When Clinical Guidance Becomes “Gender Ideology”

An Evidence Audit of the HHS Report on Gender Affirmation, Authoritarianism, and Political Violence

By Jenn Burleton

Emerging Analysis · August 2026

Last reviewed: October 1, 2026

Why the 13-Item Version Matters

The 13-item Brief Left-Wing Authoritarianism Scale is the shorter of two abbreviated forms—25 and 13 items—derived from the original 39-item Left-Wing Authoritarianism Index. In their initial validation, Costello and Patrick found that both abbreviated forms closely approximated the longer measure’s overall psychometric properties and relationships with external criteria.

The authors nevertheless advised that the full instrument may be preferable when left-wing authoritarianism is itself a principal focus of the research, and that the 25-item version may be better suited to analyses requiring more reliable distinctions among the measure’s dimensions. That guidance is relevant here: HHS describes left-wing authoritarianism as a construct “at the center” of its research but administers the shorter, 13-item form.

This choice does not invalidate the analysis. It does, however, mean that the results are best interpreted as associations involving a brief overall measure of left-wing authoritarianism—not as a detailed assessment of anti-hierarchical aggression, anti-conventionalism, and top-down censorship as distinct dimensions.

Appendix B Simplifies the Scale’s Factor Structure

HHS Appendix B assigns each of the 13 items to one of three subscale headings: Anti-Hierarchical Aggression, Anti-Conventionalism, or Top-Down Censorship. The published LWA-13 measurement model is more complicated. Costello and Patrick report that a handful of retained items serve as indicators of more than one factor, reflecting both general left-wing authoritarianism and variance associated with different dimensions.

The authors also found some loss of discriminant validity among the abbreviated subscales, although the differences from the full instrument were modest. They nevertheless concluded that, overall, the LWA-13 and LWA-25 closely reproduced the performance of the original index.

HHS does not report or interpret separate scores for the three dimensions; its analyses use an overall LWA score. Accordingly, Appendix B’s categories should not be mistaken for a complete representation of the published measurement model, but this simplification does not by itself invalidate the report’s total-score analyses. It would become more consequential if the three dimensions were analyzed or interpreted as independently measured constructs.

HHS Does Not Report Sample-Specific Reliability for the LWA-13

For its newly developed Clinical Gender Affirmation Scale, HHS reports an internal-consistency estimate (Cronbach’s α = .95), mean, and standard deviation. For the LWA-13, the report identifies the instrument but provides neither a sample-specific internal-consistency estimate nor overall score descriptives.

Reliability is not a fixed property that automatically transfers from one study to another. The LWA-13’s previous validation supports its general use, but it does not substitute for reporting how consistently its items functioned in the present sample. Given the central role assigned to LWA, an estimate such as Cronbach’s alpha or McDonald’s omega would have been informative.

This omission does not establish that the LWA-13 was unreliable, nor does it by itself negate the reported associations. It prevents readers from evaluating the measure’s internal consistency and overall score distribution among these 1,208 respondents.

The LWA-13 Includes Violent and Coercive Content

Two LWA-13 items are particularly relevant to HHS’s later analyses of political violence. One explicitly asks whether respondents would enjoy seeing perceived oppressors suffer violence. Another endorses replacing the established order “by any means necessary.” Agreement with these statements contributes to the overall LWA score.

The report’s political-violence outcomes are separate survey questions, so the relationship is neither mathematically circular nor a part–whole correlation. There is, however, conceptual overlap: a predictor incorporating violent or revolutionary-coercive sentiment is being related to willingness to justify other specified acts of political violence.

This association is also consistent with the scale’s original validation, in which political violence was one of the external criteria used to evaluate the abbreviated LWA measures. HHS’s results therefore show that respondents scoring higher on this broader bundle of authoritarian, censorious, coercive, and partly violent attitudes were more likely to justify the particular violent acts presented in its survey. They do not represent the unexpected discovery that a wholly nonviolent psychological disposition is associated with political violence.

The overlap does not invalidate the correlations, but it limits their novelty and should inform their substantive interpretation.

What the LWA Results Can—and Cannot—Show

The HHS findings support a limited but meaningful conclusion: within this sample, respondents with higher LWA-13 scores also tended to endorse CGAS statements more strongly and to express greater justification for each of the six specified acts of political violence. These associations remained statistically significant after political identity was included in the reported models—and, in the CGAS model, after age was included as well.

The findings do not establish that left-wing authoritarianism caused either response, that participants possess a clinically identifiable authoritarian personality, or that the scale’s three dimensions operated independently in this sample. Nor can an opt-in, nonprobability survey establish the prevalence of left-wing authoritarianism in the U.S. population. The results concern associations among scores on particular self-report measures collected from these respondents at one point in time.

Clinical Gender Affirmation Scale (CGAS)

The HHS report introduced a new five-item measure that its authors named the Clinical Gender Affirmation Scale (CGAS). Unlike the LWA-13, the CGAS was developed specifically for this study rather than adopted from a previously established instrument.

Participants rated five statements addressing the perceived psychological consequences of non-affirmation, support for transgender identity expression, use of chosen names and preferred pronouns, and parental support for children’s transgender exploration. Responses were combined into an overall score, with higher scores representing stronger agreement with the five statements.

Because the CGAS is newly constructed, the interpretation of every analysis involving it depends on the content of its items, how they were derived from the cited guidance, whether they measure a coherent construct, and whether the scale’s name accurately characterizes that construct. The following sections examine each of those questions.

What the CGAS Actually Asked

The HHS report says the Clinical Gender Affirmation Scale was created specifically for this research to measure endorsement of what the authors describe as the “official clinical consensus positions” of the American Psychological Association, American Psychiatric Association, World Professional Association for Transgender Health, and American Academy of Pediatrics. The scale contains just five items, each rated on a seven-point agreement scale.

Participants were asked to rate their agreement with the following statements:

1. Failure to affirm a transgender person’s gender identity increases the risk of serious psychological distress.
2. People should be supported in expressing their transgender identity.
3. Using the name a person chooses as part of their gender transition is a necessary form of support.
4. Using a transgender person’s preferred pronouns (e.g., he/him, she/her, they/them) is a necessary form of support.
5. Parents should support their children’s transgender exploration.

These five responses form the CGAS score. Higher scores directly represent stronger agreement with this particular bundle of propositions. Interpreting that score as a measure of the broader construct “gender ideology,” however, requires construct-validity evidence beyond showing that the five responses are internally consistent.

CGAS Does Not Measure Medical Treatment

HHS describes the CGAS as measuring “official clinical consensus positions” concerning transgender people and medical interventions. Yet none of its five items asks respondents to evaluate or endorse a medical intervention.

The items address psychological distress associated with nonaffirmation, support for transgender identity expression, chosen names, preferred pronouns, and parental support for gender exploration. They do not ask about puberty suppression, hormone therapy, surgery, diagnostic assessment, treatment eligibility, informed consent, treatment risks, or clinical outcomes.

Accordingly, CGAS scores cannot by themselves demonstrate support for gender-affirming medical treatment. They measure agreement with five predominantly social and interpersonal propositions derived from clinical guidance—not respondents’ views about the medical interventions often central to public debates over gender-affirming care.

How Faithfully Do the CGAS Items Represent the Source Guidance?

HHS states that the five CGAS items were “derived directly” from clinical guidance, and Table 1 identifies passages associated with each item. The questionnaire statements are not direct quotations, however. They synthesize and sometimes narrow or strengthen language appearing in the displayed source excerpts.

This distinction is relevant to the scale’s content validity. Establishing that an item originated in professional guidance does not, by itself, establish that its reformulation preserves the guidance’s meaning, scope, and degree of certainty.

All five items have identifiable roots in the cited professional guidance, but they are not verbatim reproductions of it. Some consolidate distinct concepts under “affirmation,” add normative language such as “necessary,” or narrow broader discussion of gender exploration to specifically “transgender exploration.” The CGAS therefore measures agreement with HHS’s reformulated propositions—not direct agreement with the source documents themselves. That distinction matters when the resulting score is subsequently characterized as measuring “clinical gender affirmation” or, more broadly, “gender ideology.”

What Does High Internal Consistency Establish?

The five CGAS items address several kinds of propositions: psychological risk, social support, chosen names, preferred pronouns, and parental responses to gender exploration. HHS combines the responses into a single score labeled “clinical gender affirmation” and reports very high internal consistency: Cronbach’s α = .95. On the seven-point response scale, the reported CGAS mean was 4.87 (SD = 1.93).

Cronbach’s alpha describes how strongly a set of item responses covaries. The reported value shows that responses to these five items were highly internally consistent in this sample. It does not mean that the items are interchangeable or establish that they measure only one underlying construct.

Alpha alone cannot establish unidimensionality, construct validity, discriminant validity, or the accuracy of a scale’s name. A very high value can also reflect closely overlapping items or a shared response to their common political, moral, or social framing.

HHS does not report a factor analysis or comparisons with alternative measures designed to determine whether the five items measure a distinct construct rather than a broader attitude or ideological orientation. The reported alpha therefore supports the narrower conclusion that the five responses form a strongly internally consistent composite; it does not establish CGAS as a validated measure of a distinct construct called “clinical gender affirmation.”

The measure was introduced and analyzed in the same sample. No replication in an independent sample is reported.

The Name Implies More Clinical Specificity Than the Items Measure

The name Clinical Gender Affirmation Scale may suggest that the instrument measures a specifically clinical construct. Yet CGAS was administered to members of the general public, not clinicians, and its items do not assess clinical decision-making, diagnostic reasoning, treatment practices, adherence to clinical protocols, or patient outcomes.

“Clinical” describes the provenance of the selected source material rather than the respondents, behaviors, or decisions being measured. Deriving propositions from professional guidance is not the same as measuring clinical competence, clinical practice, or support for particular medical treatments.

This does not make the name inherently illegitimate: the source material has a genuine clinical and professional origin. The most concrete description of CGAS, however, is a five-item self-report attitude measure constructed from HHS’s reformulations of selected guidance.

From Clinical Guidance to “Gender Ideology”

The report does not stop at describing CGAS as a measure of agreement with five propositions derived from clinical guidance.⁷ It interprets those responses through a broader political category: “gender ideology.”

This interpretive vocabulary predated the report’s public introduction of CGAS: the administration had already adopted “gender ideology” as a policy category. The available public record, however, does not establish precisely when development of the scale began.

The survey itself does not ask whether participants endorse “gender ideology,” and no CGAS item uses that phrase. Participants respond to concrete statements about distress, support, names, pronouns, and parental responses to gender exploration. The broader ideological label comes from the report’s interpretive framework and is extended in HHS’s public-facing description of the research.

The interpretive progression is:

selected clinical guidance → five reformulated survey items → a single CGAS score → “clinical gender affirmation” → “gender ideology” → “radical gender ideology.”

Each step adds interpretation. The farther the language moves from the original survey items, the more important it becomes to distinguish what participants actually endorsed from what the report and HHS subsequently say those endorsements represent.

This distinction affects what the statistical results are understood to represent. The report estimates CGAS’s association with LWA and, in separate models, CGAS’s association with political-violence responses. Public summaries can compress those findings into a broader linkage among authoritarianism, political violence, and “gender ideology”—even though the report did not model all three together and did not independently define or directly measure “gender ideology.”

The most defensible description therefore remains the narrowest one: CGAS measures agreement with five researcher-constructed statements derived from selected clinical guidance. Reinterpreting that score as endorsement of a broader ideological system requires additional evidence.

Early post-release framing. Within days of publication, HHS and several conservative media and advocacy outlets—including Breitbart, Washington Examiner, Washington Stand, Decision Magazine, New York Post, and National Pulse—were describing the findings in language broader than the five-item CGAS itself. HHS used “radical gender ideology”; other coverage used formulations including “trans beliefs,” “trans ideology,” and “transgender ideology,” frequently alongside claims about political violence.

That is early evidence consistent with the hypothesized process of construct reification, not yet evidence that CGAS has become an established scientific or political category.

What Happened After ROGD Was Published?

Littman’s ROGD paper was published on August 16, 2018, as an exploratory parent-report study.

Within two weeks, conservative media coverage was already giving the study a more expansive framing. The Daily Wire described the research in its headline as concerning teens “imitating their peers by turning trans,” although the article also quoted Littman’s caution that the design could establish neither causation nor prevalence. Fox News described the study as finding that social media and friends could influence teenagers to change their gender identity. On August 31, The Federalist presented the findings as evidence of possible “social contagion”; on September 4, an article published under the byline Jamie Shupe called the research “pioneering” and a “blockbuster study” and asked whether it provided “formal academic validation.”

At the same time, professional organizations were warning against treating the label as settled science. On September 4, the World Professional Association for Transgender Health (WPATH) stated that ROGD was not a recognized medical entity and described it as an acronym for a proposed phenomenon that might or might not warrant further investigation. WPATH specifically cautioned that “official-sounding labels” could lead clinicians, community members, or scientists to form premature absolute conclusions about adolescent gender-identity development.

Advocacy uptake followed quickly. On September 19, Parents of ROGD Kids thanked Littman for “giving us the words” to describe what they believed they were observing and said her study had “meticulously” quantified the phenomenon.

The relevant parallel for this audit is the speed of rhetorical elevation. Within weeks, an exploratory parent-report hypothesis was being presented in media shorthand as “social contagion,” a “blockbuster study,” and possible “formal academic validation.” That does not make ROGD and CGAS scientifically equivalent. It shows how quickly a newly named research construct can acquire broader political meaning before its evidentiary status is settled.

What Was HHS Trying to Learn?

In practice, the survey uses CGAS to test whether agreement with five selected propositions derived from clinical guidance is associated with left-wing authoritarianism, political identity, and judgments about political violence. It does not evaluate whether the underlying guidance is medically accurate, whether it improves or worsens patient outcomes, or how clinicians apply it.

CGAS’s immediate research function is therefore not to evaluate gender-affirming care. It converts agreement with selected propositions from that framework into a variable whose political and psychological correlates can be analyzed.

Why HHS commissioned this particular inquiry remains only partly documented. Before publishing the report, the department had adopted opposition to “gender ideology” as a policy objective; its public release subsequently framed the findings in explicitly ideological terms. That chronology makes institutional purpose a legitimate subject of scrutiny, but it does not establish the private intentions of the officials or researchers involved.

The public record does not identify who requested the research, what statement of work governed it, how the investigators were selected, what deliverables were specified, or what internal discussions preceded publication. Those records are needed to understand how a federal health agency selected and structured this political-psychology inquiry.

What the CGAS Results Can—and Cannot—Show

The CGAS provides a numerical summary of how strongly respondents agreed with five researcher-constructed statements derived from selected professional guidance. The survey can test whether scores on that bundle were statistically associated with responses on other measures in this sample.

The safest interpretation remains correspondingly narrow: a higher CGAS score means greater agreement with these five particular statements. It does not independently establish that the items measure a distinct or stable construct, and the cross-sectional design cannot establish causal direction or temporal sequence.

The report interprets that score as “clinical gender affirmation” and relates it to the broader category of “gender ideology”; HHS’s public release extends the framing to “radical gender ideology.” Those successive labels introduce interpretive claims beyond what the numerical score itself establishes; neither internal consistency nor statistical association independently validates those broader constructs.

CGAS and Left-Wing Authoritarianism

After defining CGAS and LWA separately, the report asks whether the two scores are associated. They are.

In the full sample of 1,208 respondents, CGAS and LWA had a Spearman correlation of ρ = .54, p < .001. In practical terms, respondents who scored higher on the five-item CGAS also tended to score higher on the 13-item Brief Left-Wing Authoritarianism Scale.

The report also fit a simple regression predicting LWA scores from CGAS scores and reported β = .56, R² = .315, F(1, 1206) = 554.60, p < .001. In this one-predictor model, the fitted relationship accounted for 31.5% of the variation in LWA scores within the sample. That percentage describes model fit; it does not mean that CGAS caused 31.5% of respondents’ authoritarianism.

These are substantial statistical associations. What they mean, however, depends on what the two instruments contain—which is why the measurement audits above matter.

What Does a Correlation of .54 Mean?

A Spearman correlation of ρ = .54 indicates a substantial positive association between respondents’ rank ordering on the two measures. Respondents with higher agreement across the five CGAS propositions generally also had higher aggregate scores on the Brief LWA Scale, which contains coercive, censorious, anti-conventional, and explicitly Left-coded propositions.

The coefficient does not mean that 54% of respondents who agreed with the CGAS items were authoritarian; a correlation coefficient is not a percentage. It also does not establish that CGAS causes LWA or that the two scales measure the same construct.

The narrow conclusion is that, within this sample, higher values on one score tended to accompany higher values on the other.

Why the 31.5% Figure Needs Careful Interpretation

The report emphasizes that CGAS “accounted for approximately 31.5% of the variance in LWA.” That figure comes from a simple regression using CGAS as the sole predictor of LWA; the model did not control for political identity or other covariates.

An R² of .315 means that the fitted relationship with CGAS accounted statistically for 31.5% of the observed variation in measured LWA scores within this sample. It does not mean that CGAS explains 31.5% of left-wing authoritarianism as a psychological phenomenon.

Because both variables are questionnaire scores containing politically and morally related content, their shared variance could reflect genuine psychological overlap, overlapping ideological content, political identity, general response tendencies, or other unmeasured characteristics. This model alone cannot distinguish among those possibilities.

The 31.5% figure is therefore a statement about covariation between two measured scores—not evidence that clinical gender affirmation causes or explains nearly one-third of authoritarian psychology.

What Happens After Political Identity Is Included?

The report also asks whether the CGAS–LWA relationship is merely a byproduct of ordinary left–right political identification. To examine that question, the researchers fit an ordinal regression with CGAS as the outcome and LWA, political identity, and age as simultaneous predictors. This analysis included the 1,192 respondents who could be placed on the left–right political spectrum.

More left-wing political identification was the strongest predictor of higher CGAS scores (β = .512, p < .001). LWA nevertheless remained positively associated with CGAS after political identity and age were included (β = .326, p < .001), while age had a much smaller positive association (β = .068, p < .01).

This is an important result: the CGAS–LWA association was not entirely captured by the particular political-identity measure used in the model. It also matters that political identity remained considerably more strongly associated with CGAS than LWA did (β = .512 versus β = .326).

Statistical adjustment has limits, however. Including political identity separates the association represented by that particular variable; it does not remove the political and ideological content of either scale, rule out other shared influences, or establish that the remaining association reflects a causal authoritarian process.

From Statistical Prediction to “Ideological Extremity”

The report moves beyond describing CGAS and LWA as statistically associated to a stronger interpretive claim: CGAS may capture “a form of left-wing ideological extremity,” while “gender ideology” may function as an “ideological accelerant.”

That is a major inferential step. The analyses show that higher CGAS scores were associated with higher LWA scores and, in separate later models, with greater justification of political violence after political identity was included. Those results do not themselves establish that CGAS represents ideological extremity or that gender-affirming clinical guidance accelerates authoritarian psychology.

The report’s proposed mechanism is more specific: what it calls “gender ideology content” may supply people already higher in LWA with targets, moral urgency, and justification for coercive action—in effect, telling authoritarian psychology who the enemy is, why the situation is urgent, and when force may appear legitimate.

That mechanism is a theoretical interpretation of the observed associations, not something demonstrated by this cross-sectional survey. The authors acknowledge that the data cannot determine whether CGAS-consistent beliefs influence LWA, whether people higher in LWA are more likely to endorse those beliefs, or whether another factor contributes to both. They call the present correlations a “discovery sample” and say a separate experiment is being designed as a preregistered “confirmatory causal test.”

An accelerant is not a neutral metaphor: it implies direction, temporal sequence, and intensification—the precise elements this survey does not establish. The distinction is straightforward: the survey establishes association; the “ideological accelerant” mechanism remains a hypothesis.

What Else Could Explain the CGAS-LWA Association?

The report interprets the association remaining after political identity and age are included as evidence consistent with a specifically authoritarian psychological process. Other explanations remain possible, however, and the cross-sectional design and limited set of covariates cannot distinguish among them.

Ideological sorting. CGAS contains propositions associated with contemporary progressive politics, while the Brief LWA Scale contains explicitly Left-coded political content. A single left–right political-identity variable may not fully capture issue-specific ideology, coalition affiliation, or ideological intensity. Respondents located within the same broader political coalition may therefore score higher on both measures even when authoritarianism is not the only shared factor.

Measurement overlap. The instruments do not duplicate one another’s items, but neither is politically neutral. LWA includes hostility toward conservatives, concern about right-wing extremism, censorship of right-wing websites, and other politically specific content. CGAS measures positions that have also become politically polarized. Their association may therefore reflect some shared political and moral content in addition to authoritarian disposition.

Broader moral or social orientation. Commitments involving equality, protection of marginalized groups, social inclusion, or opposition to perceived oppression could influence responses across both instruments without being synonymous with authoritarianism.

None of these alternatives disproves the reported CGAS–LWA association. They show why the analyses do not identify the psychological mechanism responsible for it.

What This Audit Establishes

The report’s significance lies in two distinct developments: it identifies limited associations among specific questionnaire scores, and HHS interprets those associations within a federal “gender ideology” framework that predated the report’s publication. The sections below distinguish what the CGAS research adds to that framework from what the public record still does not establish.

The Federal “Gender Ideology” Framework Predated the Report’s Publication

The report did not introduce “gender ideology” into federal policy. By the report’s public release in August 2026, the Trump administration was already using the term as a political and administrative category, had directed agencies to act against what it called “gender ideology,” and had incorporated that framework into health policy and grant administration.

HHS was also using the explicitly evaluative term “sex-rejecting procedures,” which outside advocates had proposed and urged federal officials to adopt. The administrative vocabulary through which the CGAS findings were presented was therefore already in place.

This chronology does not establish that the policy framework preceded the report’s commissioning or the development of CGAS; the public record does not disclose those dates. Nor does it prove that the statistical conclusions were predetermined. It does establish that the report entered an existing policy environment in which opposing “gender ideology” was an explicit governmental objective.

What the CGAS Report Adds to That Existing Framework

What the report adds is not the political category itself, but a new quantitative and psychological layer around it.

Researchers constructed a numerical score from agreement with five propositions derived from selected clinical guidance. In separate analyses, CGAS scores were associated with a 13-item measure of left-wing authoritarianism and with justification ratings for six selected acts of political violence. The report interprets these relationships using concepts such as “ideological extremity” and an “ideological accelerant,” while HHS’s public release extends them into claims about “radical gender ideology,” left-wing authoritarian beliefs, and political violence

The resulting interpretive sequence is shown below:

The survey establishes the CGAS score and the statistical associations described above. The transitions from that score to “clinical gender affirmation,” “gender ideology,” and “radical gender ideology” are interpretive steps beyond the propositions respondents actually rated.

What the Public Record Still Does Not Show

The public record does not establish:

• precisely when the research was commissioned or when development of CGAS began;
• who within HHS originated or approved the specific research question;
• what statement of work, objectives, deliverables, funding mechanism, or other commissioning arrangement governed the project;
• how this particular research team was selected to conduct the work; or
• what internal HHS review, communications, or decision-making preceded publication and public rollout.

The available record also does not establish that HHS predetermined the results, directed the researchers to produce particular findings, commissioned the work to support a predetermined policy or enforcement objective, coordinated the research with the May 2026 counterterrorism strategy, or allowed any outside organization to design or control the project.

Burleton Education has likewise found no evidence that CGAS itself has been cited to terminate grants, restrict funding, deny program eligibility, or support regulatory action. Existing records document administrative and funding consequences associated with the administration’s broader “gender ideology” framework; they do not establish that CGAS itself has been used to determine those consequences.

Burleton Education has requested additional commissioning, funding, and communications records through pending Freedom of Information Act requests. In the meantime, additional public records and reporting can narrow some questions without resolving the central gaps identified above. Claims presented as documentary fact must therefore remain bounded by the available record, and Burleton Education’s editorial assessments should be revised if new records materially support, qualify, or contradict them.

---

Does the Report Cross Into “Junk Science”?

The audit above has addressed two related questions: what the HHS report actually establishes and how its findings may function within a broader federal policy framework. One question remains before reaching a final assessment: when do scientific limitations, interpretive overreach, and public presentation collectively become a misuse of scientific authority?

Preliminary, imperfect, or politically consequential research is not automatically “junk science.” Nor should the label be applied merely because findings conflict with Burleton Education’s views. If it is used here, it must follow from explicit criteria applied consistently to the report’s design, measurement, interpretation, and public framing.

Burleton Education assesses each criterion separately for the research/report itself and for HHS’s public framing and use of the findings. These are evidence-based editorial assessments, not scientific measurements or scores. The criteria can overlap and are not assumed to be independent or equally consequential. The final assessment therefore depends on the documented pattern across the criteria rather than on a numerical or quasi-numerical threshold.

---