One of the accidental biases I think is introduced to people by their schooling is the idea that all problems have solutions. This is not a deliberate bias or anything, but an accidental one introduced by the fact that the entire process of being educated is basically being walked down a cultivated garden trail in which you are presented with a series of carefully-designed problems where the answers are intended to help your understanding of some topic. But in the real world, problems don't always have answers. If you can't get the data to answer a question, the correct answer is to not answer the question. If you can't gather enough data to separate the error bars, then you don't have an answer to the question. All you can say is that the effect, if it exists, isn't large, and you're not too sure about the effect. If what you have is not sufficient to make a call... stop.

Once you see this clearly, the idea that one can be literally unable to gather enough data, but somehow one is obligated to just create an answer even so is just... weird.

And one of the major criticisms of the current scientific system is that you can't publish a paper that says "We looked for X and didn't find it", so there's massive incentive to just force an answer out. And when you force an answer out, whereever the answer came from... it isn't the data. It's something else. Varies depending on the topic, of course, but it's certainly not the data.

(By contrast, in life we must frequently create answers to questions we don't have enough data to scientifically answer the question. That's the nature of the ride we're on. But it is improper to force answers in the scientific domain.)