn about physiology experiments and data analysis in lecture and may have become more sophisticated in their thinking with the passage of time. The strength of our approach is that students enrolled in lecture alone served as the control group for students taking the laboratory and lecture together. Improvement scores in the lecture-only group were attributed to all factors other than the laboratory curriculum (e.g., seeing the evaluation for a second time). Thus any difference between the improvement scores for laboratory + lecture students and those for students taking only the lecture course can be considered a consequence of the laboratory experience. We tested for such differences between groups with Student’s t-test.
Evaluation Tool: Investigation of the Effect of Step Cadence on Heart Rate
Scenario/Hypothesis.
Three Biocore 324 students, Terry, Tonya, and Taliz, did their group research project on the factors that influence heart rate during stepping on an exercise step. Taliz suggested the topic because she teaches a step aerobics class and wanted to know what routines are likely to elicit adequate (but nonlethal!) heart rate responses in her students.
The students hypothesized that the cadence (rate) of stepping would affect heart rate. To test this hypothesis, they performed the experiment described below. Read how they conducted their study, examine the data they collected, and then answer the questions about the meaning of the findings.
Methods.
The group selected Taliz to be the subject for the experiment, as she was in the best aerobic shape of the three. A metronome was used to provide the appropriate beat for stepping. Heart rate was monitored with a pulse plethysmograph that was attached to Taliz’s index finger.
The experiment used four different step cadences: 92, 98, 102, and 108 steps/min. Taliz performed three trials at each of the four step cadences (12 trials in all). During each of these trials, she started by standing still for 30 s (to get a preexercise value for heart rate), and then she stepped up and down on the step at the appropriate cadence for 2 min. A trial did not begin until Taliz’s heart rate while standing varied 3 beats/min or less over a 30-s period.
Data.
Table 3 shows the order in which the trials were conducted. It also gives Taliz’s average heart rate during the last 5 s of each standing and stepping period.
Analysis.
[Be sure you explain the reasoning behind your answers. Show all calculations and logic on this sheet OR the attached blank sheet.]
Create a graph to help you visualize the effect of step cadence on heart rate (do this on the attached graph paper).
What is the nature of the relationship between step cadence and heart rate?
On the basis of your analysis, what advice would you give Taliz about the step cadences she should use in her routines if she wants to keep the heart rate of her students between 110 and 120 beats/min during the 45-min class?
If you wanted to predict heart rate from step cadence with great precision, how would you improve the experimental design?
Evaluation Scoring Criteria
Each of the four questions on the evaluation was scored on a scale from 0 to 3 points, using increments of 0.25 point. For each question, the scorer assigned points for elements of answers that reflected key aspects of the expected answer. In some cases, students could also lose points for adding incorrect elements to otherwise correct answers. The general criteria used in scoring each question are described below.
Question 1: Create a graph to help you visualize the effect of step cadence on heart rate.
For question 1, students needed to create a well-labeled and appropriately sized scatterplot illustrating the relationship between step cadence and heart rate (an example is shown in Fig. 2). Students could have shown a separate line for heart rate while stepping and while standing or a line representing the difference between stepping and standing heart rate at each cadence. Their graph could have included individual points for each trial or the mean (of the 3 trials) for each cadence and error bars representing the variability.
View larger version (18K):
[in this window]
[in a new window]
|
FIG. 2 Example of a well-designed graph in response to question 1. Effect of cadence on stepping and standing heart rate (HR) during the last 5 s of the measurement period. Each point represents the average of 3 trials at that cadence for the single subject (Taliz). Error bars, ±1SE.
|
|
Question 2: What is the nature of the relationship between step cadence and heart rate?
For question 2, students needed to note the nonlinear (higher cadences had disproportionally higher heart rates), positive relationship between step cadence and heart rate. They also should have noted whether their answer concerning stepping heart rate took into account the increase in standing heart rate as trials progressed.
Question 3: On the basis of your analysis, what advice would you give Taliz about the step cadences she should use in her routines if she wants to keep the heart rate of her students between 110 and 120 beats/min during the 45-min class?
For question 3, students needed to state in detail the limitations in their ability to generalize from the data given. They could have said that a person similar to Taliz in fitness level and other characteristics should be able keep her/his heart rate between 110 and 120 beats/min with cadences of 
98–102 if their stepping duration was 2 min. Students needed to say that the data from the experiment do not allow predictions about heart rate based on step cadence for stepping periods much greater than 2 min or for people much different from Taliz in fitness level, body proportions, gender, etc. Students could also have said that they could not make reasonable predictions because of problems with the experimental design (see below).
Question 4: If you wanted to predict heart rate from step cadence with great precision, how would you improve the experimental design?
For question 4, students had to point out the significant flaws in the design of the experiment and discuss how they would account for other influences on heart rate. For example, the trials were not performed in random order. In fact, the cadence was systematically increased throughout the experiment, potentially confounding the effects of fatigue, body temperature change, etc., with cadence effects. A better design would have ensured that test conditions were constant (initial standing heart rate, time of day, humidity, temperature, etc.), used trials long enough in duration to ensure that subjects had achieved a steady state, used a more precise instrument for measuring heart rate (e.g., a heart rate monitor), and included a larger range of step cadences. Students should have noted that more trials per cadence would have improved the precision of predictions based on the data. They also needed to suggest increasing the number of subjects, but they had to appreciate that subject characteristics (such as fitness level, gender, body proportions, previous stepping experience) could increase variability if not accounted for (without necessarily saying how they would account for such variation).
|
RESULTS AND DISCUSSION
|
|---|
Although it was not possible to assign students randomly to the laboratory + lecture or lecture-only groups, we did look for evidence of comparability between the two groups. Students in the laboratory + lecture and lecture-only comparison groups did not differ significantly (P = 0.68) in the final grades they received for the previous lecture course in the sequence (Cellular Biology). In addition, the two groups also did not differ significantly in their initial scores (from the beginning of the semester) on any of the four evaluation tool questions (Fig. 3). When the improvement scores (final minus initial scores) for the two groups were compared, the laboratory + lecture group had a statistically significant (P = 0.05) improvement score for the hardest question, question 4 (Fig. 4). This finding suggests that participating in the laboratory course did improve students’ abilities to meaningfully interpret data and critique flaws in experimental designs.
View larger version (17K):
[in this window]
[in a new window]
|
FIG. 3 Average group initial (beginning of the semester) score for each question on the evaluation tool. The maximum possible score for each question was 3 points. Error bars, ±1 SE. P values represent significance level of Student’s t-test for differences between groups.
|
|
View larger version (16K):
[in this window]
[in a new window]
|
FIG. 4 Average group improvement score (final score minus initial score) for each question on the evaluation tool. The maximum possible score for each question was 3 points. Error bars, ±1 SE. P values represent significance level of Student’s t-test for differences between groups.
|
|
We noted that some students in the laboratory + lecture group scored less well on question 1 at the end of the semester compared with the beginning. We believe that this may be due to the exclusive use of Microsoft Excel for creating graphs in the laboratory course. Students in the laboratory course may not have fully understood how to create graphs by hand because they could get by using computer software, or they may have had little patience, at the end of the semester, for making a graph by hand when they knew there was an "easier way." In any event, we now allow students to use the computer to generate graphs only after they have shown they know how to make appropriate graphs by hand.
Qualitative data from student evaluations show that students value the chance our laboratory course gives them to plan and conduct their own experiments, and many look back on this experience as the highlight of the four-semester sequence (see representative comments listed in Table 4). Students particularly value the Proposal+ project that takes up the last five weeks of the semester because they conduct research on a novel question. Often what is novel about their experiments is fairly incremental (e.g., the subject populations being tested), but now and then students come up with truly original ideas to investigate. Some research teams definitively answer their experimental question and present this final project as a scientific paper. Those who are not able to do this succeed in developing a better method to test their original idea, and they present their efforts as a pilot-tested proposal for future work. This strategy is particularly affirming for the numerous teams who spend five weeks solving one important problem after another but do not obtain enough useful data in the end to draw conclusions about their hypothesis.
To summarize, we developed an experimental analysis evaluation tool and used it to assess whether an inquiry-based organismal physiology laboratory course helped students improve their ability to analyze data and experimental design. The evaluation showed that students enrolled in both laboratory and lecture improved more than students taking only the lecture course for the most complex question. We conclude that our new inquiry-based curriculum is accomplishing its primary goals. We have described our evaluation tool in considerable detail so that others interested in measuring experimental analysis abilities in their students may use it.
|
Acknowledgments
|
|---|
We thank Kathleen Daniels for the many hours spent scoring student responses to the evaluation tool. We are also grateful to the students in the Organismal Biology course who participated in this evaluation.
Address for reprint requests and other correspondence: M. J. Myers, Dept. of Biology, College of St. Catherine, Mailstop #4173, 2004 Randolph Ave., St. Paul, MN 55105 (E-mail: mjmyers{at}stkate.edu).
Received for publication June 10, 2002.
Accepted for publication December 9, 2002.
|
References
|
|---|
- Howard Hughes Medical Institute. Beyond Biology 101: The Transformation of Undergraduate Biology Education. Chevy Chase, MD: Howard Hughes Medical Institute, 1995.
- Kolkhorst FW, Mason CL, DiPasquale DM, Patterson P, and Buono MJ. An inquiry-based learning model for an exercise physiology laboratory course. Adv Physiol Educ 25: 45–50, 2001.[Abstract/Free Full Text]
- National Research Council. National Science Education Standards [Online], 1994. http://www.nap.edu/readingroom/books/nses/html (May 24, 2002).
- National Science Foundation. Shaping the Future: New Expectations for Undergraduate Education In Science, Mathematics, Engineering, and Technology (NSF 96–139) [Online], 1996. http://www.ehr.nsf.gov/ehr/due/documents/review/ 96139/start.htm (May 24, 2002).
- Springer L, Stanne ME, and Donovan SS. Effects of small-group learning on undergraduates in science, mathematics, engineering, and technology: a meta-analysis. Rev Educ Res 69: 21–51, 1999.
- Sundberg MD, Armstrong JE, Dini ML, and Wichusen EW. Some practical tips for instituting investigative biology laboratories. J Coll Sci Teach 29: 353–359, 2000.
This article has been cited by other articles:
|
|
|
|
 
P. A. Marshall
Using Saccharomyces cerevisiae to Test the Mutagenicity of Household Compounds: An Open Ended Hypothesis-Driven Teaching Lab
CBE Life Sci Educ,
December 1, 2007;
6(4):
307 - 315.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. A. Rol de Lama, J. P. Lozano, V. Ortiz, F. J. Sanchez-Vazquez, and J. A. Madrid
How to engage medical students in chronobiology: an example on autorhythmometry
Advan Physiol Educ,
September 1, 2005;
29(3):
160 - 164.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. B. Luckie, J. J. Maleszewski, S. D. Loznak, and M. Krha
Infusion of collaborative inquiry throughout a biology curriculum increases student learning: a four-year study of "Teams and Streams"
Advan Physiol Educ,
December 1, 2004;
28(4):
199 - 209.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
