This post is part of the series about publishing your first paper. Now that you’re hopefully on board with the idea of why it’s a good idea to publish your PhD work, you get to decide what to publish from all the work you’ve been doing.
Creating a clear message
So you and your supervisor think you have something interesting which is worth publishing? Great!
Just like with presentations, it’s important to create a clear message. Therefore, the first step is defining what you want the key message of the paper to be.
What is the most important finding from your work?
Thinking about this question will give you a clear picture of what you want the publication to focus on and the rest of the paper can revolve around this theme.
It is inevitable that during your experiments there will have been complications: certain things won’t have worked, equipment may not have been available or you might have needed to change your setup. Even with the best of intentions, the path to achieving publishable results is rarely straightforward. For most of us our experiments likely look something like this:
It’s really important to remember that you don’t need to include every experiment, just the ones which help to create a coherent story for the message you’re making.
That’s not to say that you can’t include things which went wrong. You certainly can and should, especially if they help to show the significance of your main results. But what you should try and avoid is unnecessarily confusing the reader with extra experiments you did on the side which don’t really add to the key message.
Bias in only publishing positive results is a problem yet including all your results isn’t the answer.
Key content
Remember that you’ll need to have enough content to convince the editor and reviewers that your work is worth publishing. And there are two main parts to this:
- Present something new
- Back it up with solid foundations
Presenting something new
Unless it’s a review paper you’re writing, the first thing you want to do is ensure you are presenting something new.
How do you know if you have something new and worth publishing? Well for starters scan the literature and discuss how your findings sit within the field with your supervisor.
To publish you don’t need to make an earth-shattering discovery, but you do need to advance the field.
In addition to creating or finding something completely novel, you could for example advance the field by:
- Using a new technique to analyse existing data
- Systematically compare testing methods
- Improving understanding of current mechanisms or approaches
For the first paper of my PhD work we used a technique which had been developed for very expensive and specialised synchrotron facilities and applied it to lab equipment in combination to trying out a variety of sample preparation techniques from previous studies.
The foundations for a strong paper
Once you think that you have something worth publishing, you need to ensure that the framework is there for your work to be taken seriously and not rejected for glaring oversights or omissions.
Without solid foundations, your paper collapses.
The content of the foundations vary depending on your research area. In general you could categorise them as:
Significance: Can your study help progress the field?
- Can it be widely applied or does it only work under very limited conditions?
- Factors that may come into play include: how representative your testing conditions are, economic viability, how easily it could be implemented.
- You should write your discussion carefully to both highlight what is good about your work but also truthfully acknowledge limitations.
For my PhD work, clinical relevance was a consideration. How could the project actually help people? An obvious improvement to the paper I published would have been to use human tissue rather than cow.
Validation: Can I trust the results?
- When you’re presenting something new, reviewers (and readers) want to be sure the work is rigorous.
- If you’re showing a new technique, can you prove you get the same results with a more established technique?
- Consider including a “sanity check” of some kind to show that the system reacts as would be expected. This could be by presenting testing results under several conditions or by.
For my paper validation came in two parts:
- Including a control measurement to give context to the values we presented;
- Using an established technique (histology) and directly comparing between the two which would help convince readers of our results:
Validation also ties in closely with both Statistics and Comparison.
Statistics: Was it a fluke?
- You need to demonstrate that the work really does show what you’re saying it does.
- To get an idea of how much testing you need to do, look back at previous studies and check their sample size.
- You may need to include a larger sample size than you have already tested but this depends a lot on the technique. After that you’ll ideally want to do some statistical testing to ensure what you’re claiming holds true.
One of the main comments we had to address by reviewers for my paper was sample size. They simply didn’t like that we had tested so few samples. We addressed the comment by both including more data and revising the text to highlight the limitations and the need for even more testing.
Comparison: How does it compare to previous studies?
- You want to be able to show that your study progresses the field, and this includes showing performance against other work. Ideally you’ll set up the study from the start to be able to compare your results to previous literature.
- If you can test your samples/technique using exactly the same protocol as previous papers, it makes it much easier to convince reviewers.
- The alternative approach to allow comparison is to test other samples/techniques yourself in your study.
In my paper we compared our findings to other papers in the discussion.
If you address each of these, you’re in good standing. I should point out: it’s not that you won’t be able to publish your work if it fails one of these, you just may not be able to publish it in what may be your first choice journal.
Some of these questions can be answered in the text of your paper with some good justification and referencing, others will require the right experiments to be carried out if you are to satisfy the reviewers.
In summary, start by thinking about what you want the main message to be and work back from that, including experiments which strengthen this message. You may think of extra experiments which you could carry out but should you spend your time on them? This brings us on to the next point: deciding when you’re ready to publish.
I hope this series is helpful to you if you’re starting on your journal of publishing. Is there anything that you think I’ve missed?
Be sure to subscribe to stay up to date with this series and other science content:
