194 / Academic Writing for Graduate Students Constructing a Research Paper II / 195
Outlining the Structure of the Text
A final option is to consider whether you need to explain how your
text is organized. This element is obligatory in dissertations, but is
only included in RPs under certain circumstances. One such circumstance
arises when your text is unusual in some way, such as not
using the IMRD format. Another arises if you are working in some
new field. Cooper (1985) found, for example, that outlining the RP
structure was quite common in computer technology. Ask yourself
whether your anticipated readers need to have the organization of
the RP explained.
Here is a useful example of a textual outline, well-motivated by
the unusual structure ofthe paper. Notice how it uses a good variety
of sentence structures. The paper is about currency rates in the
European Common Market and was written by one of our students.
The plan of this paper is as follows. Section II describes the current
arrangements for regulating exchange rates within the EC.
In Section III a theoretical model is constructed which is designed
to capture these arrangements. Experimental parameters are
then tested in Section IV. Finally, Section V offers some suggestions
for the modification of the current mechanisms.
(Pierre Martin, unedited)
Task Eleven
Below is a textual outline by another one ofour students. Notice how
this time it lacks variety. Can you rewrite it?
The rest of the paper is organized as follows. Section 2 presents
the theoretical concept. Section 3 presents the empirical specification,
the implementation of the model. Section 4 presents the
results of statistical and other computational analyses. Section 5
summarizes the findings and provides a brief discussion concerning
the shortcomings of the methods employed. Finally, an appendix
presenting the detailed algebraic works is presented at the
end of the paper.
(Abdul Malik, unedited)
Task Twelve
Mow write, or rewrite, an RP introduction of your own.
Discussion Sections
It is not so easy to provide useful guidelines for writing Discussion
or Conclusions sections. (We will not distinguish between these two
terms, since the difference is largely conventional, depending on
traditions in particular fields and journals.) See what is done in your
own field.
The problem is that Discussions vary considerably depending on
a number of factors. Not all these factors are understood, but one
important one is the kind of research question—or questions—that
the study attempted to answer. Another factor that leads to variation
is the position of the Discussion section in the RP. By the time
readers reach the Discussion, authors can assume a fair amount of
shared knowledge. They can assume (if not always correctly) that
the reader has understood the purpose ofthe study, obtained a sense
ofthe methodology, and followed along with the results. Authors can
use this understanding to pick and choose what to concentrate on in
the Discussion. As a result, they typically have greater freedom than
in the Introduction.
Overall, if Results deal with facts, then Discussions deal with
points; facts are descriptive, while points are interpretive. Effective
Discussion sections are similar to effective lectures, which, as Olsen
and Huckin (1990) note, are based on points, rather than on facts.
Further, authors of Discussions have some flexibility in deciding
which oftheir possible points to include and then which to highlight.
Discussions, then, should be more than summaries. They should
go beyond the results. They should be
Academic Writing for Graduate Students
Constructing a Research Paper II / 197
AND, if possible, some
combination of these.
more theoretical
or
more abstract
or
more general
or
more integrated with the field
or
more connected to the real world
or
more concerned with implications
or applications
As Weissberg and Buker note, "in the discussion section you should
step back and take a broad look at your findings and your study as a
whole" (1990, 160).
We have said that Discussions can be viewed as presenting a
series of points. Typically, they are arranged as in table 22.
Move 1 is usually quite extensive, andMoves 2 and 3 are often quite
short. At this point, you might want to observe that Move 1 and the
later moves seem self-contradictory. Why, you may ask, build up
something in order to apparently attack it later? However, if we
remember positioning, we can see that authors can present themselves
very effectively by both
1. highlighting intelligently the strengths of the study
and
2. highlighting intelligently its weaknesses.
Indeed, Moves 2 and 3 can also be used to identify and open up
future research space for authors and their colleagues. However,
TABLE 22. Discussion Moves
Move 1
Move 2
Move 3
Points to consolidate your research space (obligatory)
Points to indicate the limitations of your study (optional but
common)
Points to identify useful areas of further research (optional
and only common in some areas)
this is less likely to happen, according to Huckin (1987), in areas
here there is fierce competition for research grants.
Task Thirteen
We have noted in this task nine points we would like to make in the
Discussion section ofthe paper on sentence connectors. They are not
yet in order. We believe that they fall in the following categories:
Move 1
(Consolidation)
Move 2
(Limitation)
Move 3
(Further research)
Six Points (3 results, 1 methodology, 1 centrality,
1 literature comparison)
Two points
One point
Into which category does each point fall? Fill in the blanks with the
labels. The first one has been done for you. Review the Methods,
Results, and Introduction sections of our mini-RP, if necessary (see
Unit Seven and Unit Eight Task One). The first one has been done
for you.
1. Move 2 (Limitation)
This is a very limited study restricted to a single field.
2.
Position varies from one connector to another (+ example[s]).
3. _
Sentence-connectors are quite common in academic writing (average
of 2 per page).
4. __
Our survey shows unexpected differences in the frequency with
which individual connectors are used (+ example[s]).
198 /AcademicWriting for Graduate Students
5.
Further research in this area might produce materials of
greater help to writers, especially nonnative speakers.
6.
Our survey shows that 25% of connectors do not occur at the
beginning of sentences.
It is important to conduct surveys to establish where connectors
actually occur in sentences.
8.
We are not yet in a position to offer explanations for choices of
connector positions.
Twenty-five percent noninitial seems higher than the grammar
books would predict, but lower than Morrow (1989), who found
53% noninitial in an economics journal (although Morrow used
a broader definition of connector).
As we can see, the heart of a Move 1 typically consists of statements
of results followed by a follow-up of some kind. The follow-up
might take the form of examples, comparisons with other work,
conclusions that might be drawn, or commentary on whether the
results are expected or unexpected.
Task Fourteen
Please write our Discussion section for us. Refer to the Results section
before Task Five for details of our study.
Opening a Discussion Section
As we have already suggested, there are many options in opening a
Discussion. Consider the case of the following data. We studied Discussion
openings in 15 articles from a small U.S. regional journal of
Constructing a Research Paper II / 199
atural history research. We found great variation. Four sections
open with the main results. This was the largest category, but still
less than 30% of the total. Three begin with a discussion of the
literature. Here are two examples.
a Graikowski et al. (1986) recovered . . . toxin from . . . and found
that . .. suffered 100% mortality when . . .
b Food shortages, social stress . . . within . . . are causes of dispersal
among . . . (Fritz and Mech 1981, Messier 1985, Mech
1987, Packard and Mech 1980).
Two sections start in a more dramatic way by offering a general
conclusion.
c. Apparently, we are witness to the early phases of a classic population
explosion.
d. From this data, it is clear that . . . are not major consumers of
commercially importantfish-speciesin . . .
The remaining types ofopening occur only once in the sample. We
were surprised, for example, to find only one opening that reminds
the reader of the original purpose.
e. The objective of the survey was to quantify the number of...
within . . .
In another case, the author opens with a summary.
f. This report brings together all known records of... since 1959.
In another, the authors raise the level of discussion by referring to
theory.
g The interrelationship of bird populations and the environment
is extremely complex.
One author starts with a comment about methodology.
• There is a bias associated with using either ground or aerial
counts, exclusively.
200 / Academic Writing for Graduate Students
Another author begins his Discussion section by highlighting the
special importance of his research site.
i. . . . is one of the few sites in North America where the presence
of a significant number of migrating . . . has been documented.
And in the final case, the author actually begins by discussing the
limitations of the data.
j. The census figure of... is expected to be an underestimate of
the total population of...
This small survey shows some of the many strategies that can be
adopted for opening a Discussion section. The choice of strategy
clearly depends in part on how the authors view their work. We will
briefly comment on the last three cases. In h the author begins with
a methodology critique of previous work, because one of his main
points is that he has taken the trouble to "combine both aerial and
ground surveys." In i the researcher begins by stressing the point
that the location of his research site offers exceptional advantages.
Finally, take the case of j. It might appear that the author of/ has
adopted a very risky strategy, but in this particular context it is not.
It soon emerges that carrying out a complete census of this particular
species would be very difficult. Therefore, the author presumably
felt on safe ground when he opened in this way. Indeed, he can go on
to claim that his numbers are much larger than anybody else has so
far been able to report.
Task Fifteen
Survey and classify the openings of at least six Discussion sections
from a journal in your field. Bring your findings to class.
Language Focus: Levels of Generalization
In the Results sections, statements may be quite specific and closely
tied to the data.
As can be seen in Table 1, 84% of the students performed above
the 12th-grade level.
Constructing a Research Paper II / 201
Seven out of eight experimental samples resisted corrosion longer
than the controls.
On the other hand, in the Abstract or in a Summary section, space
restrictions may lead to a high level of generality.
The results indicate that the students performed above the 12thgrade
level.
The experimental samples resisted corrosion longer than the con-
trols.
In the Discussion, we usually expect something in between these
two levels. One common device is to use one ofthe following "phrases
of generality."
Overall
In general
On the whole
In the main
With . . . exception(s)
Overall, the results indicate that students performed above the
12th-grade level.
The overall results indicate . . .
The results indicate, overall, that . . .
In general, the experimental samples resisted . . .
With one exception, the experimental samples resisted . . .
Limitations in Discussions
We saw in Introduction Move 2s (see page 185-89) that extensive
"negative"language was a possible option. In contrast, Discussion
Move 2s tend to use less elaborate negative language. The main
reason is obvious; it is now your own research that you are talking
about! Another reason is that many limitation statements in Discussions
are not so much about the weaknesses in the research, as
bout what cannot be concluded from the study in question. Producing
statements ofthis kind provides an excellent opportunity for the
202 / Academic Writing for Graduate Students
writer to show that he or she understands how evidence needs to be
evaluated in the particular field.
Task Sixteen
Complete four of the statements in set A. Base two on the mini-RP
on sentence connectors and two on studies you have been involved
with in your field. Complete at least one statement from B.
A. Limitations of Research Scope
1. It should be noted that this study has examined only . . .
2. This analysis has concentrated on ...
3. The findings of this study are restricted to ...
4. This study has addressed only the question of...
5. The limitations of this study are clear: . . .
6. We would like to point out that we have not . . .
B. Limitations in Conclusions. Below are some typical openings
for statements that firmly state that certain conclusions
should not be drawn.
1. However, the findings do not imply . . .
2. The results of this study cannot be taken as evidence
for . . .
3. Unfortunately, we are unable to determine from this
data . . .
4. The lack of ... means that we cannot be certain . . .
We said earlier that Move 2s are optional in Discussions. If you
feel it is unnecessary to comment on your work in either ofthe above
two ways, a useful alternative is to place the limitation in an opening
phrase.
Notwithstanding its limitations, this study does suggest . . .
Despite its preliminary character, the research reported here
would seem to indicate . . .
However exploratory, this study may offer some insight into . . .
Constructing a Research Paper II / 203
Paragraph
Paragraph 2
Paragraph 3
Fig. 11. Shape of a longer Discussion
Cycles of Moves
Finally, we should point out that many Discussion sections run
through the Move 1-2-3 (or part of it) sequence more than once.
Commonly, each cycle occupies one paragraph. Further, the more
research questions there are to be discussed, the more this cycling is
likely to occur. Such cycling can also occur in Introductions, but it
tends to be less common, especially in shorter RPs.
If you wish to write a longer Discussion, follow the shape recommended
in figure 11. Begin with specifics and then move towards the
more general.
Task Seventeen
Write or rewrite a Discussion section for your own research. If you
are working with others, collaborate with them.
Acknowledgments
Acknowledgments have become an integral part of most RPs. Indeed,
one famous professor of our acquaintance reported to us that
he always reads the Acknowledgments section of an RP first. When
we asked him why, he replied, "Oh, the first thing I want to know is
who has been talking to whom." While we do not think that this is
204 / Academic Writing for Graduate Students
standard reading behavior, it does show that Acknowledgments can
be more than a display of necessary politeness.
Acknowledgments occur either at the bottom of the first page,
following the Discussion, or sometimes at the end. They provide an
opportunity for you to show that you are a member of a community
and have benefited from that membership. Here we list some of the
common elements in Acknowledgments.
1. Financial support
Support for this work was provided by (sponsor).
This research was partially supported by a grant from (spon-
sor).
This research was funded by Contract (number) from (sponsor).
2. Thanks
We would like to thank A, B, and C for their help . . .
I wish to thank A for his encouragement and guidance
throughout this project.
We are indebted to B for . . .
We are also grateful to D for ...
3. Disclaimers (following element 1 or 2)
However, the opinions expressed here do not necessarily reflect
the policy of (sponsor).
The interpretations in this paper remain my own.
None, however, is responsible for any remaining errors.
However, any mistakes that remain are my own.
4. Other versions
An earlier/preliminary version of this paper was presented at
(conference or seminar).
5. Source
This article is based on the first author's doctoral dissertation.
Constructing a Research Paper II / 205
This paper is based on research completed as partial fulfillment
for the Ph.D. requirements at (university name).
Notes
1. We believe that, if permitted, Acknowledgments should be written
in the first person—/ for a single author and We for coauthors. It
is possible to find phrases like "the present authors," but we consider
them too formal for this situation.
2. As far as we can see, financial support tends to come first,
followed by thanks. Disclaimers seem optional. Mentions of other
versions and sources (if used) seem to come either at the beginning
or at the end. (But note that, in theses or dissertations, it is customary
to open with thanks to supervisors, advisors, committeemembers,
etc.)
Task Eighteen
Write a suitable Acknowledgments section for one of your pieces of
work. If necessary, invent some forms of assistance to expand the
section.
Titles
Although the title comes first in an RP, it may sometimes be written
last. Its final form may be long delayed and much thought about and
argued over. Authors know that titles are important, they know that
the RP will be known by its title, and they know that a successful
title will attract readers while an unsuccessful one will discourage
readers.
What then are the requirements for good RP titles? In general, we
suggest the following three.
1. The title should indicate the topic of the study.
2. The title should indicate the scope of the study (i.e., neither
overstating nor understating its significance).
3. The title should be self-explanatory to readers in the chosen
area.