The Crumbling
Posts
Ted Cruz's "Woke" Grant List Misleads the Public About Federally Funded Science

Ted Cruz's "Woke" Grant List Misleads the Public About Federally Funded Science

A story about how lazy data mining further erodes the public's sense of reality

Sophia Florence & Damian
March 03, 2025

In October 2024, Ted Cruz and others in the U.S. Senate Committee on Commerce, Science, and Transportation published a large document that they suggested showed rampant abuse of government funding by the Biden administration. They claim that money is being funneled into divisive “DEI,” or “diversity, equity, and inclusion,” instead of standard scientific research. They highlighted over 3,400 research grants, and created a list of hundreds of words and phrases, such as “race,” “Latinx,” “equity,” “underserved,” “Black,” “female,” and of course, woke classic, “climate change.”

Each grant was given labels of supporting either concepts of race, social justice, environmental justice, gender, or status. The document explaining the results of their searching, which Cruz takes full credit for, gives some detailed description of these categories, shown below.

Image 1 - The guidelines for how Cruz and his collaborators labeled projects

His accusation, explicitly in the document, is that the grants are given to work that only focuses on these categories, with no actual scientific merit. However, these are National Science Foundation grants. Anyone with passing familiarity with that agency should realize instantly that something isn't adding up here.

NPR and other outlets covered this.Their articles examined some specific proposals, but we felt that something was missing - a real thorough look at the data. Cruz used a (clearly very unsophisticated, judging by the results) Python program to scrape the data, so why not use one of our own to do a more thorough analysis?

We also chose a wide variety of grants to look over. Our new research staff member, Damian, went through a large set of random grants manually. We both found a lot of evidence that this is just lazy work or deliberate disinformation...well…the kind of work you'd expect from Ted Cruz.

Without further ado.

Damian’s Comments:

I only got through about 20 spot-checks in the list and I started to feel like I was beating a dead horse. None of the grants are specifically “DEI” research. Nothing here is “gender studies”. The conflation of climate science with “wokeness” is specious at best and absurd.

It’s impossible to conclude that this is a serious investigation of fraud, waste and abuse. The only way to read this is as an attack on our nation’s scientific infrastructure at large. It is an escalation and broadening of right wing attacks on science that have occurred for decades, firmly in the tradition of the second Bush administration and the Reagan administration.

The shoddily constructed methods of this “investigation” are not designed to ferret out waste and abuse. They are designed to indiscriminately catch researchers, scientific institutions and whole areas of study in a dragnet. It’s clearly an attempt to cast scientists as ideological enemies, undermine public trust and discipline higher education.

The difference between this document and its predecessors is that the target of the attack is not on “narrower” avenues of research, like evolution or stem cells. The target of the attack is on the people who produce science and our understanding of the life support system, the environment, that undergirds all human activity.

This is plain in the text of the methods. Cruz and his cronies downloaded over 32,000 grants awarded over the Biden administration from USAspending.gov, which tracks government grants. They assembled a list of “DEI keywords” quite possibly from their own imaginations and then say they supplemented them with other words from dictionaries maintained by the National Association of Counties and the University of Washington. I wasn’t able to track down those sources online, assuming they actually exist.

They then ran a Python script against all of the grants, checking for hits against their list of terms. They culled grants that only got hit by a single term and then further filtered the remaining grants using the Excel qualifier function to isolate key phrases. They removed any remaining keyword or key phrase hits. They claim this confirmed the presence of “DEI” in the grants. As a final check they manually checked around 1,400 grants for errors, leaving them with 3,483 grants as DEI grants.

From the jump this is basically the same censorious approach that social media surveillance and internet filter companies market to schools. It’s keyword and keyphrase logging without care for context, definition or purpose.

While Cruz and his cronies may have been able to avoid the embarrassing mistake of flagging black hole research as “woke” their approach still flags things like “studying desert climate” for agriculture or “diverse cybersecurity applications” as woke.. Building a massive defense research and industry complex in El Paso is woke because the grant application is aware of local demographics. Studying the health of women is woke because women exist. Using particle colliders to study the high atmosphere and building new astronomical tools are woke. Trying to study hurricanes with lasers is woke because climate. By this definition, Nixon’s NEPA law is woke because it mandates environmental justice considerations. It’s all woke folks.

The other thing these keywords catch are things like workforce development initiatives, educational initiatives, the pipeline into STEM jobs. In 2022, Congress passed the CHIPS act which among other things, mandated the creation of a NSF Diversity office, increased women and minority STEM participation, and increased outreach to minority-serving institutions. If you were a large institution seeking a grant during or after 2022, your grant writers would have been pitching how your big research center would meet one or more of these goals. This means that even if your research center wasn’t specifically focused on gender inequities you’d probably include the language to increase your chance of getting the grant.

So now we’re in 2025, the CHIPS act is law, Trump and Elon are pulling the copper wires out of the federal government and Ted Cruz is attacking institutions for trying to meet the Congressionally-passed policy goals that build out the scientific workforce. So now his dumb keyword pulling catches Cornell’s push for a rural high tech research, development and tech incubator center in the rust belt. That means that encouraging kids to like math is woke. In Cruz’s own state rural geoscience, or cybersecurity scholarships or the big El Paso natsec project are all too woke.

It’s important to remember that Cruz’s efforts should not be taken in good faith. It’s no different than dragging embryonic development researchers in front of the House science committee, FOIAing the emails of disinformation and climate researchers, or the Cold War blacklisting of anti-nuclear war scientists.

Sophia’s Comments:

Whichever staffers prepared this either failed to do the requisite homework for browsing through this kind of data, or they intentionally obfuscated what was actually happening. The methodology section of their report claims to have sorted out things which did not meet certain criteria, and there is evidence that they did try to sort out things that were not truly “DEI.” (Apparently anything related to green energy or climate change is part of “DEI” now?) But it’s very clear that they missed some essentials.

For starters, many of these had obligations from the federal government which appear to not have been paid at all. Usaspending.gov splits the money into two buckets.. An amount obligated, which is what the federal government has said that they will pay, and an amount outlayed, which is what they’ve actually paid. Our method repeated their research, while including and focusing on outlayed amounts, and found that not only did 190 of these grants show not having paid anything at all (0% of obligated amount), only 262 of them actually appear to have paid the full amount obligated. Of the total amount obligated across all of these grants, which Ted had listed at over $2 billion, just under $759 million, or roughly 35% appears to have actually been paid.

Out of states receiving funding, his home state of Texas received the third largest amount of money, both in actual outlays and obligations, compared to other states. Texas’ outlays totalled over $54 MILLION, which is roughly 7.16% of the total amount of outlayed money for all grants. While a lot of blue states with high populations and large institutions benefitted from this heavily, so did traditionally red states like Georgia, Florida, Indiana, and Missouri.

Image 2 - Percentage total share of paid grant money by state

As far as the substance of the claims made in the document, as Damian noted before, many of these grants don’t really meet the qualifications of being “woke” or “DEI.” One notable mention is for MIT to create a public access cloud computing service. Only two of Cruz’s list of bad words from the document in the grant proposal are “sustainable,” which is not referencing the environment, and “socioeconomic,” which notably isn’t about serving marginalized groups or anything like that in this context. “Socioeconomic” only appears because it’s pulled from the description of an NSF program called “Strengthening America’s Infrastructure,” or SAI.

SAI focuses on something we’ve been hearing a lot about lately, the efficiency of work in the public sector. You know, the thing that certain extremely rich and extremely divorced folks have a borderline-obsessive fixation on, but somehow they can’t imagine efficiency as anything other than “cancel federal science funding.” Another SAI program grant was awarded to help scientists communicate with the public has no DEI or environmental aims.

Another is a research grant to fund work in applied discrete mathematics. Its only mention of something “woke,” is that it will emphasize recruiting diverse groups of students. The grant was awarded to a program in a small city in Pennsylvania, so it’s not likely that they would have had to put much effort, if any, in on this task. The grant authors aren’t promising to even achieve that recruiting goal.

This is common in the fairly large number of these grant proposals that we’ve read: the commitments to anything remotely resembling DEI or environmentalism are just bare-bones commitments, and often reflect requirements the NSF asked for, in specific kinds of grant proposals.

One of the most egregious things I found in here was targeting a piece of research set to take place at Northern Kentucky University, whose description only included three of the over 695 words Cruz highlighted were “status,” “trauma,” and “traumatic.” This caught my eye, because these words aren’t really “woke” or “DEI” or any of that. One of them, “status,” is used in so many ways that I can’t begin to imagine why it was included, and the other two are often used in medical settings. That's precisely what this grant is for. It focused on building a system that uses eye-tracking to detect cognitive abnormalities and traumatic brain injuries. The description went on to talk about helping people in rural areas, who may not have good access to healthcare. This work would also be incredibly helpful for military veterans who also have an incredibly high incidence of TBI’s and other cognitive abnormalities from their service. I can’t overstate how absurd I find it that a Republican, or anyone really, would want to defund this kind of research, or to call it “woke.”

Other notables include the following:

Another from SAI, with no other terms from the naughty words and phrases list, except “socioeconomic,” which focuses on disaster preparedness for infrastructure, especially in the case of earthquakes. (And another which meets almost identical criteria)
One that seeks to help both urban and rural areas get access to 5G cell phone service with a focus on energy savings and lowering costs
A study on insect genetics that was targeted for words like “female,” and “systemic,” which are specifically referring to sex within insects and systemic changes in the biology of those insects (and another that got flagged for almost identical reasons, which focuses on mate choice in animals and how that impacts biodiversity)
This one, grabbed for the word “barriers,” which is strictly about migrating animals. No mention of the word “barrier” or “barriers” or anything else outside of strict biology context. 10/10, very dry, very boring.

It’s noteworthy that, out of those 695 keywords and phrases mentioned, only 311 appeared at all. This makes the inclusion of the whole list in the report feel incredibly misleading, especially since that fact goes entirely unmentioned. It’s likely that someone reading through this report would see that list of words and infer that they all appeared at some point in the flagged grants. While their methodology said that they removed grants that the qualifier formula that they used identified as having erroneous instances of the words in the list (which of course isn’t shown because why would they show that they’re doing shoddy research), they clearly didn’t, as evidenced by the issues that we found in the week and a half that we took to work on the research for this.

Image 3 - The most unqualified researchers alive talking about qualifiers

The last thing we’ll say about this is probably the most important thing: DEI is actually a good thing! Unless you are wildly dedicated to having an underclass in society that’s specifically based on racial or gender criteria (which of course the party of equal opportunity would never do /s), you would want to train all different kinds of people to accomplish all different kinds of work. It’s sort of required by a high-tech capitalist economy.

Our specific way of funding education in America, often by local property tax revenue, leaves out kids in both urban and rural settings. To wonder out loud in public why those kids aren’t reaching the same goals as our precious suburban children indicates a skill issue that I simply wouldn’t admit to publicly.

These grants from NSF and other federal agencies very often provide training to underserved minorities wherever they are because that’s what equal opportunity literally means. You have to actually fund opportunities in places where they wouldn’t exist, because people can’t afford to leave. They can’t afford to do much else. I’m from a place like that originally, and so perhaps that’s why this is raising my blood pressure to the point where I feel like I'm literally going to pop.

I think we all should recognize that the things Republicans are calling “DEI,” especially in the context of government funding, are specifically about helping people have equal opportunities. That’s all it really is. If they don’t want that, then they should just admit it so the rest of us can get to work.

If we made any mistakes in this document, or for access to spreadsheet files and methodology for our deeper research into this “research,” feel free to write to us at [email protected]

In order to further verify specific numbers in this database for further work, we will be reaching out to some grant recipients. This will help understand if the usaspending.gov database is working correctly and up-to-date. If we find that we are in error anywhere, or that things need an update for any other reason, we will add an update here above the article for clarity.