Labor Saving Techniques
Given that a major goal of the course re-design project is to substitute technology "capital" for faculty teaching "labor," in what particular aspects of the course and its delivery are you finding that you are able to do this?
A lot of the work we are doing now will save time in the future; however, we are not seeing the benefits of the labor saving aspects of the course because we are involved both in teaching the course and in implementing the re-design. We have faced considerable technical issues in trying to work within WebCT and in trying to get WebCT and other online course management tools to do what we need. Our approach has been to develop software tools to automate repetitive and time-consuming activities, including some of the interactions with WebCT, so future offerings of the course will have faculty time savings.
We are presenting one of our tools, Labrador, at a WebCT conference in July 2002. This tool has saved considerable faculty and TA time in retrieving student files submitted though WebCT and disbursing them to the plagiarism detector (Moss) and to the TAs for grading. Another tool in development creates an improved environment for creating quiz questions and automatically uploading them in the correct format to WebCT or allowing older quiz questions to be revised much more easily than within WebCT. We also have a newly developed Web crawler tool that can go through the chat and discussion sessions and remove all student names and nicknames (substituting anonymous identifiers) so chats can be disseminated to the class for additional discussion without revealing the identity of the participants. This tool will be used to share interesting chats and to correct misconceptions that arise during chat sessions.
Because of the way WebCT is configured to handle course material from term to term, bringing assignments forward to the next term has been time-consuming. We are developing software tools to assist in this process as well as learning how to circumvent some of the problems in using WebCT. The main time savings compared to the traditional course has been in having everything be electronic: handing out assignments, having students submit assignments, having TAs grade assignments, submitting the assignments to the plagiarism software, and returning the graded assignments to the students. To achieve these savings, we have created software tools which should yield savings in the future.
The real advances have been in “soft” areas such as mentoring students: the re-design has resulted in a shift in activities from delivering content in a lecture format to interacting with students in both the “lecture” and lab sessions. Since content is handled online, we have been able to experiment with large group activities, such as pair-and-share, even in the lecture sessions. We are still experimenting with the best format for the lecture, particularly since students are asking for additional class hours (with additional course credits) so they can interact more significantly with the instructors.
The re-designed course also appears to be motivating the students to read the book and do the work. We believe this is more important to effective learning than having the instructor deliver content. It is not that we are making content delivery wonderful (though we are trying to that as well) so much as that we are getting the students more active in acquiring content.
Discussion groups and chats within WebCT provided good opportunities for mentoring. In class, student discussion is limited by time constraints: students need to think about answers to questions before responding or being given the answer. The online capability allows students to think about their questions before submitting them, knowing that they can get an answer at any time. Thus, a student can go home, think about something for 2-3 hours, and then ask deeper questions and expect to receive an answer fairly quickly. These questions normally would not arise in classroom.
The changes in course delivery allow faculty and TAs to interact more with students during the extended laboratory times. However, monitoring student interactions on discussion and chat groups or replying to large numbers of email messages has increased the overall workload. As the class size increases, however, there may be some economies of scale.
There is also a downside: If a single instructor is handling over 50 students, the email gets to be too much. Some students try to interact through email exclusively even though it is not best way to handle all matters, particularly those that involve repeated email messages going back and forth.This can happen when students are arguing for more points on an assignment/test. We plan to create a “Corresponding TA” position and train the person to answer questions and refer students to the most productive channels for getting their problems resolved, thus relieving the instructor of this large burden.
Use of WebCT for sharing files between faculty and staff was more useful than anticipated. It allowed multiple people to easily access the same material and work on it for the course. Having bulletin board areas in WebCT was very useful for communicating important messages to students. It was more efficient and reliable than email and easier to deploy.
We made use of the online facilities to have faculty and TAs hold online office hours in the evenings when many students are working on their assignments. Students could ask questions online both privately and publicly. TAs and faculty could answer from the comfort of their homes, thus providing expanded services without inconveniencing faculty and staff by having them in the office at night. This provided a time savings for students in their studying as well.
As part of the re-design, we were able to give a quiz in every lab using WebCT. The quizzes were automatically graded with feedback to the students as soon as the entire class had taken the quiz. To keep the quizzes from being too high pressure, each one constituted only one of 30 small assignments feeding into 10% of grade. However, by making the quizzes part of the grade, students took them seriously.
A test bank was created for each quiz with 13 multiple choice questions; each student was asked 7 of them. Since later sections seemed to perform at similar levels to earlier ones, students did not appear to be organizing to get answers. Thus cheating was not a problem. However, creating a reasonably large database of well-tested, unambiguous multiple choice questions of similar difficulty is a time-intensive process and is ongoing. Once a reasonably large test creation database exists, this will be an area where there will be a significant a time savings for future faculty.
There were also technical issues in creating quizzes in WebCT: WebCT does not handle quizzes and self-help material in the same way, so making answers available was a problem; WebCT did not always know right answer; formatting quiz questions was a problem, particularly with fixed-width fonts (needed to make programming code easily readable), especially in the Respondus exam-creating application—material can not be graded if pasted in as an image, and it is awkward to make it work in HTML. We are planning to work on these technical issues to make the self-assessments really useful to the students.
At the moment, the primary feedback to the students is a raw score. We would like more detailed feedback on where students need to focus their studying to improve their programming skills. WebCT and Respondus quizzes do have the option of providing additional feedback to correct or incorrect answers if feedback is given immediately, but other tools would have to be created to track and give feedback based upon multiple questions or instruments of assessment.
Online quizzes provided the opportunity to give extra advice to students about wrong answers. Having all of the material online made it easier to look over student work, know what was turned in and what wasn’t, and assess the student’s strengths and weaknesses with respect to mastering the course material. Instructors could look up records directly and spent less time fussing with paperwork, which left more time to talk to students about more substantive issues.
December 2002 Update: Because of Convocation in the fall term, a lecture was lost during the term. The instructors had the students read the lecture notes as a pre-lab exercise and then take a quiz online about the notes. The quiz was taken at home with “open notes”: the goal was to make sure the students went over the material and learned it. This approach worked very well, and it raises the question of whether a similar approach can be taken for other lecture sessions, keeping in mind that some students feel too much is being put on their shoulders and “not enough is being done by the instructors.”
Program in Course Redesign Quick Links: