This work was supported by the Howard Hughes Medical Institute (contract grant sponsor). The cost of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We developed web-based modules addressing fundamental concepts of introductory biology delivered through the LON-CAPA course management system. These modules were designed and used to supplement large, lecture-based introductory biology classes. Incorporating educational principles and the strength of web-based instructional technology, choices were made about knowledge presentation, representation, and construction (W. A. Nelson, D. B. Palumbo (1992) J. Educ. Media Hypermedia 1, 287–299). Knowledge presentation focused on big and connecting ideas. Knowledge representation provided students the opportunity to interact with concepts in several ways using multiple representations. For knowledge construction, we facilitated students' active and meaningful interactions with content using interwoven high-level questions. Students' extended responses to a questionnaire indicated that these modules influenced the students learning in meaningful ways. (For access to demonstration modules, go to demo.lon-capa.org/cgi-bin/signon.pl?hhmi.)
Engaging students in large, undergraduate science courses includes a 2-fold challenge. In order for meaningful learning to occur, courses must first engage the students' attention, and second the students must engage in the course content . Of course, even the most interesting and organized lecture, with many “cool” examples, does not capture the attention of all of the students all of the time. Finding ways to have students become active participants in their own learning is likewise difficult. Assigning homework often is not practical because of the time it takes to grade. Assigned readings from the textbook are, by the very nature of textbooks, dense and often difficult for students. The end result is that many instructors are unsatisfied with their students' achievement and would welcome ways to help students prepare for and perform better on standard assessments.
We developed web-based biology modules to supplement a lecture-based presentation of introductory biology. This paper will describe the design of these modules and the ways in which they engage students' attention and promote student learning by enabling them to interact with the content of introductory college biology.
We did not want to produce an “on-line textbook” because textbooks generally do an excellent job of providing foundational information. So we posed a simple question: What can web-based instructional materials do that a textbook/lecture usually does not do? The answer was 3-fold: 1) focus on developing the connections between the fundamental ideas of introductory biology (knowledge presentation); 2) give students several ways to interact with fundamental ideas, including animations, cartoons, and interactive exercises (knowledge representation); and 3) place questions within the content to increase student interactions with the content (knowledge construction). All three functions were facilitated by the web-based course management system, LON-CAPA.11
DESCRIPTION OF TECHNOLOGY
In addition to being able to present content in any web-based format, LON-CAPA generates individualized question sets and quizzes that promote student collaboration on assignments, while reducing opportunities for blanket copying (www.lon-capa.org). Of the many question types available, we used primarily multiple choice, multiple true/false, and mix-and-match. To a lesser extent, and as a development tool, we used on-line essay questions (graded by a person) in order to learn more about students' thinking and understanding of particular topics.
LON-CAPA provides students with immediate feedback on the correctness of their responses (except for essay questions). Course instructors can set the number of chances students have to get a question correct and typically allow enough tries so that students are able to correct mistakes prior to the assignment due date without penalty. Feedback specific to a unique question can be programmed into the system. The system records students' responses and performance, thus making it well suited for use in courses with large enrollments. The assessment records are available on-line to both the instructor and the individual student.
DEVELOPMENT OF MODULES
The curriculum development group consisted of Michigan State University faculty in the College of Natural Science. With the exception of those members who brought skills in technology to the project, all were trained in the biological sciences. The two course instructors were housed in the Microbiology and Molecular Genetics and the Zoology Departments. Biologists with appointments in the Division of Science and Mathematics Education had knowledge of current issues in science education, with particular expertise in assessment and student-centered learning at the tertiary level. Two postdoctoral associates working on the project were trained in science research and were interested in the use of technology in teaching and in expanding their understanding of student learning through tertiary science education research. We believe this combination of expertise was essential for achieving the desired results.
This group used three principles to guide module development:
Develop materials that could be used to supplement large lecture classes;
Use instructional technology in ways that draw on its strengths and unique attributes;
Use educational research to inform curriculum development.
The LON-CAPA course management system allowed us to interweave content (information) pages and question pages. The resulting modules were thus embedded with questions that required students to engage the content as it was presented. Often, these questions were modeled after the type of questions a Socratic lecturer might ask while teaching. We also designed questions that helped students interpret animations and other graphics as well as questions that helped students develop or tie together complex ideas. Such questions required students to apply and synthesize information presented and, in doing so, enabled the instructors to assess an overall understanding of the module objectives.
To date, we have developed a set of 24 modules to be used in the course “Introductory Cellular and Molecular Biology” and have completed 11 modules in the companion course, “Introductory Organismal Biology.” All of the modules are supplemental in nature and designed to augment rather than replace face-to-face sessions and print materials. The Cellular and Molecular Biology modules have been used in two introductory biology courses, one a large university-wide introductory course and the other a smaller course offered through a residential science program. Course instructors awarded points in various ways to students for completing the questions interspersed throughout the modules. Over 1000 students have used these materials over three semesters.
Instructional technology can be used in three ways: to present knowledge, to represent knowledge, and to help students construct knowledge . Knowledge is typically presented to learners in organized collections of information and ideas. Representations of knowledge include the signs, symbols, and words that stand for ideas or phenomena . Certain standard representations are widely used within a discipline . Learners construct new understandings as they work with new ideas by connecting them to their previous experiences and their existing framework of understandings .
Knowledge presentations and representations are both building blocks and tools for learners, but understanding itself comes through construction—the active mental engagement with material [1, 4–6]. Each individual learner infers, elaborates on new information, and generates relationships among ideas. Thus, when we say that instructional technology can be used to aid students in constructing understanding, we mean that it must actively engage and guide students in making sense of subject matter content.
How, then, did we have the modules fulfill these three roles for instructional technology? For knowledge presentation, we needed first to determine what to present. We were aware that “the problem with a major emphasis on ‘covering the material’ is that many of the facts that constitute this material appear to most students as disconnected nuggets of information .” Therefore, we chose to design our modules around a few key ideas that connect subtopics rather than trying to cover all aspects and all details.
As with knowledge presentation, we chose to concentrate our efforts in knowledge representation on those areas where students have difficulties. Kozma describes how students using standard representations often focus on surface features without understanding what they represent . Unlike experts, students cannot move easily between standard representations of the same idea. Therefore, we developed resources that helped students make connections between ideas or events and their standard representations as well as between different standard representations of an idea. In addition, we developed nonstandard representations, mostly analogies, for some particularly challenging and abstract ideas. Besides engaging students, these novel representations were meant to help students link abstract ideas to more familiar and well-understood experiences .
We found, as did Nelson and Palumbo, that while knowledge construction is probably the most important goal for instructional technology, it is the most difficult to achieve . Students need to do more than have an opportunity to look at or look over material, they need to interact with it and get feedback on their developing ideas. We achieved this in two ways. First, we interspersed graded questions throughout the modules. Attaching points to the completion of the questions encouraged students to engage with the material. Then, in order to increase the quality of this engagement, a significant fraction of the questions required more than identifying or transferring information. Many studies have shown that students' achievement improves when the teacher's questions require the use of higher-order thinking skills by students such as inference, elaboration, or application [9, 10].
PRESENTATION OF KNOWLEDGE: GRAPHIC DESIGN AND CONNECTING IDEAS
When designing the presentation of knowledge in the modules, we considered both issues of graphic design and learning theory. In consultation with a technology specialist, we incorporated the following ideas about effective graphic design into our web page templates:
Consistent, highly visible headings: one designating the unit topic and one or two indicating the subtopic;
Page width constrained to 500 pixels to ensure convenient viewing on most monitors;
Large, sans serif font for ease of reading on-screen text;
Short pages to minimize the need for scrolling; and
No links to external pages that might confuse students' navigation
As described in “Conceptual Framework,” the design of our modules was informed by studies that showed that students often fail to connect individual pieces of information  and many textbooks fail to help them distinguish important concepts from details . We deliberately focused the modules on big, connecting ideas. For example, in the modules on photosynthesis and respiration, the presentation includes both content and question pages that make frequent reference to diagrams showing the overall cycles, reactants, and products. The molecules in the Calvin cycle are represented as Lego® building blocks where each block represents a carbon atom (Fig. 2). The goal was to help students see both pathways as cycles in which the carbon backbones of basic molecules are altered and not as a plethora of difficult-to-memorize chemical intermediates.
In the genetics modules, we emphasized the connections between genes and protein production. Ordinarily, students do not make these connections; instead they assume that proteins are not produced from recessive genes (unpublished data). Therefore, in all of the genetics modules, whether we are talking about Mendel's peas, cat coat color, or human genetic diseases such as muscular dystrophy, students are asked to make connections between protein production and particular alleles.
In order to focus on big ideas, we relinquished many of the details to the textbook. This is evident in a comparison of the number of words in the textbook and the modules used to present the same topics. To illustrate, we counted words on three randomly chosen pages from a standard introductory biology textbook  in two different chapters, “Respiration” and “Genetics.” Each chapter contained 21 pages; the average number of words per page was ∼500 and each page included 1 or 2 figures. The corresponding content in our modules was presented in 15 pages, of which eight pages included text and a diagram or animation and the remaining seven contained interactive questions. This suggests that the textbook used ∼10,000 words to cover what the modules addressed in just over 1000 words. What then did we give up to achieve such brevity? There are some topics that we did not cover at all. For example, we did not mention the methods by which humans are screened for genetic diseases, the genetic hazards of inbreeding, or the impact of nature and nurture on phenotype. We left such illustrative examples to the lecturer and textbook. For some topics, we actually gave fewer representations than the textbook. For example, the modules explored only one case of incomplete dominance and co-dominance, whereas the textbook had four. This complemented the textbook treatment by focusing student questions on a single case that required them to make connections across ideas. Other topics were simply addressed with fewer words than the textbook by not providing detail for all ideas and instead asking students to make inferences. In other modules we simplified the language and, therefore, sometimes the science, though faculty reviewers provided input to ensure that accuracy of content was not sacrificed.
REPRESENTATIONS OF KNOWLEDGE
An important part of an expert's understanding of biology is facility in the use of the standard representations of the discipline. These representations are considered “standard” in that people conversant in the field commonly use them. Students, on the other hand, often have difficulty using and moving between the compact, coded information of these representations [1, 3]. As defined by Bloom, the use of multiple representations requires “comprehension,” particularly the ability to “interpret” and “translate” .
One way to bring novices closer to an expert's understanding is to engage students' with novel representations of standard representations based on ideas already familiar to them . For example, to comprehend the process of respiration, students need to understand oxidative phosphorylation. This in turn requires a working knowledge of the electron transport chain. There are standard ways of representing the interrelationship of molecules and membranes of the electron transport chain, one of which is the cartoon shown below (Fig. 3).
This representation is packed with information, and students don't have a framework to help them make sense of it. To help students learn the important ideas embedded in this representation, we designed an analogy for the electron transport chain module. We used two animations to accomplish this. The first animation is “Electron Man” (Fig. 4), a comic book story in which Electron Man takes steps to generate energy to save his city. The second animation is that of the standard representation shown in Fig. 3. These two representations were followed by a series of questions to help students unpack the analogy and determine the significance of the electron transport chain.
Another approach to helping students understand standard representations is to explicitly present how different representations are related to each other . In the module on Mendelian genetics, we asked students to work with a variety of representations of genotypes and phenotypes. Biologists can look at a standard Punnett square and determine the genotypes and phenotypes of the parents. They know that the letters within the Punnett square represent predicted genotypes of offspring and can visualize the corresponding phenotypes. They recognize that letters on the outside of the square represent the meiotic products or gametes of the parents. A similar glance at a Punnett square by novices does not present this rich network of information. For example, we found that a large fraction of introductory biology students could not identify the gametes produced by meiosis in two parents despite the fact that they could draw the appropriate Punnett square on which the gametes are represented (unpublished data).
To help our students come to this understanding in our modules, we crafted an animation that linked the steps and products of meiosis directly to Punnett squares (Fig. 5). Meiosis was depicted with the usual bowtie representation of condensed chromosomes labeled with upper and lower cases letters representing different alleles (first representation). In the animation, the resulting meiotic products or gametes from each parent moved into the appropriate place on the Punnett square (second representation). The letters representing the expected genotypes of the offspring then appeared in the Punnett square (third representation). The letters representing offspring genotypes metamorphosed into the corresponding phenotypes (fourth representation). The phenotype pictures moved out of the Punnett square and sorted into the predicted ratios (fifth representation). Subsequent questions asked students to translate information from one representation to another. Work with more than one representation provided students with opportunities to construct a broader and interconnected understanding of scientific knowledge. By comparison, the standard textbook associated phenotypes and genotypes in Punnett squares for both mono- and dihybrid crosses, but did not present transmission of traits from parents all the way “back” to meiosis.
KNOWLEDGE CONSTRUCTION: INTERSPERSED QUESTIONS AND HIGH-LEVEL QUESTIONS
In addition to knowledge presentation and representation, knowledge construction plays an important role in fostering understanding [1, 2]. It has been widely accepted that knowledge construction takes place through students' inferences, elaboration on new information, and generation of relationships among information [1, 4]. In order to develop a deep understanding of the information, students need to actively engage in the process of knowledge construction rather than simply memorizing and recalling information as presented [1, 6]. In particular, high-level questions can facilitate the knowledge construction process . Studies demonstrate that questions associated with instructional material significantly influence students' achievement by guiding and stimulating their learning .
In our modules, we built the on-line learning environment to help students actively engage in the process of knowledge construction. To accomplish this goal, we developed modules with two main features: 1) content that is divided into small logical chunks interspersed with questions; and 2) graded questions that often required higher-level thinking.
We divided the content into manageable pieces and posed questions that required students to interact with the material before proceeding. Consideration of content density is critical in developing on-line materials. In contrast to the textbook material, on-line modules should present a smaller amount of content on a page to help students understand that content and provide them with incremental logical chunks to reduce cognitive load . According to student survey results, 94% of the respondents thought that the content, sequence, and layout of our modules were appropriate.
We also included two levels of questions in the modules, higher-level questions (Level 2) as well as lower-level ones (Level 1). Both levels of questions were designed to encourage interaction between students and learning materials along with instant feedback. Question types that we used were multiple choice, multiple true/false, and mix-and-match. We developed the foils for some questions based on students' preconceptions to facilitate active knowledge construction.
Level 1 questions required students to use the information on module page(s) directly or with minimal interpretation. These questions correspond to the knowledge level of Bloom's taxonomy . Level 1 questions were designed to guide the students in their examination of the details of the presented information. Level 2 questions, corresponding to the comprehension and application levels of Bloom's taxonomy, promote more active interaction with the materials and higher-order thinking. They require students to: 1) apply the information and concepts from the page(s) to new situations; 2) draw inferences and elaborations from the information; and/or 3) synthesize information from several pages.
Examples of the two levels of questions are presented in Fig. 6. Both of the questions are from the Mendelian genetics module. The page preceding each of these two questions is a content page describing the “terminology of genetics,” in which terminology was represented in an animation. Answers to the question on the “Vocabulary and Standard Cartoons” can be found on the previous content page(s). However, the answer to the question of “Why Are Gametes Haploid” is not directly presented in any of the pages. Students need to analyze and apply presented information to answer this Level 2 question.
Table I compares questions in our on-line modules with those in the textbook  used in our biology courses and shows the different percentages of Level 1 and Level 2 questions. Questions in the textbook are found at the end of the chapter and are typically Level 1 questions. On the other hand, our supplemental and complementary modules placed more emphasis on Level 2 questions. Thirteen representative modules were analyzed as well as three related sections from the textbook. For comparison purposes, we defined a question page for the modules as a web page that included a question along with some supporting text. For the textbook, a question page was equivalent to a single question in the end-of-chapter problem sets or in supplementary text materials.
The percentage of question pages in our modules varies from 25.0 to 83.3% with an average of 57.1%. In the textbook, the average percentage of question pages is 71.1%, where again each question is equivalent to one question page. Of the module questions, 58% were Level 2 while only 22% of the textbook questions were Level 2. As we intended, the on-line modules asked students more high-level questions than does the textbook. These questions were expected to better facilitate students' engagement in a higher level of knowledge construction [10, 13].
At the end of the semester, students in two different introductory biology courses were asked to respond to an unstructured, open-response question about the modules and how they affected their learning. The characteristics of the two courses were slightly different. One was a large, university-wide introductory course with over 300 enrolled students. The primary instructional mode was lecture, and 11 modules were used as graded homework assignments. The other was a much smaller course housed in the university's residential science college with ∼100 enrolled students. The instructional mode in this course was also lecture, and three modules were used as graded homework assignments.
Despite the differences in class size and number of modules used in the two courses, the responses to the questionnaire were remarkably similar. All responses were coded to determine to what extent students identified the design features of the modules as being helpful in their learning of biology. Overall and across both courses, the percentage of students who felt that the modules were helpful was 75%. They expressed views similar to: “I did find [the modules] to be helpful and [they acted as] another supplement to the things we were learning in class” and “[the modules] approach gave me a better understanding on how each process worked.”
Some students offered more specific responses. Representative responses identified as aligning with the three design principles (presentation, representation, and construction) are as follows:
“I thought [the modules] were a good way to stress the important topics of the chapter.”
“[The modules] allowed me to learn the material by having me make connections.”
“The explanations, diagrams, and movie/slide shows helped a lot to visualize what was happening.”
“The pictures and movie clips help to visualize how things work.”
“[The modules] make you apply information that you have learned into more difficult questions.”
“It helps me to sort through and further comprehend the material I am reading.”
Each category was mentioned by 20–25% of the students.
Of the 25% of students who responded with negative comments, the majority raised issues with the course itself rather than the modules. Of those students who did express frustrations with the modules, responses tended to focus on the challenging questions. For instance, one student said “sometimes the questions are difficult and I can't (for the life of me) get the correct answer.”
Table Table I. Comparison of the types of pages and questions in standard textbook and modules
Level 2 questions
Electron transport chain
We thank Scott Harrison, Gail Richmond, Randy Russell, and XuanLi Yao for conceptual and technical contributions to the modules. We also thank the graphic design work of Elizabeth Anderson and Marlene Cameron and the expert technical support of Jiatyan Chen.
The modules utilized a course management system developed at Michigan State University by Gerd Kortemeyer and colleagues called LON-CAPA. LON-CAPA is a distributed open-source Learning Content Management and Assessment System that enables instructors to create educational materials and to share such learning resources with colleagues across institutions in a simple and efficient manner. This innovative content management system allows great flexibility in the design and use of web-based course materials. More information about LON-CAPA is available at www.lon-capa.org.