WRITING SKILLS

Always bear in mind that your title will be read perhaps hundreds of times more than your paper. Your choice of title will influence how many people will read your paper. Indexing services such as BIDS will present your paper to completely different readers depending on what you put in your title.

A good title will:

Be specific

NOT 'Influence of pesticides on insect reproduction'

Concisely define the contents of the paper in as few words as possible

NOT 'Preliminary investigations into the possible influence of dieldrin on parameters describing clutch size in the beetle, Prostephanus truncatus (Coleoptera: Bostrichidae)'

BUT 'Influence of dieldrin on clutch size in Prostephanus truncatus (Coleoptera: Bostrichidae)'

Not contain abbreviations, chemical formulae or jargon

When searching an index people generally look up the word rather than a formula or abbreviation, e.g. Water instead of H2O. It is always safer not to abbreviate.

Has correct use of syntax

back to contents

An abstract is a mini version of your paper. It will allow others to assess the contents of your paper quickly and to decide whether to read it or not. An abstract should therefore include brief statements of:

The main aims of the project

The main methods employed

A summary of results obtained

The main conclusions

Make sure you include your main findings in the abstract and do not "save them 'till last". A paper is not a detective novel designed to create suspense. State your findings at the beginning and many more people will read the paper.

back to contents

A good introduction will

State the problem you are attempting to solve

Indicate the source of inspiration for your study

Summarise related work. A reader should be able to assess your work without being required to read additional papers.

Indicate where your research fits into previous work.

Define any technical terms or abbreviations you intend to use later.

Justify your choice of methods.

Clearly state your aims.

An introduction should be written largely in the present tense.

e.g. "Little is known about the reproductive behaviour of dung beetles .....", "We have chosen this method of study because ..."

back to contents

This section must accurately describe what you have done. Enough should be included to allow others to repeat your work (if this is possible). Methods should be written in the third party past tense.

For example:

'Samples of dung were removed each day ...'

NOT 'I removed samples of dung each day ...'

Materials

Exact specifications of materials must be listed to allow your experiment to be reliably repeated.

For biological material, list sources, culturing methods, genus and species. Note the strain if appropriate. Inform your reader what age, sex, physiological state you used in your experiment.

For example:

'Male flies were acclimatised at 25°C for two days before testing.'

Be specific about quantities of chemicals and dimensions of equipment. Chemical names are preferable to trade names unless brands differ and this may affect the results. In such cases list the chemical name, trade name and manufacturer.

Always use S.I. units when specifying quantities, concentrations etc.

Methods

Methods can be followed most easily if presented in chronological order. Types of methods, e.g. assays, habitat sampling methods, may be grouped together in order to add to clarity.

Any new method must be described in full. Always be exact in dimensions and details. A method that has already been described in an accessible journal need not be described by you in detail - refer the reader to the publication instead.

Name any statistical tests used, but only describe them in detail if they are uncommon ones.

Remember the rule: Methods should be described unambiguously and in sufficient detail to allow the reader to repeat the work exactly.

back to contents

Avoid repeating details of Methods in the Results section.

Limit any mention of methods to the minimum needed to let the reader know what results you are presenting.

Use the past tense, i.e. such and such was found.

Data should generally be summarised and put into a format where the main findings are easily accessible. You may use tables, graphs and/or descriptive statistics. The reliability of these results can then be further described using derived statistics.

When to present data in the text

When there are only a few values estimated that can easily be written as a sentence.

When the data are very repetitive and can easily be summarised in a written form.

When there are few points of interestin the data.

When to use a table

When the exact figures are important to the result

When there is no obvious pattern to the data

Present your data EITHER in a table OR in the text, not both.

How to set out a table

The table caption should be self-explanatory without reference to the text.

Data should be arranged so that like forms of data form the columns (rather than the rows).

For example:

Table 3. Mean oxygen uptake (µl/egg/d) of eggs of C. maculatus (a Yemen strain) on untreated cowpeas. Groups of 5 beans in a flask, each with a single egg, were measured. A total of 10 flasks was used. The ratio of mean uptakes compares the uptake of a Yemen strain with that of a Brazil strain at the same age.

When to use a graph

When the data show an obvious trend or pattern of difference between treatments.

When a graph conveys more information than a table.

Make rough drafts of both tables and graphs to help you judge which is more clear.

You often save time by plotting rough drafts by hand rather than using a computer package.

back to contents

Discussing the significance of your results is the most challenging part of writing a paper. Take some time to think before writing this section and it could save you a lot of waffle.

A good Discussion will:

Summarise the main findings of your results from an overall perspective. What trends, patterns become apparent when you step back and take all your data as a whole?

How do your data fit with existing work? Do your data agree with previous findings and/or hypotheses?

Does your work suggest any new hypotheses?

Can you suggest improvements on your work or further work that could clarify the subject?

Were any of your data points unusual or conflicting with the rest? How could this have arisen?

How reliable are your data?

back to contents

When writing you will often want to include other people's findings. Let the reader know your source of information by a system of referencing. There are many systems of referencing and the convention varies widely between journals.

Always record references in full on index cards even if you only intend to use part of the information; you may need to publish references in another format later. It is easy to take information out and a nightmare to go back to all the references to add last page numbers for instance. The Harvard System is a recommended method that retains all the information usually required for any system of referencing.

The Harvard System

In the text:

References are cited in the text as a 'name and year', e.g:

Red frogs were found to jump more often on hot days (Higgins and Brown 1994).

Many papers have a long list of authors. All names are always listed when there is only one or two authors, e.g:

(Higgins 1994), (Higgins and Brown 1995)

Most journals have an 'et al.' rule when there are three or more authors listed on a paper. If no rule is specified, authors should be listed the first time you refer to a paper with three authors, e.g: (Higgins, Brown and Peacock 1993)

If the paper is referred to again later in the text it can be shortened to: (Higgins et al. 1993)

If the paper has four or more authors it should be cited as:

(Higgins et al. 1992)

even when first mentioned. If you need to cite two or more papers by the same author/s from the same year simply label them (Smith 1980a) and (Smith 1980b).

In the reference section all references cited in the text should be written out in fuller form in alphabetical order at the end of your document in the reference section.

Examples in Harvard system format:

Bell J. (1987) Doing Your Research Project Buckingham: Open University Press.

Collinson D. (1992) Plain English 2nd edition, Buckingham: Open University Press.

Møller H., R. H. Smith and R. M. Sibly (1989a) Evolutionary demography of a bruchid beetle. I. Quantitative genetical analysis of the female life history. Functional Ecology 3:673-681.

Møller H., R. H. Smith and R. M. Sibly (1989b) Evolutionary demography of a bruchid beetle. II. Physiological manipulations. Functional Ecology 3: 683-691.

Note that the titles of books and journals are often underlined or italicised.

back to contents

©University of Leicester, 1997. Writing Skills for Postgraduates and Other Research Scientists, Lucy Birkinshaw and Professor Robert Smith. Based on a pamphlet originally prepared by Robert Ash for the Faculty of Social Sciences. Web-editors: Ted Gaten and Magnus Johnson

Last updated: May 2003
Biology Web Maintainer
This document has been approved by the head of department or section.