Section I
Chapter 1:
Scientific and Technical Communication in Context
Part 1;
Part 2;
Part 3
Chapter 2:
Reading Scientific and Technical Texts
Chapter 3:
Writing Scientific and Technical Texts
Part 1;
Part 2;
Part 3
Chapter 4:
Conducting Research
Part 1;
Part 2
Chapter 5:
Understanding Audiences
Part 1;
Part 2
Chapter 6:
Persuasion and Critical Thinking
Part 1;
Part 2
Chapter 7:
Participation and Policy
Part 1;
Part 2
Section II
Chapter 8:
Definitions, Descriptions, and Instructions
Part 1;
Part 2
Chapter 9:
Correspondence
Chapter 10:
Job-Finding Materials
Chapter 11:
Proposals
Part 1;
Part 2
Chapter 12:
Technical Reports
Chapter 13:
Scientific Articles and Abstracts
Chapter 14:
Oral Presentations
Chapter 15:
Formatting, Designing, and Using Graphics
Part 1;
→ Part 2
Grammar Handbook
Section III
Chapter 16:
Opening
Geoff Cooper:
"Textual Technologies"
Discussion
Chapter 17:
Opening
Steve Fuller: "Putting People Back Into the Business of Science"
Part 1; Part 2
Discussion
Chapter 18:
Opening
William Keith: "Science and Communication"
Discussion
Chapter 19:
Opening
Sujatha Raman: "Challenging High-Tech War"
Discussion
Chapter 20:
Opening
Dale L. Sullivan: "Migrating Across Disciplinary Boundaries"
Discussion
Chapter 21:
Opening
Tobias, Chubin, Aylesworth: "Restructuring Demand for Scientific Expertise"
Part 1; Part 2
Discussion
Bar graphs show comparisons among quantities. Each bar's length and area corresponds to a distinct value - number of cars, inches of snow - allowing the viewer to make comparisons among quantities: the longer the bar the greater the number. Like line graphs, bar graphs are plotted on a horizontal (X) and vertical (Y) axis within a coordinate plane. And like line graphs the horizontal axis usually represents time, the vertical axis quantity. The bars themselves can be drawn either horizontally - to show changes in quantity at a specific time - or vertically - to show changes over time. Keep in mind, however, you can clearly label each axis and bar to represent different variables and units of measure.
In constructing bar graphs, follow these basic guidelines:
Include variables or units of measure at regular intervals along the axes;
Determine the proportion of the horizontal and vertical axis in relation to lengths of the lines of the bars. For vertical graphs draw the vertical axis approximately three-quarters the length of the horizontal axis. If the vertical axis is the same length or longer than the horizontal axis the values will appear exaggerated. For horizontal graphs, a longer horizontal axis diminishes the real difference among values.
Make all bars equally wide; make the space between bars one-half the bar width;
Label each bar and axis clearly;
Use grid lines or tick marks when necessary to clarify the relation among lines and variables in the coordinate field. As with line graphs, if you include grid lines or tick marks, keep them in the background so as not to compete with the primary data. Maximize the amount of data to the amount of ink used.
Use varying shades of a muted color (like gray), and/or labels to distinguish bars. Use cross-hatching, vertical lines, dots and bright colors with restraint. Beware of moiré effects.
As with all graphs, bar graphs possess limitations you must consider. The resolution among values is dictated by the number of bars used and the relative difference among values. For example, let's say you want to demonstrate a steady, monthly decline in the price of a gallon of regular unleaded gasoline over a year long period. If the price drop is incrementally small and constant - one-quarter of a cent a month - the overall significance of a three cent drop in price vanishes among 12 bars representing essentially the same value. Accordingly, if you must adjust the baseline (either the horizontal or vertical axis) to accurately represent the data with bar length, you sacrifice text space and degrade the resolution among the values.
Common variations of the basic bar graph include the deviation bar graph, the grouped bar graph, the subdivided bar graph, the 100-percent bar graph and the pictograph.
The deviation bar graph shows quantities that digress from an established baseline. Deviation bar graphs are often used to highlight gains and losses; the bars on the positive side of the baseline represent gains; the bars on the negative side of the baseline represent loses.
The grouped bar graph (also know a the multiple bar graph) compares quantities by placing sets of two or three bars next each other. Each group may represent variables over different time periods. Bars are distinguished by labeling, shading and/or cross-hatching. Each set of bars is separated by at least one bar's width. However, in order to avoid moiré effects, consider using varying shades of the same color and providing as much white space as possible between sets of bars.
The subdivided bar graph divides each bar into two, but no more than three, proportional quantities. Each section of the bar is distinguished by labeling, shading and/or cross-hatching. Generally, each quantity represented by sections of the bar is generally added to give a total.
The 100-percent bar graph, like the subdivided bar graph, shows proportional quantities. Each bar extends to 100 percent; each section of the bar represents a percentage of the total. Each section of the part is distinguished by labeling, shading, and or cross-hatching.
Pictographs (or pictorial graphs) deviate from simple bar graphs by replacing bars with symbols (isotypes) representing quantities. Each symbol is assigned a value - each symbol of a cup of coffee equals $500 in gross sales - provided in a key, or which is given in a statement substituting for the quantity scale on the graph. Pictographs appeal to wide audiences and are easy to design given the hundreds of predrawn symbols and pictures (called clip art) available in computer software packages. Here are some guidelines to consider in designing pictographs:
Arrange the pictograph horizontally rather than vertically;
Select symbols directly corresponding to the object. Images of coffee cups, for example, should not be used to represent coffee beans;
Make all symbols the same size and same space apart;
Make symbols represent whole number; do not use a portion of a symbol (either dividing the symbol or reducing its size) to represent fractions. If you want to convey more technical information, consider using a different type of graph.
Versions of the pictograph use different size symbols to compare the quantity or magnitude of given variables with or without an X-Y axis. Be careful not to ask the viewer to make judgments about volume based on the representation you provide.
Charts convey abstract relationships - such as chronology, causality, authority, and hierarchy - among parts of a system, steps in a process and positions within an organization. While the design of most charts follows established conventions, in contrast to tables and graphs charts embody the creativity of the designer in meeting the needs of an audience. We will look at the most common charts - the pie chart, the flow chart, and the organization chart.
Pie charts - also called pie graphs, pie diagrams or divided circle graphs - compare the proportion of parts of a quantity - such as output, income or time - to a 100-percent whole. Pie charts occupy a prominent place in popular newspapers and magazines because the familiar design effectively shows the proportion of quantities to the whole and is easily understood by the layperson.
In designing pie charts, consider the following conventions:
Begin by locating the first radial line at the top of the circle - twelve o'clock - and arrange each section in clockwise in decreasing order. If you arrange the information in another way - according to subject matter or comparing each half of the circle - explain your design to the viewer;
Make each slice accurately display a proportion of the whole by converting the percentage each slice represents into degrees dividing the circle. One-percent equals each 3.6°ree; segment of the circle;
Compare no less than two and no more than five items. Advice on the number of slices into which the pie should be divided varies from five to eight. As a guiding principle, keep in mind the more equal and greater number of the quantities represented, the more difficult it is for the viewer to distinguish proportions;
Label each slice horizontally and include the value it represents;
Combine small segments into a larger segment labeled "Other" or "All Other." However, you must catalog for the viewer the items you include under that heading.
Use varying shades of a muted color (like gray) to distinguish slices. Use cross-hatching, vertical lines, dots and bright colors to highlight sections with restraint. Beware of moiré effects;
Separate -when necessary - one of the slices of the to lend emphasis;
Do not use several pie charts to represent a number of small data sets. Viewers have difficulty compare quantities in several distinct graphs.
A flowchart delineates the sequence of steps in an open or closed system, procedure or process. Open systems, procedures or processes have a start and a finish. A flowchart starting with a representation of the amount of copper mined in the United States and ending with its industrial, commercial uses is an example of an open system flow chart. Closed systems, procedures or processes begin and end in the same place. A flowchart showing the steps in using a software program is an example of a closed system flow chart. Flow charts are also used in illustrating manufacturing processes or how a bill in the United States Congress becomes a law.
Typically, flowcharts utilize labeled rectangles and circle linked together to show progressive stages, but pictorial symbols such as clip art images make a flowchart more interesting. You can achieve graphic emphasis in flowcharts - and other graphic aids as well - by varying the width of lines. Tufte (1983) suggests data graphics are enhanced "... by the perpendicular intersections of lines of differing widths (185). Keep in mind, however, the more ink you use the more importance the viewer lends the information.
A version of the flowchart with which many of you are familiar is a decision chart. A decision chart maps different routes to a set of results depending on yes or no answers to given questions. Related to decision charts are instructional charts. Found frequently in tutorials accompanying software programs and software manuals, instructional charts ask the user to make yes or no choices in response to given prompts. In response to the user's answers, the instruction chart displays appropriate commands and their consequences.
An organization chart displays the structure of an organization. Many organizations are arranged hierarchically and organizational charts offer an effective ways to show the chain of rank, authority and responsibility. Show graphic emphasis by increasing the ink to information ratio. The power of a particular position is highlight by enlarging the rectangle including the position title - and, considering audience, a brief description of the title - widening the width of the lines of the rectangle, and widening the width of lines connecting certain positions. The selective use of color and typeface variations also lends graphic emphasis to the organization chart.
Diagrams illustrate the relationship among physical components of objects. By providing a more direct correspondence between an object and its demonstration - often in three dimensions - diagrams reduce the level of abstraction found in tables, graphs, charts and words. However some diagrams employ symbols - such as schematic diagrams - to represent the relation among components. But more specialized diagrams are meant for specialized and professional audiences. Generally, audiences find diagrams more accessible and easier to follow than other graphic representations. Common diagrams include cutaway diagrams, exploded diagrams, schematic diagrams and block diagrams.
Cutaway diagrams present a three-dimension cross-section of an object by pictorially removing part of the exterior to reveal what is underneath. If, for example, you could peel back the plaster on your wall, you would see the electrical wiring, insulation and framework underneath. Cutaway diagrams provide viewers an orientation to the internal components of an object in reference to its external characteristics.
Exploded diagrams are similar to cutaway diagrams in exposing the internal components of an object, but the components are shown separately in actual physical relation to each other. Exploded diagrams are an invaluable component for repair manuals of all types.
Schematic diagrams use symbols to illustrate concepts (such as resistance to conducting electricity) and components of an object or system. The purpose of schematic diagrams is to provide a true representation from which an object - a circuit board for instance - is built or repaired. Widespread use of schematic diagrams is usually reserved for engineers and other professional. But you will likely run across schematic diagrams in many of the consumer electronics products you purchase.
Similar to schematic diagrams are block diagrams which mapping the relations among components in a system or mechanism, but convey less information. Block diagrams often substitute for longer, more complex prose explanations of the components of a system. (Insert figure)
Photographs and slides, compared to the graphics and visual images previously discussed, provide the most accurate representations of objects and phenomena. The availability, declining cost and ease in using instamatic and 35mm cameras allows you to take pictures without consulting professional photographers. Basic photography, and some developing skills have become necessary for practitioners in many technical and scientific fields. Some industrial laboratories and firms, however, employ professional technical photographers
Scientific and technical literature is almost exclusively a black-and-white medium. Black-and-white photographs are known as continuous tone images because they contain black and white tone of varying intensity. As with color graphics, the use of color photographs increases the costs and adds time to the production of a publication. Cost and time in using photographs is a product of the camera's abilities; instant cameras develop film automatically - 35mm film requires developing and processing.
Generally good photographs have the following features:
A "glossy" or smooth finish which reproduces well during the printing process;
A sharp contrast. Pictures in bad focus, with glare or that are over- or under-exposed should not be used;
A clear background. A cluttered background in which items are seen other than the one you want to emphasize distract the viewer. However, while this criteria applies to technical artifacts that, in some cases, are easily removed from their environment, the same cannot be said of photographs of natural phenomena. More care needs to be taken to frame objects in the photograph properly. To frame an object properly, the picture should be taken at a distance to realistically show the relationship among all parts. Certain aspects of photographs with too much, or extraneous detail, are highlighted by superimposing small arrows (also called callouts) on the picture. Photographs are also cropped, labeled, enlarged or reduced to emphasize an object or a portion of an object.
A clear angle and sense of scale. Objects can appear larger - a photograph taken looking up at an object - or smaller - a photograph taken looking down at an object - than their actual scale as compared to the photographer. When necessary indicate the angle at which you took the picture. Give the object a sense of scale by framing it with another object which indicates size - a ruler at the bottom of the photograph - or with which people are familiar - a tree, a hand, or a coin.
Slides are frequently a medium around which classes and oral presentations in science and technology are organized. The criteria above apply to taking slide film, although the emphasis is on color.
In the following section, Dr. Chris Ruckman, a mechanical engineer formerly with the United States Navy, offers advice for those of you contemplating using, or buying computer graphics software packages. Dr. Ruckman categorizes and explains the kind of software used in graphing numerical data and in generating technical illustrations. He also provides helpful criteria to consider in matching your graphing needs with the requirements of your profession and the particular job.
With modern graphics software you can create publication-quality charts and illustrations no matter how modest your artistic skills may be. But choosing the right software can be tricky. Your software must be appropriate not only for the task you expect to encounter, but also for your budgets of money and time. Questions abound. Will basic, easy-to-learn software packages suffice, or will the versatility and power of more advanced tools be worth the additional expense and training time? Can a single multi-purpose package provide everything you need, or will separate component packages make more sense? For incorporating graphics with a text can a basic word processor do the job, or will a full-blown page layout package be more appropriate? The paragraphs below describe five basic categories of graphics software, and offer some practical advice to help you decide what software will be right for you.
Categorizing software is difficult because products evolve rapidly, and because many packages address multiple categories. But even considering the widely varying hardware systems (desktop computers, graphics workstations, multi-user mainframes, etc.), some generalizations about the functions of graphics software is possible. There are three basic categories of software for manipulating and graphing numerical data:
Spreadsheets: Spreadsheets are computerized ledgers used to manipulate tables of data; with mathematical functions, built-in programming languages and other features, they are far more capable than the paper spreadsheets after which they were modeled. In addition to their analysis capabilities, spreadsheet programs can produce the line plots, bar charts, and pi charts most frequently used in technical illustration. many also include basic drawing and annotation tools, making them adequate for a variety of technical illustrations. Some will find everything they need in a good spreadsheet; for others, the selection of plot types and the control over formatting options will seem limited.
Graphing Packages: Picking up where spreadsheets leave off, graphing packages provide a wider array of graph types and increased control over the printed output. Specialized chart types not found in spreadsheets may be invaluable to your specific application. For example, polar plots, which graph data in polar coordinates, are useful in many scientific disciplines but often are not implemented in spreadsheets. Curve-fitting, spectral analysis, contouring, and other date analysis procedures are often automated in graphing packages, saving you time.
Advanced "graphic data analysis" packages: When the sheer volume of data overwhelms the illustrative capabilities of traditional charts and graphs, advanced presentation formats can point out features of the data that would not otherwise be apparent. Animation, multi-dimensional color-mapping, and other sophisticated effects once possible only on high-end graphics workstations have migrated to more affordable computer systems, although in terms of hardware/software costs and training time such techniques are still expensive. Keep practical matters of production in mind; an animation cannot be photocopied, nor can it be printed in a technical journal.
For creating technical illustrations, two other categories of software are available:
Bitmap-based illustration packages: A bit-map based "paint" program treats a drawing as a collection of dots (a "bitmapped image"). Simplified versions, with provisions for drawing lines, shapes, text, and other elements, are often bundled with other software at no extra cost. They are easy to master, and their basic capabilities are suitable for uncomplicated illustrations. Revising the bitmapped images, however, can be difficult and time-consuming. For example, once a line or shape is drawn it cannot be moved or resized except by erasing it and redrawing it.
Vector-based illustration packages: In a vector-based drawing program, objects can be moved or resized much more easily than in a bitmap-based program. This flexibility allows speedy revisions and sophisticated effects. For complicated graphics such as exploded diagrams, cutaway diagrams, and dimensioned drawings, vector-based illustration packages are more capable than bit-map based packages. Although they usually cost more and require more training time, the added flexibility is essential for producing finely detailed technical illustrations.
Of course, there are many special-purpose programs that do not fit into the categories named above. Software packages for computer aided drafting (CAD), statistical analysis, mathematical or numerical analysis, structural or fluid analysis, etc., can usually produce graphics for use in documents. Other helpful programs focus on specific needs with graphics such as organization charts, flow charts, or molecular diagrams. If you find yourself spending too much time on any one aspect of your graphics needs, investigate special-purpose programs. They could make your life easier.
Try to consider all your computing needs; your word processing and graphics packages should work as a team. Technical writers have benefited from the emergence of modern graphical user interfaces, primarily because they have made it easier to integrate graphics with text. For example, if a chart is "embedded" within a document, you can revise the chart by invoking a separate "window" for the program that related the chart; meanwhile, the chart itself is stored in the same file as the document. Embedding streamlines the revision process, since graphics are typically modified several times during the life of a document. Embedding also simplified document tracking and long-term archiving by keeping everything, including graphs and figures, in a single file rather than in separate files.
By automatically handling such chores as axis scaling and label positioning, graphing software allows you to concentrate on finding the but format of the data. Use the time and effort saved to advantage. Instead of simply accepting the first plot that pops up on your screen, experiment with different data formats. Look critically at all the plots used in a document. If you find yourself mentally superimposing two different plots, or using your thumb as a ruler to compare scales between them, try plotting different combinations of data or using a different type of chart. You will improve readability, and you may reveal features that would otherwise have gone unnoticed.
In fact, plotting data with a computer can be so easy that a few words of caution are in order. Many graphing programs are capable of producing advanced, complex date formats with little effort on the part of the user; for example, consider the impressive-looking 3-dimensional, color-mapped surface plots that appear in advertisements for graphing software. But in reality, few situations needs such complicated data formats. Complex graphics tend to obscure individual data points. Where possible, make the format simple and familiar so that your reader concentrates on the date itself rather than the format. On the other hand, a complex format such as surface plot can show vast amount of information in a compact space, allowing for a broad overview of the data. If you need a complex format to illustrate a trend or idea, be sure to explain the format to your audience before you describe the actual data. You may even need a combination of simple formats (to describe individual data points) and complex graphics (to describe the overall view).
By the same token, you should also exercise caution in adding color to you documents. Color printers, historically expensive to buy, use and maintain, are becoming more affordable and convenient. But printed color graphics cannot be reproduced easily in photocopies, and many technical journals do not accept color graphics simply because their printing facilities cannot accommodate color. In addition, color can be distracting when its sole purpose is to advertise that the author has access to a color printer! Although color loses impact when overused, judicious use of color can enliven a document and enhance the transfer of information.
If you do frequent oral presentations, you will find presentation software invaluable. Presentation software is like a word processor that has been optimized for outlining, formatting and revising oral or on-screen presentations. Title and border formatting, bulletted lists, and so forth are done automatically, freeing you to concentrate on larger problems - organization, focus, flow and timing. Audience handouts, presenter's notes, and other conveniences are usually included. Close integration with word processing, graphing, and illustration software makes it easy to bring existing work into a presentation.
A last word of advice, especially if you are just beginning a career: invest time and effort to really learn about your software. Only if you budget time for software training will you take full advantage of the powerful features built into your software. An example is automatic list numbering in a word-processor: learn the skill once, and you will probably use it for the reset of you career. The same is true for graphics software. Realize that time spent learning is time well spent.
Guidelines for designing graphics ordinarily assume the human brain works like a computer. On this model, visual perception is the result of the brain decoding and assigning meaning to the sequenced information encoded in visual representations. Sense data, in itself, contains no meaning; the brain constructs images and their meanings. This is the leading view of experimental psychologists and philosophers and dates to the seventeenth century. Given this model, our job as graphic designers is to encode information and provide the most efficient way for viewers to decode it.
Trends in sophisticated graphic representation - computer simulations and virtual reality - suggest the more accurately the visual representation mirrors our daily perceptual world - a world of movement, background and texture - the better we respond. Can we attribute a better responses to visual aids to better encoding? Certainly. But correlating visual perception to computer decoding is not the only way to design graphics. As the task of the natural sciences is to study natural world - a moving, changing, three-dimensional world - then perhaps visual representation should reflect this dynamism. In pondering visual perception and representation information, ponder what world you want to represent.
1) Many well-known scientists and inventors illustrated their concepts such as Newton, Leonardo da Vinci, Thomas Edison. Look at books in which the illustration of scientists, or illustrations of scientific experiments are included. What do the illustrations reveal about the concepts of early science and technology? Do the drawings of the scientists and inventors accurately depict what was to become their invention or discovery?
2) Find a copy of a journal or newspaper that uses graphics heavily. Using a specific as story or journal article as a basis, give an analysis of the effectiveness of the graphic. Consider whether you think the graphic is necessary, if it produced professionally, if it conforms to criteria for effective graphics presented in this chapter and if the graphic is explained in the article.
3) Using whatever techniques you think are most effective, draw an illustration depicting the development and evolution of a technology over time. Determine which advances were the most significant. Examples could range from the development of the bicycle, the internal combustion engine, the fountain pen, or the personal computer.
4) Create an organizational chart for the job at which you work, your fraternity or sorority, or local government organizations.
5) Select a graphic you think that represents "flatland." Revise the graphic to include a three-dimensional, textured representation.
Becker, Richard A., William S. Cleveland and Alan R. Wilkes. " Dynamic Graphics for Data Analysis" in William S. Cleveland and Marylyn E. McGill. eds.Dynamic Graphics for Statistics. Belmont, CA: Wadsworth, 1988.
Cleveland, William S. and Marylyn E. McGill. eds.Dynamic Graphics for Statistics. Belmont, CA: Wadsworth, 1988.
Cleveland, William S.The Elements of Graphing Data. Monterey, CA: Wadsworth, 1985.
Cleveland, William S. and R. McGill. "Graphical Perception and Graphical Methods for Analyzing Scientific Data,"Science 229, no. 4716, 1985, 828-833.
Cleveland, William S. and R. McGill "Graphical Perception: Theory Methods and Application to the Development of Graphical Methods,"Journal of the American Statistical Association Vol. 79, no. 3 & 7, 1984, 531-53.
Cochran, Jeffery K., Sheri A. Albrecht and Yvonne A. Green. "Guidelines for Evaluating Graphical Designs: A Framework Based on Human Perception Skills."Technical Communication, First Quarter, 1989, 25-32.
Ford, Brian J.Images of Science: A History of Scientific Illustration. New York: Oxford University Press, 1993.
Gibson, James J.The Ecological Approach to Visual Perception. Hillsdale, NJ: Lawrence Erlbaum, 1986.
Gibson, James J.The Senses Considered as Perceptual Systems. Boston: Houghton Mifflin, 1966.
Mitchell, John.Writing for Technical and Professional Journals. New York: John Wiley and Sons, Inc., 1968.
Reed, Edward S.James J. Gibson and the Psychology of Perception. New Haven: Yale University Press, 1988.
Tufte, Edward R.Envisioning Information. Cheshire, CN: Graphics Press, 1988a.
Tufte, Edward R. "Comment on Dynamic Graphics for Data Analysis" in William S. Cleveland and Marylyn E. McGill. eds.Dynamic Graphics for Statistics. Belmont, CA: Wadsworth, 1988b.
Tufte, Edward R.The Visual Display of Quantitative Information. Cheshire, CN: Graphics Press, 1983.
Supplimentary References
Bethune, James D.Technical Illustration. New York: Wiley, 1983.
Blanchard, Russel W.,Graphic Design. Englewood Cliffs, NJ: Prentice-Hall, 1984.
Lefferts, Robert.How to Prepare Charts and Graphs for Effective Reports. New York: Barnes and Noble, 1982.
Sanders, Norman.Photographing for Publication. New York: Bowker, 1983.
Bar Graphs
Charts
Diagrams
Photographs and Slides
Computer-Generated Graphics
A Final Word
Exercises
References
Chapter 15: Part 1