Finding Data with Impact

BREAK THROUGH

WITH DATA

Drive your

mission forward

by building credibility

& engaging your audience

Flickr – Jason Mrachina

@WCA4KIDS #NFPSUMMIT

www.wca4kids.org/blog

@wca4kids



WHO? @WCA4KIDS #NFPSUMMIT


PROCESS

1 Finding Data with Impact

2 Presenting Data with Punch

3 Driving Advocacy with Data

1 Finding Data with Impact


2 Presenting Data with Punch

3 Driving Advocacy with Data

www.flickr.com/bionicteaching

Make data approachable.

Flickr - colorblindPICASO

Jaym

ug.

com


www.flickr.com/bionicteaching

ACK!!!

Awwww....

Flickr - colorblindPICASO


or lose your audience.

Flickr - micadew

Flickr - kengo

Finding Data Sources THE GOOD, THE BAD, AND THE UGLY


Reliable

Flickr – popofatticus

Neutral Flickr – Tim Evanson

Raw Flickr – popofatticus

Saying something

doesn’t make it true.


US Census •Decennial Census

•American Community Survey

•Population Estimates


Other Sources Child Data

nyskwic.org datacenter.kidscount.org

Education

data.nysed.gov ocrdata.ed.gov

Health health.westchestergov.com/statistics health.ny.gov/statistics/

Economy bls.gov


SOURCE EVERYTHING

& bu i ld c red ib i l i t y.


DATA TRAPS

THERE ARE TWO KINDS OF STATISTICS,

THE KIND YOU LOOK UP & THE KIND YOU MAKE UP.

-REX STOUT, DEATH OF A DOXY


Source: Opinions Dynamic via Fox News

2012 Presidential Election – GOP Candidates

70%

63%

60%

Back Palin Back Huckabee Back Romney



Recognizing Limitations DON’T USE DATA JUST BECAUSE IT ’S THERE!


The Problem with Surveys •Leading questions

•People like to say they’re good

•Majority of surveys are based on samples


What’s a Sample? •A sample is a subset of a population.

•It is ideally a representative of the population.


Children Under Age 5 in Poverty

in Greenburgh Town

Guess

Male ???

Female ???

One Year Estimates

0

170

Five Year Estimates

99

92

2013 One Year Estimates

2009-13 Five Year Estimates

American Community Survey, Table B17001


Margin of Error the amount by which you expect your results to vary, if a different sample were taken.


One Year Estimates

Margin of Error

Male 0 ±193

Female 170 ±233

Five Year Estimates

Margin of Error

Male 99 ±71

Female 92 ±78

Children Under Age 5 in Poverty in Greenburgh Town



2009-13 Five Year Estimates

American Community Survey, Table B17001

-193

-63

193

403

0

170

-300

-200

-100

0

100

200

300

400

500

Male Female

One Year Estimates


28 14

99 92

-300

-200

-100

0

100

200

300

400

500

Male Female

Five Year Estimates

170 170


Smaller Samples

LARGER MARGIN OF ERROR

-193

-63

193

403

0

170

-300

-200

-100

0

100

200

300

400

500

Male Female

One Year Estimates


Smaller Margin of Error LARGER SAMPLES

28 14

99 92

-300

-200

-100

0

100

200

300

400

500

Male Female

Five Year Estimates

170 170


Less “confident” estimates •Communities with smaller populations

•Subgroups (ex. Gender, Race, Age)

•Surveys that sample across shorter time periods


Choosing Data with Impact


Choose Data THAT TELLS YOUR STORY


Numbers alone ARE MEANINGLESS.


40.719612, -74.006130, 5.000000


Poverty Rate


American Community Survey

Table S1701 @WCA4KIDS #NFPSUMMIT

What’s the poverty level? For one person?

For a family of four?


$11,770 2015 Poverty Guidelines

US Department of Health and Human Services @WCA4KIDS #NFPSUMMIT

$24,250 2015 Poverty Guidelines

US Department of Health and Human Services @WCA4KIDS #NFPSUMMIT

MEDIAN INCOME FAMILIES WITH CHILDREN UNDER 18

Low

$23,019

High

$250,000+


Finding Data with Impact

Data & Analytics

nfpsummit process

presenting data

data sources

male female

data traps

data approachable

sources child data

flickr colorblindpicaso