Yearly Archives: 2025

The Hockaday School Teaches Science With a Purpose

At The Hockaday School, middle school students combine science and social impact in a special project beginning in the seventh grade. 

Students visit the Joppa neighborhood in Dallas and partner with Joppy Momma’s Farm to test soil and water to offer remediation solutions. 

The eighth graders grow microgreens in class as part of their investigation of Earth’s biosphere and ecosystems. As students grow the greens, they learn the science of plant growth, practice scientific inquiry skills, and gain a deeper understanding of nutrients in food. Students study the stages of germination, the conditions necessary for germination, how to use various laboratory tools to collect data, how to formulate scientific conclusions based in evidence, and how to evaluate uncertainty in experimental results. Students explore how and why a seed germinates, including why water and exposure to air are important as well as how a germinating seed gets energy before it can get energy from photosynthesis.

Partnering with Hockaday’s Dr. William B. Dean Institute for Social Impact, students then donate the microgreens to Joppy Momma’s Farm, where farmers sell them to restaurants including Cafe Momentum and Asian Mint, owned by Hockaday alumna Nikky Phinyawatana, class of ‘96. 

Proceeds from the sale are used to expand the farm, which produces fresh, nutritious foods and makes it more readily available to the neighborhood. 

“It’s important for students to understand the connection between what they are learning and the real world,” said director of innovation and collaboration Laura Day. “This partnership does that and creates a social impact.”

Science Department chair Peggy Cagle sees the project fitting seamlessly within the curriculum.

“By growing microgreens for Joppy Momma’s Farm, the students are doing science with a purpose,” Cagle said. “Not only do they learn about the science of plant growth and practice inquiry skills, but they also see how their learning can have a positive impact. Our students also love watching the microgreens grow, then tend to them meticulously, and they are excited about the opportunity to visit local restaurants and taste their own products.”

Gender balance in computer science and engineering is improving at elite universities but getting worse elsewhere

The share of computer science and engineering degrees going to women has increased at the most selective American universities over the past 20 years and is approaching gender parity, while the proportion has declined at less selective schools. Those are the main findings of a study my colleague and I recently published in the journal Science.

Jo R. King and I analyzed over 34 million bachelor’s degrees awarded by nearly 1,600 American universities from 2002 to 2022 – data covering almost all bachelors-degree-granting institutions in the U.S. We wanted to identify which factors best predict parity among men and women in physics, engineering and computer science majors.

We focused on the ratio of how many physics, engineering and computer science degrees men earned out of the total degrees they earned across all majors, relative to the corresponding ratio for women. A university’s average math SAT score among admitted students emerged as the strongest predictor of the relative ratio – the two ratios compared – and its importance has grown over time. SAT scores range from 200 to 800.

At universities with average math SAT scores of about 770 – which tends to be the standard at only the most elite, math-focused U.S. schools, such as the Massachusetts Institute of Technology, the California Institute of Technology and Harvey Mudd College – men earned these degrees at 1.5 times the rate of women in 2022, an improvement from 2.2 in 2002.

In schools with math SAT scores around 560, the ratio hasn’t changed – men earned physics, engineering and computer science degrees at five times the rate of women in 2022, the same as 20 years earlier. At less selective schools with math SAT averages around 450, the ratio was even more uneven. Men earned these degrees at 7.1 times the rate of women in 2022, more than double the 3.5 ratio from 2002.

Overall, across all schools, men earned degrees in these fields at 4.4 times the rate for women in 2022, a slight improvement from the 4.6 rate in 2002. This rate is very different from the near parity in gender found in other science, technology, engineering and math fields.

Previous studies showed that individual student-level factors influence success in these fields. These include test scores in non-STEM subjects, confidence in one’s STEM abilities, intentions to major in physics, engineering and computer science and career aspirations.

We examined two additional nationally representative datasets to test whether these personal factors explained the patterns across schools. They did not. Even when men and women began college with similar achievement, interests and aspirations, those at less selective schools showed much wider gender gaps.

The most selective universities, such as the Massachusetts Institute of Technology, saw the greatest improvement in closing the gender gap in computer science and engineering.
AP Photo/Charles Krupa

Why it matters

The college majors of physics, engineering and computer science are some of the most gender-imbalanced majors in U.S. colleges today.

This divide has serious consequences for gender equality and economic opportunity. Male and female graduates in technical fields, even from less selective schools, earn higher salaries. When women at these schools don’t complete these degrees, they miss valuable economic benefits.

Our findings raise particular concerns for women of color. They more often attend schools where gender gaps among STEM majors continue to widen. This pattern compounds existing inequalities and limits diversity in fields where different perspectives drive innovation.

Our research suggests that current efforts to increase women’s participation focus too heavily on elite schools. While these universities show important progress, I believe it’s important to adapt successful strategies to all types of institutions.

What still isn’t known

Although we have identified a new pattern and have shown when and where gender gaps are improving or worsening, we cannot say exactly why some institutions have been more successful than others in closing these gaps. We need more research to understand what specific practices or environmental factors at high-performing schools are making the difference.

We also need to better understand how early educational experiences and social factors influence women’s choices about pursuing physics, engineering and computer science degrees at different types of institutions. We do know that gendered stereotypes linking men to math, physics, engineering and computer science develop early.

Additionally, more research is needed on the role of faculty diversity and mentorship programs in supporting women’s success in these fields.

Finally, women are severely underrepresented in physics, engineering and computer science fields around the world. We don’t know whether similar patterns regarding university selectivity exist internationally.

The Research Brief is a short take on interesting academic work.

3 ways Trump’s EPA could use the language of science to weaken pollution controls

Environmental issues were conspicuously absent from the 2024 U.S. presidential campaign, but moves by President-elect Donald Trump’s first administration and his leadership picks for his next administration offer clues to what may be ahead.

They point to a second Trump administration likely loosening regulations on industries, particularly oil, gas and petrochemicals, giving them wider permission to pollute.

Some actions will be overt. But history suggests this administration may also try to use the language of science – terms such as transparency, citizen science and uncertainty – to weaken environmental and health protections and write regulations more favorable to industry.

Those ideas surfaced during the first Trump administration and in conservative agendas such as Project 2025. Project 2025 was written by former Trump administration officials, including several people Trump has tapped for his next administration. Trump distanced himself from the project during the campaign but now says he agrees with many parts of it.

As president and a candidate, Donald Trump proposed loosening restrictions on fossil fuel industries. The phrase ‘Drill baby, drill,’ on a screen behind him during a talk on Oct. 14, 2024, reflects that.
AP Photo/Alex Brandon

I followed the first Trump administration closely as a researcher involved in the Environmental Data & Governance Initiative, or EDGI. The group was founded in 2016 to document Trump’s efforts to dismantle the Environmental Protection Agency. During Trump’s first administration, we archived climate and environment datasets used by scientists, advocates and policymakers who were worried these might be hidden by the administration. We also tracked how the Trump administration changed climate language on agency websites.

EDGI also interviewed agency staffers facing political pressure and explained the potential impact of policy shifts and rule changes.

Here are three ways the second Trump administration could try to use the language of science to write policies that sound beneficial but could have profound effects on environmental health.

1. ‘Strengthening transparency’ to block use of health data

When you hear words such as “transparency” or “open source,” they probably sound positive – the idea is that all of the parts can be seen and checked.

But would you want your health records open for anyone to see? The privacy of health care records was at the heart of a debate over a policy the first Trump administration created called “Strengthening Transparency in Pivotal Science Underlying Significant Regulation Actions,” or the “secret science” rule.

The rule could have prevented the government from considering important health research in setting pollution limits.

Decades of health data gathered from people across the United States have shown how air pollution from power plants and other sources can contribute to cancer and other illnesses. That data has provided the evidence for regulations that have cleaned up the nation’s air and water for a healthier environment.

But the raw data from those studies can’t be made public because it involves people’s personal health records. The rule the EPA finalized during the last weeks of the first Trump administration required the agency to give less consideration to studies if the underlying data wasn’t publicly available. A court vacated the rule on Feb. 1, 2021.

I expect Trump’s EPA will try again to require that the agency’s rules have a basis in published raw data. The Project 2025 agenda calls for “true transparency” to be a defining characteristic of the EPA, including “the establishment of open source science.” That would limit the use of private health data or data whose use is licensed by companies. This would make it harder to develop rules protecting public health.

2. Boosting public scrutiny of the EPA

The author of Project 2025’s chapter on the EPA was Mandy Gunasekara, who served as chief of staff to Trump’s EPA administrator in the first administration. Apart from transparency, Gunasekara has also pitched “citizen science” as a way to “deputize the public to subject the agency’s science to greater scrutiny.”

At its best, citizen science is an important way for the public to ensure research reflects their interests and experiences. At its worst, citizen science is used to delay meaningful actions.

Who benefits from “deputizing the public” to scrutinize EPA science depends on who has access to information and the resources to engage. Wealthy industries and private interests may gain a greater voice, while the communities most affected by pollution remain sidelined, particularly if government makes the EPA’s science difficult to find.

Project 2025 also calls for resetting the composition of the EPA’s advisory boards – and even suspending some of them. These boards consider feedback from industry, academia and communities. Similar actions during Trump’s first administration reduced the number of academics and representatives of nongovernmental organizations on these boards, while increasing the number of industry consultants.

3. Using uncertainty to avoid regulation

Uncertainty is another important scientific term that the first Trump administration used to promote deregulation, particularly for chemicals.

When the EPA studies chemicals, there are uncertainties around the health effects at different levels and types of exposure. A precautionary approach assumes that chemicals have adverse effects at low doses and that those effects grow as exposures increase or accumulate. Many scientists consider precaution a safer bet when not enough is known about the chemicals’ effects.

Yet some chemicals may not actually cause harm until they reach a certain threshold. In the view of the chemical industry, that means a “better safe than sorry” approach can be wrong. Instead, the industry says, chemical regulation should be based on the best available science. However, the best available science on chemicals is often inconclusive. In the absence of a precautionary approach, the industry’s argument for the “best available science” could actually mean less justification for regulation.

Project 2025 proposes that, on his first day in office, Trump should issue an order to “reject precautionary default models and uncertainty factors” that “drive flawed and opaque decisions.”

The consequences could include EPA analyses that understate the risks of toxic chemicals when research is still emerging, such as with PFAS.

What’s next?

Our team at EDGI is working with partners to again identify federal web pages and datasets vulnerable to removal, modification or attrition. That allows us to sound the alarm if these resources for tracking and addressing climate and environmental change go missing. We believe watchdog strategies made political appointees hesitant to order more changes during the first Trump administration.

I don’t think Trump’s next EPA will be straightforwardly “anti-science.” I believe, however, that it will use language that appears to boost research openness and citizen participation but that is aimed at undermining policies that protect human health and the environment.