Most of my work is “exploration science”, so there aren’t that many experiments involved. We sometimes put different materials on the ocean floor and leave them there to see what species colonise them, and that goes wrong perhaps one in four times – usually (we think) because something big and unknown comes along and carries off the whole experiment!
I’m not sure, I’ve done a lot of experiments over the years I’ve been in the lab, and I’ve never really counted up the ones that have gone wrong, or not worked. I think it would be pretty depressing!
Thankfully many more work than those that don’t, but with my embryos, I know that not all of them are not going to work, as it’s only successful about 14% of the time. That might seem pretty low, but when I started on this 4 years ago, it was only 5% of the time, so I am getting it to work much more. I will say that I don’t make as many mistakes as I did when i was first working in the lab, so not counting my embryos, if I was to do 3 experiments a day, 5 days a week, thats 15 expts a week, on average, maybe one or two a month won’t work, so that’s not that bad I think!
Probably more than went right! Usually with either experiments or the kinds of calculations that I do, it takes several goes to get everything working as it should be and to get some useful results. Nothing works first time!