Hacker News Reader

Comment by mhh__ | Hacker News Reader

> In the age when the scientific papers quality is falling

[Citation needed]?

Most academic researchers (https://www.nature.com/articles/533452a) believe that we're facing a phenomenon called the "reproducibility crisis" or "replication crisis", where experimental papers regularly get published even though the experiment does not work as described. This leads to more general doubts about the quality of papers in general - if you know you can't trust that experiments will reproduce, should you trust that survey data is accurate or that analyses are correct?

That isn't the point I was making. The replication crisis is a real problem, but what I am objecting to is the far more serious allegation that overall the quality of academic research is dropping.

Science is very very difficult, a journal will try to only publish things it thinks are true however papers (especially in the more informal channels some journals have/less prestigious journals) which are merely interesting or could-be-right are published all the time. That mostly isn't a bad thing - something has to be published properly to be considered properly and fleshed out, dead ends are still ends.

Common knowledge does mot need citations. Am an academic.

It's not though. And for a topic such as this where people in academia are going to be obviously biased, citation is all the more important.

It's also common knowledge that even comparing the academic workings of two sub fields is non-trivial

But reproducibility is essential for science, and it has been discussed on HN articles frequently in the past few years that this is very difficult nowadays. Let alone the maligned incentives for funding academia (you are biased to have “big breakthroughs” that catch headlines so you can get more funding, rather than getting out the truth), etc.

Just take a look at this search query for “academia” on HN:

https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...

HN is full of armchair polemicists and contrarians (on academia I am not one of those, but am for other matters), so I wouldn't consider that an unbiased sample.

> in the past few years that this is very difficult nowadays.

I strongly disagree. It has always been as difficult as today (if not more), it's just that these days were aware of it.

Am an academic too. I don't find this to be common knowledge, but a common misconception. Would like to see a citation

Ha. Try to publish that.

I'd bet you a pint you could publish it without any real issue as long as you weren't glib or facetious in your presentation.

Reply on news.ycombinator.com