The People Lab
LightNews LightNews
banner
thepeoplelab.bsky.social
The People Lab
@thepeoplelab.bsky.social
760 followers 140 following 150 posts
Research lab led by @elizabethlinos.bsky.social at Harvard Kennedy School.

We build evidence on the people in government and the communities they are called to serve.
Posts Replies Media Videos
thepeoplelab.bsky.social
Great conversations, thoughtful sessions, and RCT tattoos—what more could you want from a day on evidence-informed policymaking?

Thanks to @j-palna.bsky.social, LEO, and King County for hosting the Evidence Matters Convening and for bringing so many good ideas together.
November 13, 2025 at 4:15 PM
thepeoplelab.bsky.social
Special thanks to Heidi Reinson, and Helelyn Tammsaar for their incredible hospitality and for making the visit so impactful!
September 9, 2025 at 8:19 PM
thepeoplelab.bsky.social
Shoutout to our talented Du Bois Scholar, Jonora, for an outstanding final presentation! Her thoughtful lit review is already helping guide a new research venture at The People Lab. Grateful for your summer with us and excited for what’s next! 🙌
August 8, 2025 at 4:12 PM
thepeoplelab.bsky.social
Our latest n=9 field test involved waffle cones. Results: instantly raised happiness scores by 200%. 🍦 #SweetIncentives #NudgeAndFudge #ConeformityBias
June 25, 2025 at 3:32 PM
thepeoplelab.bsky.social
A huge thank you to Terrance Smith for joining us and the Bloomberg Center for Cities at Harvard University to share his insights on trust and innovation in local government!
February 26, 2025 at 10:53 PM
thepeoplelab.bsky.social
🎉 We're thrilled to have Noe Negri join The People Lab as our Assistant Director of Operations! With over 12 years of experience, she is already making everything run smoother and bringing operational excellence to our team. Welcome aboard, Noe! 🌟
December 6, 2024 at 10:03 PM
thepeoplelab.bsky.social
Congrats to our very own @maxspohn.bsky.social for winning 3rd place in the Best Student Poster Award at @sjdm-tweets.bsky.social! 🏆 His research shows how partisans misjudge out-group ideological complexity, fueling distrust. Insights offer ways to reduce animosity. maxspohn.com/IdeologicalC...
November 26, 2024 at 6:34 PM
thepeoplelab.bsky.social
Forget spreadsheets—today, we crunched the numbers on Sasha cuddles. Stats say 11/10! 🐶📈🐾

#EvidenceBasedPawlicy #RandomizedControlTails #NudgingWithNoses
November 13, 2024 at 5:37 PM
thepeoplelab.bsky.social
Big thanks to @danhonig.bsky.social for linking rigorous research with practice in "Mission Driven Bureaucrats." The government workforce thrives on empowerment and trust!

Learn more about Dan's work here -> danhonig.info
October 29, 2024 at 2:35 PM
thepeoplelab.bsky.social
How can governments break the status quo and nurture passionate public servants?

📅Find out on 10/28 as @danhonig.bsky.social discusses "Mission Driven Bureaucrats" with @elizabethlinos.bsky.social. Excited to co-host this event with HKS BSC!

Register here -> bsc.hks.harvard.edu/engage/event...
October 21, 2024 at 7:46 PM
thepeoplelab.bsky.social
Big thanks to the City of San Diego's Performance & Analytics Department 🐼 for hosting us last week! We're excited to collaborate on new projects that range from how to improve equitable engagement with city services and how to support front line workers! Exciting things ahead!
October 15, 2024 at 8:08 PM
thepeoplelab.bsky.social
So grateful to have leaders from top government fellowship programs: Foster America, Fuse Corps, Govern For America, and Public Rights Project visit us at TPL to talk about the future of talent in government and how to build trust in institution! 🏛️
May 2, 2024 at 9:25 PM
thepeoplelab.bsky.social
🔍 Simplified letters increased account registration by 9%, while those emphasizing social belonging saw an 11% boost. #HigherEducation #Nudges
April 25, 2024 at 2:46 PM
thepeoplelab.bsky.social
Check out our January 2024 newsletter, sent out today!

It includes a note from @elizabethlinos.bsky.social, a snapshot of research and publication highlights from 2023, and information on new jobs and ways to partner with us.
mailchi.mp/a12b497fd897...
January 12, 2024 at 8:46 PM
thepeoplelab.bsky.social
Our research suggests that government outreach still needs to focus on building #trust and #credibility if it is to be effective. 
Learn more about our research here: 6/6 www.peoplelab.hks.harvard.edu/project/does...
A screenshot of the project page for The Formality Effect on The People Lab's website.
November 27, 2023 at 8:22 PM
thepeoplelab.bsky.social
4️⃣. If the #FormalityEffect surprises you, you’re not alone. 351 scholars and practitioners surveyed prior to our experiments overwhelmingly predicted informality would be more effective. 5/
Results from a survey of 472 professionals recruited via social media, professional and academic networks. The cells indicate the percentage of participants who completed the full survey (N=351) who predicted that government communications with the corresponding attribute would be most effective at encouraging recipients to take the requested action. The participants were presented with binary choices for each attribute, and all attributes were presented in a random order. Column 2 reflects the subset of respondents who completed the survey and indicated that they worked in academia; column 3 reflects the subset of respondents who completed the survey and indicated that they worked in government.
November 27, 2023 at 8:21 PM
thepeoplelab.bsky.social
3️⃣. When we tested the effect of formal vs. informal communications in three real-life contexts, we found that residents were as much as 45% more likely to take a requested action if they received the formal communication. 4/
A table shows the results of field studies across three policy domains: business self-certification, local government service, and earned income tax credit. Formality effects of 25%, 45%, and 28% were found, respectively.
November 27, 2023 at 8:20 PM
thepeoplelab.bsky.social
2️⃣. Across three online experiments, we found that formal communications are considered more credible and the sender is assumed to be more competent, trustworthy, and more likely to be from the government. 3/
A table shows the results of online experiments that found that the sender of a formal communication is assumed to be more competent, trustworthy, and more likely to be from the government than the sender of an informal communication.
November 27, 2023 at 8:19 PM
thepeoplelab.bsky.social
🚨 NEW PAPER IN NATURE HUMAN BEHAVIOUR 🚨 

Across a range of contexts, we find a counterintuitive #FormalityEffect: Formal government communications are more effective than informal ones at shifting residents’ behavior. 1/ 🧵

With @karminker.bsky.social, Chris Larkin, Lindsay M. Moore.

rdcu.be/drRnl
November 27, 2023 at 8:16 PM
thepeoplelab.bsky.social
4️⃣Having a higher initial share of White male co-workers can also negatively affect the promotion rates of #BlackWomen. We show that Black women are 20.3 percentage points less likely to be promoted on time than White women, who have the highest #promotion rate in the firm. 5/
Similarly, the specification in Panel (b) is the same as in Table VI Column 3, except we interact, respectively, with the share of White male coworkers and the share of White female coworkers, instead of the share of White coworkers, and we control for the other race shares separately by gender. We therefore no longer control for the overall share of female. Plotted coefficients are the marginal effect of having either more White female (on the right) or White male (on the left) coworkers. For instance, for Black Male the plotted coefficient on the right panel is the sum of the coefficient on % White Female Coworkers, its interaction with Black, with Male, and with both Black and Male. Similarly, for Black Female the plotted coefficient on the right panel is the sum of the coefficient on % White Female Coworkers and its interaction with Black. All standard errors are robust and clustered at the office × year level. 95% confidence intervals are reported.
November 13, 2023 at 5:02 PM
thepeoplelab.bsky.social
3️⃣. We find that *both* White men AND White women negatively affect turnover rates for Black women. A one-standard-deviation increase in #BlackWomen's initial share of White co-workers is associated with a 10.6% increase in their 2-year turnover rate. 4/
This figure separates out the effect of having more White male (on the left) or White female (on the right)coworkers on turnover (Panel (a)) and promotion (Panel (b)). The specification in Panel (a) is the same as in Table V Column 3, except we interact, respectively, with the share of White male coworkers and the share of White female coworkers, instead of the share of White coworkers, and we control for the other race shares separately by gender. We therefore no longer control for the overall share of female.
November 13, 2023 at 5:02 PM
thepeoplelab.bsky.social
1️⃣. We first document a large racial turnover and promotion gap. Controlling for other factors, Black employees are 32% more likely to turn over within two years and 26% less likely to be promoted on time than White employees. 2/
A graph shows the fraction of employees who leave the firm by race. The x-axis is number of days since hire.
November 13, 2023 at 4:59 PM
thepeoplelab.bsky.social
🚨NEW WORKING PAPER🚨: Studying 9,000 new hires in a professional services firm, @elizabethlinos.bsky.social, Sanaz Mobasseri, and Nina Roussille find Black women's retention and promotion rates decline as their initial share of White coworkers increases. #econtwitter 1/
This paper investigates how having more White coworkers influences the subsequent retention and promotion of Black, Asian, and Hispanic women and men. Studying 9,037 new hires at a professional services firm, we first document large racial turnover and promotion gaps: even after controlling for observable characteristics, Black employees are 6.7 percentage points (32%) more likely to turn over within two years and 18.7 percentage points (26%) less likely to be promoted on time than their White counterparts. The largest turnover gap is between Black and White women, at 8.9 percentage points (51%). Drawing on conditional random assignment of new hires to initial project teams, we then show that a one standard deviation (14.0 percentage points) increase in the share of White coworkers is associated with a 10.6 percentage point increase in turnover for Black women.
November 13, 2023 at 4:56 PM
thepeoplelab.bsky.social
As part of the Bloomberg Harvard City Leadership Initiative's Talent Management course, we are thrilled to spend this week working with city leaders from across the world on some of the most pressing public sector workforce challenges.
November 1, 2023 at 6:22 PM