This week we look at how the roles we play shape our lives.

You want to support my work but don't need a keynote from a mad scientist? Please become a paid subscriber to this newsletter and recommend to friends!

Research Round-up

Cultural Transmission: Inheriting Human Capital

Our parents and peers define roles we will play for the rest of our lives.

Modern socioeconomic gaps stem less from financial inheritance and more from the cultural transmission of human capacity. For example, individuals with PhDs are much more likely to have parents with PhDs. And this gap is even greater for PhDs in economics. Highly educated parents disproportionately invest in the educational success of their children—a rational economic choice if I’ve ever heard one, but a choice that crowds out exceptional students lacking such inheritance.

Data from Norway further reveals how financial inheritance, while impactful for the ultra-wealthy, plays a surprisingly small role in overall societal inequality.  Even with counterfactual scenarios removing financial inheritance entirely, the overall distribution of wealth remained largely unchanged. Again, it’s the cultural transmission of knowledge and opportunities, not just money, that drives the chasm.

The fortunate few inherit powerful societal roles to play throughout their lives; the unfortunate must (and can!) carve their own roles from cultural enrichment and defiant imagination. The next research points at how.

Dungeon & Dragons for Diplomats & Doctors

The random circumstances of our birth can assign us lifelong roles—both internalized and externally reinforced—that limit who we are. Here’s new research on breaking out and becoming more.

Limited cultural enrichment can trap young people throughout the world into limiting social roles. In contrast, students from regions with greater cultural diversity and a history of intermingling are more likely to become social brokers in new communities, bridging gaps and strengthening connections. This intentional exposure to diversity fosters social adaptability and openness-to-change, both substantial predictors of income, health, and wellbeing across a lifetime.

Fighting internalized limits can be as “simple” as the power of imagination…and a little old school role-playing. Young girls who pretend to be successful female scientists persisted significantly longer in challenging science activities than girls simply told about successful women. In fact, these girl scientist Groundlings completely erased the persistence gaps with boys.

Let’s be clear what I’m saying here: D&D + diversity = dreamy destiny!

Weekly Indulgence

Quick: name all of the amazing people working in AI! I bet that list includes a bunch of guys that never coded a model in their life. We’re going to change that at Ellevate Network’s 2024 #MobilizeWomen

Come hear me rant, rave, and wax nerdy for one of the keynotes—then I’ll panel up on women breaking through in AI.

Best of all…this conference is free! Grab your ticket today: https://ringcentr.al/4cFUPyN

Stage & Screen

• June 5, Online: Mobilize Women Summit I'll talk about starting my company to end postpartum depression.
• June 18, Stockholm: Hyper Island is host an AMA for me in Sweden!
• June 19, Stockholm: Buy tickets for the Future of Talent Summit and so much more!
• June 20-21, Amsterdam: TNW ...well, I don't know exactly what I'll be talking about, but it will be huge!
• June 21, Leeds, UK: Society of Otolaryngologists What else: changing education for doctors in an AI-rich world.

Find more upcoming talks, interviews, and other events on my Events Page.

If your company, university, or conference just happen to be in one of the above locations and want the "best keynote I've ever heard" (shockingly spoken by multiple audiences last year)?

<<Please support my work: book me for a keynote or briefing!>>

SciFi, Fantasy, & Me

Of course I'll say something about Dune Part 2, but who's going to watch it or not because of my comments. So...next week.

This week I watched the first 5 episodes of Dark Matter. I didn't remember having read the book back when it came, but with a refresher from Wikipedia I recall the criticism that both the character development and science were rather shallow. In the Apple TV series, they have given the characters more time, which I appreciate. The science, however, is still the quantum mechanics version of "you only use 15% of your brain." Despite that, it was a fun book that is turned into a slowish but still engaging show.

If you want a Dark Matter with no pretense of science but just some adventurous antihero shenanigans, give the entirely unrelated 2015 series of the same name a try.

Excerpt: "Network Effects"

A common narrative in the gig economy is that it increases economic dynamism by providing new opportunities to people struggling within the traditional economy. The greatest pride at ShiftGig was when a ShiftGiggler was “stolen” away by a client for a full-time job. ShiftGig’s clients would never have brought in a young woman with no college degree or previous work experience, but when that same woman spent three months side-by-side with their full-time employees they asked her to stay. ShiftGig’s entire office would buzz for days every time this happened.

While stories like this were an enormous emotional lift at ShiftGig, anecdotes are not science. Unlike Uber where a gig worker either drives or they don’t, ShiftGig had a hierarchy of job titles with increasing pay matching increasing responsibilities. If gig work provides a dynamic lift for workers, then ShiftGig’s internal pay ladder would be the perfect opportunity to track that growth. So, I had the data science team set about measuring the historical impact of gig work on the lives of ShiftGigglers. We made three important discoveries.

First, the flexibility of gig work is transformative for single mothers struggling with traditional employment. A common story for this population was an endless cycle of new job, family emergency, firing… new job, family emergency, firing. The ability to opt in or out of a specific gig when they expected to be available freed these working moms from that cycle of trading off between family and work. Nearly a third of ShiftGig’s workforce came from this population and the majority of them would not otherwise have been able to earn a steady income.

Second, there was a wild disparity in work hours within the workforce. A very small number of ShiftGigglers captured nearly all the shifts, while the majority didn’t get any gigs and left the app. ShiftGig desperately wanted to bring more workers into their system, but unreliable workers meant clients would cancel contracts. So, the community service managers, who handled the relationship with the client, would horde shifts for the workers they knew were the most reliable. Workers who joined ShiftGig’s network early captured all of the opportunities to demonstrate their reliability, making it difficult for new people in the network to connect to shifts. Even in the lowest-skilled labor market, there was still something of a “creative–service” split along the single dimension of reliability.

Third, and most disheartening, after two years completing shifts for ShiftGig, workers’ average earnings increased by only $0.02/hour. While single moms found work they might never have had and a small number of ShiftGigglers landed full-time jobs, the majority of people in the network found themselves no better off than where they started. Despite its internal pay ladder, for the average worker the fabled dynamism failed to emerge.

Interestingly, the wild disparity in work hours mirrors a phenomenon I’ve seen many times across many domains: network capture. One of the clearest examples of this was at the MOOC giant, Coursera. As they were searching for strategies to monetize their growing student population, one obvious market was job placement. More than 30,000 students enrolled in a single session of co-founder Andrew Ng's machine learning course. This is an enticing population for employers desperate to find talented data scientists and machine learning engineers. Internal research at Coursera had identified that the single best predictor of employment success for its students was their participation in online discussions. The best future employees by far started as peer tutors, answering other students’ questions about the course material. Coursera just needed to identify those peer tutors (using some of Andrew’s own natural language processing algorithms) and potential employers would pay them for recommendations.

Read more about how none of this worked out for the better when How to Robot-Proof Yourself hits the shelves!

Vivienne L'Ecuyer Ming