Q Cheng
qiqicheng.bsky.social
Q Cheng
@qiqicheng.bsky.social
Quantitative Research Associate @ #BeeWell | Uni of Manchester 🐝
Exploring adolescent mental health & wellbeing 📊
Causal Inference | Applied Econometrics 🌍
Great story in @theguardian.com sharing the latest @beewelluk.bsky.social @uomseed.bsky.social research looking at prevalence & impact of isolation in schools.

More detail in here 👉 tinyurl.com/yffvjp4x

@erthornton.bsky.social @neilhumphrey.bsky.social @olademk.bsky.social @qiqicheng.bsky.social
October 23, 2025 at 9:48 AM
Reposted by Q Cheng
"If we get this right, it will be the most important thing we have done in this Bill - but, at the moment, there is no measurement,” said Lord Gus O’Donnell in yesterday’s Children’s Wellbeing and Schools Bill debate.

🔗 www.ourwellbeingourvoice.org
🎥 Watch his speech below 👇
September 17, 2025 at 2:19 PM
🏳️‍🌈 We know LGBTQIA+ youth urgently need effective mental health support. Our study provides an independent evidence that "Free2B," a 1:1 mentoring service, significantly improves their mental wellbeing. A very promising intervention! doi.org/10.1002/jad....
#LGBTQ #YouthMentalHealth #AcademicSky
September 11, 2025 at 8:59 AM
Reposted by Q Cheng
LGBTQ+ young people in the UK face a severe, unequal mental health crisis.

Services like Free2Talk offer life-changing, cost-effective help, but they need greater support. 1/7

pbe.co.uk/publications...
August 27, 2025 at 8:03 AM
Reposted by Q Cheng
Thank you to everyone that got involved in our event yesterday to explore the latest #BeeWell findings from Greater Manchester!

It was fantastic to work with partners and young people to think about next steps to take in response to the latest data.
April 17, 2025 at 12:52 PM
Reposted by Q Cheng
To celebrate the launch of our latest #BeeWell headline findings in Greater Manchester (GM), we are sharing a new blog on youth-led action in response to #BeeWell, from across every local authority in GM!🧵

@bigchangecharity.bsky.social
April 1, 2025 at 10:09 AM
Reposted by Q Cheng
link 📈🤖
Quasi-Bayes in Latent Variable Models () Latent variable models are widely used to account for unobserved determinants
of economic behavior. Traditional nonparametric methods to estimate latent
heterogeneity do not scale well into multidimensional settings. Distributional
restrictions all
February 22, 2025 at 3:06 AM
Reposted by Q Cheng
“The inclusion of young people with SEND’s voices is not just a matter of good practice – it is a fundamental right...”

A new #BeeWell blog highlights the importance of involving young people with special educational needs and disabilities (SEND) in decision making.
February 6, 2025 at 9:52 AM
Reposted by Q Cheng
New #beewell open access registered report on the longitudinal links between sleep, physical activity and wellbeing in adolescence: link.springer.com/article/10.1...
Longitudinal relationships across sleep, physical activity, and mental wellbeing in early-to-mid-adolescence: a developmental cascades investigation - Quality of Life Research
Purpose Sleep (SL), physical activity (PA), and wellbeing (WB) are three factors linked to positive development in adolescence. Despite theoretical support and some empirical evidence of developmental...
link.springer.com
January 28, 2025 at 7:01 PM
Reposted by Q Cheng
We still have a relatively poor understanding of the relationship between evidence and policy. Program evaluation in particular is often motivated by a desire to make policy better. But how effective is program evaluation itself?Michelle Rao's JMP tackles this question. www.michellerao.com/research
November 27, 2024 at 5:42 AM
Reposted by Q Cheng
1. Leave the data in levels -- no differencing. Run the pooled OLS regression across the two time periods:

Y(i,t) on W(i,t), W(i,t)*[X(i) - X1_bar], 1, D(i), X(i), D(i)*[X(i) - X1_bar], f2_t, f2_t*[X(i) - X1_bar]

Coef on W(i,t) = D(i)*f2_t is the same ATT estimate.
Consider the simplest DiD setup with panel data. T = 2, controls X(i) don't change over time. D(i) is the "ever treated" indicator, f2(t) the second period time dummy. W(i,t) the time-varying treatment W(i,t) = D(i)*f2(t). The X(i) appear flexibly to allow selection and heterogeneous trends.
November 21, 2024 at 3:42 PM