gebelo

from matplotlib.colors import LinearSegmentedColormap

bins = [0,1]

# Maps values to a bin.
# The mapped values must start at 0 and end at 1.
def bin_mapping(x):
    for idx, bound in enumerate(bins):
        if x < bound:
            return idx / (len(bins) - 1.0)

gebelo

def zptile(z_score):
    return .5 * (math.erf(z_score / 2 ** .5) + 1)

zptile(0.95)

# excel says: 0.8289438737

0.8289438736915181

via: http://stackoverflow.com/questions/2782284/function-to-convert-a-z-score-into-a-percentage

gebelo

SELECT dob, SUBSTRING_INDEX( `dob` , '/', 1 ) AS mo,
SUBSTRING_INDEX(SUBSTRING_INDEX( `dob` , '/', 2 ),'/',-1) AS dy, 
SUBSTRING_INDEX(SUBSTRING_INDEX( `dob` , '/', 3 ),'/',-1) AS yr,
CONCAT(SUBSTRING_INDEX(SUBSTRING_INDEX( `dob` , '/', 3 ),'/',-1),"-",SUBSTRING_INDEX( `dob` , '/', 1 ),"-",SUBSTRING_INDEX(SUBSTRING_INDEX( `dob` , '/', 2 ),'/',-1)) AS dt
FROM table

"dob"	"mo"	"dy"	"yr"	"dt"
"7/19/1970"	"7"	"19"	"1970"	"1970-7-19"

gebelo

2010-2014, ACS 5-year sample
Source: IPUMS.org

Nativer American Alone, by Tribal Identification	
302: Apache	 70,940 
303: Blackfoot	 28,256 
304: Cherokee	 279,721 
305: Cheyenne	 14,386 
306: Chickasaw	 23,110 
307: Chippewa	 114,423 

gebelo

The correct 'link':
<div class='map-navigation'  data-zoom='12' data-position='<%=d.tractlat%>,<%=d.tractlon%>'>Zoom</div>

The 'link' that zoomed but then reloaded page:
<div class='map-navigation'><a href='' data-zoom='12' data-position='<%=d.tractlat%>,<%=d.tractlon%>'>Zoom</a></div>

The zoomer function:
document.querySelector('.map-navigation').onclick = function(abc) {
    var pos = abc.target.getAttribute('data-position');
    var zoom = abc.target.getAttribute('data-zoom');

gebelo

Bonus example:

The IPUMS SDA query interface allows you to collapse variables into broader categories. I do this often with occupations -- individual occupations are usually too small to carry a meaningful sample size, but if you look at the occupational hierarchy in the docs, you can see how you might tile up individual occupations into broader but still meaningful categories.

So try this.

1) Pick any sample to query

2) put sex as your column