More Essays  | [Bill's Home]

Plagiarism. How to Spot it, and how to get rid of it!


Plagiarism is increasing, mainly because of the WWW. It is so easy to find a relevant site and download whole pages of material, and then publish it as your own. With a trained eye you can easily spot it. Before I start on this I would like to explain my viewpoint on plagiarism:

"Plagiarism occurs at many levels, and is often unavoidable. In most cases it involves unintentional copying of material, typically, these days, from the WWW. This often shows that a good deal of research has occurred, and that material which has been copied has been placed mistakenly into a final document. I have found that many students, under pressure of deadlines, will not have enough time to fully edit, and review their work. Thus once identified, the student should be given the opportunity to properly reference their material, unless it can be proven that the student intentionally plagiarized the material (which I believe happens in very few circumstances). The direct copying of programs, designs, and so on, and identifying it as one's own work is obviously more serious matter than simply copying background material. Again the student should be given the opportunity to properly reference the sources of the material, if exact matches are found.

It is key to remember that direct copying of material has occurred ever since the start of writing, itself, and it is not anything new. The thing that has changed is that it is now extremely simple for text to be found, in a fraction of a second. The strange thing is that many academics view the direct copying of diagrams as acceptable, but not the direct copy of text. It's all strange to me, and I would never pass judgment on anyone without strongly considering their viewpoint. In fact the only way that we could be fair on anyone, is that if we were to open-up every single essay that was ever submitted, and check every single line that was written. I bet that would open-up a few problems.

Most academics base their work on others, so it is natural that work that we have read will appear in some form or other. In fact it should be encouraged, especially if it is well written It's the skill of refining it and properly referencing it, that can take a while to develop. So ... my option is ... give the material back, and let the author trace the reference, and add it. Simple as that! The next time, I'll bet, it won't happen. To me, that's called LEARNING, and it's the foundation that academia is built on.

So to any academics out there ... be fair on your students, and give them advice on how they could make their work better. If possible, ask them to revise their work, and resubmit. It's a difficult lesson to learn, but once it has been learnt it will be learnt for life"

- Bill, 26 May 2002.

Here are a few that can be used to spot that you've taken material from another source:

Essay reads like a user manual. This is very typical in software development projects, where whole chunks of the user manual are pasted into the document.
Worst offenders:
Visual Basic, Visual C++, and anything to do with Java.

TIP: Read the material that you have written, and if it doesn't sound like you, re-write it (or get rid of it, completely).

Diagrams are beautifully produced. This is a particularity easy one to spot as intricate diagrams are inserted, alongside in-line Word diagrams. As a tip, a diagram taken from a WWW site should always be referenced, or should be redrawn, if possible.
Worst offenders:
OSI model, Ethernet networks and Cisco diagrams (especially pictures of routers and switches).

TIP: Re-do your diagrams, and interpret you own design. If you have to use the diagram, put a reference to the original source beside it.

Text is full of Americanisms. This might just be okay if the person is from the USA, as they tend to use these phrases, but they are unacceptable for a non-US person to write in a report (especially in a formal one).
Worst offenders:
grunt work and to shave time (Yuk!). Does anyone know what grunt work means?

TIP: Carefully review your terms. If possible get rid of business metaphors, and Americanisms (unless, of course, you are American).

Change of style. A key in assessing a report is to judge its consistency. Plagiarism can be easily spotted if the style changes from a vibrant, well-written text, to a boring, simple style, and then back again. After you've read a few reports, you can spot this one, easily.
Worst offenders:
Changes of punctuation (commas and semicolons appear, as if by magic, and then disappear again), changes from English to American spelling (organise to organize) and changes of style (especially from passive to active verbs - the program was written in Visual Basic to Visual Basic was used to write the program).

TIP: As before, read the material out-loud, or get someone else to read it for you. In fact the best person to read your report is someone who does not have a technical background. They'll soon spot changes of style.

Material covered has nothing to do with the subject, but covered extremely well. This is one of the best ways to spot plagiarism, as often a writer will typical find a nice site, and then copy whole batches of material. Unfortunately it tends to be unrelated to the actual content of the report, as the material was written for another purpose.
Worst offenders:
OSI model, Java theory (objects and things), and networking theory.

TIP: If it's not relevant, get rid of it. A reader cannot spot the bits that have been taken out, but they can certainly spot the bits that have been added, and that are not relevant. Readers can often end up confused, if the report has no general focus. If possible, plan your report out and try to define the general areas which you need to cover in the theory sections. Then, as plainly as you can, write your own interpretation of the theory. Add references to any material which you have taken from.

No narrative, at all, in the text. It is extremely difficult to read a report if it doesn't actually tell a story. Thus material that has been taken from other sources can sometimes to spotted if there is no narrative in the text, and it just seems to move from place to place, without any links.

TIP: Define your aims at the start of the report, and then in each chapter define the aims of that chapter, and why the material has been included, and how it fits together.

Essay reads like a product advertisement. This is a very typical one, these days, but it is easy to spot as there a lots of detail in the report, especially relating to the minimum and recommended specification. When you've read many reports, you'll know that there's nothing worse than reading lots of superficial detail (32MB of his, and in-line pipeline cache, of that). Often you'll typically find that there is a complete lack of criticism of the system being described.
Worst offenders:
Linux, Intel Pentium, Novell NetWare, and Microsoft Windows.

TIP: Get rid of superficial information. If you have to include it, put it in the Appendix. If possible try to put the other side of the argument.

Text looks like a WWW page. This is an easy one to spot as there are lots of table boxes that you would find on a WWW page, such as:

Bert Mike Ken
1 2 3
5 5 6

The font is typically Arial, and the border is typically the standard HTML table border.

Worst offenders: Linux, Intel Pentium, Novell NetWare, and Microsoft Windows, again. In fact there a lots of offenders here.

TIP: Reformat your tables in Word. A good tip is to convert the table to text (with Tab spaces), and then reformat the table, by selecting Text-to-Table. Also, put a reference to the table, if you have taken it from the WWW. Remember, marker will often give more marks for report with proper references, as one without them, as it shows the depth of research, and show how your work fits with others. It also gives students the opportunity to quickly find related material, if they have to use your report.

Text reads like a WWW page. This one typically goes along with No.8 (text looks like a WWW page), but a writer will typically try and hide the WWW page by reforming it to be more consistent.
Worst offenders:
Linux, Intel Pentium, Novell NetWare, and Microsoft Windows, again. In fact there's a lots of offenders here.

TIP: Reformat the page with your own style sheet. Beware, in Word, of format styles such as HTML Preformatting, and so on. If possible convert the styles back into something like Normal, Heading 1, and so on.

Hypertext links still embedded. This is a very typical one, where the author has forgotten to take the hypertext links out, and they are not properly referenced. Hypertext links work on a WWW page as the user can simply click on a link, and go there. In a report it is not really possible to do this, and no-one would ever write the document like that. Here's the difference:

This difference can be found at

as opposed to:

This difference can be found with myprog [ mydoc].

Worst offenders: Typically these come from personal pages, as many of the larger companies will not embed too many links in their pages, as they prefer to use a more structured approach (with menu items, and links outside the main text).

TIP: Find all your hyperlinks (typically with a search of www.), and review them. I often change the style for Hypertext links so that they are not viewed with underlines (and are not blue).

Other worrying signs:

Essay goes from zero to 100 pages, in the matter of a day.
No drafts, just the final version.
Lack of references in the References section.

Here are a few tips to avoid it:

Rewrite in your own words. If you find a good WWW site, do the following: read the material; look away; and rewrite it. Your own words are always better than some else's, as it's your own style.

Gid rid of superficial information. Many WWW sites, especially ones which are selling things will give detailed specifications on their products. These might be fine if you're look to compare one product with another, but is totally boring in an essay.

Put detailed information in the Appendix. You must watch that detailed does not get in the way to the main material, so put detailed information in the Appendix (especially code listings).

Add proper references to material that you find. The amazing thing is, that proper use of references is likely to generate higher marks, than no references, so put as many references into your report as you can. It shows that you've done a good deal of research on your subject area.

And finally ....