Profile

Cover photo
Andrew Wheeler
Works at John Finn Institute of Public Safety
Lives in Albany, NY 12222
58 followers|467,245 views
AboutPostsPhotos+1's

Stream

Andrew Wheeler

Shared publicly  - 
 
Python post. Counting neighbors within a buffer using KDTree's in sklearn.neighbors. Example counting shootings within 1km of schools in DC.
1
Add a comment...

Andrew Wheeler

Shared publicly  - 
 
Was working on some nearby neighbors in space calculations. Was writing code to do all pairwise comparisons for around 4,000*100,000 and then counting the neighbors within X meters. Wasn't working out so well.

I updated the code to make a KDTree using scipy and numpy for the 100,000 set. Building the tree only took seconds, and then counting the neighbors took around a minute.
1
Add a comment...

Andrew Wheeler

Shared publicly  - 
 
This scatterplot is more informative than a few p-value violin charts going around.
This graph is from the Open Science Framework attempt to replicate 100 interesting results in experimental psychology, led by Brian Nozek and published in Science today. About a third of the experi...
1
Add a comment...

Andrew Wheeler

Shared publicly  - 
 
Expansion of NYC subway over time
New York City Subway Over Time [GIF] [464x600] CLICK HERE FOR MORE MAPS! thelandofmaps.tumblr.com
1
Add a comment...

Andrew Wheeler

Shared publicly  - 
 
In NYC, calls about noise and blocked driveways are most frequent in zones between racially homogenous neighborhoods.
1
Add a comment...

Andrew Wheeler

Shared publicly  - 
 
This kills me, as I can not make any sense of this argument (and I've experienced it in rural PA/NY as well).

1 - Who are you kidding that there is any economic benefit of tourism?
2 - Clearly such development will bring in more jobs to local individuals than tourism, even in the best case tourism scenarios

Seriously, how about showing how much tourism $$$'s there actually are before saying that something will decrease them. 
2
Add a comment...
Have him in circles
58 people
Rose Bellandi's profile photo
Guillaume Filion's profile photo
Jeremy Miles's profile photo
Janet Stamatel's profile photo
Matt Asher (Statistics Blog and Sun Falls Press)'s profile photo
Jonny Hughes's profile photo
Jeremy Shifton's profile photo
vlad simion's profile photo
Gautam Savaliya's profile photo

Andrew Wheeler

Shared publicly  - 
 
I wanted to give more answers for words I regularly misspell.

I gave seperate, but exersize and hetereoskedasticity I loathe as well.
1
Add a comment...

Andrew Wheeler

Shared publicly  - 
 
As a colorblind person who specializes in making maps for data visualization,I was inspired byBlake’s recent blog post on Introducing a Color Blind Filterto take a careful look at the...
1
Add a comment...

Andrew Wheeler

Shared publicly  - 
 
I'm starting to get over these studies, mainly because the overall idea has a large flaw in it.

So the idea is, race is not entirely discrete, and that darker skinned individuals suffer more racial animus than lighter skinned individuals. Ok, pretty straight forward and no argument there.

The second part though, looking at the correlation between skin tone and particular outcomes, and then saying it is evidence of racial bias, is misleading. This is problematic, because people are the accumulation of their entire experiences. The outcomes of adults are likely the product of a bunch of stuff going on in there prior history.

For example, the same theory would potentially lead to worse educational experiences, as teachers are likely to be racially prejudiced against darker skinned students. Thus employment outcomes in adults may occur even without racial discrimination due to the earlier education disparities.
Age-old stereotypes interact in complex ways when new arrivals look for jobs.
1
mkrasmus's profile photoAndrew Wheeler's profile photo
2 comments
 
The darker skin discrimination is not only a US thing, so in general immigration is not a reasonable cop-out for the problem. 

Being a social scientist and mostly doing observation research I am sympathetic to "it is the best we can do, and there will always be confounds". I'm ok living with the uncertainty - I'm not ok with people making interpretations that require cognitive dissonance though to be true. 

Part of my issue with this design is that it hides behind complexity. On its face it appears like a neat little identification strategy, but it isn't, and is subject to the same critiques of inferring racial discrimination from more traditional discrete measures. 
Add a comment...

Andrew Wheeler

Shared publicly  - 
 
new blog post on why I don't like music playing in the workplace
1
Add a comment...

Andrew Wheeler

Shared publicly  - 
 
My study on 311 calls and crime in D.C. focuses on more general crime, but the lessons are similar. There is only a very small correlation between "broken windows" (and other measures of physical disorder) and crime.
Why community policing should focus on helping to resolve personal and domestic disputes, not signs of physical decay.
1
Add a comment...
People
Have him in circles
58 people
Rose Bellandi's profile photo
Guillaume Filion's profile photo
Jeremy Miles's profile photo
Janet Stamatel's profile photo
Matt Asher (Statistics Blog and Sun Falls Press)'s profile photo
Jonny Hughes's profile photo
Jeremy Shifton's profile photo
vlad simion's profile photo
Gautam Savaliya's profile photo
Places
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Currently
Albany, NY 12222
Previously
Bloomsburg, PA 17815 - Canton, PA 17724
Links
Other profiles
Contributor to
Story
Introduction
researcher in criminal justice 

Work
Employment
  • John Finn Institute of Public Safety
    Analyst, 2009 - present
Basic Information
Gender
Male
Andrew Wheeler's +1's are the things they like, agree with, or want to recommend.
Trying again
www.statschat.org.nz

This graph is from the Open Science Framework attempt to replicate 100 interesting results in experimental psychology, led by Brian Nozek an

Experiments in simulating colorblindness
www.mapbox.com

As a colorblind person who specializes in making maps for data visualization,I was inspired byBlake’s recent blog post on Introducing a Colo

Mapping the American Commute
www.undertheraedar.com

One of my summer projects this year has been attempting to map the American commute, following earlier work on a similar subject. Put simply

The Land Of Maps - New York City Subway Over Time [GIF] [464x600]...
thelandofmaps.tumblr.com

New York City Subway Over Time [GIF] [464x600] CLICK HERE FOR MORE MAPS! thelandofmaps.tumblr.com

Music and distractions in the workplace
andrewpwheeler.wordpress.com

I was recently re-reading Zen and the Art of Motorcycle Maintenance, and it re-reminded me of why I do not like to listen to music in the wo

Neighbors Complain Most to 311 Along Fuzzy Racial Boundaries, NYU Study ...
www.citylab.com

In NYC, calls about noise and blocked driveways are most frequent in zones between racially homogenous neighborhoods.

Contrary to Broken Windows Theory, Crime and Neighborhood Disorder Stem ...
www.citylab.com

Why community policing should focus on helping to resolve personal and domestic disputes, not signs of physical decay.

Is Bayesian A/B Testing Immune to Peeking? Not Exactly
varianceexplained.org

Since I joined Stack Exchange as a Data Scientist in June, one of my first projects has been reconsidering the A/B testing system used to ev

waiting is torture, but it's not so bad if there are mirrors or trees
punkrockor.wordpress.com

Operations research is the discipline of making better decisions. We have to solve the right problem to better inform decisions, and sometim

Some pattern math – regular patterns | Imagico.de
blog.imagico.de

There are now also various options to generate regular patterns in my pattern generator tool. The difficulty here is that from the map rende

Unattributed quotes and the propagation of myths about homicide in America
nonparibus.wordpress.com

"One study of a high-crime community in Boston found that 85% of gunshot victims came from a network of just 763 young men—or less than 2% o

the place of reproducible research
scatter.wordpress.com

The ongoing scuffles over reproducible (or is it replicable? or robust?) research always seems to miss one point particularly important to m

Late Again?
chroniclevitae.com

A faculty member wondered why her students were always late. So she asked them.

is parenthood really worse than divorce? demographic clickbait in the wa...
scatter.wordpress.com

The following is a guest post by Aaron Major. If you’ve seen, or heard about, the Washington Post piece on having a baby being worse than de

Using local Python objects in SPSSINC TRANS - examples with network stat...
andrewpwheeler.wordpress.com

When using SPSSINC TRANS, you have a wider array of functions to compute on cases in SPSS. Within the local session, you can create your own

Fraud detection in retail with graph analysis
www.datavizualization.com

Guest blog post by Jean Villedieu Fraud detection is all about connecting the dots. We are going to see how to use graph analysis to identif

A Rare Albino Raccoon Is Found in a West Virginia Dumpster - CityLab
www.citylab.com

Some people say finding one is less likely than getting hit by lightning.

L.A. Puts Its Water Supply Undercover With Shade Balls - CityLab
www.citylab.com

It’s part of a $34.5 million project to protect the city’s water supply from the sun.

That Pareto Guy? Turns Out He’s Pretty Smart. — Medium
medium.com

A database without context is like a gigantic buffet at a cheap resort: It looks huge, but there’s really very little me…