| ▲ | jraph a day ago | ||||||||||||||||
> There isn't room for everything I get your point, but this is not specific to null results. > It only takes an individual No no no. The desirability of null results need to be recognized and somewhat consensual, and high impact journals and conferences needs to accept them. Otherwise, there's no reason researchers will work to publish them. 1. I don't publish anymore: I'm not a researcher anymore. I didn't encounter the case during the short time I was one (I could have, though. Now I know, years later. I suspect it would have been difficult to convince my advisors to do it). I hope this doesn't matter for my points to stand on their own. Note that I think null results ARE NOT failed research. This is key. 2. Ideally, null or positive result alike, the experiments and the studies need to be solid and convincing enough. Like, there needs to be enough salt and not too much, the dinner needs to be tasty in both cases. If the dinner doesn't taste good, of course you don't publish it. There is something wrong with what you've done (the protocol was not well followed, there's statistical bias, not enough data points, I don't know) It feels like we are talking past each others, you are thinking I'm talking about failed research, but I'm talking about a hypothesis you believed could be true, you built an experiment to test it, and found no correlation in the end. This result is interesting and should be published, it's not failed research. As it happens, I attended a PhD defense less than a month ago where the thesis lead to null results… The student was able to publish, these null results felt somewhat surprising and counter intuitive, so it's not like it's impossible, it just needs to be widely seen as not failed research. | |||||||||||||||||
| ▲ | 9rx a day ago | parent [-] | ||||||||||||||||
> The desirability of null results need to be recognized and somewhat consensual If it is interesting you should also find it interesting when you read it 30 years in the future. You don't need other people. It's a nice feeling when other people want to look at what you are doing, sure, but don't put the cart before the horse here. Publish first and prove to others that there is something of value there. They are not going to magically see the value beforehand. That is not how the human typically functions. It's not like you have to invent the printing press to do it. Putting your work up on a website for the entire world to see is easy peasy. Just do it! > Ideally, null or positive result alike, the experiments and the studies need to be solid and convincing enough. No need to let perfect become the enemy of good. Publishing your haphazard salting experiment isn't apt to be terribly convincing, but it gets you into the habit of publishing. Eventually you'll come around to something that actually is interesting and convincing. It's telling if someone isn't willing to do this. > The student was able to publish, these null results felt somewhat surprising and counter intuitive, so it's not like it's impossible Exactly. Anything worthy of the major leagues will have no trouble getting formally published. But not everything is. And that's okay. You can still publish it yourself. If you want to play baseball, there is no need to wait around for the MLB to call, so to speak... Just do it! > you are thinking I'm talking about failed research [...] it just needs to be widely seen as not failed research. Yes, I am talking about what is widely seen as failed research. It may not actually be failed research in a practical sense, but the moniker is still apt, especially given that you even call it that yourself. I guess I don't understand what you are trying to say here. | |||||||||||||||||
| |||||||||||||||||