ASSIGNMENT MARKING SHEET

This is a mock-up of the marking sheet I used for type B assignments in 1995. Details might change, but the principles will persist, unless I find good reason to change them. It seems to work well enough in practice.

There are three slightly different sheets, for loners, group members, and groups. The difference is explained in the material on assignment.

LONER

NAME :

Item Assessed	Part of item	Maximum mark	Notes..................................................................................	Your mark
Content	Design and method	10
	Description	10
	Evaluation	10
	Future	10
	Real-time and control	10
	Sources	10
Presentation		10
		TOTAL
		OUT OF

MEMBER OF GROUP

NAME :

Item Assessed	Part of item	Maximum mark	Notes..................................................................................	Your mark
Content	Design and method	10
	Description	10
	Evaluation	10
	Future	10
	Real-time and control	10
	Sources	10
Presentation		10
		TOTAL
Weight :		OUT OF

Yours

Your weight

Group

Group weight

Your share

Weighted sum

RESULT ( % )

(

/

*

+

/

*

*

) / (

) =

GROUP

NAME :

Item Assessed	Part of item	Maximum mark	Notes..................................................................................	Your mark
Content	Design and method	10
	Description	10
	Evaluation	10
	Future	10
	Real-time and control	10
	Sources	10
Presentation		10
		TOTAL
Group weight :		OUT OF

The pattern isn't rigid; if some category isn't appropriate for a particular topic for some reason, I miss it out. That's why the "TOTAL" is left open. The same pattern goes for members of groups and groups themselves; all the sheets are much the same in their bodies, allowing for material to turn up either in the group or individual parts, as appropriate.

The complicated piece at the end of the MEMBER OF GROUP sheet is where I combine your individual mark with the group contribution. The weights are usually worked out by counting pages - notice that you don't win by writing pages of rubbish, unless the group material is even worse.

I give marks under several headings. The topics covered by the headings can turn up anywhere in your report – the list isn't intended to be a list of titles of separate sections of your report, though in some cases it's obviously convenient to group things in that way. For example, while it might be a good idea to have a section in which you discuss the overall implications of what you've done, little bits of evaluation can very well turn up elsewhere.

I distinguish content from presentation. Under content, I've classified your approach to describing the device, work to be done and how you did it, and other matters concerned with the machine itself; under presentation, I've included more literary matters which go to making the reports readable – and also, under reports, I assess ( by counting ) the effectiveness of your weekly reports through the mail. Here are some comments on the headings of the Content section.

• Design and method covers your approach to the assignment. Under this heading I give credit for things like sensible experiments, ingenuity ( when appropriate ), and general evidence of applied common sense.

• Description covers your account of the assignment. The description should be self-contained – the reader can be assumed to have relevant general computing knowledge, but not any foreknowledge of matters specific to the topic of the assignment.

  Descriptions of the machinery, software, or whatever that you're using should give enough detail to make sense of any following experiments or discussion, but not a lot more. If you need more, give a reference to some appropriate document, or put additional material into an appendix.

  A description of what you originally intended to do should be there somewhere. If you can, say why it's a good thing to do, and make some reference to any appropriate previous work ( which will sometimes, but not necessarily always or only, be earlier 473 reports ).

  Descriptions of what you did – experiments, software development, calculations – should make clear what you did and why you did it.

• Evaluation covers your assessment of what it all means : results, conclusions, review of your original aims. Did what you expected actually happen ? Is that good or bad ? Can it be improved ? ( and, if so, how ? )

• Future covers your recommendations as to what should or could be done to continue your investigation. Now you've done your bit, what new avenues of exploration are open to someone else ? Don't let your imagination loose; these suggestions should be plausibly accomplishable in future assignments.

• Real-time and control is my attempt to determine whether you got all you could out of the assignment in support of the material of the 473 course. That's what it's for, after all. It's true that the matters which turn up will be on the small scale side, but they can still be illuminating if you discuss them carefully.

• Sources covers the connections between your descriptions and other publications. I expect appropriate references adequately specified : however it's done, there should be enough information unambiguously to identify the document.

  The source for anything coming from anywhere but you or reasonably accessible common knowledge ( evaluated at stage 4 in the Computer Science department ) should be cited. The general rule is to give enough information for someone else to find the same information if need be ( or to avoid fruitless searches if you learnt it from a chat with a mate ).

  It is usual, and helpful, to collect together the sources cited into a list, which conventionally comes at the end of the document. That makes the reference easy to find when you're actually looking for it, rather than reading through the text.

I mark presentation, because it's important. That doesn't mean impeccable typesetting, though; it just has to be clear, easy to read, to the point, and such like. Excellent presentation is possible with a typewriter, pen, and ruler; some people can even manage it with handwriting. I correct spelling and punctuation mistakes if I see them; if nothing else, it shows I've read your reports. There are no marks explicitly for spelling, but poor spelling is poor presentation. I think it's important that someone should point out your mistakes so that you have a chance to learn how to do it right; probably no one else will, so I'd better. ( Indeed, I might be the only person in the department who's old enough to be able to spell properly anyway. ) Perhaps I should remark that I do read the text as well as correcting the spelling; that isn't always obvious, particularly in the better reports where I don't feel it necessary to add many other red marks. I might or might not comment on grammar or style ( "sentences" without verbs, split infinitives, etc. ) : if I don't, that's only because it would take too long to do adequately, not because I haven't noticed, and certainly not because I don't care.

I might also take into account the steadiness of your progress through the time of the assignment, as determined by your mail reports. In previous years, I've just counted the reports, but I'd prefer to be more critical if I can find a good way to do it.