|
Graphing and the Visual
Presentation of Data Part 1: Introduction
and Basics
Overall Apperance This section might also be titled "miscellaneous", but the following points can affect the overall appearance of the graph. When including extra text or graphics, ensure that they are of sufficient quality and resolution so they do not become pixelated. Enlarging or distorting non-scalable images will eventually result in pixelation, but scalable images will do quite well in most circumstances. On occasion, importing a file from one program/format to another will also result in poor quality, so it pays to check your material before use or submission. There are situations where the graphing software or the printer may limit the resolution or quality of hardcopy output. When this happens, either settings, drivers, or hardware needs to be changed. If the data is going up on a presentation, the problem will then lie with either the computer or the projector's settings and limitations. A relatively uncommon occurrance is the production of printing artifacts or defects on hard copy output. Whether one-time or recurring, they are not acceptable and should never be permitted in work submitted to others. The output of dot-matrix and many inkjet printers falls on the line of acceptability: ensure there are no smears or mis-alignments. If the print is fuzzy or unclear, a better machine should be found and used. Many graphs benefit from--and some basically require--printing in "landscape" orientation. Some type of binding is necessary to keep the information produced on hardcopy organized, typical types include staple, 3-hole punch, tape, and spiral. When this is preferrable, make sure the binding is at the top of the graph, so the graph faces outward towards the viewer and away from the binding mechanism. Once you have developed a favorite or best format for graphs or spreadsheets, make a copy of the file and have it to use as a "cut and past" template for future data. As an engineer, one of my biggest pet peeves is having to do work twice! You should be creative (within reason) in making graphs look professional. It is not a waste of time to spend an hour or much more formatting a report or presentation. The more viewers there will be, the more time should be spent, as people notice the little mistakes and inconsistencies...it is this attention to detail that shows you mean business and care about the message and information. The goal is to impress your audience with apearance as well as your content. While many people would like to be impressed by your content and not worry about your layout, it is not easy for anyone to ignore errors or bad presentation materials. It is imperative that you remove as many distractions to your viewer as possible.
Critiquing Your Graph Just like any other important effort your first attempt is a (sometimes
very good) rough draft. Once you have put your graph together, check out
what needs to be changed and do so. Then leave it aside for a couple hours
or so and come back to it with a fresh mind. Make sure it answers the
following questions: For example, let us take a look at the # People Wearing Ties and Revenues vs. Week of 2004 graph below as a rough draft of a presentation slide to the Executive Committee for Salary and Employee Morale.
To show the Committee that more sales occurred when employees were dressing "business casual" is pretty straight forward, but keeping the past material in mind, we can come up with the following items to work on: Title Left y-axis (#1) X-axis Right y-axis (#2) Plot area Series Audience / Clarity Implementing changes to eliminate these issues results in a slightly more polished version, the Effect of Employee Dress on Sales Revenues: FY 2004 graph.
Exercises Tell the students to take a graph they have already created and "polish it up". They should spend 10-15 minutes making it a complete story unto itself, perfecting every aspect. Optional: on the day it is to be collected, have students trade with another person, who should write 3 good aspects and 3 poor aspects of the graph on the back. Have them trade with a second person. After submission, review highlighted aspects of this document, as seen fit (a quick paging through the submissions will give a good idea of what needs to be emphasized). Return graphs at next meeting, with thorough comments. Give students a raw data file (CSV text to add an extra step), have them create a graph from the data, and interpret as much as possible with what they have in front of them. Have students find a temperature profile (country, average planet, whatever) spanning at least a few decades and plot it versus energy usage by that same group/area/country. Take a known data set and alter the data...see what the students do when plotting it. Provide a graph which has poor formatting and many errors. As the students to make constructive comments on their copy, have them assign penalties and compare the students' weighing of how how important different parts of a graph are.
Web Resources Below are a few links I have come accross while trying to find other resources on the web. No other site I can readily find seems to cover how to put a "serious" graph together and format it, hence part of my drive for this document. Some of the sites below cover fundamentals, some interesting sub-topics covered above, and some are just interesting. Incidentally, if you conduct a Google search for "how to draw a graph" you will come up with many links on "graph theory". This is an area of mathematics dealing with the network of relationships between points. If you are interested in graph theory, take a look at links 2 through 5 under "Other Interesting Stuff". Links are listed in the order in which they have been encountered. They are the better materials I have found so far, check comments for other notes.
References Murgio, Matthew P., "Commuications Graphics", Van Nostrand Reinhold Company, New York, 1969. [notes] Stokinger, Mr., Lecture in 5th Grade Science Class, Diamond Middle School, Lexington, MA, 1985. [no notes available] Tufte, Edward R., "Envisioning Information", Graphic Press, Cheshire, CT, 1990. [notes] Tufte, Edward R., "The Visual Display of Quantitative Information", Graphic Press, Cheshire, CT, 1983. [notes] Tufte, Edward R., "Visual Explanations: Images and Quantities, Evidence and Narrative", Graphic Press, Cheshire, CT, 1990. [notes]
Corrections, Contributions, and Copyright I hope this document has helped provide an idea of what is acceptable for data presentation in today's professional environment. This is a "living" document, so if you have any suggestions for additional topics or issues that should be addressed, send me an email at JDM[-at]JustinDanielMeyer[dot-]com (this email is obscured so web "spiders" and "bots" can't glean it for spam lists-just read into it). This goes for any corrections, too! If you would like to use or re-print some of this material, feel free to do so, in accordance with the reproduction agreement listed below. If you find this material helpful, please spread the word. Encouragement is nice, but even better is telling others about resources from which they may learn. This document is copyrighted 2005 by Justin Daniel Meyer. Material on
this page is subject to this reproduction agreement.
Figure Creation Notes For the Author Open Excel full screen (running at a resolution of 1600px by 1200px), and set graph size to 150% (margins must all be 0.75"). Make sure no items are selected of have pop-up names then press Alt-PrintScrn to do a window capture. Past to image program and crop to 1350px wide by 930px tall: 35px - 143px ....................(-35px) - (-143px) 1384px - 1072px....................(-44px) - (-62px) Convert this to "full screen" reference, to avoid changes in future Excel window sizes. This is saved as *_lg.png. For this page, a 50% reduced version is made (smoothing on) and saved as *_sm.png.
or - export slides
from ppt (full/automated sizing), then convert and reduce as needed.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Copyright (c) 2005 by Justin Daniel Meyer. Material on this page subject to this reproduction agreement. |
