- Home
- Articles
- Reviews
- About
- Archives
- Past Issues
- The eLearn Blog
Archives
To leave a comment you must sign in. Please log in or create an ACM Account. Forgot your username or password? |
Create an ACM Account |
The text-based interaction of e-learning arguably creates more work for an instructor. That, coupled with the lack of face-to-face contact with students, makes teacher absenteeism easier to rationalize (and harder to notice). Online instructors often go absent from their classes for spans of time simply not tolerated in the traditional classroom. The irony is that the current model of e-learning sets student needs and instructor workload in opposition-online students need interaction with their instructors far more than their face-to-face counterparts. It may seem politically expedient to ignore such a problem. However, instructors, students and universities would all benefit from practical solutions to this dilemma.
Student satisfaction in online courses is highly correlated with interaction with the instructor (Shea, Swan, Fredericksen, Pickett, 2001; Trippe, 2001). Online students need frequent support from the instructor to allay feelings of isolation and insecurity. Devoid of the visual cues and interaction of the traditional classroom, students need instructors present throughout of the course; to give complete and explicit directions and to provide ongoing feedback.
The current model of asynchronous e-learning emphasizes a one-to-one style of communication that has many educational advantages, but also disadvantages. In contrast, the face-to-face lecture hall lends itself to "one to many" communications. A few vocal students may directly engage the instructor, but most students benefit, passively receiving "one to many" communications. In e-learning, however, each student is expected to participate and the online instructor faces an overwhelming workload when responding individually to every student. If each online student in a class of 25 posts four times a week and the instructor responds individually to each posting, the instructor posts at least 100 times a week. No wonder many full-time faculty at research universities shy away from teaching online.
Because reading and replying one-to-one with text is much more labor intensive than the oral exchanges in a face-to-face class, e-learning courses usually require more work than face-to-face classes (Connone-Syrcos and Syrcos, 2000). Professors teaching online courses have many other responsibilities. Full-time faculty write articles and grant proposals, handle administrative tasks, and teach other classes. And adjuncts usually hold a full-time job. During a busy semester when deadlines come up and personal lives intrude, these other demands sometimes seem much more urgent than posting up to 100 times a week.
Sources of Data
Those who oversee online instructors repeatedly field student complaints about instructors absent from e-learning courses for considerable periods of time. Unobtrusive examinations of these problem e-learning courses revealed that students were receiving little feedback from the instructor. The instructor would often say that she had been monitoring student postings and would post if the students went off track. These instructors need to be reminded that online students have a psychological need for instructor presence. In most problem cases, instructors were up-front, acknowledging other pressing professional or personal demands which would soon ease up, allowing them to once again participate more fully. However such intermittent absences are distinctly unfair to students. An instructor absent from a third of the lectures would not be tolerated in the traditional classroom. "AWOL" online instructors do not represent the majority, but they certainly represent a significant minority, perhaps a large as a quarter of all online instructors.
Why are many online instructors intermittently absent? Is the volume of online work and other job demands overwhelming? Is it unrealistic to expect online instructors to communicate regularly one-to-one with students? Pilot studies conducted by master's students presented strong evidence that online instructors and students have divergent perceptions about instructor-student interactions. A graduate student, Michelle Ferreira (2000) conducted a pilot study of 20 students and their two professors and concluded, "A large disparity exists between what the students perceived … versus what the professors were actually doing [to create social interactions]." Because of the limited number of participants in the pilot study, little can be generalized. We (a professor and a graduate student), conducted a comparable, but more extensive study, this time involving 59 online students and 25 online instructors from a number of universities in New York state. The investigators first conducted email and phone interviews with open-ended questions and then created online questionnaires.
Virtually all 25 online instructors said they went online frequently, often twice a day, to read and answer all student questions and thus provide plenty of timely feedback. However, more than one-fourth of the online students surveyed said that they received little or no feedback. Apparently there is a problem: 25 percent of students is a sizable percentage, more than should be ignored. Students mentioned online discussion areas as particularly notorious for a lack of instructor presence.
The conventional wisdom in e-learning is that online discussions belong to the students. Online discussions are best stimulated, not by strong statements, but by instructors answering discussion postings with non-obtrusive Socratic questions. It is the instructors' responsibility to facilitate the collaborative learning process by stimulating student-to-student dialogue (Palloff, Pratt, 1999). Experience suggests this is valid, however with multiple work demands it may be easy to slip from online facilitator to invisible observer. From the online survey section of the study, 95 percent of the instructors said they spent a lot of time replying to student queries, yet 85 percent of the students said they felt insecure, isolated or confused at least part of the time. There is a discrepancy here. Something is clearly wrong. According to the survey, the majority of instructors often respond to student questions by email. In fact, email is the norm for asking and answering questions for many instructors and students. Other students, not seeing the instructor's emailed response to other students, frequently ask the identical question a number of times. Therefore email, with its one-to-one communication format, isolates students and creates more work for the instructor.
The pedagogical model of today's e-learning classes stresses student autonomy and initiative. Leaving the teacher-driven, face-to-face classroom to enter a student-driven online class may be difficult. Students in the study said work and family demands keep them from clarifying ideas with their classmates. Some instructors said that they had trouble getting students to participate in group activities.
Face-to-face courses usually have a bell-shaped normal grade distribution, while online courses often have a bifurcated U-shaped grade distribution (Smith, Ferguson, & Caris, 2002). Because many students are new to online learning and unfamiliar with its demands for self-reliance, many students disappear after the first few weeks. The remaining students likely earn good grades. This may stem partly from the structure of online educational environments and a teaching model that fails to properly address online student insecurity.
One challenge for studies like this is reaching students and instructors in a number of different online learning organizations (Taveras, 2003). Because of issues of privacy, administrators are loath to release email lists. However, the common-sense analysis of current college e-learning and these two studies make it apparent that although instructors feel they provide sufficient feedback, a significant proportion of online students feel otherwise.
Another problem is that studies investigating shortcomings of online learning are politically incorrect. E-learning's ability to reach new student populations is a positive societal force and a booming business. No one wants to rock the boat. However if there are problems with instructor-student interaction in online classes, then commercial and educational needs are best served by addressing those problems instead of ignoring them.
Possible Solutions
There is a structural flaw with online learning that results in intermittent (but significant) instructor absences. What is to be done? There are two approaches to addressing the flaw: adapt to the existing model and infrastructure of e-learning, and redesign the infrastructure and model.
Adapting: The first step in correcting these flaws in online learning is to acknowledge them and then bring them to the attention of administrators and e-learning instructors. It is not enough for the instructors to read students' postings and respond only when students go off track. If students in online courses don't feel a strong instructor presence, students start to feel insecure and participate less in the course. However, students also need to understand that taking an online class does not mean less work, but more. Online education is time-consuming and demanding. Administrators need set the tone and spell this out to professors and students.
Administrators can set up rules for professors to follow in their online courses. Professors may generally follow rules, such as "Post at least three times a week." But there will never be universal compliance. It is difficult to effectively enforce rules about participation. Our department debated the possibility of an "Early Warning System" for instructor absences. Partway through the semester, online students would be polled for quality of interaction. The information would be available to the director of online courses. However this solution smacks of "big brother" and violates the spirit of academic freedom.
A friendlier approach combined with rules may yield better results. Mentors to e-learning instructors can offer practical suggestions to help improve instructor-student interaction. A formal "Questions" area in each instructional module can bolster instructor presence. The first thing that the instructor should do on entering the online course is answer questions about current assignments. Nothing breeds online student insecurity faster than unanswered questions; nothing alleviates it faster than answers.
Many online students email their professors voluminously. Professors often respond to the same question many times with different students. A practical rule to save instructor time is to limit all non-confidential student questions to public forums within the online course, so that the other students will also see the instructors' response.
A simple and linear e-learning course design can go a long ways to minimizing student insecurity and making the instructor's job easier. Another simple rule is to keep only one module open at any one time. This means that students and instructor need only focus attention on that one current module.
Small things can have big effects. If the instructor changes the course announcement every time they log in, including the content, color and font style, students will sense instructor presence.
Online instructors can maintain course presence more effectively if they understand the function of instructor postings. Through content analysis of instructor postings from hundreds of online courses, Blignaut and Trollip (2003) derived a taxonomy with two broad categories, postings with academic content and postings without. Postings with academic content are of three types:
Postings without academic content fall into three categories:
By understanding the functionality of different types of postings, online instructors can improve course quality and student satisfaction.
There may be other useful frameworks for analyzing online courses. We suggest analyzing online course communications in terms of three types of visibility: first topological visibility, how many people are technically allowed by the software to see a posting, second probable visibility, how many are likely to see a posting, and third, actual visibility, how many actually see a posting. Topological visibility includes three common categories:
Topological visibility puts a hard limit on how many persons in the class can view posting. However it does not determine how many people will be likely to see a particular posting, nor how many will actually see a posting. All one-to-class postings are not equally likely to be seen (probable visibility). An instructor announcement on the opening screen will likely be seen each time someone enters. An instructor posting on an obscure branch of a threaded discussion will likely be less viewed.
Student reading patterns (i.e., how many and which students view which postings) is actual visibility. Online course management systems could record, analyze and provide a visualization of who reads which postings. The authors suggest the "iconic footprint," a simple technique for visualizing student reading patterns. At the beginning of the semester, the instructor and each online student could use a paint program (pixel editor mode) to create their own logo or iconic footprint to be automatically affixed to every posting they read. By glancing at the collection of iconic footprints on a posting, students and instructor can instantly see who and how many have read a posting. Not only would digital footprints provide a simple way to visualize reading patterns, it would give online students an increased feeling of individuality and identity. Online course-management systems should also statistically measure the actual visibility for different categories of postings over the whole semester. Such increased instructor accountably may also have disadvantages.
Radical approaches: Even if instructors adapt to the current e-learning situation, the structural flaw will remain. In e-learning courses, there is typically a ration of one instructor to 25 students. Students expect one-to-one interaction with the instructor. Labor intensiveness of the text format and the instructor's competing work demands make this awkward. Perhaps we need to fundamentally change the topology of e-learning interfaces.
One possibility is a system of hierarchical peer-reviewed group assignments developed by David Hanson and Troy Wolfskill (Borman & Washington, 2003; Wolfskill, 1998) as an online component in face-to-face chemistry courses. For problem-solving assignments with multiple solutions, students are put into groups of five. Once every student in the group submits a possible solution, the group then discusses and selects the best one, finally submitting it to the whole class. The online environment is "hard-wired" so no student can participate in the group discussion without first submitting an individual assignment. Once every group has submitted their solution, the whole class, including instructor, discusses the group solutions and selects several as outstanding.
Neural Networks (NN) and Artificial Intelligence (AI) also have some potential for helping with problems of instructor presence. Within the e-learning environment, an AI or NN program could monitor for signs of "healthy" or "unhealthy" patterns of participation and report back to the instructor. This may sound futuristic, but it could be as simple as searching for unanswered questions in open modules and alerting the instructor. Semantic searches could find clustered "distress" phrases: "help…," "What's going on…," "Does anyone know…," and "How do…" AI/NN could measure course vitality by examining reading and posting patterns. Even the simple depth of the discussion tree is a sign of discussion vitality (Swan, 2001).
Threaded discussions reach a certain threshold size, they are notoriously cumbersome to read and comprehend. An AI program could reorder a threaded discussion, grouping related postings together to make them comprehensive. The instructor could annotate this unified version of the discussion and post it as a discussion summary. This would help address one common student complaint, lack of instructor presence in threaded discussions.
However, automated AI approaches also have their dangers. Currently, because instructions are written before the course opens and are sometimes so complete, online instructors already think that their students need only intermittent attention. With more automated monitoring systems, if nothing shows up on the artificial intelligence "radar," some instructors may be even more likely to disappear. It is important for students to feel an instructor presence in the course, not just when students are puzzled, but all the time.
Conclusion
The physical remoteness of e-learning makes interaction with the instructor the most important factor in student satisfaction. It is that same physical remoteness which makes instructors a little more lax and a little less accountable. Academe needs to acknowledge the problem and start seeking honest answers. The software infrastructure necessary for e-learning makes instructor accountability technically easy. However, prevailing attitudes about academic freedom make tacking the problem politically complicated.
New models of e-learning emphasizing peer evaluation might ease the laboriousness of instructor-student one-to-one text communication. New more intelligent software tools might make the online instructors' job more lively. For example, suppose providing instructor feedback to student online postings was as simple as a few strokes of a magic marker circling and annotating text passages, math notation, and diagrams. This would be possible if online documents used a metaphor of clear plastic "transparent overlays," one text, another graphic, etc. This is certainly technically feasible, but it challenges entrenched notions of what an online document is and forces educational institutions and course management companies to revisit "sunk costs," huge economic investments in existing technical solutions.
Imaginative instructional design and new technical tools provide some hope for the future. However, the ultimate answer to lack of instructor presence is the human solution: a more conscientious and dedicated instructor.
References
1. Blignaut, S. & Trollip, S. (2003). A taxonomy for faculty participation in asynchronous online discussions, published conference proceedings of Edmedia 2003, June , Honolulu, Hawaii.
2. Borman, S. and Washington, C. (2003). Nontraditional teaching: Online conference focuses on methods that enhance or replace conventional lecture format, Chemical and Engineering News, March 10, 2003, Volume 81, Number 10, CENEAR 81 10 pp. 45-47, ISSN 0009-2347, http://pubs.acs.org/cen/education/8110/8110education.html
3. Connone-Syrcos, B. and Syrcos, T.P. (2000) Computer-mediated Communication in Distance Education, in book, International Perspectives on Tele-learning and Virtual Learning Environments, G. Orange and D. Hobbs (Eds.), Ashgate Publishing Co., Burlington, Vermont, 2000
4. Ferreira, M. (2000). For a Distance-Learning Class to be Successful, What Types of Student-Centered Instructional Interactions are Necessary for Positive Learning Outcomes?, unpublished paper submitted for course credit in EST 571, at Stony Brook University
5. McKlin, T. (2003). Artificial Neural networks to help online learning, published conference proceedings of Edmedia 2003, June , Honolulu, Hawaii.
6. Palloff, R M; Pratt, K; (1999). Building Learning Communities in Cyberspace, Effective Strategies for the Online Classroom, Jossey-Bass Publishers, San Francisco: CA
7. Shea, P., Swan, K., Fredericksen E., & Pickett, A. (2001). Student Satisfaction and Reported Learning in the SUNY Learning Network, in book, Elements of quality online education, Volume 3 in the Sloan-C series, The Sloan Consortium New York, NY
8. Smith, G. G., Ferguson, D., & Caris M. (2002) Teaching Online versus Face-to-face, Journal of Educational Technology Systems, 30, (4), 337-364
9. Swan, K. (2001). Immediacy, Social Presence, and Asynchronous Discussion, , in book, Elements of quality online education, Volume 3 in the Sloan-C series, The Sloan Consortium New York, NY Taveras, M. (2003). A survey of instructor/student perceptions with respect to social interactions in Asynchronous On-line courses, unpublished Master's Thesis, Stony Brook University
10. Trippe, A. (2001). Student Satisfaction at the University of Phoenix Online Campus, in book, Elements of quality online education, Vol. 3 in the Sloan-C series, The Sloan Consortium New York, NY
11. Wolfskill, T.(1998). http://www.chem.sunysb.edu/hanson-foc/lucid.htm
To leave a comment you must sign in. |
Create an ACM Account. |