Plant Physiol. (1999) 119: 1177-1186

A Computer-Assisted Personalized Approach in an Undergraduate Plant Physiology Class¹

Nancy N. Artus² and Kenneth D. Nadler^*

Department of Botany and Plant Pathology, Michigan State University, East Lansing, Michigan 48824

	ABSTRACT

Top Abstract Introduction Methods Results & Discussion References

We used Computer-Assisted Personalized Approach (CAPA), a networked teaching and learning tool that generates computer individualized homework problem sets, in our large-enrollment introductory plant physiology course. We saw significant improvement in student examination performance with regular homework assignments, with CAPA being an effective and efficient substitute for hand-graded homework. Using CAPA, each student received a printed set of similar but individualized problems of a conceptual (qualitative) and/or quantitative nature with quality graphics. Because each set of problems is unique, students were encouraged to work together to clarify concepts but were required to do their own work for credit. Students could enter answers multiple times without penalty, and they were able to obtain immediate feedback and hints until the due date. These features increased student time on task, allowing higher course standards and student achievement in a diverse student population. CAPA handles routine tasks such as grading, recording, summarizing, and posting grades. In anonymous surveys, students indicated an overwhelming preference for homework in CAPA format, citing several features such as immediate feedback, multiple tries, and on-line accessibility as reasons for their preference. We wrote and used more than 170 problems on 17 topics in introductory plant physiology, cataloging them in a computer library for general access. Representative problems are compared and discussed.

	INTRODUCTION

Top Abstract Introduction Methods Results & Discussion References

Information technology is increasingly used in a variety of ways to aid instruction in the biological sciences. Laboratories or lectures may include computerized simulations of life processes (Jaffe and Lynch, 1989; Hutchings et al., 1994; Hall, 1996; Beneski and Waber, 1997), many tutorials have been written (Blakely, 1988; Dewhurst et al., 1989; van Geloven, 1994), some publishing companies now offer CD-ROM study guides to accompany their textbooks, and many instructors use a World Wide Web site to supplement their courses because the Web is a vast and rapidly growing source of information (Somerville et al., 1997).

Information technology is rarely used for homework sets in the biological sciences, however. Regular assigned homework is recognized as a valuable tool in classes in the physical sciences: it is an active rather than a passive form of learning, it challenges students to keep up with material, and it encourages students to consult on an individual basis with the instructors. In the physical sciences, several computer programs have been developed to create individualized problem sets in which random numbers are used to generate unique data for each student (Castleberry and Lagowski, 1970; Kashy et al., 1993; Spain, 1996; Hodges, 1994). It is difficult to imagine a class in mathematics or physics that lacks regular homework; few would argue that calculus can be mastered without practice in solving problems. Biology, by contrast, seems a more qualitative subject in which much of the material taught is conceptual rather than numerical. Thus, some biology instructors may regard homework as unnecessary or even inappropriate for the teaching of conceptual material. In addition, many instructors are reluctant to give regular homework because of the time required to grade it, or even because it is difficult to control the possibility of students copying each others' answers.

We believe that homework is as valuable a teaching and learning tool in the biological sciences as it is in the physical sciences. Ideal homework assignments should be individualized (similar in form but different in detail) so that students can be encouraged to work together to discuss concepts but to do their own work to arrive at correct answers to their particular problems. Cooperative learning has been used in college physics, astronomy, chemistry, biology, earth science, and business classrooms to improve learning (Mazur, 1997). Some desirable learning outcomes thus promoted include higher achievement, increased retention, more frequent and higher-level reasoning, critical thinking, more time on task, and improved attitudes toward teachers (Johnson et al., 1991). A cooperative-learning environment develops when students study together to solve similar but nonidentical problems. Individualized homework also encourages each student to self-organize and reorganize knowledge; in the words of Pines and West (1986): "Knowledge is not acquired passively."

We adapted the CAPA software system to develop individualized, networked problem sets for our large-enrollment, junior-level plant physiology course at MSU. The CAPA system, originally developed for physics classes at MSU (Kashy et al., 1993, 1995), is now in use for chemistry and physics courses at several universities around the country (see http://www.pa.msu.edu/educ/CAPA/). CAPA generates a wide variety of quantitative and qualitative problems. The problems within each homework assignment take on the same form and cover the same principles for all students, but they are unique for each student. Each student receives a printed assignment with quality graphics. They also can access their assignments and submit answers on line, receiving immediate feedback. If the answer is correct, the computer immediately rewards the student; if it is incorrect, the student may resubmit an answer without penalty until either a predefined limit on the number of tries is reached or the due date and time for the problem set is reached. Hints for many problems are available on line, and after the due date, on-line explanations are also available, if provided by the instructor. While they are on line, students may view earlier assignments or a summary of their scores to date. CAPA keeps track of the number of tries made by each student on each problem, thus allowing the instructor to identify difficult and/or poorly constructed problems and address these in subsequent lectures. In another file, CAPA records the times, durations, and locations of student log-ins. CAPA's Grader program generates a summary of scores for each student, which frees the instruction staff to arrange individual interactions with students and to give weekly homework assignments in a large-enrollment course.

In this paper, we describe the types of plant physiology homework problems we have created with CAPA, present the results of surveys of students' responses to and acceptance of CAPA in our plant physiology course, and present a comparison of student performance on examinations between current students (with CAPA) and students from previous years (without CAPA), highlighting the positive effects of regular homework assignments.

	MATERIALS AND METHODS

Top Abstract Introduction Methods Results & Discussion References

CAPA was originally developed at MSU (Kashy et al., 1993). The CAPA software operates on any UNIX server connected to the campus ethernet system. The main CAPA host can be accessed from either a PC or a Mac running X-Server software. We used the Windows-NT (Microsoft, Redmond, WA) operating system and eXceed (Hummingbird Communications, Ltd., North York, Ontario, Canada) X-server software. CAPA software consists of four main modules (three have been described by Kashy et al. [1993]) designed to create and preview problem sets, parse and print sets, handle remote sessions by students who access CAPA, and generate grade reports for the class or for individual students. Assignments are prepared by making a prototype source file with the Quizzer program. Templates are available for most problem types. The qualitative problems contain multiple parts, so for each of these parts, up to four variants are written that are randomly selected and permuted by Quizzer for individual assignments. Random numbers within a defined range are used to generate data for quantitative problems. Students connect to the host computer via the Internet and access assignments or grade summaries by entering the course acronym, their student number, and the unique CAPA identification number assigned by CAPA to that student's assignment. This process secures access to individualized assignments.

Student Enrollment and Survey of Student Evaluations of CAPA in Plant Physiology

At the end of each semester, an anonymous survey was distributed to each student to evaluate their acceptance and opinions about the plant physiology sets we constructed with CAPA. The nature of problems in plant physiology and physics (for which CAPA was originally designed) are sufficiently different that we decided to assess student attitudes toward both the system and our homework assignments. The frequency of each response and the Pearson correlation coefficients (item intercorrelations) were determined by the scoring office in the Department of Computer Sciences at MSU.

Statistics

	RESULTS AND DISCUSSION

Top Abstract Introduction Methods Results & Discussion References

CAPA Homework Assignments in Plant Physiology

View larger version (49K): [in this window] [in a new window]   View larger version (50K): [in this window] [in a new window]   Figure 1. Two versions of a sample set of selected CAPA homework problems in plant physiology. (Figure continues on next page.)

Selecting All Correct Statements or Answers (problem 1): In this type of problem, several statements (or "items"; there are seven [A-G] in this example) are listed and the student must select all of the correct statements. For any student, from one to all of the statements listed may be correct; or there may be only one correct selection for one student and seven for another. For each item listed, there are up to four possible variants. For each student, CAPA randomly selects one variant of each item and then permutes the order of the items. Thus, with seven items and four variants of each item, as in problem 1, we can generate 4⁷ × 7 factorial, or approximately 86 million, unique versions of this particular problem. In addition, the "choose" function was used in the text of problem 1 to randomly select one of three equations from respiration (the oxidative decarboxylation of pyruvate, isocitrate, or -ketoglutarate), which further increases the variation between problems.

Selecting Answers in Sequence (problem 2): A correct answer for this type and the following three types of problems is a sequence of entries, the number of which corresponds to the number of items (six entries for problem 2). The choices for entries may be A/B, true/false, greater than/less than/equal to, or any set of answer variants defined by the instructor. For example, in problem 2 the student selects between F for floral and V for vegetative; a correct answer would take the form FFFVVF. Again, the choose function was used in the text of problem 2 so that each student is randomly assigned any one of several problems on long-day or short-day flowering plants, together with the appropriate critical day length.

Matching (problem 3): The student matches each item (1-5) to each property (A-D). There are up to four variants for each item, but the properties are the same for each student. Matching does not need to be one-to-one, because the properties may be used more than once or not at all.

Matching to Labels in a Figure (problem 4): CAPA randomizes the labels (A-D) on a figure and permutes the order of the items (1-8). The author of the problem may choose to draw the figure with a graphics program or scan an existing figure.

Ranking (problem 5): The student must list the items in the appropriate order. The ranking may be in time (as in this example), magnitude, or order in a biosynthetic pathway.

Free Form (problem 6): There is no template for writing this type of problem as there is for the other problems. Problems requiring a calculation are often written in free form. Random numbers within limits defined by the instructor may be used to produce a unique problem for each student.

The problems in the sample sets (Figs. 1 and 2) test several levels of knowing or categories of the cognitive domain (Bloom et al., 1956). The first and fifth problems in set one (Fig. 1) test for a base level of cognition ("knowledge"): The student names, defines, identifies, or orders information. The third, fourth, and sixth problems require a higher order of understanding (the third or "application" level): The student relates a precursor or monomer as a building block to various products, relates an element of a phloem-tissue diagram to structure and function, or computes a cell pressure potential, hence turgidity of the cell. The second problem in set one demands that the student analyze the photoregime and determine the relationship between photoperiod and the critical photoperiod, predicting flowering outcome; this represents the fourth ("analytical") level of knowing. The problems in sample set two require the highest order of understanding ("evaluation"): The student hypothesizes (level five or "synthesis") an order of intermediates in the carbon dioxide fixation pathway based on labeling kinetics and then evaluates which subset of the pathways proposed as answers would result in those kinetic data. More than one of the answers is correct. The data in problem 2 in set two are necessary to evaluate which of several possible pathways is correct; this tests understanding at the evaluation level.

View larger version (21K): [in this window] [in a new window]   Figure 2. Two sample problems on 14CO2 fixation in photosynthesis. The hint for the first problem states, "14C is an isotope of carbon that allows tracking of carbon. The compounds that are detected with 14C label after feeding 14CO2 are those that are made in the CO2 assimilation pathway. The order that they appear in is the order that they occur in the pathway." The hint for the second problem states, "Consider the reaction X + Y  Z. If Y is in short supply, X will accumulate and the level of Z will diminish." The figures are redrawn with permission (Mohr and Schopfer, 1995).

Because a variety of qualitative questions can be written with CAPA (Kashy et al., 1995), it is readily adaptable to the biological sciences. We have written more than 170 problems for our third-year, university-level course (Botany 301) covering 17 areas in plant physiology (Table I). In all of these subject areas, most or all of the problems we wrote are qualitative in nature. Quantitative problems were written for the areas of plant water relations, transpiration, enzyme kinetics, and Mendelian genetics. Our problems have been cataloged in a computer library and are available for use at other institutions that adopt CAPA (see http://www.pa.msu.edu/educ/CAPA/ for CAPA overview, institutions with CAPA licenses, and extensive contact, system requirement, and licensing information, documentation, and a sample set).

There are several features of CAPA (immediate feedback, multiple tries, and the hints) that, when combined with a well-planned lecture, can be used successfully with challenging problems. The problems presented in Figure 2 require the student to develop the pathway for photosynthetic carbon fixation from Calvin and Benson's original data. The hints available for these problems (see legend to Fig. 2) guide students by describing the logic required to solve the problems. When these problems were used in the Spring 1997 semester, 69 of 76 students (91%) who attempted these problems answered both correctly. We set the limit to answer each question correctly at 10 tries. The average number of tries to solve problem 1 was 2.0 ± 1.8, and 44 students (58%) answered it correctly on the first try. The average number of tries to solve problem 2 was 2.3 ± 1.8, and 37 students (49%) answered it correctly on the first try. Thus, the students did very well on a set of challenging problems that we had been reluctant to include in previous traditional homework sets.

Student Assessment of CAPA in Plant Physiology

View larger version (32K): [in this window] [in a new window]   Figure 3. Student responses from two semesters to survey questions 14 and 15 on CAPA. A, Question 14 is, "As a learning tool, I found CAPA ... ... ... " B, Question 15 is, "Rate your experience with CAPA." Response choices available are (1) quite negative, (2) somewhat negative, (3) indifferent, (4) somewhat helpful, and (5) quite helpful. The percentage of the responding students in each semester selecting each choice is shown.

View larger version (25K): [in this window] [in a new window]   Figure 4. Student responses from two semesters (gray bars, Fall 1996; striped bars, Spring 1997) to survey question 16, "If I had a choice between CAPA assignments used in BOT 301 [Botany 301] and hand-graded essay homework, I would be ... ... ." Response choices available are (1) strongly in favor of hand-graded homework, (2) moderately in favor of hand-graded homework, (3) indifferent, (4) moderately in favor of CAPA, and (5) strongly in favor of CAPA. The percentage of responding students selecting each choice is shown.

Many students reported anecdotally that they devote several hours to each CAPA homework set. Our log-in data (not shown) suggest that the students make multiple attempts on most problems, indicating increased time on task. We also observed a generally improved student attitude toward the instructors and an increase (1 order of magnitude) in the number of students seeking the instructor's assistance in answering CAPA homework problems compared with typical homework sets. These effects were observed previously in physics courses using CAPA (Kashy et al., 1993). We believe that there are three major reasons for the increased student time on CAPA homework: Students know that they have the opportunity to rework answers multiple times, they are not penalized for incorrect answers, and they are not judged while they are learning. An important element of learning is making mistakes and then correcting them. By eliminating ranking during the learning process, CAPA is eliminating a discouraging learning environment for many students.

The student responses were analyzed for correlations between answers to the survey questions. For example, if a student answered yes to question 1, was the probability greater than 95% that he or she would answer yes to question 5? Students rating CAPA as a helpful learning tool (question 14) were likely to consider the on-line and "multiple-try" features of CAPA helpful, and they felt rewarded when the computer told them that they were correct. Students who had a positive experience with CAPA (question 15) liked these same features as well as the immediate feedback and the explanations that are available on line after the due date. Students who used CAPA in a previous course (question 1) were likely to prefer CAPA homework over hand-graded homework. (The converses are also true: The small subset of students who thought CAPA was not a useful tool did not find the on-line and multiple-try features useful, etc.) The only significant negative correlation occurred between questions 3 and 10, suggesting that students who worked individually rather than in groups were more likely to view the explanations given after the due date.

Student Performance

Performance on CAPA homework correlated with performance on lecture examinations in the Spring 1997 semester (r = 0.6136; not shown). Because questions derived from CAPA homework problems were only a part of the examinations in 1997 and 1998 (Table IV), this moderate correlation was not unexpected. The correlation between homework scores and final-examination scores in physics classes in which CAPA was used for all homework sets, quizzes, and examinations was only 0.300 (Kashy et al., 1998). There were eight students (9%) in our plant physiology course who scored very high on the homework but did not perform well on the examinations. These students may have needed more time to think through problems. There are several features of CAPA (immediate feedback, multiple tries, on-line hints) that give students at all levels of ability the opportunity to succeed if they give enough time and effort.

Our data clearly indicate the beneficial effect of mandatory homework sets on student performance on examinations. When the plant physiology course was taught in 1995, there was mandatory, hand-graded essay homework. In 1996, the hand-graded homework was optional rather than mandatory; by 1997 and thereafter, mandatory CAPA sets were used. The mean scores on equivalent examinations for the 4-year period from 1995 to 1998 are given in Table III. These data indicate that students performed better when given mandatory homework. There is a statistically significant difference between the examination scores in 1996 and the scores in 1995, 1997, and two of the three examination scores in 1998 (P < 0.003). There were no significant differences between equivalent examinations in 1995 and 1997 or 1997 and 1998, and two of three of the comparisons were not significantly different between 1995 and 1998. Thus, scores on five of the six examinations given when CAPA was in use significantly exceeded equivalent examination scores from when there was no mandatory homework. Examination scores from when CAPA was used were similar to scores from when hand-graded essay homework was required. Scores on eight of the nine examinations given when homework of either format was required were significantly higher than scores on comparable examinations given when homework was not mandatory. These data suggest that mandatory homework improves student performance on examinations.

In principle, other variables might have affected student performance on examinations in this class. Among them are student response to different instructors, changes in the laboratory component of the class and its valuation in the final grade, and changes in examination format. However, the course was taught by K.D.N. for the semesters reported, and the laboratory component and its weight in the determination of final grades were unchanged. The examination format (relative weighting of short answers or CAPA questions to essays) varied between comparable tests in different semesters and between different examinations in the same semester (Table IV), but the conceptual material covered in the essay questions did not change. Comparison of the data in Tables III and IV reveals no relation between examination format and student examination performance. In Spring 1996, when no homework was required, the three tests contained short-answer sections valued at 40, 42, and 41 points, whereas in Spring 1997, when CAPA homework was mandatory, the corresponding CAPA sections were valued at 46, 46, and 50 points; yet, student examination performance in these 2 years differed significantly (Table III). Student mean examination scores were, with one exception, not significantly different in 1997 and 1998 when CAPA homework was required in both semesters (Table III) but the examination formats were different (Table IV). Similarly, student examination performance in 1995 was not significantly different from that in 1997, but the valuation of the short-answer section of examinations in 1995 was variable, often quite different from the valuation of the CAPA section of the examination in 1997. We find no evidence that relates superior student examination performance to changes in instructor, laboratory part of the course, or examination format. We conclude that mandatory homework improves student performance on examinations, with CAPA effectively substituting for traditional hand-graded homework in plant physiology.

Moreover, CAPA has several important advantages compared with traditional hand-graded assignments. Grading time is greatly reduced, which frees the instructor to help students. There is a favorable shift in student attitude toward the instructor. Rather than viewing the instructor as an evaluator, students consider the instructor as a resource and coach, assisting them in answering the computer questions. We also found that with CAPA it was feasible to cover a broader range of subject material and include more challenging material on homework and examinations. Before we adopted CAPA for homework assignments, our students struggled with homework problems at the first (knowledge) and second (comprehension) levels of cognition (Bloom et al., 1956). With CAPA, we included problems at the highest levels of cognition and students successfully answered them. Thus, although mean examination scores generally were not significantly different with hand-graded versus CAPA homework, course standards and overall learning were increased by implementation of CAPA.

Other Computerized Learning Systems

Another novel learning and assessment tool used for more than 70 subjects at universities in Australia is BrainZone (Strassburger, 1997). This World Wide Web-based tool is used to create questions (multiple choice or short answer) that students can access from any on-line computer at any time. Students log on to the site, enter an access code and password, and select a subject and topic. BrainZone gives them a random set of questions to work through and provides instant feedback. BrainZone may be used in either a "learning mode" or a "testing mode." In the learning mode, students may repeat tests as many times as they wish, each time receiving a unique set of problems. As with CAPA, the instructor is able to monitor student progress and study habits and can identify questions that give students the most difficulty. Early evaluations of BrainZone revealed improved study habits on the part of students and improved performance on examinations. Students rated BrainZone very favorably.

To our knowledge, we have described the first application of CAPA software in the life sciences. Previous applications of CAPA have been limited to the physical sciences (Kashy et al., 1993, 1998; Morrissey et al., 1995; Thoennessen and Harrison, 1996), although recently it has been implemented in courses as diverse as Family and Child Ecology and Human Food and Nutrition (see http://www.pa.msu.edu/educ/CAPA/). CAPA is adaptable to many fields because of the variety of formats available for writing qualitative problems. Students rate CAPA positively as a useful learning tool. Our CAPA plant physiology problems are available to other institutions, and anyone interested in adopting CAPA for a plant physiology course should contact K.D.N. For other courses, see http://www.pa.msu.edu/educ/CAPA/.

	FOOTNOTES

   Received June 22, 1998; accepted December 1, 1998.

	ABBREVIATIONS

Abbreviations: BSCS, Biology Study Center Software. CAPA, Computer-Assisted Personalized Approach.

	ACKNOWLEDGMENTS

We are grateful to Dr. Ed Kashy for heading the development of the CAPA program and for his enthusiastic guidance throughout this project. We thank Nancy Davis, MSU CAPA coordinator, for much technical assistance, and Dr. Leroy Olson, MSU Computer Laboratory, for assistance with the statistical analysis of correlations and item intercorrelations.

	LITERATURE  CITED

Top Abstract Introduction Methods Results & Discussion References

Beneski J, Waber J (1997) Virtual Laboratories in the Sciences. McGraw-Hill, Hightstown, NJ

Blakely L (1988) First steps using personal computers in a plant physiology course. Computers Life Sci Educ 5: 1-8

Bloom BS, Engelhart MD, Furst EJ, Hill WH, Krathwohl DR (1956) Taxonomy of Educational Objectives: Handbook I. Cognitive Domain. Longmans, Green, New York

Castleberry S, Lagowski JJ (1970) Individualized instruction using computer techniques. J Chem Educ 47: 91-96

Dewhurst DG, Meehan AS, Williams A, Woods D (1989) A CAL program to teach the basic principles of genetic engineering: a change from the traditional approach. J Biol Educ 23: 218-222

Hall DW (1996) Computer-based animations in large-enrollment lectures: visual reinforcement of biological concepts. J Coll Sci Teach 25: 421-425

Hodges L (1994) Computer problems for large classes. Physics Teacher 32: 16-17

Hutchings GA, Hall W, Thorogood P (1994) Experiences with hypermedia in undergraduate education. Computers Educ 22: 39-44 [CrossRef]

Jaffe CC, Lynch PJ (1989) Hypermedia for education in the life sciences. Acad Comput 4: 10-13 , 52-57

Johnson DW, Johnson RT, Smith KA (1991) Active Learning: Cooperation in the Classroom. Interaction Book Co, Edina, MN

Kashy E, Gaff SJ, Pawley NH, Stretch WL, Wolfe SL, Morrissey DJ, Tsai Y (1995) Conceptual questions in computer-assisted assignments. Am J Physiol 63: 1000-1005

Kashy E, Sherrill BM, Tsai Y, Thaler D, Weinshank D, Engelmann M, Morrissey DJ (1993) CAPA: an integrated computer-assisted personalized assignment system. Am J Physiol 61: 1124-1130

Kashy E, Thoennessen M, Tsai Y, Davis NE, Wolfe SL (1998) Using networked tools to promote student success in large classes. J Engineer Educ 87: 385-390

Kleinsmith LJ (1987) A Computer-Based Biology Study Center: Preliminary Assessment of Impact. Acad Comput 2: 32-33 , 49-50

Mazur E (1997) Peer instruction. http://galileo.harvard.edu/galileo/lgm/pi

Mohr H, Schopfer P (1995) Plant Physiology. Springer-Verlag, New York, pp 177-178

Morrissey DJ, Kashy E, Tsai I (1995) Using computer-assisted personalized assignments for freshman chemistry. J Chem Educ 72: 141-146

Pines AL, West LHT (1986) Conceptual understanding and science learning: an interpretation of research within a source of knowledge framework. Sci Educ 70: 583-604

Somerville C, Flanders D, Cherry JM (1997) Plant biology in the post-Gutenberg era. Everything you wanted to know and more on the World Wide Web. Plant Physiol 113: 1015-1022 [CrossRef][Medline]

Spain JD (1996) Computer-interactive problem sets for general chemistry. J Chem Educ 73: 222-225

Strassburger L (1997) BrainZone: an assessment tool to facilitate the needs of both lecturers and students as universities move into flexible delivery. http://www.curtin.edu.au/conference/ASCILITE97/papers/Strassburger/Strassburger.html

Thoennessen M, Harrison MJ (1996) Computer-assisted assignments in a large physics class. Computers Educ 27: 141-147 [CrossRef]

van Geloven MP (1994) Computer assisted learning: an instrument to improve the quality of education? Eur J Agric Educ Extension 1: 77-82

Copyright Clearance Center:   0032-0889/99/119//10
 
© 1999 American Society of Plant Physiologists