We are a team of 12 undergraduate students in the College of Engineering and Physical Sciences at the University of Birmingham. We are working with Maple TA, an automatic marking system for mathematics, to create a bank of questions to aid the student learning experience. We shall use this blog to record our daily progress throughout the summer.
Our last week was pretty busy!
On Tuesday the team received the visit of Michael Pisapia, VP Europe at Maplesoft and Aron Pasieka, who presented their most recent project Mobius, an online course environment that includes all the key features of Maple TA – natural math notation, visualizations, assessment tools, etc. Over lunch, the team also had the opportunity to showcase some of the questions we have written as well as reporting on our experience as Maple TA question authors – Aron even gave us some hacks! Overall, this was an extremely enjoyable session.
On Friday, the team met with Professor Kai Bongs, Head of the Cold Atoms Group. He was interested to hear how we felt about independent projects and interest in possibly setting up our own spin-off companies had changed over the internship. It was great to be able to discuss collectively our initial motivations in accepting the intership, and what we feel we have got from it.
We also met with Dr Steve Quigley, EPS Director of Education to discuss the project and its outcomes. Steve seemed impressed and confident that the project has potential to have a significant impact in the student learning experience across college.
The day saw an end with a Maple TA Team Curry Night!
At the end of the penultimate week, the maple TA team focused on finishing up the tasks that had been given to us from the other departments across uni. Our projects in Electrical Engineering, Maths, Nuclear Engineering and within Physics are all very close to completion. Projects within the department (Quantum, Radiation & Relativity, Stats) are making good progress too, but at a lower priority: external affairs needed to be dealt with first.
A small team visited Martin Long to see what he thought of the revision questions that had been written for year 1; maths for physicists. Martin said he was very pleased with the result and said he was a ‘happy bunny’!
On the Friday, another group spoke to Dirk Hermans about how some of the problems seen in some of his questions for Maths students had been circumvented. We spoke about the next steps such how the questions would be stored and accessed by students.
The team working on the repository released a little surprise! A chrome extension which would assist in writing questions. It contains tools which could: re-size question boxes, add an auto preview and provides info about the different grading functions.
I’ve worked on a frequently-asked-questions document, containing a list of problems and fixes that the team came across when writing questions. They should be especially helpful for those who will start writing their own questions after our placement ends.
Just 1 week left!
Tuesday was the first day we had Physics undergraduates from outside the team coming in to test questions and go through the “student experience” of Maple TA, so we had a challenge on our hands in getting everything working perfectly. Thankfully we got a good cross-section of questions working in time for the outsiders.
On Wednesday we had another set of talks from the PhD students about their work on phase transitions. This was very interesting and almost exclusively dealt with physics that none of us knew anything about. Afterwards we had an opportunity to chat with them about what it’s like to do a PhD at Birmingham. There seemed to be a consensus among the team that this was incredibly useful and probably should be available for all Undergraduate students at the end of Year 3.
We’re starting to see the workload decrease as we focus more on refining and rigorously checking over everything we’ve written so far. It’s also looking like all the code and questions we’ve written will be polished and working perfectly by the end of next week. Now for the important matter of writing all the publicity to do with the project!
Time flies when you are having fun and those past two days were no exception!
The team has finished testing all the questions written for the School of Mechanical Engineering! Now that MechEng is out of the way, it is time to combine efforts to prepare a proof of concept demo for Computer Science.
In order to plan the weeks ahead and manage the workload, a general meeting has also taken place with Nicola. There will be more meetings happening in the next week with the various departments across college as everyone seems to be quite enthusiastic about investigating the possibility of deploying Maple TA in diverse subjects!
The most exciting event of the week, by far, was the visit from Lora Morris, the College Head of HR! The team gave her an impromptu overview of the project.
Lora said :
I was really impressed with the hive of activity and level of enthusiasm for the project; the interns were clear that the project had the potential to make a real difference to staff and students in the future as the technology is rolled out across the College. They are in the vanguard of best practice in this area.
This has clearly left the team even more motivated and we are looking forward to the upcoming days!
Friday morning marked the start of a day of many meetings (with Computer Science, Mechanical Engineering and Chemistry) as well as other activities including stress testing of the Maple TA systems and continuing to write and test questions for a variety of disciplines across the College of EPS.
The meeting with Computer Science focused mainly on converting questions from their Mathematical Techniques for Computer Science module into Maple TA, as well as discussions about the capabilities of Maple TA and the Maple engine. Other topics brought up included the use of the repository developed by Rob and Jordan as well as introducing Maple TA questions for other modules within the School of Computer Science – with an emphasis on a course on Logic.
Meanwhile, Rob and Mike C were attending a meeting with Mechanical Engineering to fix some of the issues that they had encountered with the tutorial sheet questions we had written so far – most of the problems/bugs were able to be reproduced and fixed, however a few were not and so any potential fixes had to be delayed until a later date.
Another meeting happened on Friday with Chemistry, in this talk Edna and Dan presented the Maxwell-Boltzmann distribution Maple MathApp they had been working on to Sarah, the School of Chemistry representative. This app was well received along with the adaptive question features showcased at the meeting.
After the successful meetings, the rest of the day was spent on continuing (or, some cases, starting) to write the Maple TA questions required by Mechanical Engineering, Chemistry, Mathematics, Computer Science, Metallurgy and Materials as well as some questions for our own School, Physics and Astronomy. This marked the end of a busy week in which everyone was ready for a well-deserved bank-holiday weekend!
At the start of this week, work continued on finishing off the Mechanical Engineering tutorial questions and I began allocating Computer Science questions to translate to Maple TA. The rest of the week will be dedicated to completing current questions being worked on and starting new batches of questions such as recently received requests for Nuclear Engineering and Electrical Engineering.
To allow for a break from writing questions, we attended a “Journal Talk” within the Theory department by one of the PhD students whilst eating cake!
It was by Wednesday that we realised that this week would be characterised by meetings – lots of them. However, Wednesday itself was a quiet start as most of the team continued to work on tasks that had already been started.
Edna and George started to explore the full capabilities of Maple TA’s grade book. They created some sample assignments and we each had a go so that we could simulate a class of students attempting some revision questions, to see what it looks like when a teacher views these attempts.
Later, Greg announced that we would be heading outside to do some more juggling and that he would be assigning us some very serious and definitely not made up levels. I managed to achieve level 4 and a half, although Jordan was well in the lead in the lofty heights of level 13.
On Thursday the meetings started to increase in frequency, as we met with two members of staff within the Theory Group about some Physics modules. Jordan, George and I met with Dr Rob Smith to discuss writing questions for the 3rd year Radiation and Relativity module. Additionally George, Mi and Dan met with Dr Martin Long to discuss the 1st year Maths for Physicists module.
Although the meetings were fairly brief, we each left with a clear idea of what our next step was. For Dr Long, we will be sending some appropriate questions that already exist within Maple TA to give him an idea of what the system can do. For Dr Smith we will be some converting multiple-choice questions from STACK and investigate the writing of some exam-style questions to supplement them.
Friday will be a hectic day as we have meetings scheduled with Mechanical Engineering, Chemistry and a new department – Computer Science.
Ciao for now!
Our week started with a second meeting with the School of Chemistry. This time Dan and I met with Dr Ian Shannon to discuss where Maple TA could be used to supplement the teaching delivery within Chemistry.
We were given two distinct projects both concerning experiments, which was quite exciting given that we had not yet extended Maple TA to this type of tasks!
Our first job consists on setting up an error analysis worksheet. The team agreed that this quite fit within the Maple TA framework and was a natural extension of work that had already been undertaken. What is more, it could potentially be generalized and extended to other subjects, where lab work plays an important role!
The second task we were assigned brought up some interesting challenges which we are very keen to explore, e.g. Input of student data, and validating it against experimentally reasonable values, just to name a few. We shall endeavor to pilot this and will keep you updated on our progress!
As for the rest of the team, while Dan and I were at the meeting, they have managed to come to grips with writing MathApps! Meanwhile, Jordan and Rob continued expanding our repository and Ben has joined them on this quest! As a consequence, we now have grading code that is tailored to the maths required in each question!
Tuesday saw another round of testing as well as some MathApp writing. On the afternoon, all of us met individually with Nicola for a brief chat about the project.
That is all for now.
The end of the week brought about a whole new wave of testing with questions for Maths and Mech Eng forming the bulk of the work. Fortunately, on this occasion most of the changes that had to be made were relatively small. This is probably mostly due to us being better at writing questions, and the very useful question repository!
Friday was exciting because we sent off our first round of questions to another department. 4 class tests were delivered to Mech Eng, along with a set of sample questions including randomization and feedback. These were mostly developed by Helen and Mike C.
Dan has spent the day creating a MathApp to model the Boltzmann distribution for Chemistry. It allows students to change various parameters and see the effect on the distribution. It also shows the RMS speed, average speed and most probable speed. (Latest status: it’s working!)
Today is my last day because I’m going away from next week. It’s safe to say I’m not going to miss getting up really early (well, 6 a.m – it’s early to me!) to commute to Birmingham.
I’ve really enjoyed having the opportunity to work as part of a fun and enthusiastic team creating resources that will hopefully make a real difference to the learning experience of students throughout the College.
Good luck with the rest of the project guys, I’ll be keeping an eye on the blog!
We’re back! These past few days have certainly been productive in terms of question writing. We’ve started working on Maths tutorials for Mechanical Engineering and things are going smoothly. We’ve also continued working on Maths questions and most of them are ready for testing!
This Wednesday Edna and Helen and I met with Dr. Sarah Horswell to discuss about implementing Maple TA in the School of Chemistry. We were delighted to hear that most of the types of questions she told us about could be implemented in Maple TA and possibly even be improved. She also gave us good tips on designing multiple choice questions in order to make the students pay close attention to what is being expected from them (for example, check if they forget to convert degrees Celsius to Kelvin).
It seems that the key features for Chemistry will be adaptive questions and MathApps. The former should be designed mainly to give adaptive feedback and let the students know about the mistakes they are making as accurately as possible. Again, the team will be investigating the implementation of conditional feedback. The latter have great potential in creating interactive environments such as functions which depend on parameters whose values we can vary freely.
Today, we also had a general meeting where we presented our progress to Jon and Nicola. The main topics included setting up deadlines for the things we’re working on at the moment, work on individual projects on topics we suggested when applying for the internship, investigating the possiblity of identifying questions which could be useful to more than one department (and setting up an infrastructure for that) and, last but not least, making our efforts known to the general public towards the end of the internship.
Dan, over and out.
In the afternoon Jon, Jordan and Rob had a conference call with some helpful Maple TA developers. In this call they discussed some of the queries we had about writing questions to see if they had any solutions.
Many things we had found were already known by the developers. We found that some of them were by design, it was useful to learn more about why. Other inquiries will be addressed in due time. Some of our requests were for features they had not come across before. They will therefore be investigating these further. Fortunately, they were able to give us a fix to our largest problem – the switch function – so we can now get around this.
Overall the meeting was productive and very useful for us.
That is all for week 3!
~Rob & Jordan