From Gender Inequality to Dementia Care: Emma Zang on the Social Forces That Shape Our Lives

Rick Harrison
Emma Zang outside the sociology building

Emma Zang studies how institutions, policies, and social environments shape people’s lives across the life course.

Her questions often begin with lived experience, but she uses population-level data and rigorous quantitative methods to determine how broadly those experiences apply.

As an associate professor of sociology, biostatistics, and global affairs, and a faculty fellow with the Institution for Social and Policy Studies, Zang helped launch ISPS’s popular Population Studies Workshop.

We recently spoke with her about her work and her life, and how the two intersect to advance our understanding of health and aging, family demography, and inequality. This conversation has been edited and condensed.

How did growing up an only child in the northern part of China shape the questions you study today?

My mother used to work for the Census Bureau, and then later she worked as an accountant. After she was laid off because of economic reforms, she started her own small business, a job agency mostly serving rural migrants moving to cities to find jobs. My father was born in rural China and worked as an engineer at a state-owned enterprise. Because of the economic reform, the company’s profits were very bad, so for many years of my childhood, my mother was the breadwinner of the family. 

Looking back, I think that has something to do with how I think about women’s roles and ambition. I had this idea very early in my life that women need to be independent. I learned early that economic independence gives women greater security and agency. My mom’s success gave me a lot of confidence that, given real opportunities, women can build fulfilling, responsible, and independent lives. That was the early root of my research focusing on gender inequality. 

You studied law, sociology, economics, statistics, and public policy. How did that broad academic background shape your approach?

My mom thought being a lawyer would be a good job, and she correctly noticed that divorce rates were increasing in China. She thought being a divorce lawyer would be a good career, and she was right.

I became interested in family law, partly because of my mom’s advice, but also because I was interested in family and gender issues. When I was taking law courses, I noticed a pattern. Many provisions appeared formally gender-neutral but produced gendered outcomes in implementation. In divorce cases, women frequently received less favorable property-division outcomes. That sparked my curiosity. I wanted to know more about culture, social norms, and why those outcomes happened. 

And that motivated you to take sociology courses?

Yes. Then I started using sociological knowledge to think about many questions, not only family and gender. Coming from an average family in northern China and going to college in Shanghai, I had been working so hard. And I kept wondering: Would being more successful make me happier? Intuitively, I knew that making more money or being more successful probably would not make me happier, but I was stressed about being perfect and working hard all the time. 

What clarified this for you?

I wanted something objective. I wanted evidence that could test my intuitions systematically. That is how I became interested in quantitative approaches, using nationally representative data to answer social questions. That motivated me to study statistics and econometrics. All of it originated from my desire to do quantitative social science research.

What connects your work on divorce law, sibling spillovers, fertility, remote work, aging, and inequality?

My personal life often gives me the initial question and motivation, but the research quickly becomes much larger than my own life. My undergraduate thesis, on whether upward mobility makes people happier, was motivated by the concerns I had at that time. Later, during my Ph.D., people around me started marrying and thinking about children, and I began working on divorce law in China and how it affected inequality in property division after divorce. I also did work analyzing U.S. fertility trends by educational attainment and race.

More recently, I became interested in remote work because my husband was working from home. When we first moved to New Haven in 2019, he worked in finance, and there are not many finance jobs in New Haven. Because we wanted to avoid a long commute, he arranged to work remotely before the pandemic. That made me interested in remote work even before COVID hit. Then when the pandemic hit, the question suddenly became urgent for millions of families.

Now I have multiple projects looking at how flexible work arrangements shape family dynamics and gender inequality. 

What drew you into research on dementia, cognitive aging, and caregivers’ experiences?

This one was personal too. My grandmother was diagnosed with dementia. At the beginning, some people wondered how cognitive aging was part of sociology. But cognitive aging has become an increasingly important subfield. Aging and health are so important, and there is a lot of demand for research. Dementia, in particular, is an area where the NIH and the federal government have invested heavily because there is still no cure, existing treatments help only some patients and do not reverse the disease, it is very expensive, many people have dementia, and it places a lot of burden on caregivers and social systems. 

What can sociologists do to help?

Sociologists can add a lot beyond the medical aspects. Social determinants of health can help us understand how modifiable risk factors — such as social environments, educational attainment, resources, and family dynamics — shape the experience of people with dementia and their caregivers. I started working on dementia caregiving and caregivers’ experiences, and that has led to multiple NIH-funded projects.

How would you describe the distinctive approach of your lab?

I would categorize much of my research around two approaches. The first is population-based research. Most of our questions call for observational and quasi-experimental designs rather than randomized experiments. Most of my research uses population-level data, meaning complete population registers where they are available. Or nationally representative surveys, so the estimates can represent the population we are interested in. We care a lot about external validity, in addition to internal validity. 

Meaning, you want your findings to be generalized to different people in different settings.

That’s right. The second approach is the life-course approach. We do not only look at what is happening right now or what happened three or four years ago. We look at situations from when someone was born through their life course. We care a lot about life-course exposure.

Can you give me a recent example of why that’s valuable?

My lab is launching an ambitious project to get people’s residential addresses from the year they were born up to where they currently live, working within secure, de-identified data environments. Then we can link their life histories to geographic contexts, such as policy exposure, neighborhood disadvantage, built environment, natural environment, socioeconomic resources, and county characteristics. Once we have people’s addresses, we can trace the social context a person lived through across their whole life.

What do you mean by their social context? What sorts of questions might you ask?

For example, we can examine cumulative exposure to things like policy liberalism or pollution across a person’s life course and how that affects current cognition. We can ask: What is the critical timing? Is childhood the most important period, or is later life more important? Does it matter if someone moves to better and better places, or worse and worse places, even if the overall exposure level is the same? Those are the questions we are trying to answer.

You are also using AI to build new measures of neighborhood disadvantage. What does that project involve?

We recently received funding to build a new neighborhood disadvantage index using AI to incorporate different kinds of data sources. Traditionally, most neighborhood disadvantage indexes use census data or American Community Survey data, which focus on socioeconomic resources. But now we have remote sensing data, Google Street View data, and other sources that provide much more information about neighborhoods. 

Emma Zang speaks at a lectern in a classroom

Emma Zang delivers a presentation about how she has been using AI for her research.

What sort of data might you get from Google Street View?

With Google Street View, you can see what is on the street. You can see whether there is a broken window in a building, whether there is trash on the street. We can capture a lot about the built environment and physical disorder using image data. From remote sensing data, we can see land use, tree cover, and the condition of roofs.

We are starting the project in New Haven, using three data sources: census data, Google Street View, and remote sensing, meaning satellite data. Once we have a proof-of-concept paper, we hope to add more types of data, including qualitative interviews and media data, so we can incorporate people’s perceptions of neighborhood safety and other qualitative evidence into the index. 

How is AI changing the possibilities for social science research?

I am very interested in AI right now. I have several projects exploring how AI can help us conduct better social science. So far, my experience has been very positive. AI can accelerate a lot of the tedious tasks researchers used to do, though the outputs need careful validation. It can also help us integrate multiple types of data, such as qualitative data, image data, and satellite data. 

How do you know if you can trust the data supplied through AI tools?

Some of our methodological projects are trying to detect the boundaries of AI’s capacities to do responsible research. We are not simply applying AI. We are trying to see where AI is reliable, where it is not reliable, and what the potential ethical considerations are. We are trying to make clearer the boundaries of using AI in social science research. 

One of your recent papers examines sex-selective abortion bans and infant health among Asian immigrant mothers. What motivated that study?

That project is related to my increasing interest in the lived experiences of Chinese immigrants and Chinese Americans in the United States. When I was living in China, I was more interested in things happening in China. I still am, but I have also become increasingly interested in Chinese immigrant and Chinese American experiences in the United States, especially because the U.S.-China relationship has made things more difficult for Chinese communities in recent years.

The sex-selective abortion ban paper was one of my first efforts to examine more recent discriminatory laws or policy narratives affecting Chinese and Asian communities in the U.S. The laws are abortion bans, but the narratives used to justify them have an obvious discriminatory flavor. Advocates often argued that immigrants from China, India, or other Asian countries generally bring a culture that does not value girls and women and therefore abort female babies. But if you look at the data, there is no solid evidence that Chinese immigrants in the U.S. commonly abort female babies. 

This kind of framing is harmful because many Americans have limited understanding of Chinese communities or immigrants from China. It fits into stereotypes, so people may believe it. In the paper, we found that this kind of narrative can increase stress among Asian immigrants in those states. Using a quasi-experimental design, we found effects on Asian immigrant birth outcomes, including birth weight and gestational age. The magnitude was not huge, but we did find an effect.

What does that study suggest about the wider consequences of policy?

Laws can cause harm even when they do not directly affect every person in a group. The framing of a law can make people feel socially isolated or singled out. That stress can matter. These outcomes — birth weight and gestational age — are commonly used to examine stress-related mechanisms from policy. So, the broader point is that public narratives and policy environments can affect people’s health and well-being, even indirectly.

You have also studied upward mobility, race, gender, and young adults’ health. What did that work show?

That paper is one of the pieces I am proud of because it was a collaboration with a Yale undergraduate, Melissa Tien. A lot of research looks at gender differences or racial differences, but more research now emphasizes intersectionality among race, gender, socioeconomic status, nativity, and other dimensions. In that paper, we focused on race and gender. 

The stereotypes and challenges Black men face can be very different from those Black women face, and the same is true for white men and white women. We looked specifically at health behaviors. In our sample, mental health problems, for example, are often more associated with women, especially white women, while substance use is more common among white men. For Black men, the health behaviors we saw were more related to risky behaviors and physical harm. For Black women, chronic conditions, which are relatively rare among young people, were relatively high.

Why do you want to look at these problems in this way?

These findings highlight the importance of looking at intersectionality rather than grouping everyone of the same race together or everyone of the same gender together. It is important to pay attention to heterogeneity. In social science research, we often focus on average treatment effects, but we can overlook how a policy, treatment, or context affects different groups differently.

Including location, right?

Yes. The paper also looked at county-level upward mobility context. The idea is that a “good place” is not always good for every group of people. For most health behaviors, growing up in better counties is generally good. But drinking behavior was worse in relatively affluent counties, possibly because richer places have more restaurants and bars and more opportunities for social drinking. Reality is often more complicated than our hypotheses.

Curiosity-driven research has real value, but I am not satisfied with only that. Hopefully, my research can be useful for policymaking, or at least affect people’s opinions and help people better understand social phenomena. 

—

Emma Zang

Your work crosses sociology, medicine, public health, environmental science, economics, and other fields. What have you learned from interdisciplinary collaboration?

I am a very interdisciplinary person. My Ph.D. is in public policy, and one reason I chose public policy instead of sociology was that I liked interdisciplinary training. Different fields offer different perspectives on the same types of questions. Over time, the questions that economists, sociologists, and political scientists are interested in are increasingly converging, and the tools we use are also converging. We are all using causal inference, statistics, AI, and computational social science. There is really no reason not to collaborate. 

This is one of the advantages of ISPS. There are a lot of opportunities to share ideas.

Collaboration is fun, and it is also one of the best ways for me to learn. I cannot take courses all the time anymore, but by working with colleagues in statistics or medicine, I’ve learned a lot about biological aging and new statistical tools. 

You recently helped launch the Yale Population Studies Workshop at ISPS. Why was that important to you?

I consider myself a population scientist. Demography or population science is an interdisciplinary field. Historically, formal demography focused on fertility, mortality, and migration, the three major elements shaping population composition. But modern demography has expanded to include social demography, family demography, health, aging, and many other fields. 

Population centers are common on many major campuses, but Yale did not have a population seminar or population center, which was odd. I wanted to do this for a long time, but I waited until I got tenure. After that, I talked with Alan Gerber, and he was very supportive of building something new. His request was that we have co-organizers across disciplines, which made sense.

We brought in colleagues from sociology, political science, public health, and other parts of Yale. People have been very enthusiastic. People from public health and anthropology told me they had not realized population studies could include so many different topics, including the work they were already doing. 

Your research has received attention from news outlets and policy organizations. Why is it important to communicate with the public?

I think it is very important to do translational research, especially in social science. What is the purpose of doing research? Curiosity-driven research has real value, but I am not satisfied with only that. With the resources and training I have received, I feel I should do bigger things. Hopefully, my research can be useful for policymaking, or at least affect people’s opinions and help people better understand social phenomena. 

Do you have any advice for other researchers who might want to get their work out to a larger audience?

I always say yes when a legitimate reporter contacts me. Not every researcher wants to make time to talk with reporters, and some colleagues feel nervous doing it. I do not feel that way. I feel I can connect with reporters very well, and I do not like to use jargon. I try to explain findings in clear, nontechnical language, and that helps people understand them. 

And you often relate your work to common issues.

Most of my research topics are related to real-life problems. Remote work and gender inequality affect many families. Dementia affects many families. Some of my research has been funded by the Alzheimer’s Association, and sometimes they invite me to advocacy events. When families affected by dementia come talk to me, you can see in their eyes that they really care. They appreciate the research. Those moments motivate me and make me proud of the work I am doing.

What do you see as the most pressing challenges in addressing health inequality today?

First, I think we need to get more people on the same page that health inequality is actually a problem. These days, for many reasons, some people do not think inequality is the most important issue. Health disparities research has become politicized. But the bottom line is that we want to make sure everybody has a decent life and basic health needs are met. Even for people who do not care about inequality or social justice as concepts, basic health needs are a matter of human rights. 

If you think from that perspective, in a high-income country like the United States, we should focus on disadvantaged groups who do not have adequate access to health care. We still need public narratives and policy discussions that align people around that goal.

The second issue is research funding. Cutting research funding will not help reduce health inequality or improve health outcomes for people in America. The U.S. is still a leader in health care and medical research because it attracts high-quality researchers and funds high-quality scientific work. If we cut funding for health research, we harm one of the country’s greatest strengths.

The third issue is translational work. Researchers need to communicate findings to the public and policymakers so people can see the value of investing in research. It is also important to connect better with communities. That is one goal of our New Haven-based AI project. We live in New Haven, and as an institution, we should try to contribute to the community by better understanding neighborhood needs, resources, disadvantages, and challenges.

What changes would you like to see in how society addresses aging and inequality?

There are many things I would like to see. On the medical side, I hope people can find treatments for dementia. That would be terrific. I also hope there will be better treatments for other cognitive and neurological conditions. 

From a social perspective, I hope we can de-stigmatize many so-called diseases and disabilities. Disability is shaped by impairment together with the social and physical environment. It means that a person is unable to perform certain functions given the current environment. But if we change the environment, for example, replacing stairs with ramps or making buildings more accessible, many people with physical disabilities would not have the same problems accessing spaces. I also hope there will be less ageism, so people are less nervous about getting old or having wrinkles. We need new ways of thinking about aging, gender, race, and disability that allow people to live more fully. 

Looking ahead, what do you hope your research legacy will be?

For me, personally, I deeply care about disadvantaged groups, including immigrants, women, and people from poor families. Helping them have happier lives, more fulfilling lives, and more support is my personal ambition in doing research.

That is why I am happy I ended up in sociology. Beyond the theory and methods, I feel that most people in the field deeply care about disadvantaged groups. But I also want to figure out how to really help them, not only as a matter of moral commitment. There are economic constraints, budget issues, and practical issues. I want to conduct responsible research that can tangibly benefit these groups of people.