Introduction
I have found that over time, students are becoming less skilled
in term paper writing. Things my instructors at university took for granted I knew are not
things I can take for granted my students know. I think there are two causes of this:
- The increasing ease by which low quality information can be found.
- The increasing ability of students to do simple things.
I will elaborate on the issue of low quality information throughout this portion of the website.
However, the problem is students find it easy to be able to Google websites. This is like
how the prevalence of modern junk food has meant people increasingly choose to eat poorly even though
healthy food is widely available. Because cheap, prepared junk food exists, people have forgotten how to cook.
In the same way, people have forgotten how to read quality journal articles because cheap websites are widely
available.
The main problem with using Google and other web search engines
is that valuable information is typically either complex or behind a paywall, and
Google can't get access to it. Examples of this kind of information are information in detailed
books, or academic conferences and journal articles. You have to use your university library
to get access to these resources- your university fees have paid for your paywall access. I have found
to my surprise that most students don't know how to access a university subscription database or eJournal
subscription.
With regard to doing simple things, students can Google things easily. There are tools
like Grammarly to help students write simple essays. But there's more to an essay than stringing
together a bunch of sentences. Thought and structuring are required. The analogy here is to a
car that is well painted and polished, but look under the hood and everything is rusty.
To my surprise, students know how to use the surface features of Google, but do not know how
to use Google to perform more in-depth searches. As an analogy, now that everyone
has a car, no one knows how to drive to really get the most out of their car.
A good term paper is not just a regurgitation of facts. It is a sustained argument.
The main thing your instructor will look for is your argument. That argument in turn
will be backed by not only different sources of evidence, but different types of evidence.
Given students don't seem to know how to write a term paper, I have put together this portion of my website to provide a simple introduction. This is divided into four parts: (1) Overview, (2) reading and looking things up, (3) writing
the paper, and (4) how your instructor can catch you cheating- unfortunately, the lack of competence is
also driving students to find ways to cheat.
Throughout this section of the website, I will give examples from the discipline of Information Systems.
That's because that's the topic I know.
The current contents are:
- Overview
- Reading and Looking Things Up
- Writing the Term Paper
- Catching Cheating
Any questions or comments should be directed to: The creator's email
Term Paper Writing Process
A good term paper is not done in a day. Indeed, a properly written term paper takes months to do. Broadly,
the steps required are:
- Choosing an initial topic
- Broad reading
- Structuring the essay
- Reading to write
- Writing the essay
- Polishing the essay
Choosing an initial topic. In the typical term paper, your instructor
will give you a list of topics to choose from. Here, you pick one of the topics that interest you.
Pick something you think will be valuable for you to know about. If you do the term paper well,
the knowledge you learn from doing independent reading will often stick with you for the rest of your life.
Broad reading. In this stage, you read to get a general sense of
your topic. This is where reading Wikipedia, web pages, and textbooks (not necessarily the one prescribed in
your course) are useful. You read at this point to get a broad sense of what the topic is about and to
figure out how to structure your essay. Unfortunately, most students stop reading at this point and never
do the next reading step (reading to write).
Structure the essay. Here, you read the rubric that determines
how you will be assessed. You then set up the main points of the essay. Most students
then write the essay from here, ignoring all the other critical steps. Most students also don't
realise they must not only follow the rubric, but develop the essay into a coherent argument.
Reading to write. In this stage, you read to obtain information
so you can write your term paper. This is very focused reading. You have the plan for the paper in
mind while you read, and everything you read, you do so with the intention of putting that knowledge in
the paper. This stage involves much more focused literature, including in-depth books, academic
conference papers, journals and things that are typically paywalled. You typically have at least three separate
files open while doing this: (1) the thing you are reading, (2) a document where you take notes, and
(3) a referencing tool like EndNote, BibTeX, Mendeley or Zotero.
When reading to write, you look in the paper to find evidence that either supports or contradicts your
argument. You note that specific piece of evidence down and then where you got it from. You add
the reference to your referencing tool.
Writing the Essay. Now that you have gathered the evidence,
you write the essay based on the evidence. It is entirely possible the evidence contradicts the
initial beliefs you had when you did broad reading. If so, you'll have to restructure your essay
slightly. Writing the essay involves putting your notes and references together in a coherent
form, hopefully using the structure you made under "Structure the essay."
Polishing the Essay. If you are doing things right,
this step of the term paper takes the most time! You should read your first draft and be
unhappy with it. You should be able to say, "This part isn't well developed,"
"this part is wrong," "this part needs to be elaborated on."
You then tear your essay apart, and rewrite it to be clearer. You keep doing this until
you are satisfied. Good novelists can write 10 drafts of their novels before being satisfied.
Good photographers throw away most of their work, only keeping the best.
About 80% of a good whiskey disappears before it is finally sold.
In the same way, you need to read your essay, cull the rubbish, and keep and expand on
the good stuff.
Reading and Looking Things Up
There are many written sources of knowledge. Not all written sources of knowledge can be found on the Internet.
When reading, you should always keep in mind at least three things:
- Your source is intended to achieve a particular objective.
- Your source may not be reliable.
- Your source may be biased.
Below are some examples of different written sources of knowledge.
Internet Sources
Wikipedia: Wikipedia is an
excellent place to start your reading. Wikipedia entries
are intended to provide good overviews of topics, and they often provide helpful references for you to further perform your
inquiries. However, Wikipedia articles are not necessarily reliable. Wikipedia itself will tell you it
is not a reliable source of information. It will also tell you that Wikipedia encourages
certain kinds of bias.
News Articles: Most news publishers have temporal
pressures. They are required to get news out as quickly as possible Thus, news sources
tend to be shallowly researched. The exception to this is in-depth investigative reporting,
which unfortunately is becoming rarer and rarer. Furthermore, not all news sources are
reliable. As a classic example, Fox News, one of the largest "news publishers"
in the United States has been repeatedly caught publishing stories the publisher knew
were untrue. News sources also often carry a bias. Fox News, for example, leans heavily
in favor of the Republicans, a political party in the United States.
Websites: Remember that it is effectively
costless to create a website. So, anyone can put up a website and say what they
want.
Always ask yourself why did the author
write this article? What credibility does the author have? What's the author's agenda?
Many of my students rely excessively on websites for their term papers without realising the
websites they relied on were written by consultants trying to solicit business.
Invariably these consultants oversell the technology they are advertising without critically
analyzing the actual weaknesses of the technology.
Government Report: In the interest of transparency,
democratic governments must often present information to the public. Such governments
often do so via reports on their websites. Like all organizations, governments have
agendas and reports can be misleading.
White Paper: A white paper is an article produced by a company for public
dissemination. Companies often produce white papers as a way to market themselves.
Community Forums: A community forum is a platform
for discussion on various topics. This is an information source
too many students neglect. I will often ask students to discuss the strengths and weaknesses
of a particular technology. One of the best places to discover weaknesses is to go to places
where people using the technology frequent, and read posts of these people complaining about the technology!
Of course, like all knowledge, there are biases. Community forums are often frequented by workers,
not managers. What is good for managers, might not be good for workers. Also, worker reports
might not be reliable. Someone might deliberately seed bad reviews on a community forum to
tarnish a competitor.
Non-Internet Sources
Textbooks: Textbooks have a role similar to Wikipedia.
They provide a good overview of a wide range of topics. Textbooks also tend to have
a wide range of useful references for you to look up. Textbooks are slightly better
than Wikipedia, because they are sometimes lightly
reviewed prior to publication. However, they are not a necessarily reliable source of
knowledge. For example, in Chua and Storey (2011), I highlighted errors in how
various textbooks in Database and Information Systems transformed entity relationship diagrams
to database tables incorrectly. Indeed, there isn't really something prohibiting someone
from writing and self-publishing their own textbook.
In-depth books: These can be written by a wide
variety of author types. In disciplines like anthropology, these are an important kind of publication.
Indeed, one of my favorite in-depth books is Argonauts of the Western Pacific (Malinowski, 1922), a classic
in the gift-giving literature. In other disciplines, they provide a great way of educating the public
about difficult to understand topics. An example book like this is Information Rules (Shapiro and Varian 1999).
Still other books are extended arguments. For example, Guns, Germs and Steel (Diamond, 1997) argues
the availability of steel and concentrated living conditions are the core ingredients allowing
western civilization to dominate.
Collected books: A collected book is a book where one or more
individuals edit a book, and invite authors to contribute chapters. The idea here is the invited
authors are experts in the specific topic of the chapter and so the whole book becomes a fount of wisdom.
An example of such a book is Power, Action, and Belief, a classic book in Actor Network Theory (Law, 1986).
The credibility and reliability of a collected book depends on the editor(s) who review and collate
the book.
Academic Conference Papers: An academic conference paper
is a short (frequently 10 pages or less) summary of an academic presentation. You can identify a paper
as a conference paper, because it will be identified by a conference acronym and words like, "Proceedings of
the...," "Conference," "Workshop," or "Symposium" somewhere on the document or thing related to the document
(e.g., hosting website).
Academic conference papers
present near cutting-edge thinking at the time of publication. Generally, a conference paper
is held to a higher standard of truth than books, websites, etc. This is because most academic conference
papers are peer reviewed. This means prior to allowing the person to present, one or more academics who
are presumably experts in the field looked at the paper and declared it relatively free of error. However, the short length of academic conference papers means authors are unable to discuss ideas in-depth.
Peer review is a useful way of screening for reliability and bias, but there are limits to what it can do.
For academic conference papers, reviewers are often under pressure to finish their reviews within a short
time frame. Authors cannot be made to make too severe changes because there is a conference deadline.
As a result, academic conference reviews are often not as thorough as (for example) journal paper reviews.
Nevertheless, in some disciplines like Computer Science, academic conferences are the principal mechanism
of knowledge dissemination.
Academic Journal Papers: Academic journal papers are the gold standard
for many academic disciplines. Most academic journal papers are in-depth treatises on a very narrow topic.
They are typically difficult to read, using very precise academic jargon.
The reason academic journal papers are considered quality written sources of knowledge is because they are
peer reviewed or peer editorially reviewed. Furthermore, the peer review process can stretch for a long time,
even years. This means
reviewers can demand authors make countless revisions to the work prior to publication.
Academic journal papers can be wrong. For example, E.F. Codd, invented the relational database model,
but simultaneously introduced the idea of three-value logic to relational databases (Codd, 1979). Ever since then,
database managers have had to deal with headaches associated with nulls.
Practitioner Journal Papers: These are papers written by academics that
attempt to simplify academic research for a lay audience. Most practitioner journal papers are editorially reviewed.
This means an editor, who is normally an academic, makes the call as to whether the paper is published.
In some cases, practitioner journal papers are peer reviewed, but the review process is not as rigorous as with
academic journal papers. Examples of practitioner journals are Harvard Business Review, Sloan Management Review,
Communications of the ACM, IEEE Computer, and IEEE Engineering Management Review. In Information Systems,
the journal Communications of the AIS is a special case. The target of the journal is information systems academics,
but the journal often publishes things that are not considered academic research.
Working Paper: A working paper is an academic paper (conference, journal or otherwise)
that has not (yet) been accepted in a peer reviewed outlet. Remember that it can take years for a journal paper
to get published. One can obtain copies of a paper before it gets published from the author, or from sites like arXiv or ResearchGate. Anyone can put up a working paper.
What You Should Read and Believe
Each type of written knowledge source comes with its own specific strengths and weaknesses. Thus, when writing a term
paper, it is important you not only read from many written knowledge sources, but also read from many types of written
knowledge sources. Sources like Wikipedia and textbooks provide breadth but are quite shallow. Academic journal articles provide depth, but because of the slow peer review process are often published years after the research was actually done.
Academic conference articles are more recent, but they are likewise shallow. Working papers have not been peer reviewed.
Don't get all or even the majority of your sources from the Internet. Most published
work does not end up on the Internet.
How do you know whether to believe a knowledge source? The answer is treat all
knowledge sources with skepticism. Challenge the source while reading. Ask why you should believe the source.
Does the source provide support for its arguments? What kind of support? How does the source's knowledge triangulate
with other things you have read?
Also, reading widely is the best way to get at the truth for your term paper. Reading widely doesn't mean believing everything you read. It means
taking from multiple sources and critically assessing the truth and falsity of all the knowledge you have
accumulated. The more you read and learn, the more you will be able to critically evaluate what you have read.
References
- Chua, C. and V. C. Storey (2011). "Issues and Guidelines in Modeling Decomposition of Minimum Participation in Entity-Relationship Diagrams." Communications of the AIS 29.
- Codd, E. F. (1979). "Extending the Database Relational Model to Capture More Meaning " ACM Transactions on Database Systems 4(4): 397-434.
- Diamond, J. (1997). Guns, Germs, and Steel: The Fates of Human Societies. New York, NY, W. W. Norton.
- Law, J. (ed) (1986). Power, Action, and Belief: A New Sociology of Knowledge?. London, UK, Routledge & Kegan Paul
- Malinowski, B. (1922). Argonauts of the Western Pacific: An Account of Native Enterprise and Adventure in the Archipelagoes of Melanesian New Guinea. Long Grove, IL, Waveland Press, Inc.
- Shapiro, C. and H. Varian (1999). Information Rules: A Strategic Guide to the Network Economy. Boston, MA, Harvard Business School Press.
Academic Quality Ranking and Predatory Sources
Within each written knowledge source type, there are wide variations in quality. It is important to be able to determine
the quality of the specific knowledge source type.
This is especially true for academically peer reviewed works. Not all conferences and journals are equal.
Students are expected to read and cite from quality sources and to reject poor quality sources.
It is difficult to assess quality by simple gauges like checking the publisher. Publishers
who are reputable in one field are not so reputable in others.
The best way to assess quality is to read the paper and compare what is said against the body of knowledge you already possess.
That being said, there are some external sources of paper quality.
Journal Quality Lists
For journals,
one way to assess quality is via journal rankings. Articles in journals ranked highly are generally of better quality
than those in journals ranked lowly. Unfortunately, rankings have a subjective element and not everyone agrees
on what the quality of a specific journal is. Within business schools, there are various rankings, which include:
Note these lists are specific to business schools. So, (for example) premier computer science
journals like ACM Transactions or IEEE Transactions tend not to rank highly there.
Citation Counts
Another metric used to assess journal quality is to assess how many other papers cite work from a specific journal.
Unfortunately, they are flawed in a number of ways, including:
- Citation counts measure impact, not quality. The two are related, but not identical. Some highly cited works
are cited as (for example) examples of mistakes, or because they are controversial.
- Journals can game citation counts. For example, some journals pressure authors to cite works from the
journal prior to submission.
- Certain kinds of papers attract more citations than others. Review papers, for example, tend
to be more highly cited. So, journals that are exclusively review journals (e.g., Academy of Management Review,
ACM Computing Surveys) have higher citations than journals that accept a wide variety of papers.
- Citations have an inverse exponential distribution. A few papers are highly cited, most papers are not.
It is not uncommon for 50% of the citations to a journal to be driven by a single paper.
One simple way to obtain citation counts is to look up the paper on Google Scholar.
There is normally a cited by attribute identifying how many other papers have cited that specific work. You can click
on the "Cited By" link to see all the papers that cited that paper thus expanding your reading.
Acceptance/Rejection Rates
One can get some indication of the quality of a publication outlet by how many manuscripts are rejected.
Once more, this is an imperfect measure of quality. For example, many lower-tier journals have higher or comparable rejection rates to premier outlets simply because poor quality scholars feel they have a chance at the lower-tier journals than
at the higher tier ones. The author of this website is an editor for multiple journals, including mid and high-tier
journals and rejects a near equivalent percentage of manuscripts at both types of journals. However, I reject
far more manuscripts at the mid-tier journals because there are more submissions.
Predatory Journals/Conferences
One kind of journal/conference to avoid is the predatory journal/conference. A predatory journal/conference
is a journal/conference that accepts any manuscript so long as a fee is paid. The typical predatory journal/conference
functions on a low-cost model and so these tend to publish papers exclusively on the Internet and do not have a paywall.
This means a student writing a term paper who exclusively uses the Internet to look up knowledge sources is more
likely to read and cite a predatory source than legitimate academic writing. I have marked numerous
term papers where students cite these kinds of dubious knowledge sources. Students who cite these kinds of
knowledge sources don't tend to get good grades because they demonstrate an inability to think clearly.
I don't believe in passing a student who cites a source where the grammar is so bad as to be unintelligible and the arguments have no foundation.
Note predatory journals and conferences can be published by "reputable" publishers. The Institute for Electrical and Electronic Engineers (IEEE),
for example, has been affiliated with several predatory conferences. This is because the IEEE
does a bad job of doing oversight on conferences it is affiliated with.
There are various lists of predatory publishers. The most famous (and likely out of date) is Beall's list.
Using the Library
It is impossible to write a decent term paper without using the library. This is because most credible
sources of written knowledge are either physical (e.g., books) or hidden behind paywalls the library subscribes
to (e.g., newsfeeds, journal and conference articles).
Knowing Your Librarian
If you don't know how to use the library, ask your librarian. There is a subject librarian
in your library paid for by your school fees to help you with your research. Take advantage
of this person. Ask them to explain what the resources available are and how to access them.
Books
Don't neglect the book as a source of information! Use the library electronic catalog
to find books relevant to your term paper. One of the things I look for is if you have
cited the classic book on the subject. If you haven't read the classic book(s) on a topic, I know
you didn't do a good job.
Electronic Databases and Journals
The modern term paper requires that students leverage off these to get access to academic conference
and journal papers.
Electronic databases provide you access to a large collection of conference and journal articles.
Unfortunately, particular resources will be available in one electronic database and not others.
To properly read up for your term paper, you will have to invoke multiple databases. The databases you
will often use as an Information Systems student include:
- The ACM Digital Library: This contains all publications
by the Association for Computing Machinery as well as other important publications. For example,
the Very Large Database Journal, a premier database journal is archived in this library. The ACM digital
library is one of the first places you want to search for technical information systems papers.
- ABI Inform: This database contains material of
relevance to business school students. I find the material overlaps greatly with Business Source
Premier.
- Business Source Premier/EBSCOHost: EBSCOHost is a database
of databases. Business Source Premier is the EBSCOHost database that contains material of relevance
to business school students.
- Elsevier Science Direct: This is a large European
publisher and provides access to many journals published in Europe. Premier journals
in information systems published by Elsevier include Journal of Management Information Systems,
Decision Support Systems and Information & Management.
- IEEE Xplore: This contains all publications
by the Institute for Electrical and Electronic Engineers, which includes a number of
important computing conferences and journals.
- JSTOR: This is a large repository of journals
covering business and social science. While I use JSTOR for all my "soft"
subject research needs, JSTOR is especially useful when you are doing cross-disciplinary
work. For example, when you study computer crime, you need to consult work not only
in business, but also (criminal) psychology and sociology. JSTOR conveniently puts
all the main journals of these fields together.
Take some time to get familiar with the search features of the databases. They can save you
a lot of time. For example, one thing I often require for my students' term papers is to summarize
two academic journal articles from A* journals. Students often have difficulty with this,
and will come to be for help because they can't find any articles. This is invariably
because they don't bother to investigate the database search capabilities. Often, the databases
have options like "only retrieve peer reviewed articles" or "only retrieve
journal articles" which transforms a search students took unproductive hours on to one I took 2 minutes on.
Electronic Journals provide most issues
of specific journals. There will often be times when the database tells you about a
knowledge source but the knowledge source is not in the database itself. You can go to
the library's electronic journal locator to obtain the source.
Interlibrary Loan
Your library is often a member of a library consortium where libraries agree to share resources with
each other. If you cannot obtain a resource directly from your library, see if you can request
it via an interlibrary loan. With budget cuts across libraries (often indirectly related
to the fact students no longer know how to use libraries and so don't use them), this is often
the only way to get access to critical resources.
Using the Web Search Engine
I am often surprised at how unfamiliar students are with web search engines. The below are some tips for Google.
- There is something called Google Scholar. This allows
you to search for only academic articles.
- Google News allows you to search only on news articles.
- Google Images allows you to look for pictures.
- Please read the Google search help
to learn about Google's search commands.
- You can put terms in quotes to look for an exact match. Don't just Google paired programming.
Google "paired programming"
- To omit a word in Google Search use -. To find cities called "Elmo" and not
the Sesame Street character, Google Elmo city -muppet.
- Sometimes the term isn't what you think it is. You sometimes have to get into the head
of the writer to figure out what's the appropriate term.
One way to find out what's wrong with a
technology is to look for <technology> sucks.
Section Headings
You are supposed to turn in the term paper to get a grade. The marker will grade your
term paper based on a set of criteria called a rubric. The marker has a lot of term papers
to grade. By the time the marker gets to your term paper, the marker is half-blind from
exhaustion. The easier it is for the marker
to find the things to give you points on, the higher your grade will be.
Look at the rubric. Make the things on the rubric your section headings. For example, my
typical term paper requires students to do the following:
- Take a technology and explain how it works.
- Explain the technology's strengths and weaknesses.
- Find two companies that use the technology. Explain how it is used in each company.
Highlight how the technology is being used effectively and ineffectively in each company.
- Summarize two academic journal articles in A* journals that write about this topic.
A reasonable way of structuring the essay would have the section headings:
- Introduction: This explains the technology
- Strengths
- Weaknesses
- Use in Company 1
- Effective Use
- Ineffective Use
- Use in Company 2
- Effective Use
- Ineffective Use
- Summary of Article <Article Title1>
- Summary of Article <Article Title2>
Referencing
For the typical term paper, the student provides very little intellectual contribution.
You are basically reading a whole bunch of stuff and assembling the stuff in your own way.
The content is not your intellectual contribution. The assembly is.
When you are presenting things that are not your own, you need to cite. Failure to cite
is academic dishonesty and can get you in trouble with the university. In my class, it
definitely gets you in trouble.
How to Cite
A citation has two parts, the citation and the reference. The citation is something
in the paper text that points to the reference. The reference is a textual description of
a work with sufficient information that the reader can look up the work independently.
For example:
Incorporated into relational databases is the idea of the relational algebra (Codd, 1979)
References
Codd, E. F. (1979). "Extending the Database Relational Model to Capture More Meaning " ACM Transactions on Database Systems 4(4): 397-434.
In the example (Codd, 1979) is the citation and the whole line about ACM Transactions on Database Systems is the reference.
There are different citation styles. The one most instructors require students to use is the APA reference style.
Referencing Software
Rather than learning all the intricacies of the reference style your instructor requires, do yourself a favor and
get referencing software. Examples of such include EndNote, BibTeX, Mendeley and Zotero.
BibTeX, Mendeley and Zotero are free. EndNote can often be obtained from your university for free.
The referencing software isn't perfect. But it will save you hours of time where you won't have to format your bibliography.
How Many References Should I Have?
In the typical 10 page term paper, you could easily make 60-80 references. I am not kidding.
Most of you reading this would probably think 5 or 6 references is plenty. But let's work this out.
References give the marker an indication of how many things you have read. When you cite 5 or 6
things and they are all websites, you are telling the marker you did not work very hard.
A 10 page paper will have about 5 or 6 section headings- Introduction, three structural elements (e.g.,
strengths, weaknesses, some other thing) and a Conclusion. 5 references means your entire introduction
was one cite, your entire strengths was one cite... This is definitely plagiarism.
So, how many sources will you need to get a good idea of how the technology actually works? 5? 10?
What about the strengths of a technology? 5? 10? What about
weaknesses? 5? 10? Some of those will overlap, but some won't. You'll have websites.
You'll need to cite the classic books on the subject. You'll have academic articles...
The material for your term paper is not your own. Everything you learned to write that term
paper came from stuff you read. So, just about every sentence is going to have a reference. To illustrate,
I am going to show you portions of a paper I wrote (Chua et al. 2019):
Note this is a paper with original intellectual work. Nevertheless, observe how there are citations on almost every line.
This is because I am continuously referring
to things other people said. In total, that 10 page paper had 38 references. This is because there is
original material later in the paper and so I don't cite. You won't have any original material.
Academic Dishonesty
Citing things the wrong way can lead to an academic dishonesty investigation in at least two ways.
References
Chua, C. E. H., et al. (2019). Strategic Gift Giving in Vendor Relationships: The Gift of Cognitive Regard. Proceedings of the 25th Americas Conference on Information Systems. Cancun, Mexico.
Reading Academic Journal Articles
Most students find academic journal articles hard to read. It can take hours to plow through one of these
dense, unintelligible things. Fortunately, you really don't have to read most journal articles. The trick
is to remember when you are reading journal articles that you are reading to write, not reading for the
sake of reading.
The point of reading the journal article is so you can create one or more sentences in your term paper
and cite the journal article as being a source for those sentences. So, first, you should have some sense
of what you are going to write. Then, read the abstract of the journal article. Is the article directly
relevant to what you are saying, or is it tangentially relevant? If the latter, craft a sentence for your
paper that the journal article is related to, and you are done reading! This will be for the vast majority
of journal articles you read.
If the journal article is directly relevant, you need to read further. For the typical empirical or design piece,
you read the introduction. Then read the literature
review. Then, the next bit which describes what the authors plan to do. Then skip to the section just before the conclusion. That will be the section with all the insights.
Find all the things relevant to what you want to write and put it in your paper. You don't have to read
the middle part of the paper which contains all the details of what the authors did. You just have to figure out
what happened as a result of what authors did.
Finally, look at the reference list. This will give you a clue as to what other papers you should be reading
and citing.
Of course, there are exceptions to this. I regularly require students to summarize two academic journal articles
for their term papers. This forces students to plow through two articles. Likewise, most review/survey articles
can't be easily read using shortcut strategies- those papers are shortcuts because they are pointing you to other
critical articles in the literature.
Sources of Evidence
Almost every term paper involves the student making an argument. Even a term paper that asks one to describe
a technology, strengths and weaknesses requires the student to argue that the strengths are strengths and weaknesses
are weaknesses.
There are many ways papers can argue their case. Here are some:
- Argument from authority: This kind of argument
means you are saying someone else has argued that what you are saying is true. The classic argument from authority
is the citation. The weakness with argument from authority is, of course, that the authority can be wrong, whether
generally, or for the specific case being argued about.
- Argument from anecdote: This kind of
argument arises when you give a story demonstrating your case. The story could be from personal experience,
or it could be someone else's story. The weakness with argument from anecdote is that one case may not be
generalizable to the population. However, note that argument from anecdote allows one to disprove claims like
"Women make lousy soldiers." The case of Milunka Savić disproves this.
- Deductive argument: This is the typical
math proof. You begin with the set of rules in play and show the rules in play lead to the conclusion.
In typical business environments, deductive arguments can't be used because the rules are unclear.
However, these kinds of arguments work in theoretical computing and economics arguments.
- Empirical argument: Here, you show
a systematic study conducted either by yourself (or more typically) someone else bolsters your argument.
Empirical evidence requires the observer to accept "the preponderance of the evidence"
as an acceptable threshold for winning the argument. There are famous examples of people
that refused to accept that threshold.
Except for the deductive argument, which can only be applied in very narrow conditions, every argument type
has a rebuttal. Even the deductive argument can be rebutted by showing the rules underlying the argument
are incorrect. To be convincing, the term paper must therefore employ multiple types of argument and
several sources of evidence.
Too frequently, students just make claims without support. Another common flaw is students will make
claims with the only support being a citation- argument from authority. As I showed above, authority can
be wrong!
Overview
I am going to keep this section of the website intentionally vague so students
don't develop mechanisms to make catching cheating more difficult. Instructors
who would like to know more can email me.
If you are one of my students reading this, be warned. I have built in mechanisms
in the assignments to catch you cheating. In one year, I caught 15% of my class
using ChatGPT. This wasn't "I had suspicions." This was "I obtained the evidence,
caught and reported the students."
Large Language Models
Students are using large language models and it is possible to conclusively detect their use.
However, detecting large language models requires the instructor to have prepared in advance.
What you need to do is know how large language models work and set up the term paper to trip
the large language model up.
As an analogy, imagine we have a cat person who is trying to disguise him/herself as a normal human.
The cat person performs all tasks just as well as a human of moderate ability. It also performs
better than an exceptional human under certain conditions. For example, the cat person has
extraordinary night vision.
So, set up the exercise in a room that allows so little light that the human cannot see a particular
door, but a cat person can. When the cat person opens the door, you have identified that this is
a cat person.
One interesting thing about large language models is they are mainly trained on public rather
than paywalled data sources. Insisting students actually do proper research allows one to
easily catch large language models because they don't have access to paywalled sources. In other
words, forcing students to use the library eliminates the danger of large language models!
Paying For Essays/Reusing Old Essays
People who write essays for others for profit need to solicit a certain amount of business to remain
profitable. They need to either churn out new essays quickly or reuse existing essays.
The more effortful the term paper is, the less profitable it is for the person being paid.
Actual research in a library (as opposed
to just looking things up on the Internet) eats into the profit of the essay writer.
Another simple strategy to use is called the canary trap.
You slightly change everyone's assignment requirement. This can be done both across individuals and across cohorts. When the essay
writer reuses an old essay, you can detect it, because it doesn't fit the specific essay requirement
given to the cheater. If the canary trap is sufficiently finely defined, you can identify both
the current cheater and the original source of the copied essay.
Reading to Write
Reading to write is a very active form of reading. When you read to write, you do not just read.
Instead, while reading, you actively make
notes about the things you read and you put all the notes together onto a single sheet of paper.
Your notes form the structure for the term paper.
The notes you make comprise two things- the note itself and the source for the note. The below
is a screenshot on notes that I made about crowdsourcing. The things in curly braces ({}) are references
to an EndNote bibliography. Each EndNote reference is accompanied by a little note about how that
reference is relevant to the paper I want to write. The below screenshot is an example of me
making notes while reading to write.
The notes that I make are not a summary of the thing I am reading. Instead, I make notes about
how the thing I am reading is going to be used in my paper. For example, I am going to use the
Da article to talk about influencers in crowds and herding.
Writing the article then becomes a case of rearranging the notes into coherent paragraphs like the illustration
below: