Selected measures and interpretation
This page sets out selected measures used in the plan. Anchor facts have an exact source scope. Supporting measures remain useful when their narrower meaning is stated. They do not substitute for evidence of demand or readiness, or for an area total that is not available.
Values below keep their analytical precision. The plan's narrative uses rounded figures for readability.
Employment by industry
These measures come from the Bureau of Labor Statistics Quarterly Census of Employment and Wages (QCEW), using the 2022 and 2025 annual files. Each value covers Arkansas statewide (area code 05000), private ownership (ownership code 5) and all establishment sizes (size code 0) for the named NAICS 2022 industry.
Job counts are direct published annual averages for 2025. Growth is calculated as 100 × (2025 jobs ÷ 2022 jobs − 1) over 2022–2025.
| Measure ID | Measure | Value | Period | Industry (NAICS 2022) | Role |
|---|---|---|---|---|---|
food_jobs |
Food manufacturing private jobs | 51,799 jobs | 2025 annual | Food manufacturing, 311 | Anchor |
food_growth |
Food manufacturing employment change | 3.75363 percent | 2022–2025 | Food manufacturing, 311 | Anchor |
metals_jobs |
Primary metals manufacturing private jobs | 8,860 jobs | 2025 annual | Primary metals manufacturing, 331 | Anchor |
metals_growth |
Primary metals manufacturing employment change | 9.9392 percent | 2022–2025 | Primary metals manufacturing, 331 | Anchor |
trucking_jobs |
Trucking private jobs | 31,318 jobs | 2025 annual | Trucking, 484 | Anchor |
trucking_growth |
Trucking employment change | -5.56343 percent | 2022–2025 | Trucking, 484 | Anchor |
retail_jobs |
Retail trade private jobs | 143,001 jobs | 2025 annual | Retail trade, 44-45 | Anchor |
retail_growth |
Retail trade employment change | 0.158993 percent | 2022–2025 | Retail trade, 44-45 | Anchor |
warehousing_jobs |
Warehousing private jobs | 15,624 jobs | 2025 annual | Warehousing, 493 | Supporting |
warehousing_growth |
Warehousing employment change | 4.34783 percent | 2022–2025 | Warehousing, 493 | Supporting |
All of these count covered private wage-and-salary jobs. They do not count unique people or self-employment, and they are not a complete measure of an emerging market. For the growth measures, a reclassification in the source can change counts, and the changes are not given a causal interpretation.
Employment change in combined industries
Two supporting measures add together named industries that do not overlap. Each uses Arkansas private ownership in the QCEW 2022 and 2025 annual files, calculated as 100 × (2025 sum ÷ 2022 sum − 1).
| Measure ID | Measure | Value | Period | Industries summed (NAICS 2022) | Role |
|---|---|---|---|---|---|
power_hardware_growth |
Employment change in named industries | -8.87545 percent | 2022–2025 | 3344 + 3353 | Supporting |
health_services_growth |
Employment change in named industries | 8.83432 percent | 2022–2025 | 621 + 622 | Supporting |
Both count covered private wage-and-salary jobs, with the same limits as the industry measures above. They do not isolate power electronics, clinical research or user demand.
Employment concentration
Concentration compares each industry's share of Arkansas private employment with its share of US private employment, calculated as (Arkansas industry jobs ÷ Arkansas private jobs) ÷ (US industry jobs ÷ US private jobs). Values use the QCEW 2025 annual file for the exact named NAICS 2022 industry in Arkansas and the United States (area code US000), private ownership, with the all-industry total (industry code 10) as the base.
| Measure ID | Industry | Value | Period | Role |
|---|---|---|---|---|
food_lq |
Food manufacturing, 311 | 3.50772 ratio | 2025 annual | Anchor |
trucking_lq |
Trucking, 484 | 2.54575 ratio | 2025 annual | Anchor |
These measures count covered private wage-and-salary jobs, not unique people or self-employment, and are not a complete emerging-market measure. They differ from the location quotient published by BLS, whose denominator includes all ownerships.
Rice production
| Measure ID | Measure | Value | Period | Role |
|---|---|---|---|---|
rice_share |
Arkansas share of US rice production | 45.2084 percent | 2025 annual | Anchor |
The value is 100 × Arkansas production ÷ US production, using physical rice production in hundredweight (CWT) from the same annual USDA NASS survey release for Arkansas and the United States. The source is the NASS Quick Stats crops file dated 24 October 2026, selecting the 2025 annual survey total for rice production. Production share does not measure farm income, irrigation efficiency, aquifer sustainability or adoption.
Business use of AI
| Measure ID | Measure | Value | Collection period | Role |
|---|---|---|---|---|
AI_AR |
Employer-business AI use, Arkansas | 11.7 percent | 17 November 2025 to 8 February 2026 | Anchor |
AI_US |
Employer-business AI use, United States | 17.9 percent | 17 November 2025 to 8 February 2026 | Anchor |
Both are direct published estimates from the Census Bureau Business Trends and Outlook Survey (BTOS) AI supplement table for 2026. They use question 1, scope 1: AI use in any business function in the prior two weeks, among employer businesses, taking the "Yes" estimate at state and national level.
The BTOS is an experimental survey, and its estimates carry standard errors. It provides no Arkansas estimate by sector. Results cannot be joined across the change in question wording in November 2025.
Academic R&D
| Measure ID | Measure | Value | Period | Geography | Role |
|---|---|---|---|---|---|
HERD_growth_AR |
Academic R&D expenditure growth | 33.8862 percent nominal | 2019–2024 | Arkansas | Anchor |
HERD_growth_US50DC |
Academic R&D expenditure growth | 40.7733 percent nominal | 2019–2024 | 50 states and DC | Anchor |
HERD_AR_2024 |
Academic R&D expenditures | 492.045 million nominal USD | FY2024 | Arkansas | Anchor |
The growth measures come from the National Center for Science and Engineering Statistics (NCSES) Higher Education Research and Development (HERD) Survey for 2019 and 2024, including both the full and short survey forms. They use the source's all-R&D totals with each institution counted once, calculated as 100 × (2024 expenditure ÷ 2019 expenditure − 1). These are nominal expenditures. They are not new awards, real growth or economic impact. Source imputation is kept, and the US figure covers the 50 states and DC and excludes territories.
The FY2024 Arkansas total is all R&D at standard and short survey institutions in Arkansas, from Table 5 of the NCSES data tables (NSF 26-304), converted from thousands of dollars. It includes expenditure imputed by the source. Survey spending is not a facility operating budget or a grant award amount.
Business R&D intensity
| Measure ID | Measure | Value | Period | Geography | Role |
|---|---|---|---|---|---|
BERD_intensity_05000 |
Survey-covered business R&D relative to GDP | 0.398123 percent of GDP | Calendar 2023 | Arkansas | Supporting |
BERD_intensity_00000 |
Survey-covered business R&D relative to GDP | 2.59396 percent of GDP | Calendar 2023 | All states | Supporting |
The value is 100 × R&D (millions of current dollars) ÷ GDP (millions of current dollars). R&D comes from Table 13 of the NCSES Business Enterprise Research and Development (BERD) Survey release NSF 25-354, using the Arkansas total and the all-states total. It covers domestic R&D performed by for-profit nonfarm companies with 10 or more US employees. GDP is the all-industry total (line 1) from BEA table SAGDP2 for Arkansas (05000) and the United States (00000).
This is a survey-estimated measure of research intensity across the economy. It is not a return on investment, a count of all business R&D or an Arkansas sector estimate. State estimates rely on modeling and imputation and can change with new releases and revisions.
SBIR/STTR awards
| Measure ID | Measure | Value | Period | Role |
|---|---|---|---|---|
SBIR_2015_firms |
Observed SBIR/STTR firm identities | 32 observed firm identities | 2015–2019 | Supporting |
SBIR_2015_amount |
Observed SBIR/STTR award amounts | 35.8288 million nominal USD of award observations | 2015–2019 | Supporting |
SBIR_2020_firms |
Observed SBIR/STTR firm identities | 32 observed firm identities | 2020–2024 | Supporting |
SBIR_2020_amount |
Observed SBIR/STTR award amounts | 66.8638 million nominal USD of award observations | 2020–2024 | Supporting |
These use the current national universe of SBIR/STTR award observations, selecting Arkansas as the state of award and all observations with a start year in each window. Firm counts are distinct source firm IDs. Amounts are the sum of reported award-observation amounts.
The measures do not count spin-outs, unique projects, obligations or commercialization. Equal counts do not mean the same firms appear in both windows. Ambiguous or repeated firm identities can affect the results.
Source verification
The original-source reconciliation compares selected QCEW and BEA CSV cells, NASS annual physical observations, BTOS workbook cells, and NCSES native and published totals with the values used in the analysis. The original-source provenance file records official URLs, retrieval dates and file hashes, which link each value back to its source.
This is an analytical check of scope and arithmetic. It does not replace review by subject experts or by the people who use the data.
Definitions
Employment concentration equals the Arkansas private-employment share of an industry divided by the same industry's share of the published US private-employment total. It is calculated separately from the BLS published location quotient, whose denominator includes government. A concentration of 3.5 means the industry's share of employment is 3.5 times as large. It does not mean the industry is 3.5 times larger in absolute size.
Growth equals 100 times the later value divided by the earlier value, minus one. Detailed QCEW trends use the same NAICS 2022 vintage. Spending growth here is nominal.
An R&D-to-GDP ratio uses both amounts in current dollars, for the same calendar year and compatible geography. It measures research intensity, not a return on investment.
A change in period, cohort or classification can rule out a comparison even when both values are numeric. Missing and suppressed source values are kept distinct from true zeros. A new operating measure needs a defined reporting population before an observed zero or rate can be treated as valid.