Experience Sampling of My Stress
|pen and paper, memory belt
|Heart rate, Cardiovascular, Stress
Builds on project(s)
|Show and Tell Talk Infobox
|2011 QS Europe Conference
|This content was automatically imported. See here how to improve it if any information is missing or out outdated.
Experience Sampling of My Stress is a Show & Tell talk by Ulrich Atz that has been imported from the Quantified Self Show & Tell library.The talk was given on 2011/11/26 and is about Heart rate, Cardiovascular, and Stress.
Description[edit | edit source]
A description of this project as introduced by Quantified Self follows:
Ulrich Atz was curious about measuring his stress levels. He chose three methods to do this: experience sampling, day reconstruction method, and heart rate. In this video, he describes how he went about designing his experiment, how the different methods work and the challenges of each one, and what he learned. He was surprised to discover which method worked best.
Video and transcript[edit | edit source]
I have a background in measuring happiness, but I wanted to do something that can also be measured objectively. So in this tool I’m going to talk about one, how you can measure stress and I think there are two methods basically. First, it’s going to be more subjective and you can ask yourself how stressed am I on a scale of 1 to 10 or something. And the second thing is maybe you can measure it objectively. We have seen the heart rate we have seen some sort of brainwaves and there are more other ways like skin conductivity or stuff like that.
The second question is what is the ‘best’ way to do so and this can be more precise, or least intrusive or more comparable over time. So this is a little bit up to you. And the last thing is what can you actually take away if you want to measure your stress for example. I chose to compare and do three methods and they were done simultaneously over time. The first one is experience sampling which I’m going to explain in a minute, the second one is called the reconstruction method and it’s basically just a diary, but it’s actually a little bit more sophisticated. And the last thing is heart rate, and I chose heart rate for my objectives measure because if I had been at this conference year ago then the whole findings would probably affect me one day. But this way it had to fulfil a couple of requirements it had to be affordable and have access to raw data, and it was kind of difficult to find a sensor which would fulfil all requirements as a student. So, experience sampling is basically with the idea that you measure your experience in the moment, so not retrospectively, you don’t ask how did I feel yesterday or how stressed was I last week but actually in the moment. And this has become more popular only in recent years because we have smart phones and we carry these around most of the time. So I use to apps and one is available called the Mappiness, and it was more designed for happiness, but you can see the second slider is my measure for stress. This would prompt you at random times during the day to times, five times or 10 times or how often you want to be disturbed during the day. Obviously there are some drawbacks if you are out with good friends or having a good time with your partner, then it isn’t the best time and maybe you are declining to answer it. But in general the idea is that you can be asked at random times during the day, which should give you this gold standard of experience. The second method is the reconstruction method, and as I said it’s just a diary. You sit down at the end of the day and you get questions and this should construct a measure of stress over the day. The picture is a little bit misleading because I use Excel to write down. The last thing is the heart rate and I wear a belt which is across the chest which can record data up to 24 hours. So I would recorded over days and it would just run in the background which was the aim. This is a sample output for how the heart rate data looks like and I would select one hourly time episode because it is actually very difficult to go from the heart rate to actual stress levels. Because if you do sport or any other activity then you get a spike but it doesn’t necessarily mean that you are stressed. This was a challenge in itself in how to translate the heart rate data into an actual stress data which is reliable. The next slide we come to my analysis, and the idea is that all three methods actually measure the same thing. So we should see a trend line, and you can see I’m stressed by the experience method on the X axis, on the right it’s high levels and on the left it’s low levels and the same for the daily construction method; if the high level is low it’s low levels of stress, so we see this trend line because they should measure the same. An additional dimensional of the shape; a hollow circle would mean there was an objective treatment occasion and I mean obviously stress and treatment would not go well together in everyday language, but in an academic setting you would call it treatment. All of the treatment creations could be something like an exam or a job assessment, or a job interview. In the top right hand corner this told me that it is actually working quite well and they are measuring roughly the same. Here I am putting an academic disclaimer that I don’t have time to talk about all of the stuff that would go on in a study and how you would generalize, but just to get an idea in hot the output looks like in my case. The last thing is that I did a quantitative method to compare the three, and interestingly the dire reconstruction method worked best. Now I say interestingly and surprisingly, because the experience sampling, where you measure in the moment should actually be the gold standard and you would expect that want to measure it more accurately. My hypothesis is that if you focus on measuring stress and if you have a very good understanding, and it’s just one day after you actually experience it, then you remember it quite well and this memory bios is not distorting your measurement. Lastly, what can you learn if you want to do it and you are inclined to measure your stress? I think the thing I’ve learned is that simplicity is always something you should strive for, because I recorded all sorts of additional data and it’s just no use and you are making it more complicated. I would urge you not to go for this kind of objectives measurement even if it is very tempting, because it is just a wealth of data. Of the two months I had over 1.2 million heartbeats recorded, and it is just very difficult in terms of data management, and in terms of actually recording or finding out what the stress level is and not the measurement that you are recording because your skin conductivity level may measure something else, and it may not necessarily be stress. So just as a caveat be very careful when you are allured by this idea that runs in the background and then you find out something. Secondly, planning is very important. I would say 3 to 4 weeks is a good period which you should strive for and it should be a minimum, because it is just a very short snippet of your life and it can mean anything. So this is mine in just 3 to 4 weeks or 22 days. The best method; again, it depends on the type of personality and person. If you have a smart phone and you like technology and geeky, then go for the experience sampling with this Mappiness app, which you can download from the App Store it’s free. If you are more of a certain daily routine or if you already have a diary then use that one because it integrates very well. Another thing, just be a little bit careful as this comes up very often in academic settings, be a little bit careful when you try to change something. If you feel like using it for changing or reducing your stress, then measuring it can produce all sorts of events, so be a little bit careful. At the very last, keep it simple again.
Thank you very much.
About the presenter[edit | edit source]
Ulrich Atz gave this talk.