To estimate demand for graduates in agriculture-related fields from 2025 to 2030 and be consistent with prior reports, we began by anchoring our analysis in the Bureau of Labor Statistics (BLS) job projections using data from their Employment Projections Report. These projections provide a foundational, nationally representative estimate of occupational trends, which we used as a baseline for identifying relevant job categories.

The BLS relies on the U.S. Census Bureau and its American Community Survey (ACS) that projects population into the future. Because the BLS uses the ACS to project population, the BLS employment projections do not include Puerto Rico and the other U.S. territories. Additionally, they exclude U.S. armed forces overseas and civilian U.S. citizens whose usual place of residence is outside the United States. The U.S. resident population then utilized here includes all persons who usually reside in the 50 states and the District of Columbia.

Because BLS job titles are broad and do not always capture the agricultural relevance of a position, we enhanced this dataset using web-scraping and data mining techniques, specifically by extracting job postings from Google Jobs. Google Jobs aggregates listings from multiple major job boards, giving us wide  coverage of the employment landscape.

We used generic job titles from the BLS to initiate these searches, then filtered the postings by searching within job descriptions for agricultural-related terms and keywords. This allowed us to identify which job postings had a meaningful agricultural component, even if the job title itself was generic or ambiguous.

As employer search strategies have evolved, most employers no longer “require” a degree for new positions (some research/technical positions would be exceptions.) Employers may “prefer” or “desire” a degree in the position announcement but be open to hiring a person with experience but no degree.  In addition, many employers say nothing about degree preferences in their position announcement in order to maximize their applicant pool. Hence, most position announcements no longer specify a degree requirement, though some will mention degree preferences.

To assess demand for FARNRE graduates with an associate degree or higher as accurately as possible, the web scrape of position announcements was focused on job titles aligned with the four job cluster categories. This eliminated most positions that would be open to non-college graduates, such as seasonal farm labor.

About half of the job postings scraped did not specify any degree requirement/preference. For the other half, 20% mentioned a high school degree; the other 80% mentioned an associate degree or higher. Based on these mentions, we assume at least 80% of the total annual job openings will prefer at least an associate degree, and that some portion of the remaining 20% will as well. Overall, it is prudent to assume that 90% of the total annual job openings prefer at least an associate degree.

Each job was then classified into one of four major cluster categories — Business and Management, Science and Engineering, Food and Biomaterials Production, and Education, Communication and Governmental Services — based on the frequency of keywords tied to each area. A predefined list of terms guided this classification process, and each posting was assigned to the category with which it shared the most key identifying words. While one can argue that some jobs might span more than one job cluster, a research manager, for example, each job was classified into one of the four clusters.

These classified results provided the percentage distribution of agricultural job demand across the four cluster categories. These percentages are used later to align the supply side (graduate data) to demand projections.

This approach allows us to preserve consistency with prior editions of the report while taking advantage of more modern techniques — namely, web scraping and keyword-based classification — to provide a more accurate and nuanced understanding of current job market dynamics in the agricultural sector.

In addition to classifying agricultural job postings into broad occupational categories, we also used the web-scraped data to conduct a regional analysis of job demand. By capturing the geographic information included in the job postings, we were able to determine the percentage of each generic job category (as defined by BLS titles) that appeared in specific U.S. regions. This allowed us to break down national trends into more localized snapshots of employment demand.

Within each region, we further analyzed the distribution of jobs across the four major cluster categories — Business and Management, Science and Engineering, Food and Biomaterials Production, and Education, Communication and Governmental Services. This regional layer adds significant value for stakeholders seeking to understand where agricultural employment opportunities are emerging and how those opportunities vary in terms of job function and industry focus.