Skip to main content
Search form
  • 00:01

    GREG MARTIN: So you want to write and publisha scientific paper or an academic paper,and you want to publish it in a peer-reviewed journal.I'm going to talk you through the process of howto write that paper and what it isthat goes into the various sections of the paper.So basically, papers have got a number of sections.There's the title and the abstract.There's the introduction and background.There's the methods, results, discussion, and conclusions.

  • 00:23

    GREG MARTIN [continued]: I'm going to talk to you about what goesinto each of those sections.Let's start by talking about the title.Now, the title needs to be about 15 words or less.That's not a lot of space.You got to get to the point quite quickly.The things that you want to include in the titleare, firstly, the purpose--what is the question that you addressed?Secondly, the scope-- what are the edges of this thing?

  • 00:44

    GREG MARTIN [continued]: If you addressed a particular population group, say that.If it included a particular time frame, say that.Some of the time, you want to include the methodsthat you used.So if it was a cohort study, for example,you might want to include that in the title.This is all information that is useful to the readerif they're trying to make a decision whether or not they'regoing to read your paper.Things to avoid-- don't put in acronyms.

  • 01:04

    GREG MARTIN [continued]: Don't use abbreviations.Don't ever use exclamation marks.Keep it nice and professional, and you'll have a great title.Next, let's talk about the abstract.Now, the abstract is usually about 300 words.Again, it's not a lot of space.You want to be concise.There are different types of abstracts.What I'm going to talk to you aboutis one that's called the informative abstract.

  • 01:25

    GREG MARTIN [continued]: That's the most common type of abstract.And it follows a certain format, and we're justgoing to get into that.Of course, you're going to includethe purpose of the paper, and you'regoing to have an overview of the methods,and the main findings, the conclusions,and the recommendations.Importantly, you're not going to have any discussion.So you're not going to critique or evaluate the work.You're not going to include any subjective interpretation,

  • 01:46

    GREG MARTIN [continued]: no speculation, no references to other worksor other publications.You're going to keep it nice and factual.You're going to write in the past tense,because this is reporting on a study that's alreadybeen completed, and you're going to do the whole thing in about300 words.Now let's talk about the introduction and background.When you write the introduction and background, whatyou want to get out of this is you wantthe reader to keep reading.

  • 02:07

    GREG MARTIN [continued]: To do that, what you want to do is provide context.In other words, with reference to existing literature,you're going to tell them what we knowand what we don't know in this particular subject area.And in the context of what we don't know,you're going to introduce your research question, the purposeof your study, the question that you're going to answer.And when they read that, they're going

  • 02:28

    GREG MARTIN [continued]: to want to carry on reading.Now let's talk about the methods.If you want your research to be taken seriously,you need the reader to understand exactly how itis that you undertook the researchand drew the conclusions that you did.In fact, you need the methods to be in so much detailthat, if they wanted to repeat the studythemselves to generate the same conclusionsand results that you did, they would be able to.

  • 02:50

    GREG MARTIN [continued]: Now, what your methods section lookslike depends largely on the type of researchthat you've undertaken.So method sections in qualitative researchlooked very different to methods sectionsin quantitative research.What I'm going to be talking about hereapplies mostly to quantitative research,but the basic principles apply to any paper.

  • 03:11

    GREG MARTIN [continued]: Here's what must be included.You need to state your study question, or hypothesis.You need to include your study design.So was it a cohort study?Was it a case control study?Is this secondary data analysis?Was this content analysis of interview data,et cetera, et cetera?Next, you want to talk about your data.Did you collect the data yourself?

  • 03:31

    GREG MARTIN [continued]: If so, how?What tool did you use?Has the reliability and the validity of this toolbeen established?How did you identify relevant variables?If this was routinely collected data, like hospital dataor census data, or if you accessed available datafrom existing surveys or ongoing studies, how did you access it?How was it stored?Was it anonymized or deidentified?

  • 03:51

    GREG MARTIN [continued]: Did you have permission to access it?Describe the data in more detail.In other words, if it's sample data, what was the sample size?How did you decide on that size?Describe the variables of interest,the number of observations, if appropriate,how you analyzed the data.What software did you use?What statistical tests did you apply?What assumptions did you make?

  • 04:12

    GREG MARTIN [continued]: Highlight any methodological limitations,and describe how it is that you addressed them.And if it's appropriate, state that youacquired ethical approval for your study.Now let's talk about the results.Everything about your paper revolves around the results,and for that reason, I am going to suggestthat you write the results first, especially if you'reworking with co-authors.So you may have collaborators.

  • 04:33

    GREG MARTIN [continued]: You may be working with a statistician.You may have a supervising author.What you want to do is you want to draftthe results, and the tables, and the text of the results.You want to circulate that and youwant to get agreement on the resultsbefore you write the rest of the paper.And the reason is the results can change.In other words, people might suggestthat you include certain aspects of the studyor remove certain other aspects of the study.

  • 04:54

    GREG MARTIN [continued]: If you've written the entire paper,everything else has to change if the results change,because everything revolves around the results.So get that finalized.Get it agreed.Get it ironclad.Get that signed off before you write anything else.Now you start writing.You write in a logical manner, and make explicit referenceto any of the tables and figures included.You want to place the tables and figures

  • 05:15

    GREG MARTIN [continued]: as close as possible to the first timethat they're referred to in the text.Don't discuss the results.Don't provide explanations.Don't provide any speculation.Don't interpret them.Don't explain them.Don't hypothesize.Keep the results nice and factual.Now, don't ignore any negative findings.Include any negative results, but don't try to explain them--

  • 05:35

    GREG MARTIN [continued]: not at this point.That's for the discussion.The discussion-- this is by far my favorite part of any paper,because in the discussion, you're not onlygoing to reiterate your findings,but you're going to explain them, and interpret them,and hypothesize, and speculate.This is where you bring your thoughts and ideasinto the paper.You're going to start your discussion off

  • 05:56

    GREG MARTIN [continued]: by reiterating your research question,and given your findings, state what you believe the answerto that question is.Now, remember, don't introduce any new data or any new resultsin the discussion.The discussion has to reflect on the results that have alreadybeen presented in your paper.You need to explain and interpret your resultsin the context of the literature that you

  • 06:17

    GREG MARTIN [continued]: identified in the introduction and background.In other words, are your findingsconsistent with what other people had said,and does your data fill one of the gapsin knowledge that you identified?So you want to show how your work has added somethingto the body of knowledge and understanding on the topic.You also want to reflect on the implications of your findings,maybe in terms of practical applications or public policy.

  • 06:37

    GREG MARTIN [continued]: And of course, you want to acknowledge the limitationsof your study, and you want to discussthose limitations and the implicationsof those limitations to results.And the more you do that, the more youdemonstrate that you understand the limitations,the more credibility your results will have.And in that context, you also wantto consider and discuss any alternative explanationsfor your findings.Definitely discuss any negative findings,

  • 06:59

    GREG MARTIN [continued]: and consider the impact on your conclusions,and then talk about next steps whatare the implications of your findings in termsof suggestions for future research.And finally, there's the conclusion and recommendationsection.This is where you get to tell the reader, so what?What is the point of all of this?You want to help the reader understandwhy your research matters.

  • 07:19

    GREG MARTIN [continued]: So what you want to do is you wantto state the answer to the research questionthat you asked and state any recommendations thatcan be made as a result of your findings.[MUSIC PLAYING]


Greg Martin, Editor-in-Chief, Globalization and Health, reviews how to write a scientific or academic paper in public health, including a description of, and what belongs in, each section.

Looks like you do not have access to this content.

How to Write a Scientific Paper in Public Health

Greg Martin, Editor-in-Chief, Globalization and Health, reviews how to write a scientific or academic paper in public health, including a description of, and what belongs in, each section.

Copy and paste the following HTML into your website