Jump to content

How to write scientific papers


Amr Mak

Recommended Posts

0) The standard method to put an idea on paper is using a pen or a pencil. I'll assume you were asking about a scientific publication.

 

1) Scientific papers are usually written about results, not about ideas.

 

2) The two most common programs to write papers are latex or Microsoft Word. I also use latex for posters (with the "beamerposter" package), but I imagine standard presentation software would also work well.

 

3) For structuring of the content, just look at similar publications by others.

Link to comment
Share on other sites

My advice (#3 of the above) is to read some papers in similar areas as yours. This will give you an idea of the style and content required. Of course you will develop your own style, but using a paper you like as a template is not a bad place to start. This is assuming you have content that people will want to read.

Link to comment
Share on other sites

Make sure to test your idea. In science, we want to see results.

 

So, go for a structure something along the lines of:

 

1. There is a problem in the world, and it is this: (whatever problem you're solving)

2. I have an idea how to solve that problem.

3. If my idea works, then I predict that this test will give this-and-this outcome.

4. Here are the results, and the outcome is indeed this-and-this.

5. Conclusion: my idea is good.

 

A test can use someone else's data. You don't necessarily need to do measurements yourself.

 

If you present only your idea, but no tests, it's just speculations.

Link to comment
Share on other sites

Make sure to test your idea. In science, we want to see results.

 

So, go for a structure something along the lines of:

 

1. There is a problem in the world, and it is this: (whatever problem you're solving)

2. I have an idea how to solve that problem.

3. If my idea works, then I predict that this test will give this-and-this outcome.

4. Here are the results, and the outcome is indeed this-and-this.

5. Conclusion: my idea is good.

 

A test can use someone else's data. You don't necessarily need to do measurements yourself.

 

If you present only your idea, but no tests, it's just speculations.

 

#5,

 

Is this necessary? I have no experience in writing papers but one would assume that your audience would be the ones to decide whether or not your idea is good; Aren't they [papers] supposed to be peer reviewed before publication and therefore a certain amount of criticism has already been made?

Link to comment
Share on other sites

#5,

 

Is this necessary? I have no experience in writing papers but one would assume that your audience would be the ones to decide whether or not your idea is good;

 

You have to have some confidence that your work is up to the standard and interesting to the community. In the long run it will be the community that decides. Though this may take time, or you could be working in something quite obscure and so get few citations.

 

Aren't they [papers] supposed to be peer reviewed before publication and therefore a certain amount of criticism has already been made?

 

You are talking about 1 to 3 referees for any given journal you submit to. The feedback they give you can vary a lot. Referees can miss things, or misunderstand your work or even worse they could have some objection to your work for whatever reason. You might them submit to some journal, get rejected and move on to another journal. Depending on how many times you do this, and assuming different referees every time you are only really talking about a handful of people.

 

In practice you may have shown the paper to other people, or in the case of physics and mathematics you may have placed a preprint on the arXiv. Again, the amount of feedback you get can vary lots. Even if you directly ask, you may get no reply, a simple "that is interesting", or if you are lucky lots of useful comments and suggestions.

Edited by ajb
Link to comment
Share on other sites

Writing an article is one thing - simply pick a journal appropriate for your study, open up the author guidelines and write your article accrodingly.

 

You are talking about 1 to 3 referees for any given journal you submit to. The feedback they give you can vary a lot. Referees can miss things, or misunderstand your work or even worse they could have some objection to your work for whatever reason. You might them submit to some journal, get rejected and move on to another journal. Depending on how many times you do this, and assuming different referees every time you are only really talking about a handful of people.

 

In practice you may have shown the paper to other people, or in the case of physics and mathematics you may have placed a preprint on the arXiv. Again, the amount of feedback you get can vary lots. Even if you directly ask, you may get no reply, a simple "that is interesting", or if you are lucky lots of useful comments and suggestions.

 

Agreed. Getting accepted is totally different. Most journals (at least in evolutionary biology - my field) get far more submissions than they are able to print. Your work will have to be of an appropriate topic and standard not to be summarily rejected by the editor. If the editor feels the content is potentially acceptable for the journal it'll go to review. If you aren't conducting a broadly interesting study in a topical area with current best practice methods, chances of getting past the editor in a high impact journal are low.

 

Review can be a crapshoot or it can substantially improve your study, regardless of acceptance/rejection. One thing I've learned is to reduce the amount of arm waving and speculation in my papers - people don't want half the discussion to be about things you could or want to do in the future - focus on what you study HAS found. I got reviews back yesterday for a paper. Reviewer 1 said: "This is a well crafted, easy to follow study." Reviewer 2 said : "The paper is poorly written, difficult to follow and full of spelling, punctuation and grammatical errors. It will need substantial rewriting to become acceptable for publication." Their analytical advice varies to the same extent, but in the end it is up to the editor to decide if you answered their comments well enough. In this case the article is up for major revisions, so will likely go back to the same reviewers for a second round of revision, so I need to adequately address their concerns rather than dismissing them.

 

Finally, for the last year or so, some of the journals have been "gaming" the system to reduce their published acceptance rates and the time papers are in review. I review for a journal which recently removed the "major revisions" option for reviewers in favor of "reject, encourage resubmission". That way they can state a shorter time for papers in review, state a lower acceptance rate and look more efficient and elite. It's supremely annoying for both reviewers and authors however.

Edited by Arete
Link to comment
Share on other sites

#5,

 

Is this necessary? I have no experience in writing papers but one would assume that your audience would be the ones to decide whether or not your idea is good; Aren't they [papers] supposed to be peer reviewed before publication and therefore a certain amount of criticism has already been made?

Well... you certainly need a conclusion. But maybe I should have said:

5. Conclusion: I tested my idea, and it works.

To claim it's "good" is perhaps subjective, and should indeed be left to your audience.

 

And yes, a reviewer can disagree with your conclusion, and say it does not work. Then you need to rewrite the paper, or even improve your experiment (in a worst case).

Edited by CaptainPanic
Link to comment
Share on other sites

All the above represent excellent counsel.

 

Journals are in business to provide high-quality professional communications for professional researchers on the cutting edge in their field. In order to be of value to the discipline and of interest to the intended audience articles must represent valid, significant, original contributions to the knowledge base of the discipline. That is a high standard. It is intended to be a high standard.

 

Valid: Your article must present results that are verifiable by others and have been verified by you according to the highest standards of your discipline. If the work is experimental you must present information that will convince the reader that the data is accurate and represents what you claim to represent to represent. If the work is theoretical it must contain full mathematical detail and reasoning.

 

Original: Your paper must present something previously unknown to the community. In other words it must be a discovery.

 

Significant: The results must be of real interest to the research community. That entails addressing issues that are recognized as being of importance by that community and that will be of use by others in producing new research in the future. Many papers are rejected for publication simply because they are not sufficiently significant to warrant publication -- journals have lots of submittals and limited pages.

 

The standards are a mirror of what is required for a PhD dissertation. So ask yourself if what you have to say would warrant the award of a PhD degree.

Link to comment
Share on other sites

The standards [for a scientific publication] are a mirror of what is required for a PhD dissertation.

That is an interesting statement for that a) it sounds very reasonable, and b) I have never considered that aspect myself, before.

Link to comment
Share on other sites

Finally, for the last year or so, some of the journals have been "gaming" the system to reduce their published acceptance rates and the time papers are in review. I review for a journal which recently removed the "major revisions" option for reviewers in favor of "reject, encourage resubmission". That way they can state a shorter time for papers in review, state a lower acceptance rate and look more efficient and elite. It's supremely annoying for both reviewers and authors however.

 

In the end it would not make much of a difference as currently journals are still judged by impact factors.

Link to comment
Share on other sites

In the end it would not make much of a difference as currently journals are still judged by impact factors.

 

Indeed. I don't really know what the game is and it's speculation on my behalf. I can't see any other reason to give reviewers nothing between "minor revisions" and "reject - resubmit" aside from the fact it will artificially increase the rejection rate and decrease the time a paper is in review.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.