About the club
We read published research and explain it to people who are not in the field. We also do our own analysis and publish it every quarter. This page is roughly what we would tell you if you walked into a meeting and asked what this actually is.
The weekly hour
We meet once a week for an hour and we read one paper. Nobody is expected to have read it beforehand. That is the part most reading groups get wrong: make attendance depend on homework and attendance is gone by October. The paper goes up on the projector and we work through it together, slowly, in roughly this order.
Whoever picked it gets sixty seconds to say why. They are not allowed to use the word interesting.
We look at every figure and read every axis label out loud before anyone reads a sentence of the text. Most papers can be half understood from the figures alone. The ones that cannot usually turn out to be hiding the weak part somewhere in the prose.
One person reads the methods section a paragraph at a time. Anyone may stop them at any word they do not know, and the reader is not permitted to be annoyed about it. This is where the hour usually overruns.
Everybody in the room says one thing they did not believe. It can be small. Saying you believed all of it is allowed exactly once per term.
Somebody who has not said much summarises the result in language a twelve year old would follow. If the room cannot manage that, we have not understood the paper, whatever we happen to think.
A show of hands. About one paper in four gets written up. The rest go on a shared list that we raid whenever somebody wants a project and has no idea what to do.
The paper currently in rotation sits at the top of the home page, with a note on why it is there.
What we publish
Everything on this site is one of two things. They are different jobs and they are held to different standards.
A field note takes a recent published result and rebuilds it for a reader who has never seen the field before. Usually it comes with a model you can push around in the browser, because a number you can move teaches more than a number you can only look at. The people who did the original work are named in the first paragraph, every time.
Quarterly research is the club's own work, collected into an issue at the end of each term. It is slower and it fails more often. Every number in it ships with the script that produced it, so a reader who doubts us can run the thing and find out exactly where we went wrong.
The two feed each other, which is why we keep both. Explaining someone else's paper is how you learn to read one, because you cannot paraphrase a method you have not actually followed. Doing our own analysis is how you learn why papers are written the way they are. Once you have tried to describe your own sample selection honestly, in writing, you stop being surprised that other people's descriptions are so vague.
Authorship
Every article here is signed The Science Journaling Club. This is deliberate, and it comes up every year.
A field note passes through four or five people before it publishes. Someone picks the paper. Someone drafts. Someone takes every number back to the source table and checks it. Someone redraws the figure that nobody could read. Someone reads the finished thing cold and marks the sentences that did not land. The person who wrote the first draft is very often not the person who found the error that changed the conclusion. Working out who deserved the top line of that would take longer than the article took.
There is a second reason and it is less noble. A club where the writing carries names becomes a scoreboard, and a scoreboard is a poor environment for volunteering to check somebody else's arithmetic. We would rather the unglamorous jobs kept getting done.
If you need a record of what you personally did, for an application or for anything else, ask and you will get a letter that describes it properly, with dates and specifics. That is what letters are for. The website is not that document and was never going to be.
Method
We have no benches and no instruments. There is also no budget that would change that. The club is a room and a pile of laptops.
We decided early not to pretend otherwise. A school experiment run once, on equipment nobody has calibrated, produces a number that cannot carry any weight at all, and putting error bars around it would make the situation worse rather than better. So our own work is analytical. In practice that means one of these.
Here is the argument for why that counts as real work rather than the thing you settle for. A great deal of what science needs at the moment is not another measurement. It is somebody sitting down with the measurements that already exist and asking whether they agree with each other. When we pooled those 23 acidification effect sizes, the pooled average turned out to be the least informative thing in the output. Heterogeneity came out at 98.8 percent, which says the experiments are not estimating the same quantity. That is a result, and no pipette anywhere produces it.
Analytical work also has one property that school laboratory work could never have. A stranger can check it. Every result on this site links the script that made it. If our reasoning is wrong you do not have to take our word about the equipment, because there was no equipment. You can open the code and find the line.
Joining
Come to a meeting. That is the process. There is no application and no interview, and we have never turned anybody away.
You do not need to be able to code. Close to half the club could not when they started. You do not need to have read a paper before either. For most people the first one is whatever is on the projector the day they walk in. If you arrive with no background at all, here is what your first month actually looks like.
Somewhere in there, if you want the analysis side, say so and you get paired with whoever is running the current study. We teach that by handing you an existing script and asking you to run it from scratch and break it. Nobody starts by writing one.
Honestly
Since this page is meant to help you decide whether to spend a year here, it should probably include the parts we would rather you found out later.
If that sounds like something you would rather be doing than not doing, the work is all here to look through first.