5

There are a couple of studies in which I don't agree with the analysis methods but the authors have graciously made the data available.

I'm considering working on some re-analyses of these studies using this publicly-available data. Broadly the structure of these papers is to

  1. discuss issues in the original analysis,
  2. propose statistical and causal analysis methods that remedy those issues,
  3. run the analysis,
  4. compare and contrast to the original analysis and
  5. discuss the new implications on domain knowledge due to the improved analysis (if any).

Some of the titles are already quite long, so prepending something like "Re-Analysis of "<Original_Title>"" seems horribly-verbose. Indeed, any naming method that prepends something to the original title is going to have this problem.

How should I go about titling papers that are essentially a direct response to the methods of a particular study?

Improve this question

2 Answers 2

Reset to default
7
+50

Two years ago, I published a paper that was largely a re-analysis of some of my own prior data, plus some additional data not previously published as we tried to understand the conflicts between the other data sets. The title didn't emerge until late in the process, but ultimately focused on the comparison that we performed: "Comparative analysis of three studies measuring fluorescence from engineered bacterial genetic constructs"

I would likewise advise you to not worry about the title until you know what you have actually learned from your re-analysis:

  • If you learn something novel and interesting, that will be the focus of your title, e.g., "Circadian patterns in mosquito flight trajectories." The fact that it came from somebody else's data is a mere detail for explanation in your background and methods sections.

  • If you learn nothing interesting beside the fact that original work is flawed, you might not be publishing on your own, but instead be writing a "comment" article that urges a correction or retraction.

  • If it turns out the original article was sloppy in a way that turns out not to be very interesting or to significant affect its conclusions, you might not find it worth publishing anything at all, unless it's a pattern in the field. In that case, your title will be about the pattern, not the individual papers, e.g., "Bayesian statistics produce misleading conclusions about mosquito flight trajectories."

Note that in all cases, the title depends on the results that you get, so you shouldn't worry about it until you finish your analysis. Once you have learned from your analysis, your title should tell your reader about what you have learned.

Improve this answer
1

Depending on the extent and type of analysis you intend to perform, you could consider the term "secondary data analysis". In the medical field it is not uncommon.

Improve this answer

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.