Preparing for Collaborative Data Driven Projects December 9, 2016 Lauren Hareem Erika Haynes Naveed Salomon Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 1 University of Chicago
Agenda • 5 Minutes Introductions • 20 Minutes Intro to Project Scoping (DSaPP) • 20 Minutes Intro to Data Maturity (DSaPP) • 20 minutes Intro to Data Governance (McClean) - How to get legal agreements, stakeholder buy in • 20 Minutes Q&A
38 projects
“We are used to using data to justify funding decisions. Now we can use data to improve what we do.” Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 5 University of Chicago
Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 6 University of Chicago
This May Sound Daunting ... Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 7 University of Chicago
But Most Common Machine Learning Tasks... Dimensionality Reduction Classification Regression Clustering Dimension Reduction Labeling and Using trends to Finding existing Finding important sorting into predict outcomes groups or predictors groups categories Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 8 University of Chicago
… You Actually Learned In Kindergarten Clustering Dimension Reduction Classification Regression Finding existing Finding important Labeling and Using trends to groups or categories predictors sorting into groups predict outcomes Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 9 University of Chicago
Most Projects Fall in a Few Categories • Early warning & intervention • Efficient resource allocation & targeted action • Effective advocacy & fundraising • Data-driven policy recommendation & evaluation Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 10 University of Chicago
Data are People Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 11 University of Chicago
“Predictive analytics is emerging as a game-changer. Instead of looking backward to analyze “what happened?” predictive analytics help executives answer “What’s next?” and “What should we do about it?” Forbes Magazine Why Predictive Analytics Is A Game-Changer Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 12 University of Chicago
Project Scoping Workshop Center for Data Science and Public Policy dsapp.uchicago.edu dsapp.uchicago.edu @datascifellows 13 University of Chicago
“We are used to using data to justify funding decisions. Now we can use data to improve what we do.” Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 14 University of Chicago
Scoping a Good Project is Easier Said than Done Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 15 University of Chicago
What Makes a Good Project ● A solvable problem. ● A challenging problem. ● An important problem with social impact. ● A motivated, capable, and committed partner. a. Domain/business resources b. Data understanding resources c. Commitment to implementation ● Appropriate, relevant, available data. Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 16 University of Chicago
Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 17 University of Chicago
Relevance and Sufficiency Irrelevant and Insufficient GOAL Relevant but Insufficient Relevant and Sufficient Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 18 University of Chicago
Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 19 University of Chicago
Data Analytics: Problem Formulation • Identify Goals • Identify Actions you can take to achieve those goals – Break down actions into fine-grained questions/subactions • Identify Data Sources Resources you need and have • Identify Analysis/Modeling that needs to be done Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 20 University of Chicago
Workshop • Scope an analytics-driven project for a problem your organization is facing – Identify goal(s) – Actions (persuasion for example) – Data sources (data you have, data you need to collect, relationships to get that data) – Models ● Who? (to target for each action) ● What? (to say to them) ● How? (to use different communication channels) Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 21 University of Chicago
Workshop: Identify Actions that can be taken to achieve the goal • What programs do I have access to? What would they do differently if they had more information/knew where their interventions were most likely to be effective • Be sure to name out which individuals are taking the action Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 22 University of Chicago
Workshop: Data Sources • Data sources (data you have, data you need to collect, relationships to get that data) Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 23 University of Chicago
Workshop : Changing Actions ● Who? (to target for each action) ● What? (to say to them) ● How? (to use different communication channels) Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 24 University of Chicago
Partner Data Maturity Framework Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 25 University of Chicago
Data Maturity Framework Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 26 University of Chicago
Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 27 University of Chicago
How to Arrive at Data Nirvana Commit to data-driven decisionmaking. ● Data as a first class citizen ○ Ready for tough love from data ○ Willing to take data risks ○ Understand what data supports the mission and how. ● Has the data it needs or can obtain it ○ Technical and organizational capacity ○ Outcome linked to action ○ Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 28 University of Chicago
Data Readiness Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 29 University of Chicago
Relevance and Sufficiency Irrelevant and Insufficient GOAL Relevant but Insufficient Relevant and Sufficient Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 30 University of Chicago
Data Capture population with data population of interest Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 31 University of Chicago
Data Storage / Format .csv .json API .xml High cost of use Low cost of use Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 32 University of Chicago
Data Quality Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 33 University of Chicago
Integration Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 34 University of Chicago
Accessibility Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 35 University of Chicago
Documentation DATA Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 36 University of Chicago
Organizational Readiness Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 37 University of Chicago
Leadership Buy In Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 38 University of Chicago
Individual Buy In Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 39 University of Chicago
Stakeholder Buy In Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 40 University of Chicago
Thank You! http://dssg.uchicago.edu/ http://dsapp.uchicago.edu/ DSSG 2017 Project Partner Applications Due 1/31/17! Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 41 Contact Us University of Chicago lnhaynes@uchicago.edu
Case Studies Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 42 University of Chicago
Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 43 University of Chicago
Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 44 University of Chicago
LEAD Lack of Motor Skills Impaired Attention Hearing Loss Learning Disability Lower IQ Memory Problems Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 45 University of Chicago
Constituents Elected Representatives
Center for Data Science and Public Policy dsapp.uchicago.edu @datascifellows 47 University of Chicago
Personal features and location
Recommend
More recommend
