Social Media and Nonprofits: Measurement

A new term, "Big Data fundamentalism" — the idea with larger data sets, we get closer to objective truth.”

This article summarizes a talk by Kate Crawford a researcher at Microsoft during a Big Data Conference: http://dataedge.ischool.berkeley.edu/2013/

She offers up 6 Myths:

Myth 1:  Big Data Is New

(It's been around for a while, like all disruptive tech trends - it sometimes takes years to ramp up and then fade into being part of life."

Myth 2:   Big Data Is Objective

Need to have information literacy skills - know the context of your data - it may be skewed.     This is why you can't be "data-driven" and have to be "data-informed"

Myth 3: Big Data Doesn't Discriminate

It isn't color blind or gender blind.   When people create data sets, those become fallible human tools.    Data is something we create but also something we imagine.

Myth 4:   Big Data Makes Cities Smart

It is only as good as the people using it.    Devices are proxies for public needs.   

Myth 5:  Big Data Is Anonymous

Talking about privacy issues related to public data -- with a cell phone data you can identify 50% of who made the call with two data points, with four data points identify 95%.  Finger prints require 12 data point to identify someone.

Myth 6:  You can opt out

Uses Instagram as an example when they shared TOS.   What it didn't have was a paid option so you could pay a fee and opt out.    Will there be a two-tiered system in the future for people who want and can afford to control their data?   

"Before Big Data becomes a fact of life,  need to think about how we will navigate these systems - for individuals and society. 

This tool lets you upload your spreadsheets, it analyzes your data, and auto generates visualizations.  I haven't tried it yet, but I am wondering if it does my laundry too.    Seriously, curious about whether it is any good.  

The Nonprofit Finance Fund has released the results of their 2013 State of the Nonprofit Sector Survey and it is chock full of important information and analysis tools for nonprofits.  The study had nearly 6000 respondents and the online survey analyzer allows you to do subset analysis.  This article features some highlights of the overall survey. Marion Conway has summarized the important points.

Calming Computing - book coming out in August Distraction Addiction - has a twitter of people focusing on this.

A few good tips

Profile of the work that SumAll is doing 

Here's another resource for finding data nerds who can work with nonprofits! 

We have a different sort of solution in mind – one that harnesses the untapped potential of students. As institutions of higher education dramatically expand their data science offerings, countless students are gaining new skills, but aren’t putting them to use with real-world datasets. Next month, we’re giving tens of thousands of those students from around the world the chance to do something more exciting: work with your data to strengthen your impact.

Nonprofits of all types are now invited to apply to “Introduction to Data Science,” a massively open online course to be taught next month by Prof. Bill Howe of the University of Washington. As part of the course, students will have the option to complete a project in which they work with a real dataset to address the needs of an organization. As they learn techniques in data visualization and trend identification, these students will apply what they learn to help drive your organization’s future initiatives. The insights they offer could strengthen your impact and identify areas for growth.

In a world with limited resources, it just makes sense to utilize student talents and creativity to strengthen your work. Tens of thousands of students have already registered for the course – don’t miss your chance to recruit thei

Thorough description of how ot measure reach

This presentation is a high  level overview of several resources to improve nonprofit measurement practice and several case studies on social impact assessment.    The presentation comes from Mayur Patel who is Director of Evaluation for the Knight Foundation.     

The accompanying blog post is here.

But is big data really all it's cracked up to be? Can we trust that so many ones and zeros will illuminate the hidden world of human behavior? Foreign Policy invited Kate Crawford of the MIT Center for Civic Media to go behind the numbers:

"With Enough Data, the Numbers Speak for Themselves."

Not a chance. The promoters of big data would like us to believe that behind the lines of code and vast databases lie objective and universal insights into patterns of human behavior, be it consumer spending, criminal or terrorist acts, healthy habits, or employee productivity. But many big-data evangelists avoid taking a hard look at the weaknesses. Numbers can't speak for themselves, and data sets -- no matter their scale -- are still objects of human design. The tools of big-data science, such as the Apache Hadoop software framework, do not immunize us from skews, gaps, and faulty assumptions. Those factors are particularly significant when big data tries to reflect the social world we live in, yet we can often be fooled into thinking that the results are somehow more objective than human opinions. Biases and blind spots exist in big data as much as they do in individual perceptions and experiences. Yet there is a problematic belief that bigger data is always better data and that correlation is as good as causation.

Lesson Learned: We created the graph below using NodeXL. It shows the 5 key types of participation that visitors can engage in on the ACTion Alexandria website. In this graph, each dot represents a site visitor. Color represents how many activities site visitors have performed (e.g., orange=all 5 activities, dark blue=just 1 activity). Size also indicates the number of activities site visitors have performed. Each line represents an activity the site visitor engaged in on the website. The thickness of the lines (also referred to as edge thickness) shows the number of times a site visitor has performed an activity.

The graph highlights the following things about site participation:

1. Most people only perform one activity, although a significant number perform two activities.
2. Voting is by far the most popular way that people engage with the site.
3. Blog posters post many times, but don’t tend to comment on each other’s blogs.
4. People who engage with multiple types of activities are more likely to engage in them multiple times (i.e., have thicker lines).

More help for nonprofits! https://twitter.com/morriscox/status/329358542844870657

Not sure I agree that it is an output vs outcome ...