Analysis Paralysis

Whoopsing through the data

0 notes

Gay Men Like Craigslist Personals
Wow, long time since I had an update on this thing.  Mostly I’ve been busy with job searching and creating an actual website/portfolio.
In the meantime though, I’ve also started working away at a small project for a friend of mine who’s interested in writing an article on dating in Vancouver. 
My first stop was to grab some info off of craigslist.  I know, I know: how much of craigslist is actually about dating and how much is just looking for casual sex? Then again, how many relationships (maybe gay male ones in particular?) themselves begin as casual sex encounters?  A hard metric to get accurate data on, I’m guessing.  Chances we could get a grant for field testing/surveys?
Well anyways, I had a pretty decent scraping setup already, so I figured, what the hell, why not just see what we can see.
I’m thinking maybe this set up might apply well to the major sites out there.  Ones that are more-or-less strictly categorized into casual sex vs. relationship seekers.

Gay Men Like Craigslist Personals

Wow, long time since I had an update on this thing.  Mostly I’ve been busy with job searching and creating an actual website/portfolio.

In the meantime though, I’ve also started working away at a small project for a friend of mine who’s interested in writing an article on dating in Vancouver. 

My first stop was to grab some info off of craigslist.  I know, I know: how much of craigslist is actually about dating and how much is just looking for casual sex? Then again, how many relationships (maybe gay male ones in particular?) themselves begin as casual sex encounters?  A hard metric to get accurate data on, I’m guessing.  Chances we could get a grant for field testing/surveys?

Well anyways, I had a pretty decent scraping setup already, so I figured, what the hell, why not just see what we can see.

I’m thinking maybe this set up might apply well to the major sites out there.  Ones that are more-or-less strictly categorized into casual sex vs. relationship seekers.

208 notes

Holiday Entitlements Compared
theeconomist:

Daily chart: which country’s workers have the most holiday? North Americans get less holiday than South Americans; Asians work harder than Europeans. Among the feckless workers from the old continent, those in the troubled  economies of Greece, Spain and Portugal have among the most generous  holiday allowances.

Well, we’re a pretty crappy outlier there, aren’t we.

Holiday Entitlements Compared

theeconomist:

Daily chart: which country’s workers have the most holiday? North Americans get less holiday than South Americans; Asians work harder than Europeans. Among the feckless workers from the old continent, those in the troubled economies of Greece, Spain and Portugal have among the most generous holiday allowances.

Well, we’re a pretty crappy outlier there, aren’t we.

0 notes

Working on a new x-code/iphone app inspired by my decision to quit smoking a few days ago.  
Pro: Came across this nicely designed graph & the related non-linear relationship between life expectancy and quitting smoking.  For one, the benefits of quitting accrue as time goes on, rather than right away.  Secondly, as one study puts it:
“The years of life saved from smoking cessation accumulate at the end of life, when people are more subject to chronic and debilitating diseases. For this reason, the impact of smoking cessation on mortality adjusted for quality of life (i.e., quality-adjusted life-year) often is used as an outcome.”
Just how does a researcher decide how much less quality of life exists during our later years?  Is it an average of self-reported values? My guess (from all that time I spend observing old people) would be two peaks in the data: a large group that feels their quality of life is much worse and a big group that feels their quality of life hasn’t changed all that much.
Anyways, for pretty obvious reasons, these kind of comparisons seem a little dubious (and more than a little chilling; e.g imagine applying them to other cleavages, like class).
Con: Finding it impossible to concentrate on program logic. Or pretty much anything at all for that matter.  ugh.

Working on a new x-code/iphone app inspired by my decision to quit smoking a few days ago.  

Pro: Came across this nicely designed graph & the related non-linear relationship between life expectancy and quitting smoking.  For one, the benefits of quitting accrue as time goes on, rather than right away.  Secondly, as one study puts it:

“The years of life saved from smoking cessation accumulate at the end of life, when people are more subject to chronic and debilitating diseases. For this reason, the impact of smoking cessation on mortality adjusted for quality of life (i.e., quality-adjusted life-year) often is used as an outcome.”

Just how does a researcher decide how much less quality of life exists during our later years?  Is it an average of self-reported values? My guess (from all that time I spend observing old people) would be two peaks in the data: a large group that feels their quality of life is much worse and a big group that feels their quality of life hasn’t changed all that much.

Anyways, for pretty obvious reasons, these kind of comparisons seem a little dubious (and more than a little chilling; e.g imagine applying them to other cleavages, like class).

Con: Finding it impossible to concentrate on program logic. Or pretty much anything at all for that matter.  ugh.

0 notes

Although I’m highly sympathetic to the cause, this is possibly the most useless infographic I have come across in a while.  See the whole travesty here.

Although I’m highly sympathetic to the cause, this is possibly the most useless infographic I have come across in a while.  See the whole travesty here.

1 note

The Global Occupy Movement Visualized
I managed to get my hands on some time-series data on the occupy movement from the Guardian and put together this animated interactive map of the movement’s growth during the first 33 days.
All in all, I’m pretty pleased with how this one turned out.  It could use with a few minor changes (a max and min panning, for example).  There’s also the question of data accuracy, especially with regard to the number of protestors actually attending.  
In the aggregate though, I think it does a pretty decent job of showing the movement’s slow inception followed by its sudden, explosive, and global growth.
The most irritating parts of the process were data cleanup and bashing through actionscript, while the most satisfying was probably the small python script I made to interface with google maps and pull latitude and longitude coordinates—simple, fast, and worked right off the bat.
Anyways, check out the full interactive here. 


*Bugs/Things to work on:
- panning limits- adding second, user-sub csv & processing- youtube version?- double click to center the map and zoom on location- what’s wrong with nelson? - fort nelson- when click to pan the slippy map, should turn into a hand cursor- geocoding API: should look at all results with more than one location- change the color of the text in the buttons when zoomed all the way in or out- speed selector?- sounds for every instance?- timezone value for lat.lon: to make bubbles appear from: left->right

moar ideaz:
- as an extensions: media reports of police vs protestors 

The Global Occupy Movement Visualized

I managed to get my hands on some time-series data on the occupy movement from the Guardian and put together this animated interactive map of the movement’s growth during the first 33 days.

All in all, I’m pretty pleased with how this one turned out.  It could use with a few minor changes (a max and min panning, for example).  There’s also the question of data accuracy, especially with regard to the number of protestors actually attending.  

In the aggregate though, I think it does a pretty decent job of showing the movement’s slow inception followed by its sudden, explosive, and global growth.

The most irritating parts of the process were data cleanup and bashing through actionscript, while the most satisfying was probably the small python script I made to interface with google maps and pull latitude and longitude coordinates—simple, fast, and worked right off the bat.

Anyways, check out the full interactive here. 

*Bugs/Things to work on:

- panning limits
- adding second, user-sub csv & processing
- youtube version?
- double click to center the map and zoom on location
- what’s wrong with nelson? - fort nelson
- when click to pan the slippy map, should turn into a hand cursor
- geocoding API: should look at all results with more than one location
- change the color of the text in the buttons when zoomed all the way in or out
- speed selector?
- sounds for every instance?
- timezone value for lat.lon: to make bubbles appear from: left->right

moar ideaz:

- as an extensions: media reports of police vs protestors