dr. daniel l. lau

Before describing the history and theory of the Lau Teaching Paradigm, let me open this teaching statement by highlighting my contributions to the Electrical Engineering curriculum at the University of Kentucky. Perhaps the most important of which is that I am the first faculty member to identify Digilent’s Electronics Explorer board as a valuable teaching tool that needs to be integrated into many of the department’s classes. I first discovered the board while visiting my advisors at the University of Delaware who were using the board in their introductory circuits courses. The board is a powered breadboard that includes two channel signal generator, two port power supply, four channel oscilloscope, and 32-bit input/output logic analyzer. It is ideally suited for teaching both introductory as well as advanced courses in circuits and electronic devices. In fact, the board can even be used as a digital curve tracer, replacing an ancient device still in use in our devices lab. Today, a less expensive Discovery board is being used in our first and second semester circuits lectures as well as the introductory circuits lab.

A second contribution to teaching that I’ve made is by earning my
professional engineering certification, which is a requirement for all Kentucky engineering faculty hired after 2001. Without the certification, a professor cannot teach any course that includes a design component. All new faculty are now required to receive the PEs within five years of the their start dates. For those unfamiliar, PE certification is awarded by the state having jurisdiction for the work to be performed and requires the engineer to pass both the Fundamentals of Engineering (FE) and a Professional Engineering (PE) exam, offered through the National Council of Examiners for Engineering and Surveying (NCEES). For Kentucky faculty holding PhDs, the Kentucky State Board of Licensure for Professional Engineers and Land Surveyors will forgo the FE exam, but by doing so, the faculty member will be denied certification in other states. For this reason, I completed both the FE and PE exams, making me, perhaps, the only engineering faculty, hired after 2000, to have completed the process. In fact, our own department only has one currently registered PE, Larry Hassebrook.

A third contribution that I’ve made to teaching is by co-chairing the
University’s committee to select a new learning management system when our current LMS (Blackboard) contract expires after the 2015-16 academic year. As the university has moved more and more classes on-line, the LMS has become an integral part of the teaching/learning experience. The goal of the committee is to make a recommendation to either stay with the current service provider or to identify the best replacement system. The committee has, so far, been meeting since October of 2013 and is preparing to make a recommendation to the Faculty Senate and then the Provost by November. Over this time, we have been interviewing faculty and staff from other organizations that have or are planning on making a similar switch as well as some schools which have decided to continue with Blackboard. Aside from just usability, we are also concerned with difficulties faculty may have had in making the switch as well as the technical issues that may have arisen. Also, a campus wide survey was given in the spring to measure the faculty, staff, and student moods toward the current system and to measure their interest in a change. Because the new LMS contract will span five or more years, our decision will have a lasting impact on the teaching/learning experience for years to come.

Now with regards to my approach to teaching, I would summarize that my over arching goal is, first and foremost, to establish expectations and to make sure students know what I expect of them in order to pass the course as well as to adjust my expectations based on the skills and interests of the students in class. From there, my goal is to maintain student engagement in the classroom to create an environment where struggling students feel free to ask questions inside and outside of class. Sadly, by having a child with Asperger Syndrome and from recognizing that all the symptoms used to diagnose my daughter are symptoms that I had as a child, I am left with no other conclusion that I, myself, also have the condition. And by making this realization, I now understand why I was always a very good student with a small circle of friends and a very large circle of alienated peers. It might also explain why I’ve struggled in the classroom to engage students instead of alienating them. Now that’s not to say that I haven’t had successes in the classroom as I’ve always had a small minority of students enjoy my classes immensely, especially as it comes to graduate education where I’ve recruited several middle tier students from the undergraduate program and turned them into exceptional MS degree holders, one of which is now completing their PhD.

Moving forward, I believe I have found a winning formula to teaching that focuses on encouraging student engagement. It all starts by acquainting myself with the interests and backgrounds of the students and to gauge their interest in the course topics. I do this by asking the students to write a, “bucket,” resume as the resume that they hope to write by the time they graduate and to target this resume toward jobs in signal processing. Students are given the weekend to review on-line job postings in signal processing to identify the skills being asked for by employers, and then include these skills in their resume as if they had them. I will then review their resumes looking for specific skills that show up repeatedly. So what this resume ultimately does is that it gives me a way to gauge each student’s interest in signal processing so that I can try to incorporate assignments and projects that will bolster that skill. For instance, Simulink is a skill many students cited in their resumes, and as such, I’ve incorporated Simulink projects in the class.

Now with regards to encouraging student participation, I used to try to keep students engaged by asking questions during the lectures, but for many students, being singled out with a question that they didn’t know the answer only alienated them further. Regardless of what I said or did in the classroom, struggling students would neither ask the teaching assistant or myself for help. So to fix this issue, I’ve started grouping students into teams of three or four students who are immediately tasked with creating a team name and mascot. Students organize themselves with one team in each column of desks in the classroom. The front desk has a printed logo taped to the front of the desk so that I can see each team mascot. This immediately solves my problem of not knowing student names as I can call the team name instead.

The exact make up of the team is done by giving a calculus/circuits evaluation exam at the beginning of the semester. This evaluation includes questions on basic derivatives and integrals, Euler’s formula, and a simple complex power question taken directly from the Circuits I lab. From the scores on this evaluation exam, I assign each team an equal mix of high, medium, and low scoring students in an otherwise random fashion. Aside from assigning teams, the evaluation exam is my way of further telling students what my basic expectation of their skill set is for this course. Students, knowing that they performed poorly (surprisingly poorly, in fact), now have an early opportunity to review the related material prior to the first exam. At worst, they at least figure out where they can find that material in their old textbooks so they are ready when the time comes. Now just seeing the results of this exam (only one student out of thirty remembered Euler’s formula), the evaluation also made me realize that I would need to cover fundamental material in the early lectures, if I ever wanted to build upon it in later lectures. Clearly, my expectations were too high of the students in previous semesters and was a clear point of contention.

By having teams, I have also created a system of offering points to teams that could answer my questions in class. Over the semester, the team with the most points, “wins.” The prize for winning has not yet been determined but would have some small financial value, such as an Android tablet for each member of the winning team. By tracking points throughout the semester, I can target specific low-scoring teams with my questions so that I can keep them engaged in the race all semester long. And in order to avoid having the alpha student on each team answer all the questions, I sometimes give the teams an opportunity to confer among members to come up with a single answer they all agree.

As another means of keeping students engaged in the classroom, I’ve eliminated the 50 minute lectures and replaced them reverse lectures with 15 minute lectures followed by 35 minutes of solving homework problems in class. To account for the lost lecture time, all of my class notes are available online. In addition to the reading, students are largely responsible for reviewing the material prior to class. So time spent doing homework in previous semesters is replaced with studying the lecture slides and the reading material.

I’m happy to say that the benefits of this team approach are readily apparent. First, I know for a fact that the students are meeting in their teams outside of class to perform homework. Again, these are teams that I put together, not teams where students grouped into their existing cliques. I can also see that I have higher attendance than in previous semesters as it is rare to have more than two students absent in any one lecture. Visually, students look happier and engaged in the lectures where if, at any point, I think I am losing some students attention, I simply present a question for points to get them competing. And nothing is a better motivator than competition and winning.

Now another area where I’ve struggled to keep students engaged after an exam. In the current signal processing class, students are expected to learn about signals and systems in the time domain using convolution. Then they are expected to learn about signals and systems in the Fourier domain, and then the LaPlace/Z-domain. And that’s for both continuous and discrete time. So the exams are divided such that exam one covers signals and systems in the time domain, ending with a review of convolution. The second exam looks at the Fourier transform, while the third looks at the Z-transform. While a student would clearly need to understand signals and systems in the time domain in order to understand them in the frequency domain, its easy to see a poorly performing student approaching the material as two separate subjects. And as such, if a student fails the first exam, they simply punt that material to the final. That is, a student looks at their failing grade, realize they need to know the material better, but wait until the end of the semester to actually study that material. They ignore the fact that they need that material as a foundation for the new material for the second exam.

Recognizing that I need students to stay engaged on the first exam material, I now give all exams in class on a Wednesday and return the scored exams on Friday. On the following Tuesday, students can take a make-up exam outside of class. If they do so, then the previous exam score is erased from the grade book as if they never took it. The make-up test result then becomes their exam score, even if their score was lower than the first time they took it. What this does is that it gives the student the entire weekend to study the material that they failed during the first exam. Because the second exam replaces the first, students only take the second exam if they believe they can do significantly better.

The format of the exams is modeled after the PE exam, being composed entirely of 30 multiple choice questions and being open book and open notes. The caveat here is that notes means a single composition notebook with no more than 100 sheets and no free sheets of paper. The book is the classroom textbook. Only NCEES approved calculators are allowed and absolutely no smart phones or other electronic devices. Having only 50 minutes to complete the exam, students will have to study. There just won’t be enough time for the student to search the book for answers for any except a small number of problems. And being graded on a straight scale, missing four problems out of thirty means the student gets a B. While I was hesitant to give a multiple choice exam for a signals course, I believe that the format eliminates the subjectiveness of assigning partial credit to exam solutions from previous semesters. And this subjectiveness has been another point of student alienation in the past.

As a final means of keeping students engaged all semester long, I’ve also implemented a course long project where student teams are given a three-axis accelerometer with RS-232 interface. The chip works by simply taking power from the RS-232 buss and then spews ASCII values in triplet. Simulink, having an RS232 input signal, can read these XYZ vectors in real-time, and let the students build Simulink circuits to detect features of motion. In the class, teams are placed into competitions to perform such tasks as detect motion or to physically shake the device in their hand to try to achieve the highest acceleration of any team in the class. Again, this is another opportunity to keep the students engaged in the material.

Stacks Image 627