How to quantify productivity and covariates for future analysis and optimization?

Question

I'm interested in quantifying various event variables such as caffeine intake, other dietary supplements, hobby activity, emotional state, what time I woke up, what was the weather like, and so on, so that I can predict and control procrastination periods and maximize productivity. Therefore, I need to track these variables (including productivity and procrastination). What would be a good way of logging these variables and events with minimal effort, and without introducing much bias in the measurements? I'm pretty good at recognizing events like beginning of procrastination when it happens. I plan to use statistical tools to analyze them later.

I can carry around paper and pencil, and I own an Android phone.

Answer 1

For the subjective variables like productivity and mood - as I recall from some unrelated psychology research, a good way of getting an unbiased reading of various variables throughout the day from one person is to have a small device (Android phone?) set up to ping at random times (however many times you want - once or twice a day should be fine for you) and prompt them to fill out a quick survey about the variables ("how productive do you feel you've been for the last few hours, on a scale of 1 to 10?", "how happy are you, on a scale of 1 to 10?" etc, for whatever subjective variables you want to track).

I'm not sure how one would set that up on an Android phone, though. You could just write a quick script to do it on a laptop and make sure it goes off during the times you're normally on the computer, and give it a link to an online survey you set up for yourself?

Answer 2

I've tried a few approaches - maybe they'll give you ideas…

For a while I did a regular analsis of my browsing history to show how much time I was wasting… all it took was firefox and a very small bit of excel fu and you could do this sort of thing… (apologies for lack of tabs)

>     01/03/2011    12/03/2011  22/03/2011  27/03/2011
>     Youtbue   15  86  41  48
>     Facebook  572 1230    454 244
>     bbc   82  30  34  63
>     iplayer   7   9   4   9
>     surfthechannel    0   8   0   24
>     wikipedia 171 604 127 220
>     TV Tropes 14  29  0   0
>     Imdb  2   8   2   0

Also a running total of unread g-reader items is a pretty good clue of how bored I am - this was a quick bit of python that gave out files like this:

2012-01-21 05:00:02 1327122002.99 55
2012-01-22 05:00:02 1327208402.79 50
2012-01-23 05:00:02 1327294802.05 93
2012-01-24 05:00:02 1327381202.94 414
2012-01-25 05:00:03 1327467604.0 739
2012-01-26 05:00:03 1327554003.52 1000
2012-01-27 05:00:03 1327640403.19 886

when I was quite busy and files like

2011-01-03 09:00:02 1294045202.3 0
2011-01-03 09:01:09 1294045269.72 0
2011-01-03 09:10:02 1294045802.8 1
2011-01-03 09:30:02 1294047002.54 2
2011-01-03 09:31:13 1294047073.46 2
2011-01-03 09:50:03 1294048203.02 7
2011-01-03 10:00:01 1294048801.78 1
2011-01-03 10:01:11 1294048871.22 1
2011-01-03 10:30:02 1294050602.09 2
2011-01-03 10:31:16 1294050676.94 2

When I was procrastinating…

I've touched on my productivity graph in other posts - Gamification to improve myself for an example...

and someone is bound to mention http://www.rescuetime.com/ very soon so let's bring that up right away…

Lastly I've been looking a lot at my iPhone's log files for this sort of tracking of things - but even with a jailbreak and a general familiarity with the innards of the device I've not been able to get very far with the debuging logfiles :( you might have a lot more luck with android...