Question: What experiment have you done that didn't came out the way you wanted to?

  1. I did a very lengthy (several years) experiment when I was getting my Ph.D, and the point was to look for small subatomic particles (called neutrinos) that were predicted to come from certain types of stars that explode very violently (called gamma-ray bursts). These bursts put out as much energy in about 10 seconds as the sun will in its entire 10 billion year lifetime. We saw plenty of light from these bursts, but in the end I saw exactly zero neutrinos. That was not the result I was hoping for, but that doesn’t mean the experiment was a failure! In fact, because many scientists predicted that we *would* see neutrinos, in the end we could determine that their theories were wrong, and that we would have to come up with new ideas for what made gamma-ray bursts.


  2. @laravdw, I laughed and nearly cried when I read your question. It seems to me that almost every experiment I do turns out the wrong way. Either it just doesn’t work or something unexpected happens. When its unexpected, then I laugh. Those moments can be very important. When a scientist does research, they investigate something that hasn’t been done before or isn’t very well understood. It is expected that things go wrong, it comes with the job. If things always go right, you may not be learning anything new.


  3. I’ve tried reactions to make a new chemical that haven’t worked but I never gave up and got things done in a different way. I guess I’ve been lucky but I also know that not everything goes to plan and experiments can be unpredictable.


  4. @laravdw: Great question! Highlights the need for scientists to remain objective (without bias or judgement about how something SHOULD go). We always need to be open to the fact our theories might be wrong. Be open to taking on board new information that might be unexpected from our experiments, thinking about why that might be (e.g. was there an error in the method, maybe the study design didn’t ask the question properly or is the underlying theory wrong?), then letting that information shape what we do next – which might be to ask a new question or the old one in a different way.

    There have been plenty experiments that didn’t go the way I thought they would – but they’ve all taught me something valuable. About paying more attention to detail (measure things carefully!), designing a research question thoroughly or just being open to being wrong in what I thought I knew.

    Recently, as part of my PhD I asked a range of people about their attitudes and perceptions of dog welfare and the perceived importance of different management practices (like exercise, fresh water, access to toys, etc.). I thought I’d probably see a big difference in attitudes between people who worked as dog trainers, people who cared for the dogs in kennels and the general public. Instead, I found they weren’t so different – but across all those groups, males and females showed differences in their attitudes. That was interesting!


  5. One of the most embarrassing was when I tried to show a class of children how you can make plastic from milk or cream (it used to be used to make buttons; you just boil it up with vinegar and the casein separates out as a plastic goo which will harden as it dries). I thought cream would work better, so I bought a carton of thickened cream, and it did not work at all.
    The truth is that if you know what the result will be, it is not really an experiment. Scientists often have to try several things before something works out, and sometimes learn more from the ones that fail than the ones that work.