25 May 2018

How to read a scientific paper 4.2

This artist’s impression shows how the Milky Way galaxy would look seen from almost edge on and from a very different perspective than we get from the Earth. The central bulge shows up as a peanut shaped glowing ball of stars and the spiral arms and their associated dust clouds form a narrow band.

This post is a continuation of the last step in reading the papers, reading the whole paper. I chose an article about Discovery of Gamma-Ray Emission from the X shaped Bulge of the Milky Way. You can find a full article for free at arxiv.org. And if you feeling posh, you can also buy the version that was published in Nature.
Here is a ‘triage’ average grad student uses when reading scientific papers, and I follow that ‘triage’ here. Links will lead you to the posts that covered previous steps.

1. Read an abstract.
2. Look at the images and read captions
3. Read conclusion/summary
4 Read the paper in details. part one, part two

Read the paper in details - continuation

Ninth paragraph basically says that when the new model is applied the detection level of the proposed dark matter fell suspiciously low below the set threshold level.
For you who do not get why is this important, here is a short explanation of why there is a threshold in the first place.
A good scientist tends to establish threshold before data analysis, so that he can prevent the ubiquitous need of all human, including himself, to cherry pick stuff that will confirm his held beliefs.
A typical experiment is riddled with the errors. There are errors introduced by equipment, by technique applied during an experiment, human error, and analysis error (to mention the few most common ones). Sometimes errors crop out from an unexpected source, like in my case, local construction site introducing power variation in a power supply. The bottom line is there is no way any scientist can predict is what kind of error and where it will appear.
That’s why the scientific method has techniques developed over the centuries how to deal with those unexpected errors and still get to the truth. The threshold is one of such techniques. Before a scientist starts data analysis, he has to determine what level or errors he will allow in his dataset. Only detections that go over that threshold can be considered a real signal.
This sometimes means underestimating the scope of the detected signal, but at the same time scientist makes sure that the signal above the threshold is the one scientist search and not a product of a whiny colleague jumping up and down next to the sensitive instrument to warm himself. (Yep, that happened to me, one of the colleagues found lab too cold and decided to warm himself by jumping, causing the whole lab floor to oscillate and instruments to detect those oscillations. Mildly put, I was not very happy about it.)
Anyway, the main result is stated in this paragraph. The signal that might be dark matter is almost 4 times below the threshold and thus cannot be declared a dark matter. The possibility that the signal is caused by something else is too great to be ignored.

Tenth paragraph - the last

I have to admit, I don't really get this particular paragraph. I know that it suppose to offer an explanation of why the authors consider that the observed signal is not dark matter. But, I do not have enough back-knowledge to follow the explanation. There are few references in the paragraph that might offer an explanation, but I hope appendix will spare me that extra time necessary to dig through the references. We’ll see.


This is where all people who contributed to the work with less than 20% of the total work are mentioned. If someone contributed 20% or more they are put as coauthors on the paper.
One of my colleagues shared the tip with me about whose name to put here. Basically, you put people who deserve prize but also the ones that you would rather not have as referees of your paper.


Unless I need additional explanation, I usually skip references when reading the paper. But the fact that they exist is very, very important. References in scientific paper serve not only to place the work in its correct place in the body of the scientific knowledge but also to prove that scientist behind the published results is not someone who pulls the ideas and conclusions out of his behind.
Every statement in paper needs to be backed up by some kind of evidence, either by connecting the statement to previous evidence by reference or offering a detailed explanation of evidence behind the statement.
That’s why scientists sound harsh when talking with ordinary people about popular topics. We are used to offering evidence for any statement or admit that we do not know for sure. I’m certain that you guys know how rare is such behavior in ordinary, everyday life.
In short, references in a scientific paper are very important. The quality of references directly contributes to the quality of the presented work.


Remember how I said at the beginning that Nature papers are short, super short. Well, that’s because the true meat of the papers ends up in an appendix. That’s why most of the scientists will cite Nature paper, but not really use it for their own research because the paper itself does not contain enough explanations.

Appendix - observations

And authors of this paper did the same, they put in appendix everything that is usually put in the ordinary scientific paper. Starting with the very detailed explanation of what data they used for their work, stating the characteristic of the data, which part of the data they used, and what did they do on this level to avoid the errors. In their case, the authors described what did they do to eliminate Gamma rays from the Earth sources.
This again proves the authors are good, and the work is good. There is complete transparency about what kind of data is used, and that speaks volumes to anyone who does science. It shows what kind of scientists are behind those names. And the fact that the careful consideration to eliminate errors started in this phase of scientific work demonstrates that scientists behind are someone who will take care of every detail and ensure that they did their best to get to the truth.



No comments:

Post a Comment