Learning Focused Course Transformation

At the United States Air Force Academy, a learning-focused transformation of Biology and Physics core courses was made to support deep student learning. This involves first transforming the learning goals from using terms like "list, find, calculate, describe, use, what, and when" to "explain, analyze, apply, create, predict, and evaluate". Students are also exposed to familiar and concrete settings where the knowledge can be applied. E.g. students are asked to serve as "expert witness" in a trial which requires their knowledge on gene expression, or they have to explain how spies can tap phone lines during the Cold War using Faraday's law. Class lessons include mini-lectures (about 10 minutes) and the rest of the time is mostly spent on learning experiences through activities and exploration. Instructors become learning facilitators rather than just lecturers.

The ultimate question is whether students learn just as much from activity based lessons as from traditional lecture style of delivery of content. In Computer Science education, students are invariably exposed to learning through activity based programming assignments since the nature of the discipline is mostly practical and applied. However, programming is not the only activities that students can be involved in, even though it is the most natural one. Especially if one of the learning goals is the development of abstract thinking, care must be taken not to over-emphasize the programming aspect that may monopolize the time and attention the students should spend. Programming projects can take up a lot of time and the students may end up gaining programming skills and not other skills. Here is where concise learning goals need to be articulated and the proportion of time for each learning goal is matched appropriately with the learning activities.

Reference:

Sagendorf, K., Noyd, R., Morris, D. (2009). The Learning-Focused Transformation of Biology and Physics Core Courses at the U.S. Air Force Academy. Journal of College Science Teaching. January / February 2009.

SIGCSE Presentations on Making CS More Relevant & Engaging

Here are a few highlights from SIGCSE 2009, Chattanooga, TN. These highlights are about making computer science "interesting" and "more relevant" to students that normally bypass traditional CS courses:

1. "Rediscovering the Passion, Beauty, Joy, and Awe"--panel presentation:

Dan Garcia (Berkeley): CHANGE (Obama style?) has come to computing. Let's avoid the old style of syntax-driven CS 0/1 curriculum. Let's let students choose their own projects, mix of CS courses, partners, etc. In elementary schools, parents help out after school; so, why can't we get [CS-type] moms & dads to help out in the computing club? "But, we want a winning basketball team, so not just any mom or dad will do." When constructing assignments and putting together lecture material, think relevance! We need to motivate students so that they will want to spend hours of their own free time on programming ... just for fun.

Eric Roberts (Stanford): CS enrollment at Stanford is "skyrocketing", wiping out previous CS1 losses post-dot-com. Programming continues to be a very important skill that needs to be emphasized and taught more effectively. "The best programmers are several orders of magnitude better than the average programmer." At Google, for example, they represent one-tenth of 1% of the applicant pool. The best programmers are 300 times as good/productive as Google's typical programmer.

Zachary Dodds (Harvey Mudd College): suggests a breadth-first approach to CS, including functional programming. Right now, "What is learned is the square root of what is taught."--implying not much is learned/remembered from a typical CS course.

2. Microsoft's exhibit on computational thinking. An excellent book, edited by Yan Xu (former MSc student at UBC) is: Transform Science: Computational Education for Scientists (CEfS), Special Edition. Microsoft Research was giving away free copies at the conference. Many authors, including former UBC STLF Beth Simon, contributed 1-2 page positions and short research papers on educational themes in computer science, focusing on the current mismatch between typical CS programs and typical science programs. For example, many authors claim that CS courses are not serving biology, chemistry, physics, earth & ocean science, biological engineering, etc., students very well. We need to develop new courses that take the highlights of numerous first, second, and third year CS courses, and condense them into digestible units for such non-CS students. Such highlights include topics in programming principles, an easy-to-learn powerful language (say Python), discrete math, algorithms, complexity, scripting, database topics, etc., should be made available to biology students using examples (taken from biology, etc.) that are relevant and interesting to the students. Forget about command-line driven programs that compute interest, etc.; instead, stick with real bioinformatics examples. Incorporate visualization techniques, problem-driven applications, modelling, simulation, etc.

3. Owen Astrachan created a new CS course at Duke University for arts students, theater students, varsity athletes, would-be lawyers, etc. Owen created a very interesting, non-programming, non-math, CS course that would appeal to non-CS/non-science types. And it did! 250 students enrolled in his experimental and engaging course. He got several guest speakers (some of whom were Duke alumni) to speak on their areas of expertise. The topics included case studies of network protocols, privacy issues, social issues, copyright issues, high-profile/controversial law cases, etc. Owen gave us take-home copies of his midterm and final exam. The exams included questions on IETF standards, Internet voting, DNS servers, security, Skype and security, worms, Flickr, BitTorrent, Richard Stallman's Free Software Foundation, copyright act and RIAA, IPv6, spam, cookies, iTunes, iPhone, P2P, etc.

Knowledge Transfer Assessment

There are four ways to test whether students have acquired knowledge transfer skills according to Mayer (2001):

1. Troubleshooting - by asking students why a system does not work.
2. Redesign - by asking students for a redesign of a system for a different purpose.
3. What-if - by asking students what would happen under other conditions.
4. Principle - by asking students the function of a component in the system, or why a component behaves the way it does.

In Computer Science, this is pretty easy to do. Here are some examples:

1. Troubleshooting - have students debug a program, or discover and debug another student's program.
2. Redesign - have students build another version of an application.
3. What-if - have students compare and contrast the use of different algorithms, database design, logic design, infrastructure, etc.
4. Principle - have students construct context diagram, use case diagrams, etc. and explain how one component functions within the entire system.

References:

Mayer, R. (2001). Multimedia Learning. New York: Cambridge University Press.

Too Much "Seductive Details" In Lectures

Bet that title caught your attention ... sex and death are inherently interesting (Kintsch, 1980). So if you want to spice up your lectures, inject some sex and death. But make sure they are relevant. Seductive details are high interest to students but if they are irrelevant to what the students are learning, they are considered as extraneous details, and they actually have a negative effect on student learning.

Learners have only a limited amount of processing capacity available to them for learning. Like a battery, if the energy is wasted on irrelevant material, there is just not enough energy left for the relevant material. What's more, high interest details take up more energy than low interest details, so if you add more "seductive" material in your presentation, you are leaving your students with less energy for the more important material you want to present.

Mayer et al's article (2008) concluded that increasing irrelevant details even though they are of high interest does not appear to affect learners in their understanding of material (as measured by their retention of material), but they do disrupt their construction of a coherent mental model of the to-be-learned system (as measured by their transfer ability performance).

References:

Kintsch, W. (1980). Learning from text, levels of comprehension, or: Why would anyone read a story anyways? Poetics, 9, 87-98.

Mayer, R., Griffith, E., Jurkowitz, I., Rothman, D., (2008). Increased Interestingness of Extraneous Details iin a Multimedia Science Presentation Leads to Decreased Learning. Journal of Experimental Psychology: Applied. 14(4), 239-339.

Affect and Cognition

Learning objectives can be classified into three domains: cognitive, affective, and psychomotor (Bloom Taxonomy). Affective domain refers to learning outcomes in areas such as emotions, moods, attitudes, and feelings, which has been shown to strongly influence cognitive outcomes. As an example, the production of adrenaline and dopamine during emotional experiences affects the transfer of information from short term to long term memory, as well as the level of motivation and cognitive engagement.

Computer Science education is heavy on the cognitive domain but not so for the affective or psychomotor domains. If according to Bloom that holistic education should include all three domains, then how can this be incorporated in CS education? Other disciplines of study have used fieldwork to promote development in these domains. Students go on fieldtrips and work in groups in their learning, as well as developing friendships during the off hours social functions and activities. Engineering students have their share of pranks and parties, and for computer science students, one of the most popular forms of team activity is online video gaming. But these may not be enough to attract the students to these programs or even promote these program to the extent we like to see especially for female students. Perhaps we need to revisit the "art" of computer programming as Knuth has proposed. There is the construction of a beautiful program which seems to be missing in our current CS education, where programming can give our students both intellectual and emotional satisfaction (Ershov, 1972).

Reference:

Ershov, A. P. Aesthetics and the human factor in programming. Comm. ACM 15 (July 1972), 501-505.

Student Perceptions of Their Grades

In a recent study of 278 students from an introductory biology course at the University of Minnesota where students were asked to predict their final course grades after each of the four exams given throughout the term, the following are some of the findings:

• On the first day of class, more than 90% of the students believed they would earn an A or B in the course.
• Students who earned A's and B's in the course at the end predicted they would earn lower grades than they actually earned. Students who earned C's, D's, and E's predicted they would earn higher grades than they actually earned.
• Students who earned high grades were more likely to attend class, submit extra assignments for credit, and attend help-sessions than students who earned low grades.
• On average, it is unlikely that students will significantly improve their grades deep into a semester.

Students are poor predictors of their grades and thus often fail to regulate their learning. First year students are usually not used to the expectations and demands of college level courses, which may explain the reason for their high expectations at the beginning of the course. The students need to be made aware of this and appropriate support put in place to encourage them to participate in order for them to achieve their goals. Perhaps the extra assignments / help-sessions / etc. should be made mandatory to assist first year students to transition into college life.

Reference:

Jensen, P., Moore, R. (2008). Students' Behaviors, Grades & Perceptions in an Introductory Biology Course. The American Biology Teacher. 70(8), pp483-487.

Self-Regulated Learner

A self regulated learner (SRL) is one who knows what she wants to learn and be able to monitor and adjusting her learning in achieving her goals. This involves metacognition, motivation, and strategic actions. The diagram on the right is a metacognition model by Winne and Hadwin. "Metacognition is the awareness learners have about their general academic strengths and weaknesses, cognitive resources they can apply to meet the demands of particular tasks, and their knowledge about how to regulate engagement in tasks to optimize learning processes and outcomes."

The diagram looks too simple and cryptic. Starting with the 4 phases in the bottom right, it shows that the learner goes through these 4 phases in self regulated learning. These phases are not linear, that is, the learner may switch from phase 3 back to phase 1 and then to phase 4, etc. In any case, here is a simplistic explanation of the diagram. After realizing what task is to be learned (phase 1), the learner checks with the external and internal conditions (external conditions are conditions external to the learner, like the time / resource, etc. available, and internal conditions are the motivational factors, beliefs of the learners, etc.), the learner engages in some operations to learn what needs to be learned. In the process of learning, she compares the expected Standards she have constructed (e.g. hitting the golf ball straight, or solving a differential equation), and the Products (or results) she is experiencing (e.g. the golf ball went to the left, or the answer to the solution of the differential equation is different from the answer in the textbook), and this results in the Cognitive Evaluations. The learner then may need to revise the goals and plans (Phase 2), repeat the process, or study and find other tactics (Phase 3), and through further adaptions of these learning processes, continue to evaluate the Standards (which may also be revised), and further compare with the Products of her learning.

In computing, the task that students usually encounter is in the from of creating a computer solution for a problem. The usual tactics we provide the students in their learning include: lecture / lab materials, textbook, previously solved problems, google, etc., and students may explore all these in their learning. Learning goals are useful especially if learning goals are constructed in the form of a semantic map so students can refer back the supporting learning goals so they can reassess whether they have learned these to continue. This is also part of self-regulated learning. In any case, students often find themselves "stuck" in their assignments. What other ways can we help them get "unstuck" so they can continue in the process of self-regulated learning?

Reference:

Winne, P., Perry, N. (2000). Handbook of Self-Regulation. Edited by M. Boekaerts, P.R. Pintrich, M. Zeidner. Academic Press.