Civic Engagement Research at Facebook
2 likes4,621 views
This document summarizes Facebook's research on civic engagement and political efficacy. It discusses the multi-disciplinary research team, the importance of civic engagement to Facebook's mission, and how the team conducts qualitative interviews, surveys, and analyzes interaction data to understand civic behaviors and gaps. The research is used to imagine new product ideas like helping people connect with local representatives. The process involves understanding needs, observing the current state, instrumenting the platform to measure priorities, imagining solutions, building products, and iterating based on results. While the impact on political efficacy is unclear, the research aims to engage users and ensure responsibility as the company explores this important area.
![Qualitative research is our starting point to uncover
ethnographic insights on digital civic engagement
What we’ve done so far:
• In-depth interviews with people in 3 U.S. cities
• Informed consent from all participants
What we’ve found:
• Skepticism that individual voices matter
• Limited awareness of civic activities one could do
outside of national elections or an immediate crisis
• Few know what their elected reps are doing in office
• People sometimes worry that online political
conversations online can lead to personal risk
[UNDERSTAND]
“... I don't know that
[my voice] was
individually heard, but I
think as a whole it was.”](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-8-320.jpg)
![What we’ve done so far:
• In-depth interviews with people around the globe
• Turkey, Kenya, Indonesia, Estonia, Iceland to date
• Informed consent from all participants
What we’ve found:
• Universal desire exists to connect with representatives,
especially to get a sense of their achievements while in office
• Elections are the main vehicle to get voices heard (like US)
• Perception of account security is an important driver for
people’s willing to engage in political discourse online
We’ve been extending this internationally, too
[UNDERSTAND]](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-9-320.jpg)
![We survey people on Facebook to understand
motivations, attitudes, and real-world actions
[OBSERVE]
How many had contacted their elected reps
(N=1000)
Why people contacted their elected reps](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-10-320.jpg)
![Analyzing interaction patterns on Facebook helps
us understand engagement with politicians…
[OBSERVE]
Comments on Senator Pages:
• Red lines = from constituents
• Blue lines = from out of state
Comments aren’t just from
constituents… there is also a
vigorous national dialogue](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-11-320.jpg)
![Analyzing interaction patterns on Facebook helps
us understand engagement with politicians…
[OBSERVE]
Discussion also spikes when an issue is
part of the national conversation
Trending issues are central to engagement
and drive a lot of interactions w/ reps](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-12-320.jpg)
![Analyzing interaction patterns on Facebook helps
us understand engagement with politicians…
[OBSERVE]
Constituents and non-constituents
often comment about different
things on politicians’ Pages
Even online, constituency matters
in political interactions](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-13-320.jpg)
![… and gaps in political engagement on Facebook,
which often reflect patterns in the offline world
[OBSERVE]
Political Commenting on FB Posts By Age Self-reported Contact With Reps By Age](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-14-320.jpg)
![Correlations between survey responses and
Facebook activities can help spur product ideas…
[OBSERVE]
% of political Pages “liked” vs. self-reported political efficacy
Is this causal?
Does a person’s
political efficacy
increase if they
start connecting
with more
politician Pages?](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-15-320.jpg)
![Self-declared political ideology
of US users on their Profiles
Perhaps most importantly, observational analyses
help us stay true to core principles (like fairness)
[OBSERVE]
Self-reported political ideology
of US users based on surveys](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-16-320.jpg)
![1. How interested are you in politics?
– (Very interested to Not at all Interested)
2. Generally speaking, people like me don't have the
power to influence government policy or actions
– (Strongly disagree to Strongly agree)
3. I feel confident in my understanding of politics
– (Strongly disagree to Strongly agree)
4. Political activities on Facebook, including political
discussion, can affect government policy or actions.
– (Strongly disagree to Strongly agree)
5. In which of the following ways do you feel comfortable
talking about political issues with family or friends
– Check all that apply: In person and on Facebook, Only in person,
None of the above
6. In which of the following ways do you feel comfortable
talking about political issues publicly (i.e., with people
you might not know)
– Check all that apply: In person and on Facebook, Only in person,
None of the above
7. How often in the last 12 months have you tried to help
solve a problem in your community ?
– (At least once a week, once or twice a month, once or twice a
year, Never)
8. I feel connected to my local government
– (Strongly disagree to Strongly agree)
9. Which of the following people have you contacted in
the last 12 months about some important problem or to
give them your views?
– (Check all that apply: A local elected representative (e.g., mayor,
councilperson), A state-level elected representative (e.g.,
governor, state representative), A member of congress, the
President, a political party official, a government office)
Before imagining new products, we instrument the
platform to measure our top-level priorities
[INSTRUMENT]
Ten Questions From Our Tracking Survey On Political Efficacy](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-17-320.jpg)
![Early tracking survey results at least validate this
technique as a reliable means of measurement…
[INSTRUMENT]
People who post about politics say they are interested in politics](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-18-320.jpg)
![… and already point to novel product possibilities
[INSTRUMENT]
People feel
disconnected from
their elected
officials, particularly
at the local level](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-19-320.jpg)
![From all this research we imagine new products,
such as helping people connect with local reps
[IMAGINE]
Old](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-20-320.jpg)
![From all this research we imagine new products,
such as helping people connect with local reps
[IMAGINE]
New](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-21-320.jpg)
![From all this research we imagine new products,
such as helping people connect with local reps
[BUILD/DEPLOY/ITERATE]
New
72%
increase in
conversions](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-22-320.jpg)
![Still unclear if we can have an impact on political
efficacy, but this journey is just beginning!
[BUILD/DEPLOY/ITERATE]
Election day voting reminders… … do not significantly move political efficacy
(when comparing states that have just voted
to states that have yet to vote in the primary)](https://image.slidesharecdn.com/facebooksamidhandwinter-160503100009/85/Civic-Engagement-Research-at-Facebook-23-320.jpg)
Civic Engagement Research at Facebook
- 1. Civic Engagement Research at Facebook Samidh Chakrabarti Winter Mason TICTeC April 2016
- 2. Our research team is multi-disciplinary Samidh Chakrabarti Winter Mason Monica Lee Funda Kivran-Swaine Erhardt Graeff
- 3. Civic engagement is core to Facebook’s mission “Give people the power to share and make the world more open and connected.” Facebook company mission “Give people a voice… in public affairs” Civic engagement mission
- 4. How to make existing civic interactions better? Volume of political comments in the US on Facebook 3% of all FB comments Super Tuesday Mega Tuesday April Fools
- 5. “Political Efficacy” is a North Star for “better” Political Efficacy definition /pəˈlidək(ə)l efəkəsē/ noun In political science, political efficacy is the citizens’ belief that they can understand and influence political affairs. It is commonly measured by surveys and is used as an indicator for the broader health of civil society. 0 10 20 30 40 50 60 70 80 ANES Political Efficacy Index 1956 20121984 Sources: Wikipedia (for definition). ANES(for chart).
- 6. We think hard about the values driving our work • Be Selfless: Serve people's interests first, not Facebook's interests • Be Protective: Keep people safe (including from personal risk) • Be Fair: Provide same opportunities to everyone • Be Representative: Strive towards broadly inclusive products • Be Constructive: Build empathy and defuse acrimonious polarization • Be Conscious: Know our impact (both positive and negative)
- 7. Our research process is integral to all that we build Understand people’s needs Observe current state Instrument the platform Imagine product solutions Build deploy & iterate
- 8. Qualitative research is our starting point to uncover ethnographic insights on digital civic engagement What we’ve done so far: • In-depth interviews with people in 3 U.S. cities • Informed consent from all participants What we’ve found: • Skepticism that individual voices matter • Limited awareness of civic activities one could do outside of national elections or an immediate crisis • Few know what their elected reps are doing in office • People sometimes worry that online political conversations online can lead to personal risk [UNDERSTAND] “... I don't know that [my voice] was individually heard, but I think as a whole it was.”
- 9. What we’ve done so far: • In-depth interviews with people around the globe • Turkey, Kenya, Indonesia, Estonia, Iceland to date • Informed consent from all participants What we’ve found: • Universal desire exists to connect with representatives, especially to get a sense of their achievements while in office • Elections are the main vehicle to get voices heard (like US) • Perception of account security is an important driver for people’s willing to engage in political discourse online We’ve been extending this internationally, too [UNDERSTAND]
- 10. We survey people on Facebook to understand motivations, attitudes, and real-world actions [OBSERVE] How many had contacted their elected reps (N=1000) Why people contacted their elected reps
- 11. Analyzing interaction patterns on Facebook helps us understand engagement with politicians… [OBSERVE] Comments on Senator Pages: • Red lines = from constituents • Blue lines = from out of state Comments aren’t just from constituents… there is also a vigorous national dialogue
- 12. Analyzing interaction patterns on Facebook helps us understand engagement with politicians… [OBSERVE] Discussion also spikes when an issue is part of the national conversation Trending issues are central to engagement and drive a lot of interactions w/ reps
- 13. Analyzing interaction patterns on Facebook helps us understand engagement with politicians… [OBSERVE] Constituents and non-constituents often comment about different things on politicians’ Pages Even online, constituency matters in political interactions
- 14. … and gaps in political engagement on Facebook, which often reflect patterns in the offline world [OBSERVE] Political Commenting on FB Posts By Age Self-reported Contact With Reps By Age
- 15. Correlations between survey responses and Facebook activities can help spur product ideas… [OBSERVE] % of political Pages “liked” vs. self-reported political efficacy Is this causal? Does a person’s political efficacy increase if they start connecting with more politician Pages?
- 16. Self-declared political ideology of US users on their Profiles Perhaps most importantly, observational analyses help us stay true to core principles (like fairness) [OBSERVE] Self-reported political ideology of US users based on surveys
- 17. 1. How interested are you in politics? – (Very interested to Not at all Interested) 2. Generally speaking, people like me don't have the power to influence government policy or actions – (Strongly disagree to Strongly agree) 3. I feel confident in my understanding of politics – (Strongly disagree to Strongly agree) 4. Political activities on Facebook, including political discussion, can affect government policy or actions. – (Strongly disagree to Strongly agree) 5. In which of the following ways do you feel comfortable talking about political issues with family or friends – Check all that apply: In person and on Facebook, Only in person, None of the above 6. In which of the following ways do you feel comfortable talking about political issues publicly (i.e., with people you might not know) – Check all that apply: In person and on Facebook, Only in person, None of the above 7. How often in the last 12 months have you tried to help solve a problem in your community ? – (At least once a week, once or twice a month, once or twice a year, Never) 8. I feel connected to my local government – (Strongly disagree to Strongly agree) 9. Which of the following people have you contacted in the last 12 months about some important problem or to give them your views? – (Check all that apply: A local elected representative (e.g., mayor, councilperson), A state-level elected representative (e.g., governor, state representative), A member of congress, the President, a political party official, a government office) Before imagining new products, we instrument the platform to measure our top-level priorities [INSTRUMENT] Ten Questions From Our Tracking Survey On Political Efficacy
- 18. Early tracking survey results at least validate this technique as a reliable means of measurement… [INSTRUMENT] People who post about politics say they are interested in politics
- 19. … and already point to novel product possibilities [INSTRUMENT] People feel disconnected from their elected officials, particularly at the local level
- 20. From all this research we imagine new products, such as helping people connect with local reps [IMAGINE] Old
- 21. From all this research we imagine new products, such as helping people connect with local reps [IMAGINE] New
- 22. From all this research we imagine new products, such as helping people connect with local reps [BUILD/DEPLOY/ITERATE] New 72% increase in conversions
- 23. Still unclear if we can have an impact on political efficacy, but this journey is just beginning! [BUILD/DEPLOY/ITERATE] Election day voting reminders… … do not significantly move political efficacy (when comparing states that have just voted to states that have yet to vote in the primary)
- 24. You’re the most important ones we want to engage • Research is central to everything we do in this space • Not only for product goals… but also ensures responsibility • We are in uncharted waters… and we need your help! • Please give us feedback… we are available to you! • Come to drinks reception tonight to kick off the conversation! IN CONCLUSION…
- 25. Thank You! Questions? Samidh Chakrabarti [email protected] @samidh Winter Mason [email protected] @winteram