• Question: has an experiment ever gone really wrong

    Asked by deny520dose on 16 Oct 2024.
    • Photo: Zoe Vance

      Zoe Vance answered on 16 Oct 2024:

      I’m lucky because I’m a purely computational biologist; can’t poison myself or blow anything up even if something goes super wrong with what I’m doing! So I don’t know if I have an example of an experiment going really wrong in the way you’re probably thinking. But even in this line of work we have to be pretty careful and double check everything because our results/results from the software we write are sometimes used to make important decisions, or even just used to inform work other people are going to do. If there’s a major mistake in there that causes incorrect results it could lead to a lot of wasted effort and materials, or worst case scenario, to a wrong or late diagnosis of a patient.

    • Photo: Santosh Mahabala

      Santosh Mahabala answered on 13 Nov 2024:

      Experiments are bound to give outcomes that you don’t expect. So, every experiment may have two (or sometimes multiple) outcomes. The beauty of science is being unpredictable.
      However, we always check why it went wrong and try to understand how to fix it if it needs to.

    • Photo: Stephanie Foster

      Stephanie Foster answered on 18 Nov 2024:

      I would also add that sometimes an experiment going wrong could actually lead to an exciting discovery, maybe you make a new compound that you didn’t expect that has some interesting properties or discover that a drug you were testing has an unintended side effect that is actually of benefit to people. Always keep an open mind.

    • Photo: Martin Pike

      Martin Pike answered on 10 Mar 2025:

      A few of the early (1940-50s) experiments with ‘atomic piles’ went wrong as people did not appreciate some of the complexities of critical mass, reaction rates and moderators. Some accidents were downright foolish. Having said that, nuclear reactors and experiments have killed very few people, compared to coal-fired power stations for example.
      With many things that are new, there often are aspects or effects that had not been considered beforehand, so we often learn from ‘failed’ experiments. I did an entire BSc project using (immunohistochemical) methods that I demonstrated did not work…..but would have been great if they had. Sometimes you just have to try things out.

    • Photo: Philippa Harding

      Philippa Harding answered on 11 Mar 2025:

      I do experiments all the time, and lots of them don’t go how I plan! As a scientist I am often doing something nobody else has ever tried before, so sometimes we don’t know what will happen till we try. Usually it just means I don’t get the result I expect, fortunately it has never been a disaster though!

Comments