I have learnt from experience that asking a group to come up with a policy generally doesn't work. If I am leading the development of a policy, I prefer to draft a document and then amend it during consultation. I find that colleagues are more likely to complain about wording I have proposed than they are to propose wording themselves, so starting with something is better than starting with nothing, even if the final document is nothing like the initial document. Usually, the policy documents that I write are intended to formalize what we are already doing and it is the detailed wording that can be contentious, namely exactly what is included and how prescriptive the policy is. I normally have to make my documents less prescriptive than my first draft in order to obtain reasonably broad approval. In my experience, policies within universities largely work as guides to what staff generally accept to be the right way to proceed and not as laws to be enforced, so a reasonable degree of consensus is important.

I find that a mixture of face-to-face plus email discussion works better than only one or the other. Face-to-face discussion in a group meeting is good for reaching general consensus whereas email discussion is good for pinning down details, because it allows people time to think rather than react to broad issues. Hence, I usually draft a document and circulate it for initial discussion by email, revise the draft and lead a face-to-face discussion in a meeting, then revise the draft again and circulate if for final comment. If I get the amendments right then I don't normally get many comments on the third draft, which then becomes our policy document and I publish it, probably on our staff intranet.

One of my tasks as Director of Undergraduate Studies is to deal with our undergraduate prize nominations, which are formally decided by the exam board. We have two meetings on two successive days: the preliminary meeting of internal examiners considers matters primarily only of internal interest, which includes our prize nominations; the final meeting has our external examiners present and focusses primarily on degree classifications. Our prize nominations are based on the overall yearly and final results for the current year. I download the results for all our students and process them using Excel by sorting them by appropriate averages and deleting all but the top handful of students. I treat first-year students on their own but aggregate other years because the two groups have different prizes. A day or two before our preliminary exam board meeting, I produce a document in which I recommend who we should nominate for each of the prizes available with brief reasons and I indicate whether I think the decisions are clear cut or need further consideration. I include in an appendix raw result data for about twice as many students as we have prizes available so as to give some context. I circulate this document by email and then I lead a discussion of it in the preliminary exam board meeting, which occasionally gets quite animated! Based on the decisions made by the exam board, I email details of our nominations to the Queen Mary Bursaries, Grants and Scholarships Office, which prepares all prize awards. I also inform the prize winners by email.

We recently decided to award a new prize from next academic year, at my suggestion. We have a small meeting of undergraduate programme directors soon after the exam board meetings each year, and a couple of years ago we decided that we needed to do more to encourage students to take a project module in their final BSc year. I suggested that we award a prize for the best Third-Year Project and the meeting agreed that I should propose this at our first Teaching and Learning Committee (TLC) meeting the following academic year. I broadly followed the approach that I discussed above: I emailed the committee shortly before the meeting, we discussed the issue and then I followed up with an email stating what I thought we had agreed. TLC agreed to include both our project modules, which was a modification of my proposal. The agreed policy is now in effect and included in our module and programme details for next academic year. I was asked to advertise the new prize at the appropriate time, namely when students are choosing their modules for next year, which I have done.

We provide our students with exam papers from the past three years via a Moodle database, which I set up two or three years ago and now maintain. In the past, I have used email to collect exam papers and invited examiners to send a copy either to me or to our UG Programme Administrator or PGT Programmes Officer when they submit their final version. For this year, I decided that since I want the exam papers in QMplus it would make sense to use QMplus to collect them, which should be far more efficient than using email. I discussed this with our undergraduate exam board chair, Dr Matt Fayers, who agreed to my proposal. So I set up a Moodle assignment activity in our staff intranet and spent some time reconfiguring the access to make it as secure as possible. For example, I made the section containing the assignment activity visible only to current undergraduate first examiners. (Only staff can access the intranet anyway.) I then emailed staff and asked them to upload their exam papers "as soon as possible". It was my hope that staff would do this sooner rather than later, but I left them to interpret "as soon as possible". I immediately started to receive complaints based on concerns about security and the risk of students gaining access to exam papers before their exams.

After a couple of days of receiving such emails I concluded that I had better clarify the required time-scales. Only a small handful of staff uploaded their exam papers before their exams were sat and one or two have uploaded them since. I will ask staff again to upload their exam papers once the exam period has finished. I am not planning to do anything with them for a while and I can't publish them before September (because our current policy is not to release them until after the late summer re-examination period), but I anticipate (from past experience) that it will be more difficult to collect exam papers later than it would have been at the time they were produced. It would probably have been more successful if I had asked colleagues to provide exam papers by email, because we have used email to send exam papers to colleagues within the School for some time. I will assess how successful this initiative has been later in the summer when I actually need the past papers; I may continue this approach but with a revised time-scale in future or I may revert to using email.

In the past, we had a School database specifically for recording the results of in-term assessments: mostly mid-term tests for our first- and second-year modules. For various reasons related to centralizing of IT support, we lost this database a couple of years ago. Last year, we put in-term assessment marks directly into the central Queen Mary Student Information System (MySIS), but multiple marks that needed aggregating were handled in ad hoc ways. For this year, I proposed that we use QMplus to store and aggregate in-term assessment marks, for several reasons. The main driver was that I knew that a facility called Grades Plus was in development, which would provide students with a better view of their marks than the default Moodle facilities, but our students would only be able to use this if their marks were in QMplus. It was also consistent with our general drive to focus both students and staff on using QMplus as much as possible and it would mean that students could see their component marks while they were being accumulated.

I proposed this change of policy to our undergraduate exam board chair, Dr Matt Fayers, who agreed. In order to make the change more acceptable to staff, I offered to transfer all marks from QMplus to MySIS, so that staff did not have to enter any marks twice. I emailed staff to explain the new procedure and included some brief guidance on how to enter marks into QMplus, mainly consisting of links to QMplus help pages. I learnt fairly quickly that some staff needed more guidance than this, so I wrote more detailed help documents and published them on our staff intranet. I recommended the use of an assignment activity as recommended in the QMplus help and because I use assignment activities for my own module.

Following further discussion with colleagues trying to use assignment activities, I learnt that they were unnecessarily complicated for just recording marks and also that different colleagues were setting them up differently, leading to an inconsistent experience for our students. I therefore proposed that we just use the gradebook in future and moreover that I would set up uniform gradebook entries for all modules that needed them, which I did for our second-semester modules. The process of entering marks into QMplus went much more smoothly in the second semester. The main outstanding issue is when it is appropriate for me to transfer marks to MySIS, which I propose to discuss with our undergraduate exam board chair for next academic year.

• Our QMplus Guidelines and Strategy (Francis Wright, August 2014) and Q-Review Policy and Guidelines (Francis Wright, May 2016) documents resulted from my general approach to negotiating policy documents.
• The prize recommendation document that I presented to our undergraduate exam board last year is a discussion document that illustrates how I negotiate our prize nominations. I have redacted any information that might identify students to protect their identities.
• This project prize email thread (which I have printed to PDF) illustrates how I negotiated a proposed new project prize and how I revised the final prize details to reflect the wishes of our Teaching and Learning Committee.
• I have advertised our new project prize in the "Changes from last year" box at the beginning of our Study programme guide 2016–17.
• This exam paper email thread (which I have printed to PDF) illustrates how I negotiated (not very successfully) with my colleagues to provided their exam papers via QMplus.
• This sequence of four emails that I sent to academic staff in the School of Mathematical Sciences (which I have printed to PDF and combined into a single four-page document) illustrates the four main stages of evolution of the process for entering marks into QMplus. All but the first stage represent my response to the way my colleagues reacted to the previous stage. This effectively constituted a negotiation combined with the practical development of an effective process, which we can take forward to future years.