Tables are best inserted in the portion of the report dealing with that particular experiment. make sure that all column headings include the unit, preferably in quantity-divided-by-unit format (e.g. v/m s^-1).

Graphs are also best inserted in the portion of the report dealing with that particular experiment. It's best to draw the lines by hand. Make sure that you've tinkered with the format so as to produce enough grid lines to enable you to make a decent job of gradients. Draw best-fit, steepest and shallowest lines on straight portions, so that you can work out a gradient with  ±values. Then, if you are drawing a final graph where one of the axes is the gradients of the earlier graphs, you will have something on which to base the error bars of the final graph.

All graphs should be accompanied by a formal conclusion (see Drawing Conclusions.)

Try to avoid such phrases as "It can be shown that" or "I was able to calculate that". Remember that this is a physics exam, and every calculation that you perform in the report is extra credit - particularly under criterion B.

Be scrupulous about declaring which websites, books, people etc, you got information from. There is no penalty for getting it - it all counts as research - but there is a (heavy) penalty for putting in stuff that the marker knows perfectly well you didn't get from your head and passing it off as your own. This information (called 'references' or 'bibliography' can either go at the point where it's relevant, or in a separate listing at the end.) Be particularly careful about any concepts you are using that aren't in the course. Don't assume that the reader knows what you are talking about - give a careful explanation (and references!)

Take very serious note of all red and green squiggly underlines. Set your dictionary to U.K. English as default. Read through the final printout slowly, preferably aloud, making sure that you read what you have actually typed rather than what you intended to type. I would advise you to turn off all the auto-replace features in Word: it's much more reliable if you make all those decisions yourself. You don't want (C), which is the (incorrect) code some of you use for 'measured in degrees Celsius'), suddenly turning into ©, for example, especially not if it happens between you sending it and me reading it. Watch out for common grammatical errors not necessarily picked up the green squiggly lines such as the greengrocer's apostrophe.

In equations and formulæ, scalar quantities are conventionally in Italic, while vector quantities are in Roman Bold. Thus F=ma. Candidates often put whole equations in bold. This is BAD, and I shall in future penalise it.

Note that there is a real problem with the variable x and the multiplication symbol. It's best to use x (i.e. serifed italic) and ´ (from the symbol font). x² = x x x is less clear than  x² = x ´ x. Times New Roman is not a good font for variables: Plantin is one of the best, but Palantino (or Palermo, or whatever your particular version is called) is almost as good. For web pages I have to restrict myself to Arial and TNR because those are the only two I can be sure everyone will have on their terminals. I have to increase the size of the TNR to 14 pt as it has a smaller x-height than Arial in which it is generally embedded. Using a serifed font allows you to distinguish between 1, I and l.

There should be a space between a number and its unit and between all subsequent units. It should be a half space (achieved in Word by kerning), but that would be taking things too far. Thus p = 5 kg m s^-1 (for the magnitude of momentum). There is much to be said for using non-break spaces in quantities, so that the units do not end up spread over two lines. You achieve this with [Ctrl][Shift][Space]. If there is a negative index, Word will treat the minus as a hyphen and feel free to break there as well, so it is prudent to use a non-break hyphen ([Ctrl][Shift][-]) for negative indices in order to avoid this problem (I don't think you can do the hyphen thing at all in html).

Although Modellus doesn't allow subscripts, and you find yourself writing, for example, Ffriction, in a Word document you would expect to use conventional subscripts, such as F_friction (except, of course, when writing about your Modellus model!). Subscripts and superscripts are dealt with by highlighting the relevant text, selecting Format > Font , and then checking the relevant box.

If you click on Insert > Object, scroll to Microsoft Equation 3.0, and select that, you will find yourself with a powerful equation editor. Bear in mind that it puts all the information it needs to draw the equation you create into a separate sub-folder, and you need to keep this sub-folder in the same main folder as the main file. So you can't send someone just the file: send them the sub-folder as well.

It is rarely worth drawing diagrams in Word. They generally look awful. It is much better to draw them by hand, stick them in place and then photocopy the resulting page. Black ink produces a better photocopy than pencil. I find that sticking a piece of Sellotape™ on the back of the artwork and then only smearing PrittStick™ on the Sellotape™, means that I can subsequently remove the artwork for inclusion in a later version of the document without damaging it. You can, of course, scan said diagram in, but you usually get a greyish background unless you've got a 'black-and-white' button on your scanner. Computer diagrams can have an adverse effect on presentation criteria such as CR in A2 Research. If you really can't resist the temptation please eschew Paint, which has a desperately low resolution output. Use the Word drawing tools instead.

Sometimes one takes short cuts in the ordinary course of events - texting is an extreme example of this. In physics, common examples include typing um for mm, pi for p, or degrees for º. Make sure you check your work to remove all examples. The obvious source for these special symbols is the Character Map, which may be found in Start > Programs > Accessories > System Tools. On my computers, I've right-click-dragged the icon onto the bottom toolbar and created a shortcut, so that the Character Map is never more than a click away.

Likewise, remove all abbreviations like 'no.' for number. And make sure that you've removed all references to colour in the diagrams if you are in fact submitting in black-and-white. Also remove all references to page xxxxx in your bibliographies: change the xxxx to the real page number.

You should consider numbering your equations to make cross-referencing easier both for you and for the marker/moderator.

It is a great help to the aforesaid duo if you number your pages.

Provide a cover page that declares a title, an author and the date of production. The date must be altered manually each time you make an amendment. The point of all this is to ensure that the marker knows which of many versions in his possession is the latest, and if you use automatic dating the recipient can get confused if he accesses an earlier version electronically.

Just as you have to make clear what bits you have taken from sources, so it is prudent to make equally clear what bits you have done yourself. There is no point in doing a careful analysis if the reader thinks you've just copied it out! Markers and readers only carry out a few random spot checks. Once they are satisfied that you are being honest, they'll rely on you to alert them to the bits where you want credit.

Tony Ayres, 27 March 2004