It sort of means what you mean my an experiment. Some experiments take weeks, months or years to complete. I’ve just done an experiment on 62 people. This took about 10 months. Each person’s involvement was only about 3 and a half hours (in 2 separate appointments), and I would often have 3, 4 or 5 appointments per day. Now my job is to look at all of that data, so I can run “experiments” on the data without needing the person here. Sometimes this will take a few seconds, so I run hundreds of them a day. Sometimes it will take several weeks for the computer to get all the calculations done!
Not very often now unfortunately as my students do all the hard work! When you do experiments they can be any length of time. I have experiemtns that take ten minutes (although it takes seven days preparation before and two day analysis afterwards. Some experiments can last a life time. Experiments are part of what we do but reading and learning and analysis is also important. Some days can be very long but we try and design things to make them as efficient as possible. Our work is quite hands on so we have to be here but other experiments can be done remotely.
A typical week would have preparation on Monday, then experiments on Tuesday, Wednesday and Friday, with a bit of preparation for the following week thrown in. Once you start it can be pretty much full on and when we are doing our long term experiments, we add week ends, late nights and early morning too. Experiments are fun though (when they work, because they dont always) and can be frustarting especially when they cost a lot of time and money to do. The first time you see a new result though can be really exciting and all the hard work pays off. You might be the first person in the whole world to observe this new phenomenon, thats pretty cool.
So if in terms of data collection, for the project that I am working on at the moment we collect data 2 times per week and each session lasts about 4.5 hours. But actually there is a lot more to the experiments than just collecting the data, you also have to analyse it. That take a lot more of my time and I probably spend a couple of days out of my week thinking of new ways to analyse the data, writing the code to make it happen and looking at the results.