Unfortunately, the number of pixels an image contains is not the whole story; the quality of the image and its size also depends upon the resolution of the output device.

Resolution is the number of pixels used over a given distance and is typically measured in pixels per inch or ppi.

For example; when displaying images, computer screens use a resolution of 72 ppi. Thus when seen on screen, the above image will be 200 ÷ 72 = 2.8 inches (7 cms) wide and 250 ÷ 72 = 3.5 inches (8.8 cms) tall.

Typically however, a colour printer will operate at 360 dots per inch (dpi) for normal printing and 720 dpi (or higher) for quality printing. The result is that an image which looks great on the screen might look like its made up with a series of big coloured blocks when printed, rather than a smooth photographs.

Resizing for Quality:
By resizing the image in the document you are working on, you can change its resolution without the need for expensive photographic manipulation software (like Photoshop).

For example, if you halve the image size (assuming it was saved at screen resolution (72 ppi) - most images saved from the web will, for example, have this resolution), then it will now have a resolution of 72 × 2 = 144 dpi. This will be fine for many cases, but if you need even higher resolution, then reduce the image size even more.

This does then mean that the larger the image in the first place, the better the quality of print, but also the larger the file size you need to store.

So how do you tell what resolution your original file is, in order to know if you have to resize it? Trial and error is one way – but expensive in both time and money if you then need to resize and re-print the page – so why not let the computer tell you?

Image Resolution:
First use 'Windows Explorer' or 'My Computer' to find your stored image.

The 'right-click' the mouse on the file and select 'Properties'.

Finally, select the 'Summary' tag and read the resolution.

Illustrations and sketches

Many CD-ROM sets of 'Clip-Art' are produced in what is known as 'Vector Graphic' format. Rather than the images being made up as a 'map of pixels' as in a 'bit-map image', each line of the illustration is mathematical description of the line which then prints at the resolution of the relevant output display. The result is that the image always displays at the best quality no matter what size of image you use.
So why would we bother with bit-map images if vector images print so well? Simply because we can rarely get photographic quality from vector images. They are best used for illustrations, sketches or areas where we need 'cartoon style' colour presentation or 'technical illustrations', rather than photographic images.

So remember a good photograph is best to show the appearance of an object but a diagram is better at showing how something works or how it is put together.

Typical packages which allow you to sketch your own illustrations are Adobe 'Illustrator' and Macromedia 'Freehand'. These packages are not 'quick fixes' and take some time to learn. Alternatively if time is short, you might be better producing diagrams free hand or with the aid of straight rule, and then scanning them. This will not produce such a neat drawing, but it will often be acceptable, especially if you draw it larger than you require and then reduce the overall size when you import it into your work.

Remember, a diagram nearly always contains an element of simplification. Simple line diagrams (as above) can convey how a physical system works and can be combined with other diagrams, photographs or words to describe complex sub-parts.

Graphs and charts

Graphs and charts are not as precise as tables when it comes to presenting data, but they do have greater visual impact and are an excellent way to illustrate trends. The use of a graph when presenting your data is either to simplify your results, or to enliven them, i.e. to make your results more graphic.

Types
X-Y or Line Graphs:
Typically line graphs are those drawn on ordinary two-dimensional graph paper to represent the relationship between two (or more) variables; showing how a change in one variable relates to a change in another.

Conventionally the horizontal axis of the graph is termed the 'X - axis' or the 'abscissa' and the 'Y - axis' the 'ordinate'.

The 'independent variable' is usually plotted on the horizontal axis and the 'dependant variable' on the vertical. If you are not sure which is which, a useful clue is that the independent variable is likely to change at regular intervals (time for example), whilst the dependant variable is more erratic.

Of course it is acceptable to draw your graphs by hand on graph paper, but this then makes them more difficult to embed into your work. Many software packages produce graphs, 'Microsoft Excel' in particular is available on most University computers and produces some excellent results which are easily imported into your work.

Logarithmic Scales:
Many X-Y graphs use linear scales, i.e. the interval between the graph divisions is even. However, when displaying sets of data which spread over a very large range, then it is often better to use a 'Logarithmic' (or Log) scale.

This does make it more difficult to visually relate the trend, but it does fit all the data onto a practical size of paper, or prevents the data all being 'squashed up' into one part of the graph.

Pie Charts:
A 'pie chart' is a circle which is divided into segments or slices like a pie; the size of which represents the relative proportions of the whole data.

A pie chart can be used as an alternative to bar charts (see below) and should only be used where the number of segments or 'slices' is limited and of approximately equivalent size. If you divide the chart into more than 4 or 5 pieces, or if one piece dominates the others, it will be difficult to read. A bar chart is more appropriate in these cases.

Bar Charts:
A 'bar' or 'column' chart allows for the simple comparison of data. They are effectively one-dimensional graphs in which the size of each data set is represented by the length of a horizontal or a vertical bar.

The 'bar' can also be divided to show how that particular data set is composed. This is then called a 'stacked' bar chart.

Isotypes ('pictographs'):
In an 'isotype' bar chart, symbols are used to represent data rather than a simple bar. Isotypes tend to highlight the data and make the data more interesting. The problem comes in finding a simple symbol which looks like the data it represents. The symbol, once chosen, then presents a discrete quantity, for example:

These displays are also limited, in that once the base figure is set, it is difficult to accurately break up the symbol into very small pieces.

Histograms:
'Histograms' look very much like bar charts, but are not. They are specialised diagrams to show frequency distribution – a block graph in which the area of each rectangle represents the frequency of each class or item; i.e. how the data of a sample is distributed as a whole.

In this way it is similar in function to a pie chart so that if one class of data has a frequency twice that of another, then the area of the respective column will be twice as big. Also since a histogram plots relative distributions of a whole, all the block touch each other.

INSERTING VISUALS:
File types

Photographs and Pictures:
Most photographic or image manipulation packages can save files in a number of different ways, some of which are only 'readable' by that type of software. For example the most famous – 'Photoshop' – saves files in its native 'Photoshop' (PSD) format in order to save the files more efficiently.

However, a PSD file is not recognised by Word for example, and you would not be able to import it into your work. You therefore need to save your files in a format, which is easily importable and some of the most common file types are listed below:

'Tagged Interchange File Format' or TIFF
This was designed to be a universal bitmapped image format and is used extensively in desktop publishing packages. It is an 'uncompressed' file format (see JPEG below) and therefore produces the most accurate and best image quality, but at the expensive of a very large file size.

'Joint Photographic Experts Group' or JPEG
The JPEG format is commonly used to save photographs and is one of the standard formats for displaying images over the World Wide Web. It is a 'compressed' or 'lossy' format in that it retains all the colour information of an image but compresses the file size by selectively discarding data. You can usually choose what level of compression (and hence file size vs. image quality) you want when you save the file from standard image or photographic manipulation packages.

'Compuserve's Graphic Interchange Format' or GIF
The GIF format is another compressed file format specifically designed for cross platform and Internet use.

The file format is lossless and can produce very small file sizes, but is limited to a colour palette of 256 colours.

This means that it is better suited to graphics that have limited plain colours such as illustrations or logos. It does not usually produce acceptable results with photographs.

Vector Graphics:
'Windows Meta-File' or WMF
Only suitable for Windows based systems, the WMF format and its more advanced – 'Enhanced Meta-File' or EMF format is a common format for 'Clip-Art' CD-ROM collections. As with all vector graphics, they can be resized within the Word processing package to any size without loss of image quality.

'Encapsulated PostScript' or EPS format
In the same way that 'Photoshop' etc., saves files in its own native format, so the common Vector Graphic drawing packages – 'Illustrator' and 'FreeHand' also save images in their own, more efficient, file formats. This does not help however, when you need to import them into your work. If you are using these packages, I suggest you also save your work in either WMF (or EMF) format, as noted above, or in the more universal EPS format.

Image placement

Image Size and Text Wrap:
As we noted in the introduction to this section the inclusion of a visual within your work can help explain a difficult concept or make the work more visually appealing, but it should complement your work, not dominate it.

Graphics should be sized so that they are clear and get the message across without breaking up the flow of text too much. They should also be placed as close as is practical to the relevant text, for example a small diagram or photograph placed between sections of text can make the page look interesting; but whether you place the material at a break between paragraphs, or embed the image amongst the text is a question of style.

Style:
A brash man in a Hawaiian shirt will be treated differently to a soft-spoken man in a tailored business suit. How they are treated depends on who is making the judgement and where.

It's the same with inserting visual material. Organise your materials in the relevant style of the report. If you were producing a formal laboratory report, then it would be appropriate to include your material between the paragraphs. On the other hand, if you were producing a more informal document, say an 'operational manual' in a relaxed style of presentation, then embedding the material within the text and flowing text around it might have more impact. The Dorling Kindersley range of books is an excellent example of this style of presentation and it is no more or less effective in presenting information than the more classic Encyclopaedia Britannica – just different. So try to look at examples of other peoples' work, and discuss ideas with your tutors about their preferred style of presentation.

      

It is likely that a graph may require a full page to display properly, but to do this you may have to get the reader to keep flipping backwards and forwards between the text and a diagram on say the next page. Why not consider inserting a smaller 'thumbnail' graph – which highlights the essential data trend – into the text and allowing the reader to study the main graph later if required.

Placement and Labelling:
Where information is being presented either in the form of tables, or graphically, then words are necessary to provide the link between the text and the presentation. The main body of text helps introduce the visual material and so it follows that such material should not only be referred to in the text, but should also be numbered.

Graphs and diagrams are usually labelled as 'figures' with a numbering system, which reflects the numbering system of the document. For example, the visuals in Section 3 of the report should be labelled 'Figure 3.1, Figure 3.2, Figure 3.3' etc.

Every figure should also have a descriptive title of caption because this then aids the reader's memory when trying to recall the information; and the figure number and the title is usually placed at the bottom of the diagram.

Borders and Drop Shadows:
Where the visual to be included with your work is an isolated image located on a plain background the same colour as the paper, then there is usually no problem in just inserting the image directly into the text. Where the image has a background that goes right up to the edge of the graphic, then it is often better to but a border around the graphic to separate and distinguish it from the main body of the text.

Most word processing packages can add the border for you and can help you format it in a number of ways.

Similarly, drop shadows can add visual dimension to your page preventing it from looking flat and has the advantage that it creates a more natural separation between an image and any text that is wrapped around it.

Note where the image has a plain background, a border does not look so attractive.

Here a border helps to separate the image from the text, alternatively, a drop shadow may be used to increase the separation.

Maps:
If a diagram, plan or map is drawn to scale, then the scale should be given and ideally included with the visual.

Similarly, on a site plan or map, the direction of north should be indicated.

Checklist