Discussion
By Xiaochi Liu in Scientific Writing
The main function of the Discussion section is:
- to answer the question(s) posed in the Introduction
The discussion section will also
-
explain how the specific results support the answers
-
show how the answers relate to what is already known
-
suggest future research directions
It is not a recap. There needs to be added value to the reader.
How should we start?
-
Interpret and support key findings: answer the question that you posed in the Introduction and highlight evidence
-
If desired, restate study purpose or provide brief context first (> 2 sentences)
Organization of the Discussion
knitr::include_graphics("organization of discussion.png")
Content (information elements)
-
Review of the purpose of the study: activity, research questions, hypotheses
-
Review of findings (most -> least significant)
-
compared to your expectations
-
compared to others' findings
-
-
Explanations and/or speculations on findings
-
Limitations of the study
-
Implications of your research (generalisations)
-
Recommendations for future research & applications of your research
Two main components
-
State and interpret your key findings (Answer your research question)
-
Interpret key findings & conclude based on your findings and the literature
-
Explain how you reached the conclusions: compare/contrast with the literature
-
Give the paper significance by generalizing the results
-
Indicate clearly how you have advanced knowledge
-
-
Summarize and generalize (What is new and why your results are important)
-
Include explanations any results that don’t support your answers
-
Discuss other results and relevant hypotheses
-
Discuss possible errors/limitations
-
Provide explanations of unexpected findings
-
Suggest next steps
-
Drafting
Linking Introduction & Discussion
Comparing/contrasting your findings with others
This finding is consistent with that of Smith (2000) who …
In accordance with the present results, previous studies have demonstrated that …
These results support previous research into this brain area which links X and Y.
In contrast to earlier findings, however, no evidence of X was detected.
This study has been unable to demonstrate that …
This discrepancy could be attributed to …
There are two likely causes for the differences between …
Providing closure: write a one paragraph conclusion that summarizes your interpretations
-
Interpretation of key results (the answer)
-
Significance of the work
-
Try to generalize to broader situations
-
Practical application
-
Theoretical proposition
-
Language features
Need to pay attention to verb choice
-
Attitude to findings
-
Strength of claim
A very common structure in Discussions:
- main clause (with projecting verb) + that clause (with another verb)
These results suggest that the test material is more rigid.
Our results demonstrate that X regulates Y.
It appears that X has Y.
The presence of X suggests that Y may be stabilised by Z.
Two opportunities to indicate the strength of your claim
-
Main verb (choice of verb, tense of verb)
The results indicate/establish that there is a link …
Table 9 suggests/shows a greater response when …
The latest data question/challenge the conclusions of …
The results validate/support the hypothesis that …
The test results create/suggest a basis for …
Changes in X may have influenced/distorted the …
Figure 12 depicts/clarifies the relationship between …
-
Verb in that clause (tense, use of modal or not)
-
Tense of Verb in the that clause
-
Simple past tense: applies to your study only
It is possible that microbial activity caused some immobilization of labial soil phosphorous.
-
Simple present tense: general condition
It is possible that microbial activity causes some immobilization of labial soil phosphorous.
-
It is also possible to substitute the that clause with a Noun Phrase
Previous studies have demonstrated that there was a positive relationship between propagule pressure and the establishment success of non native species.
-
-
Express probability by using
-
modal auxiliary verbs, e.g. may, can, could
-
adjectives, e.g. possible, likely, uncertain
-
nouns, e.g. possibility, suggestion
Distance yourself from the claim by
-
using impersonal constructions
appears to/seems that…
it has been said that…
it is believed…
-
attributing the belief to others
in the view of some expert…
many researchers hold that…
Weaken generalisations by:
-
using the verb tend
-
qualifying the subject, e.g. most, many, a majority of, certain types of, nearly all
-
mentioning the evidence, e.g. According to this preliminary study; Based on the limited data available
-
adding exceptions, e.g. With the exception of…; Apart from…; Except for…
-
using a weaker verb, e.g.
leads -> contributes
causes -> is associated with
proves -> suggests
Common problems
-
The answer/ interpretation of key findings is not provided in the first paragraph
-
No concluding paragraph is provided
-
The importance/ significance of the study is not clear
-
Results are repeated or summarised in the discussion section
-
Often too long and verbose or too short and underdeveloped
Other considerations
Give credit where it is deserved
-
Methods/results from other studies added important elements to your study design
-
Your work supports/improves upon work of others
Be transparent about unexpected/results that disagree with others
- Explain the differences, don’t pretend they don’t exist
Acknowledge limitations/alternative explanations of your findings
- Pre-empts a reviewer’s opportunity to point them out
Final check
Remember that you need to ensure that your Discussion connects clearly with the issues you raised in your Introduction, especially the country where you began , your research niche, and your statement of purpose or main activity.
When the first draft of the Discussion is ready, go back to the Introduction and check for a close fit.
If necessary, redraft the Introduction to make sure the issues of importance in the Discussion appear there also.