One row per slide, to glance at while the deck is on screen. The reasoning behind every answer is in the sections below. Deck: /deck/what-is-coming-with-group-work/main.pdf.
| Slide | On screen | What you fill in or say |
|---|---|---|
| 2 | Before I say anything | From the room: one person does everything; nobody knows who is doing what; done the night before; someone disappears. Leave it up, point back at it on 5 and 6 |
| 3 | The shape of it | 60% of the module, starts Week 4, a library built in a four. Say you would rather hear a topic idea early than late |
| 4 | What you hand in | Eight. Let them count the cards: three paper, three software, two record |
| 5 | How your mark is worked out | \(I\): your engagement record off the minutes, out of 100. \(Q\): the group's submission mark, shared. Both objective; neither is my impression of you |
| 6 | A strong group | With \(Q = 74\): A 86.02%, B 76.94%, C 76.94%, D 38.47%. Do A first, let them predict D. Reverse case worth giving: \(I = 100\) with \(Q = 42\) is still 64.81% |
| 7 | So where does \(I\) come from? | Five forms, Weeks 4, 5, 8, 9 and 10, handed in. Engaged at 3 of 5 gives \(I = 60\), and \(\sqrt{60 \times 74} = 66.63\%\) |
| 8 | Who teaches you these chapters? | Nobody. No lecture on any of the seven. Let the silence sit before you answer it |
| 9 | Yes: this is self study | Self study with a social component. The reading is solitary, the understanding is not |
| 10 | How a group-project week works | The reading is homework, the meeting is timetabled. Do not imply a workshop has to follow |
| 11 | Groups | Self-selecting fours. Anyone not in one by the deadline is assigned |
| 12 | This week | Form a four and register it; email me a topic idea. Print the form, complete it at the meeting, bring it to the next session. \(I\) is counted off those five |
25 minutes. Twenty is not realistic for this deck any more: there are five slides the class answers rather than listens to, and rushing those is the same as cutting them. If you only have twenty, see the note below the table.
| Slide | Content | Minutes |
|---|---|---|
| 2 | What goes wrong in group work (blank, from the room) | 3 |
| 3 | 60%, from Week 4, a library built in a four | 1 |
| 4 | The eight files, with the count blank | 2 |
| 5 | \(I\) and \(Q\), both boxes blank | 4 |
| 6 | The worked free-rider table, marks blank | 4 |
| 7 | Where \(I\) comes from: five forms, handed in | 3 |
| 8 | The chapters, and who teaches them | 2 |
| 9 | Yes: this is self study | 2 |
| 10 | How a group-project week works | 2 |
| 11 | How groups are formed | 1 |
| 12 | This week, the email invitation, and the minutes reminder | 2 |
| Questions | 1 | |
| Total | 26 |
To get to twenty, take two answers on slide 2 instead of filling the workspace, and move briskly through 3, 4, 11 and 12, which are statements rather than discussions. Do not cut slides 5 to 9. Slides 5 to 7 are where the marking scheme lands and 8 and 9 are where the self-study expectation lands, and those two things are far harder to fix in Week 6 than to state now.
The deck. There is a write-on deck at /deck/what-is-coming-with-group-work/main.pdf. Five things on it are blank and filled in from the room: all of slide 2, the file count on slide 4, the \(I\) and \(Q\) boxes on slide 5, the four final marks on slide 6, and the \(I\) arithmetic on slide 7. Answers for all of them are at the end of these notes.
Open with the fact that the group project is 60% of the module and starts in Week 4.
Say explicitly that choosing a topic is not something they have to do alone, and that you would rather hear from a group early than see a project that was never going to work. The invitation is on slides 3 and 12, with your email address on the closing slide.
Show one worked example project from the example projects listing. Spend no more than a minute on it: the point is that students see a finished library, a README and a three page paper, not that they understand the mathematics in it.
Go through the eight files. Emphasise that every component is graded, so a student who writes the paper and the README earns marks just as a student who writes the library does. This is the first of two moments in the class where a worried student relaxes.
Introduce the geometric mean. Resist deriving anything. Put up the free rider table and let them read it. The question to ask is "what happens to D?" and then "why?". This is the second moment: a student who has been in a bad group before will be watching for exactly this.
Then the self-study block, slides 8 to 10, which is the part of this class most likely to be misheard. Put the chapters on screen and ask who teaches them. Let the silence sit for a moment before answering it. The answer is that nobody lectures these chapters: students work through them on their own, and the meeting is where they report back and find out what they misunderstood.
Say the words self study out loud rather than implying them. Then say, equally plainly, that it is not independent study in the sense of being left alone: the reading is solitary, the understanding is not. Slide 10 makes the division concrete: the reading is homework, and the meeting is timetabled, so they are not being asked to find a free hour between four diaries. Be careful not to imply that a workshop has to follow the meeting. Building the library is theirs to arrange, alone or together, and the point to make is that none of that is the meeting.
Close on the three actions for the week, the first of which is forming a group of four.
Do not let the room start packing up before the box at the bottom of the last slide. The group meets in its own timetabled slot, not in the class hour, so the pattern to state is: print the week's form, complete it at the meeting, bring it to the next class session. A group that never quite gets round to meeting is the group that loses a week of evidence, and it is entirely avoidable with one sentence here. Say that \(I\) is counted from those five forms, which ties the reminder back to slide 7 rather than making it sound like an administrative detail.
Send the following message before class:
Hi all,
In tomorrow's class I will spend the last twenty minutes giving you
an overview of the group project, which starts in Week 4 and is
worth 60% of the module.
I will cover what you build, the files you hand in, how the mark is
calculated, and how groups are formed. You do not need to prepare
anything.
The one thing to know in advance: you will need to be in a group of
four, and groups are self-selecting. If you already know who you
want to work with, that is one less thing to do next week.
Please get in touch if I can assist with anything,
Vince
Send the following message after class:
Hi all,
The slides from today's overview of the group project are on the
website, along with the full assessment page:
https://vknight.org/cfm/topics/what-is-coming-with-group-work.html
Action this week: form a group of four and register it. Anyone not
in a group by the deadline will be assigned to one at random.
Please get in touch if I can assist with anything,
Vince
Slide 2, what goes wrong in group work? Whatever they offer is the right answer, and the list is better coming from them. What usually comes up: one person does everything, nobody knows what anyone else is doing, the work is done the night before, and someone disappears. Keep the list on screen, because every item on it is something the design of this project is trying to handle, and you can point back at it on slides 5 and 6.
Slide 4, how many files? Eight. Let them count off the three cards on the slide: three for the paper, three for the software, two for the record. Counting it themselves is what makes the number stick, and it gets them reading the file names rather than skipping past them.
Slide 5, what goes in each box?
The important step here, and the one worth drawing out explicitly, is that both components are objective. Neither is my impression of you or your group's impression of you. \(I\) is counted off the minutes the group itself wrote and handed in, and \(Q\) is the mark the submitted artefacts earn against published criteria. Say this plainly. It is the answer to the fear that sits behind most of what they listed on slide 2, which is that group marks are arbitrary or that the loudest member decides who gets credit. It is also what makes the minutes worth taking seriously: they are not administration, they are the evidence.
Slide 6, the four marks. With \(Q = 74\):
| Student | \(I\) | \(\sqrt{I \times Q}\) |
|---|---|---|
| A | 100 | 86.02% |
| B | 80 | 76.94% |
| C | 80 | 76.94% |
| D | 20 | 38.47% |
Give them A to work out first, then let them predict D before you fill it in. What happens to D is that a strong group cannot carry them: D fails. Make the reverse point too, since it is the one that reassures the students who are dreading this: a student with \(I = 100\) in a weak group scoring \(Q = 42\) still gets 64.81%, well clear of the group's own mark. Two further worked scenarios are on the geometric mean page if anyone wants them.
Slide 7, where does \(I\) come from? This is the slide that makes the previous one actionable, so do not hurry it. \(I\) is not an impression of anybody. It is counted off the five minutes forms the group fills in and hands in, one in each group-project week: Weeks 4, 5, 8, 9 and 10. A form that is never handed in is not a late submission, it is an absent record, and it costs every member of that group, not just the one who was quiet.
The arithmetic blank: recorded as engaged at 3 of the 5 gives \(I = 3/5 \times 100 = 60\), and with a group mark of \(Q = 74\) that is a final mark of \(\sqrt{60 \times 74} = 66.63\%\). Students who were alarmed by D on the previous slide find this reassuring, and it is worth letting them have it: the scheme is not trying to catch anybody out, it is trying to make the record match what actually happened.
Tell them now that the group meets in its own timetabled slot and brings the completed form to the next class session, so that nobody is waiting for the meeting to be arranged for them.
Slide 8, who teaches you these chapters? Nobody does, in the sense they mean. There is no lecture on any of the seven topics listed. They work through the chapters of Python for Mathematics themselves, and the weekly meeting is where each member reports what they covered, which is the first standing item on every agenda.
Expect this to be the least welcome slide in the deck, and expect at least one student to ask why they are being taught nothing. The honest answer is worth giving: the tools weeks teach a topic at a time with worked examples, and the group-project weeks are where they practise picking up a topic without that scaffolding, which is the thing the final year will require of them. The textbook has tutorials, how-to recipes, exercises and full solutions, so the material is not thin, it is just not delivered in a lecture.
The follow-on point, which is slide 9, is that this is self study with a social component. The reading is solitary; working out what it meant is not. A student who reads a chapter, does not understand it, and says so in the meeting is doing exactly what the structure expects. A student who reads nothing and says nothing is the failure case, and the minutes will show it.
Slides 3 and 12, the invitation to email. Not a question to the room, but worth saying rather than leaving on screen. Groups that pick a topic in silence are the ones that arrive in Week 8 having built something too small to carry a three page paper, or too large to finish. A five minute conversation in Week 3 fixes both, and almost nobody starts it unless they are told twice that they may.