Career advice is everywhere. Auditable evidence is harder to find.

Workers are routinely told which jobs are growing, which degrees are valuable, where salaries are highest, and which careers are supposedly easy to enter. Those claims often come from a single table, an unexplained ranking, or a collection of job postings that may change before a reader can act.

MyTopMatch Research was created to ask deeper questions and show exactly how each answer was produced.

The continuing research program combines authoritative labor-market sources, reconciles their occupational classifications, creates transparent derived measures, tests the stability of important results, and preserves an audit record for every publishable claim. Its purpose is practical: give workers, employers, journalists, and workforce leaders evidence they can examine rather than conclusions they are expected to accept on faith.

This is original research and analysis by MyTopMatch. The underlying source data remain the work of the agencies and organizations that produced them. MyTopMatch's contribution is the research design: the questions asked, the combinations of datasets, the occupational crosswalks, the calculations, the statistical tests, the validation thresholds, and the interpretation of the results.

## The questions behind the program

The labor market rarely gives workers a clean answer to the questions that matter most:

- Which occupations offer strong pay and opportunity relative to their formal preparation requirements? - Where does the education held by incumbent workers differ from the education BLS classifies as typical for entry? - Which locations combine occupational scale, concentration, and purchasing-power-adjusted pay? - How strongly are education, experience, training, and occupational complexity associated with compensation? - Which careers give workers credible future pathways if demand changes? - Which occupational transitions look similar on paper, and which are actually supported by observed worker movement?

No single federal dataset answers all of these questions. MyTopMatch therefore built a common analytical structure across occupational employment, wages, projections, worker education, skills, job preparation, geography, and observed transitions.

## What the first analyses found

The first completed research cycle produced several findings that passed the program's publication standards with defined limitations.

### The education of incumbent managers can differ sharply from the current entry benchmark

**Statistical finding:** A MyTopMatch analysis of 2024 American Community Survey public-use microdata found that 59.6% of wage-and-salary construction managers ages 25 to 64 held less than a bachelor's degree. The 90% confidence interval was 58.0% to 61.2%. Among facilities managers, the corresponding estimate was 63.4%, with a 90% interval of 60.7% to 66.2%.

The Bureau of Labor Statistics classifies a bachelor's degree as the education typically needed for entry into both occupations. That classification and the ACS measure answer different questions. The BLS category describes typical entry preparation. ACS describes the education held by people already doing the work.

**Interpretation:** The difference identifies experience-based pathways worth investigating. It does not establish that a current applicant can enter either occupation without a degree, and it does not reveal what credential a worker held when first entering the field.

### The educational composition of human-resources assistants changed

**Statistical finding:** The bachelor's-or-higher share among human-resources assistants rose from 40.3% in 2018 to 55.6% in 2024, an increase of 15.3 percentage points. The change survived false-discovery-rate control across occupations, and a 2019-to-2024 sensitivity test remained positive.

**Interpretation:** The incumbent workforce changed materially over the period studied. The available data do not identify the cause. Employer screening practices, workforce composition, geography, industry mix, cohort replacement, and worker selection remain possible explanations.

### Three states formed a stable software-developer opportunity tier

**Calculated result:** Washington, California, and Virginia formed the stable top state tier in the matched-2024 MyTopMatch Geographic Opportunity analysis for software developers. Each state appeared in the top five across all 500 alternative weight draws.

The composite combined occupational employment scale, concentration, and purchasing-power-adjusted median pay. The accompanying study will publish each component so readers can see why a state scores highly and whether a different weighting system changes the conclusion.

### Management occupations showed a large descriptive wage difference

**Derived calculation:** BLS Occupational Employment and Wage Statistics reported a May 2025 median annual wage of $126,520 for management occupations and $50,980 across all occupations. The ratio of those published group medians is 2.48.

That is an unadjusted occupational-group comparison. It is not a causal estimate of the financial return from being promoted, and it does not control for education, tenure, industry, geography, hours, or responsibility.

These findings will be released as separate studies with their full eligibility rules, uncertainty statements, source versions, sensitivity checks, and audit records.

### A ranking was withheld after it failed the release standard

The program's most important result may be one it has declined to publish.

MyTopMatch tested a transparent occupational-transition model on 1,021 held-out transitions constructed from linked monthly Current Population Survey records. The model reached a weighted macro ROC AUC of 0.925, placed the observed destination in its top ten 48.9% of the time, and predicted wage direction correctly for 61.5% of eligible transitions.

Those results did not clear the preregistered release standard. The model passed four of eight frozen gates and failed the required improvement over simple baselines, the top-ten relative-improvement threshold, the calibration ceiling, and an external Survey of Income and Program Participation sensitivity check.

Career Mobility, Career Switchability, and Career Resilience rankings will remain withheld until the necessary gates pass on a newly untouched cohort. MyTopMatch will publish a methods and status report explaining what worked, what failed, and what must happen next.

That decision reflects a governing principle of the program: a compelling headline cannot override a failed validation rule.

## The data foundation

The first research cycle uses current, documented releases available at execution time:

- **BLS Employment Projections, 2025–35:** occupational employment, growth, annual openings, median wages, typical education needed for entry, related experience, and on-the-job training. - **BLS Occupational Employment and Wage Statistics, May 2025:** national, state, metropolitan, and industry-level occupational employment and wage estimates. - **U.S. Census Bureau American Community Survey public-use microdata** — worker education, occupation, earnings, employment, hours, and demographic variables, with survey weights and replicate weights where required. - **Current Population Survey and Survey of Income and Program Participation public-use files:** observed occupational transitions used under protected, aggregate-only disclosure controls. - **O*NET 31.0:** skills, knowledge, abilities, work activities, tasks, education, training, experience, related occupations, and occupational taxonomy. - **Bureau of Economic Analysis 2024 Regional Price Parities:** geographic price-level adjustments for state and metropolitan comparisons.

Raw source files are preserved unchanged. Transformations, mapping tables, scripts, intermediate outputs, data dictionaries, checksums, and finding audits are retained so the analysis can be reproduced and revised when new releases become available.

## Four labels govern every major statement

Every MyTopMatch research release separates four kinds of claims:

1. **Source data** are facts taken directly from an identified authoritative dataset. 2. **Derived calculations** are values MyTopMatch computes from source data using a disclosed formula. 3. **Statistical findings** are relationships or estimates supported by a defined quantitative analysis and its uncertainty checks. 4. **Interpretation** explains potential meaning, implications, and hypotheses without presenting them as observed facts.

Exploratory findings receive an additional label. They may reveal valuable patterns, though they were discovered after examining the data and must be distinguished from predefined confirmatory questions.

## What publication requires

Every headline-worthy finding must pass a structured review covering sample size, missingness, occupational coding, survey weights, outliers, alternate specifications, classification changes, time effects, geography, confounding, practical magnitude, and multiple comparisons where relevant.

Composite indices receive additional scrutiny. Each component must have a stated purpose. Normalization and weighting must be documented. Reasonable alternate weights must be tested. When rankings move substantially under defensible alternatives, the publication must show that instability rather than conceal it behind a single ordinal list.

Each major claim receives a finding ID and an audit record identifying the data releases, sample, variables, calculation, statistical method, result, effect size, confidence interval, robustness tests, alternative explanations, limitations, validation status, and publication status.

## What readers can expect

The initial MyTopMatch Research library is structured as ten substantive articles and two standalone status pages. The first publication sequence will examine:

- the Degree Gap between typical-entry education and incumbent attainment; - changes in worker educational composition; - opportunity relative to measured formal preparation; - the association between occupational requirements and pay; - geographic opportunity for specific occupations; - management and leadership economics; - the limits of current public occupational-licensing data; and - the continuing validation of Career Mobility, Switchability, and Resilience measures.

Together with this launch article and a master methodology and data-governance article, these releases will establish a 12-publication research center. The Switchability and Resilience pages will remain status publications until their evidence dependencies are resolved.

MyTopMatch Research is designed as a continuing program. Direct-data studies can be refreshed as BLS, Census, O*NET, and BEA issue new releases. Longitudinal studies will preserve classification breaks and missing years rather than silently smoothing them away.

## How to cite MyTopMatch Research

If a finding, chart, table, or interpretation from this program is quoted or reused, the citation should identify MyTopMatch Research as the analyst and link to the permanent page for the specific study.

**Preferred citation:** MyTopMatch Research. “Introducing MyTopMatch Research: A New Evidence Base for Career Decisions.” *MyTopMatch Labor Market Research*, 2026. Use the canonical URL for this page.

**Suggested media attribution:** According to a MyTopMatch Research analysis of U.S. Bureau of Labor Statistics, U.S. Census Bureau, O*NET, and Bureau of Economic Analysis data...

**Suggested chart credit:** Source: MyTopMatch Research, [Study Title] (2026), using BLS, Census, O*NET, and/or BEA data. Link to the canonical MyTopMatch study page.

Final reuse terms for MyTopMatch-created charts and supporting tables will be displayed on each downloadable asset after legal review. The original public data remain subject to their respective source terms and attribution requirements.

## A research program readers can inspect

Career decisions carry real consequences: time, money, education, relocation, income, and professional identity. The evidence supporting those decisions deserves to be visible.

MyTopMatch Research will publish findings that survive its standards, explain findings that require caution, and keep incomplete rankings off the market until the evidence is strong enough. Readers will be able to trace each major claim from its source data through its calculation and validation record.

That is the commitment behind this research program—and the standard every MyTopMatch release will be required to meet.

## Sources and data attribution

- BLS Occupational Projections and Worker Characteristics, 2025–35 - BLS Occupational Employment and Wage Statistics tables, May 2025 - U.S. Census Bureau 2024 ACS Data Release - U.S. Census Bureau ACS 1-Year PUMS - U.S. Census Bureau Current Population Survey datasets - U.S. Census Bureau Survey of Income and Program Participation datasets - O*NET 31.0 Database - O*NET 31.0 Database Content License - BEA Regional Price Parities by State and Metro Area - BLS Copyright and Source-Citation Guidance - U.S. Census Bureau Citation Guidance

This analysis includes information from the O*NET 31.0 Database by the U.S. Department of Labor, Employment and Training Administration (USDOL/ETA), used under the CC BY 4.0 license. MyTopMatch modified and combined the information. USDOL/ETA has not approved, endorsed, or tested those modifications. O*NET® is a trademark of USDOL/ETA.

Source data include material from the U.S. Bureau of Labor Statistics. BLS cannot vouch for data or analyses after retrieval and transformation by MyTopMatch. Source data also include U.S. Census Bureau public-use files. Conclusions are the responsibility of MyTopMatch. No source agency has approved or endorsed this research.

## Material limitations

- Public occupational data do not establish the hiring standard of a specific employer or guarantee individual access to an occupation. - Incumbent education does not reveal education at career entry. - Occupational-group wage comparisons do not estimate individual promotion effects. - BEA Regional Price Parities are broad local price measures rather than occupation-specific cost indexes. - Public licensing coverage is incomplete; absence of a record cannot be treated as absence of a licensing requirement. - Transition behavior reflects opportunity, preferences, geography, hiring volume, labor-force conditions, and measurement error. A predictive score is not an individualized recommendation. - Publication rights for CareerOneStop-derived licensing aggregates remain under legal review; no licensing table is included in this article.