How to write the Discussion part 1


In this video, we’ll look at the content
of the Discussion section. The discussion section is the 4th and last
part of the research paper. It comes right after the results.
In the Results section, you explained all of the facts about the trends in your data.
Now, in the Discussion, you are going to write about the meaning of those results.
In particular, you need to answer some questions. First, how did your actual results compare
with the results you expected based on the hypothesis?
And how do your results compare with the results of other researchers?
This is where you will need citations again. Many of them might be the same ones that you
used in the introduction. Then, how might you explain any unexpected
results? Here again, you need to use citations to support
your ideas. Next, how might you test those potential explanations?
And finally, based on your results, what question or questions would you logically want to ask
next? The discussion section of the paper is probably
the most challenging part, because here you have to show the depth and
clarity of your scientific thinking. In most discussion sections, authors start
by reviewing the main results of the paper. Next, you should attempt to explain any unexpected
results and the processes behind the patterns that
you found. You should also evaluate the explanations
you proposed by weighing the pros and cons in the light
of other people’s research, and considering any limitations with your
experimental design. And finally, you should end your scientific
paper by a proper conclusion that brings together everything that has been
discussed and highlights the possible applications of
the study, or possible future research. OK, let’s take a look at the Discussion
section of this paper on the effect of jungle sounds and classical
music on the behavior of gorillas in the zoo. OK, here is the first paragraph.
Notice how the beginning of the Discussion reviews the main result.
The statement of your main result at the beginning of the Discussion
should tie up with the statement of purpose of the study,
that you made at the end of the Introduction. Here is how the gorilla paper does it.
This is last sentence of the Introduction. And
this is the first sentence of the Discussion. So you see we’re using the same words here. The next paragraph summarizes the results, but unlike the Results section,
they can interpret the meaning of the results here.
So, you can see that they say “behaviors indicative of relaxation” and “behaviors
typically associated with stress”. In the Results section we can only talk about
what we measured, so we can’t talk about “relaxation”,
but in the Discussion we are free to interpret the results.
OK, the next part here goes into the critical interpretation of the meaning of the results.
But first, let’s talk about bringing your results together.
The discussion is not simply a re-writing of the results section and then adding your
opinion. The discussion section explores the main results
of your experiments and therefore one paragraph of this section
can use the information contained in several figures or tables to make a single point.
This is from a study on the growth of radish and lettuce plants under white light (which
is the FL) and Red and Blue LEDs (that’s the RB).
The blue box is from the Results, and describes the trend that is shown in the figure down
here. The green box is from the Discussion.
Here the author combines information from this figure and this table in order to write
his interpretation of the results. OK, let’s go back to the paper on gorilla
behavior. The third paragraph discusses surprising things
that they found in the results and how they compare with other researchers’
findings. You can see the citations down here.
When discussing unexpected results, always start with a scientific explanation
before you discuss problems with the experiment. Even though you’re a beginner at research,
your results are still your results. They are facts.
It’s easy to blame yourself and only discuss limitations or problems with the experiment,
but a lot of the time, there are good scientific reasons for the problems you have,
and those are much more interesting to readers. Also, regardless of the possible causes, you
should suggest a possible future experiment to solve the problem you had.
So, let’s look more closely at this. The blue box is from the Results section.
It says that none of the treatments show effects that are statistically distinct from the control
experiment. Some of these non-significant patterns are
described more in detail in this sentence. Now the green box is from the Discussion.
Here the non-significant patterns are checked against pre-existing published results.
You can see the citations here. And even though the results are not statistically
significant, these findings MAY be relevant because other
researchers have gotten similar results. The fourth paragraph discusses the mechanism
behind why the results were they way they were.
OK, one possibility is that the music is not especially good or bad for gorillas,
it just covers up the zoo noise which is bad. That’s what they say here: as a mask.
The other possibility is that music is actually good for gorillas.
Notice that they consider all possible explanations, not just the one they like best.
After proposing an explanation for your data, you need to assess the validity of this explanation.
Has a similar pattern been observed before? How was it explained?
Do your results agree with these interpretations? And if not, who do you think is right?
Also, you need to evaluate the accuracy and the precision of your results.
Are there any possible extraneous variables that might have affected your data?
How much did that affect your results? And could you propose a way to reduce this
effect if someone repeats your experiment in the future?
OK, after you have interpreted and critically examined your results, it’s time to finish
the paper. Here’s the last paragraph.
Now in this paper, they actually call it “Conclusions” but you don’t have to do this.
You could just finish with the Discussion. And this would be the last paragraph.
Again, they review the main points here. And they talk about again why this research
is interesting or important. It also states the limitations of their study
here. A small group of animals and a short period
of time. And then they finish off with recommendations
for further research. So, even if your findings are weak or non-significant,
or there were problems with your experiment, try to highlight the positive outcomes of
your research. Review the main point so that readers can
remember it. And be positive!
Highlight what you think the reader would find useful or valuable.
And don’t apologize for problems or mistakes with your experiment. Readers don’t care
about that. A good way to finish is with this: “further
research is needed to… whatever”. This could be useful to any future scientist
willing to follow your path. Now YOU don’t have to do the further research;
somebody else can do it. You just need to point it out.
Finally, you might think of the Discussion as kind of like the opposite of the Introduction.
In the Introduction, you move from the wider research field,
and narrow down your topic to the details of your own study.
And in the Discussion, you begin by reviewing the details of your experiment,
and then expand to discussing the wider research field that it is connected to.

36 Replies to “How to write the Discussion part 1

  1. As a beginner in this field, and writing my first SCI research paper for PhD…. it quite difficult to write discussion in a concise way….
    This Video really helped me, and I want to say BIG thanks to Steve Kirk… I will download this video for future

  2. Thank you for this video! I find the way discussion sections are written in papers relatively easy to understand and copy, my only problem is that I have to produce ten pages of discussion, not just two. I find it incredibly hard to still make everything sound relevant and concise instead of just bringing up uninteresting points and going off on tangents. Reading so many "short-form" discussions makes it difficult to switch into a "long-form" writing mode.

  3. Thanks so much…I was stuck for very long time on this for my 1st manuscript writing. This is a very clear guideline to help me expand my results. Is the slides available for download?

  4. Cheers. My supervisor checked my dissertation discussion and I had done the completely wrong thing. This video explains everything clearly.

Leave a Reply

Your email address will not be published. Required fields are marked *