Loading page content
Preparing content…
SOC 19-1012
Agricultural and food scientists use experiments, data and life science to solve animal, food, crop and soil problems; it suits people who first test, with one reviewable small experiment, whether they accept precision, field constraints and responsibility for results.
Model impact index for three equally weighted representative core tasks; not a measured share of working time or a job-loss probability. floor((50+50+50)/3+0.5)=50; fermatmind-ai-impact-v1; 2026-10-18.
In one sentence: a role that uses life science and statistical methods to turn specific 'farm to table' problems into reviewable conclusions; precision, field constraints and responsibility for results all have to be accepted together.
The U.S. wages and employment on this page use the BLS national combined occupation group basis (wages for 2025, month 5 (May); projection 2025—2035), and do not cover the self-employed or some agricultural workers; mainland China has no precise national wage distribution for this occupation, so industry-sector figures serve only as a broad reference and cannot be treated as an individual's starting pay or median.
People more likely to adapt: those willing to work with samples, data and field conditions over the long term, able to accept failed experiments and keep the failure record, and not averse to the idea that 'a conclusion only extends to certain conditions'. If you want only conceptual discussion and a good-looking result every time, the gap in experience is usually larger.
The BLS combined occupation group 19-1010 contains three directions: animal scientists 19-1011, food scientists and technologists 19-1012, and soil and plant scientists 19-1013; they study the efficiency, quality and safety of agricultural and food production. Scientists focus on research design and interpretation, technicians may also analyse results and prepare reports; agricultural engineers focus on equipment, facilities and engineering systems, and each position's authority must be checked separately.
The tasks of agricultural and food scientists fall into three categories: designing and running experiments (question, hypothesis, control, sampling, analysis); turning results into executable plans (process, quality standards, technical reports and recommendations); and recording and collaboration (raw data, versions, review and cross-team communication). The weight of the three varies with the animal, food, soil and plant directions.
Career direction comparison
The table below compares the three directions within this occupation with three adjacent occupations, using the same comparison dimensions: direction, core work and output, key differences, and what choice it suits better.
| Career direction | Core work and output | Key differences | Best-fit choice |
|---|---|---|---|
| Animal scientists (O*NET 19-1011.00) | Studies the genetics, nutrition, breeding and husbandry management of farm animals, and produces feed nutritive values, breeding and disease-control plans, and technical advice for producers. | The objects are economic animals and their products (meat, milk, eggs and so on), which require animal facilities, ethics review and biosafety conditions. | People willing to work with livestock and feed sites over the long term and able to accept animal experimentation and biosafety restrictions. |
| Food scientists and technologists (O*NET 19-1012.00) |
AI impact on this career
Editorial suggestion: trial the organising of literature, records, code and image candidates on fictional data; this page obtained no adoption rate, hours saved or substitution effect for this occupation. Technical capability, organisational adoption, number of positions and authorisation responsibility are checked separately. AI Impact Score 50% is an editorial estimate. Model impact index for three equally weighted representative core tasks; not a measured share of working time or a job-loss probability. floor((50+50+50)/3+0.5)=50; fermatmind-ai-impact-v1; 2026-10-18.The three tasks, ratings and sources are listed in the FermatMind method source record on this page.
FermatMind AI Impact Score estimation method v1
FermatMind / 费马测试 · 2026-10-18
Chinese mainland salary reference
Under the Industrial Classification for National Economic Activities (GB/T 4754-2017), positions related to this occupation are spread across sectors such as agriculture, forestry, animal husbandry and fishery; manufacturing; and scientific research and technical services; the National Bureau of Statistics does not list a separate wage for 'agricultural and food scientists', so this page does not pass off industry figures as the occupational median.
The China side of this page uses a broad proxy by industry sector and unit type (agriculture, forestry, animal husbandry and fishery; manufacturing; scientific research and technical services, and so on); it is not a national wage distribution for this occupation, and industry figures must not be treated as the occupational median.
Insufficient evidence: no national monthly or annual wage distribution for the precise occupation 'agricultural and food scientists' in mainland China was obtained; the National Bureau of Statistics publishes only by industry sector and unit type.
Insufficient evidence: no by-city or by-region wage sample for this occupation was obtained, and job-platform data were not included as a verifiable source.
U.S. pay reference
The median annual wage for agricultural and food scientists in the United States in 2025, month 5 (May), is $81,970; the lowest 10% are below $48,730 and the highest 10% are above $142,300. Officially published values for the 25th and 75th percentiles of the combined occupation group were not obtained in this round.
| Position in the wage statistics | Annual wage reference | Conversion boundaries | How to understand this position |
|---|---|---|---|
| Lowest 10% | Below $48,730/year | Official annual wage, not individual monthly income | The lowest 10% earn below $48,730/year on the national combined occupation group basis; this does not represent a new graduate's starting pay. |
| Median (50%) | $81,970/year | Official annual wage, not an individual's actual pay |
Occupation fit guide
IR is the interest code for animal scientists 19-1011, and food 19-1012 is IRC; it is not a unified label for the combined group. Observe whether you are willing to repeat research, recording and field tasks; ability or hiring cannot be inferred from interests.
You may be better suited to： Fit signals come from work behaviour: whether you are willing to ask where the data came from, accept failed experiments, patiently verify detail, and go back to the sampling process when field observations conflict with model results. Investigative and Realistic interests may help, but a personality label, MBTI or a single assessment cannot determine competence.
These situations call for caution： Situations calling for caution: expecting every experiment to give a good-looking result, being unwilling to keep failure records, treating statistics and records as a burden, or being unwilling to go into farms and processing sites to understand the real conditions.
Do one entry exercise： Do one entry exercise: complete a small experiment with public data (question, samples, control, raw data, analysis code, verification, deviation and a one-page recommendation), and deliberately keep the failure and uncertainty parts.
Work pressure, risks and occupational boundaries
What most needs deciding in advance is the responsibility boundary for data and conclusions: sampling bias, the sample chain, extrapolation and mistaking correlation for causation all affect real production and safety decisions. Risks differ markedly between directions: the food direction is closer to release and regulation, the animal direction involves animal health and biosafety, and the soil and plant direction involves seasonality and resource issues.
Evidence scope: the O*NET 19-1011/19-1012/19-1013 task and work context material is U.S. single-occupation respondent data; the Food Safety Law applies to food production and trade in mainland China; industry operating standards and organisational rules differ. The risk list is an editorial synthesis and contains no probabilities of occurrence.
Typical scenario：A single field trial covers only 3 plots and 1 variety, yet the team plans to write the conclusion as applying to the whole production area.
Roles most affected：Research design and field or production-line sampling positions
Entry, qualifications and career development
The typical U.S. entry point is a bachelor's degree in agriculture or a related science, with coursework covering biology, chemistry, statistics, experimental design and the target direction; some independent research positions require a master's or doctorate (O*NET 19-1011 education survey: 43% doctorate, 22% master's, 13% bachelor's, as respondent shares). Before starting, confirm whether experimental results can be independently reviewed, rather than looking only at education.
Career prospects and moves to related occupations
The BLS combined group had employment of 37,400 in 2025, with growth of 6% over 2025—2035, a net increase of about 2,400 and about 2,700 openings per year including growth and replacement. The OOH also lists animal 6%, food 6% and soil and plant 7% separately; this is a national projection, not a personal guarantee or an AI cause.
United States national (combined occupation group 19-1010) · BLS combined group 19-1010: the three directions of animal, food, and soil and plant
About 37,400 people (2025)
Employment
The demand mechanisms include improving agricultural production efficiency and product quality, responding to pests, diseases and changes in climate and weather, safeguarding food safety, and meeting demand for agricultural products; these mechanisms explain why research is needed, but they do not automatically translate into each organisation's budget.
Turn animal, food, crop or soil problems into experiments: design sampling and controls, record and analyse data, verify models, and then turn the results into technical reports and recommendations usable by production, quality or regulation.
This occupation is responsible for research design and interpretation of conclusions; technicians carry out sampling and testing; agricultural engineers are responsible for equipment, facilities and engineering systems; data scientists provide general methods but do not automatically take responsibility for agricultural experiments or food release.
Choose a real task from this page for a small practice project or a conversation with someone in the field. Use the planning guides to set a date, an output and feedback questions before deciding whether to continue.
The core of the work of agricultural and food scientists is turning animal, food, crop or soil problems into testable experiments and data-based conclusions, and taking responsibility for where those conclusions can be used. When judging fit, first look at whether you are willing to give time to sampling frames, controls, replicates, raw records and failed experiments, and only then consider interests.
Three small self-test tasks: rewrite 'does a certain feed additive increase milk yield' as an experiment with a control, replicates and metrics; use one public dataset to complete cleaning, a baseline and an error statement; and explain the uncertainty of a conclusion to a non-research colleague. Doing them incompletely is normal; the point is to observe your own real reaction to verifying detail, admitting what is unknown, and going back to the field to check. (Editorial suggestion, not a hiring standard.)
Carries out sampling and testing in the laboratory, at processing facilities or in the field, recording conditions and anomalies to standard methods.
Cleans and analyses experimental or field data, tests results with statistical models, and keeps scripts and versions.
The boundary is set by the research object, the output and the authorisation: scientists focus on research design and interpretation; technicians may sample, test, analyse results and prepare reports; engineers focus on equipment and facilities. Whether a data scientist carries domain research responsibility depends on the actual position, and a job title neither automatically grants nor excludes food release authority.
| Dimension | Primary track | Comparison track |
|---|---|---|
| Core purpose | Answers 'whether, why and under what conditions' production and food safety questions with experiments and data. | Turns reviewable conclusions into plans or standards usable by production, quality or regulation. |
| Main objects and basis | The objects are animals, food, crops, soil and their production and processing; the basis is experimental data, statistical methods and official standards. | Does not count adjacent occupations' duties (equipment design, system development) as core tasks of this occupation. |
| Core process | Question and hypothesis → Sampling and experiments → Data cleaning and statistics → Verification and boundary statement → Translation and monitoring. | AI can assist with organising and first-pass screening, but sampling design, anomaly determination and method validation remain the responsibility of people. |
| Main outputs | Reviewable experimental records and datasets, technical reports, formulation, feeding, cultivation or testing specifications, and quality and risk recommendations. | Output is judged by whether it can be independently reviewed and implemented, not by the number of report pages. |
| Responsible party | Research conclusions are the responsibility of the researchers who designed and signed them; release, recall and regulatory handling are decided by authorised positions under local regulations. | Editorial note: because organisations divide authorisation differently, written duties should be checked before starting the job. |
| Authority and approval | Formulation release, food safety conclusions, veterinary drug use and production changes require authorisation and records; holding the data does not automatically confer approval authority. | Specific authority is set out in the employer's rules, industry norms and local regulations. |
| Boundary of conclusions | Conclusions hold only within the given breed, region, process and time range; correlation must not be written as causation, nor a single experiment extrapolated into a general rule. | Uncertainty and failure records should be kept, not deleted because a conclusion looks unfavourable. |
Turn a production or quality problem into a testable hypothesis, specifying the target population, control, replicates, metrics and exclusion conditions.
Sample by standard methods and record conditions, compare treatment with control using statistical models, and keep raw data, scripts and versions.
Verify conclusions with independent data, state the applicable breed, region, process and time range, and separate correlation from causation.
Turn conclusions into formulation, feeding, cultivation, processing or testing specifications, and state the preconditions for implementation and the monitoring metrics.
Keep the sample chain, calibration and handover records; formulation release, safety and regulatory conclusions are signed by authorised personnel.
Explain results to production, quality, farmers or regulators, stating which conclusions are reliable and which are still unknown.
Before starting the job, check at least six things: which of the three directions the position belongs to; whether the employer is a research institution, an enterprise or a testing and regulatory body; the contract and identity conditions; the experimental facilities and safety training; the ownership of data and results; and who signs off on release, recall and safety conclusions.
Look at the task description in the original job posting, not the job title.
Check project funding, contract term, renewal conditions and arrangements after the project ends; this page obtained no share of project-based positions for this occupation.
O*NET 19-1011 education survey: 43% doctorate, 22% master's, 13% bachelor's (respondent shares, not the education distribution of the whole industry).
| Studies ingredients and formulations, processing and preservation, packaging and sensory quality, and produces process parameters, quality standards, shelf life and safety assessments. |
| The objects are food ingredients and processing, and the work is closer to the production line and to regulations (labelling, additives, microbiological limits). |
| People interested in food processing and regulation and willing to collaborate frequently with production and quality teams. |
| Soil and plant scientists (O*NET 19-1013.00) | Studies crop growth, soil fertility and contamination remediation, pest and disease control and nutrient management, and produces cultivation and fertilisation plans and soil monitoring recommendations. | The objects are crops, soil and field systems, with strong seasonality, much outdoor work, and sampling and spatial variation as the key difficulties. | People willing to accept seasonal field work and attentive to resource and ecological constraints. |
|---|
| Data scientists (adjacent occupation) | Builds cross-industry data pipelines and models and produces analysis tools, metrics and forecasts; does not take responsibility for agricultural experimental design, food release or biosafety. | The methods are general statistics and machine learning, without biological or regulatory background; responsibility for business conclusions lies with the business side. | People who would rather focus on modelling and engineering and do not want responsibility for samples and field work. |
|---|
| Agricultural engineers (adjacent occupation) | Designs agricultural machinery, irrigation and facility systems, producing equipment and engineering plans with a focus on energy, structures and automation. | Leans towards physical systems and engineering design; agricultural scientists lean towards biological mechanisms and experimental conclusions. | People who want to work on equipment and facilities and prefer engineering design problems. |
|---|
| Agricultural and food science technicians (adjacent occupation) | Carries out sampling, testing and recording by standard methods, producing test data and reports; does not take responsibility for research design or final interpretation. | An execution and support role, where methods are set by a supervisor or standard, and scientific and interpretation responsibility lies with scientists. | People who want to enter laboratories and the field first to build operating experience before deciding whether to pursue further study. |
|---|
Suggested order of choice: first decide the research object (animal / food / soil and plant), then look at the form of the output (plans, standards, test data or engineering systems), and finally confirm who signs off on conclusions. The three directions share experimental design and statistical skills; moving to an adjacent occupation requires building up that occupation's own output and compliance evidence.
Once the direction is decided, check the tasks and entry requirements that genuinely exist in each direction.
Editorial ratings use the existing occupational responsibilities and AI task analysis on this page. They are not adoption rates, measured effects, job-disappearance or personal-unemployment probabilities, or official scores from Stanford or another external institution. No measured time weights or future robot deployment are assumed.
Technically, AI (including large models, image recognition and statistical learning) can take on literature search and summarising, log and record organisation, code and model drafts, and first-pass screening of image and sensor anomalies. These outputs can be used to generate candidate options, but they require professional judgement and verification.
Whether an organisation adopts it depends on whether the data represent the target object, whether labels are reliable, whether metrics are clearly defined, whether review and deviation monitoring are arranged, and whether the safety and compliance requirements of the laboratory or production environment are met. Technical feasibility is not the same as having been deployed.
Insufficient evidence: this page obtained no independent causal measurement of AI's effect on the number of positions in this occupation. The BLS2025—2035 growth of 6% is an aggregate projection and cannot prove that AI has reduced positions or that it will not reduce them.
| Work area | Task | Current change | Human control point |
|---|---|---|---|
| Hypothesis and experimental design | Turn a production or quality problem into an experimental plan with controls and replicates, and determine the target population, sample size and metrics. | AI can assist with searching the literature and generating candidate designs, but there is no empirical evidence that it can replace decisions on sampling frames, blinding and exclusion conditions, which remain the researcher's judgement. | Confirm whether the sampling represents the target population, whether control and blinding are feasible, and record the design version. |
| Sampling and testing | Collect and process samples by standard methods, record conditions, times and anomalies, and retest where necessary. | Image and sensor models can provide a first-pass screening of anomalies, but validation cases under the target variety, equipment and field conditions are lacking (this page obtained no evidence of this task's realised effect). | Isolate anomalous results and retest first; method versions and calibration are confirmed by personnel with the relevant qualifications. |
| Data cleaning and statistics | Organise the raw data, handle missing values and outliers, and compare treatment with control using statistical models. | Existing tools can assist with data cleaning, code generation and visualisation; whether an organisation deploys them depends on data governance and review arrangements. | Check numerators, denominators, units and periods, and whether model assumptions and uncertainty have been reported. |
| Models and prediction | Build and validate predictive or classification models for yield, quality, disease or risk prediction. | Editorial suggestion: compare models with a baseline and independent validation data under the target region, variety and equipment conditions; this page did not verify how widespread adoption is or the size of the effect across settings. | Ensure the validation data are isolated from the training data, and keep the failure thresholds and stop conditions. |
| Records and compliance | Organise experimental records, the sample chain, calibration and versions into a reviewable file. | AI can assist with format organisation and summarising, but it does not change record responsibility or retention requirements. | The person responsible for the records confirms that the content is accurate, and the original files and revision traces are kept. |
| Translation and recommendations | Turn conclusions into formulation, feeding, cultivation or testing specifications, and give the preconditions for implementation and the monitoring metrics. | There is a lack of evidence that AI can replace judgements about process conditions and the applicability of regulations. | An authorised position confirms the implementation scope, risks and any items requiring approval. |
| Communication and training | Explain conclusions, boundaries and follow-up actions to production, quality, farmers or regulators. | AI can assist with generating drafts and multilingual versions, but the accuracy of external wording requires a final human edit. | Check the data, basis and legal boundaries before the spokesperson releases anything externally. |
The five evidence cards each explain the source, the research object, the conclusion and the usage limits; they support technical capability, task and regulatory boundaries, and are not AI substitution rates or predictions of job disappearance.
The table below compares how the main task directions within this occupation differ in AI impact.
AI primarily changesEditorial trial directions: organising literature and conditions, and first-pass screening of candidate options and images; errors and review time should be recorded, and no speed-up effect for this occupation has been verified.
Human responsibility remainsSampling frames and control design, anomaly determination, method validation and the boundaries of conclusions.
AI primarily changesOrganising record formats, summarising and retrieval can be automated; versions and fingerprints can help leave an audit trail.
Human responsibility remainsResponsibility for the accuracy and completeness of records, record-retention requirements and review sign-off.
AI primarily changesEditorial trial directions: cleaning, code and chart drafts; these must be verified against the original data, the method and the baseline, and no speed-up can be promised.
Human responsibility remainsIsolating the validation set, choosing metrics and failure thresholds, the boundaries of extrapolation, and the decision on whether to adopt.
The five steps below are an editorially suggested process for separating technical capability, human review and authorisation; it is not a mandatory process laid down by the source.
First write down the task objective, data sources, the permitted scope of use and what is forbidden, and confirm whether food safety, animal health or personal information is involved.
Limit the tool to permitted steps, such as literature summarising, a data-cleaning draft or image first-pass screening, and do not let it produce release or safety conclusions directly.
A person with the relevant training reviews input representativeness, output error and anomalous samples, and records the review conclusion and any unresolved questions.
Keep the data version, model or tool version, evaluation metrics, failure samples and the reasons for decisions, so that conclusions can be independently reviewed.
An authorised position confirms whether to adopt it and how to implement it; release, recall and regulatory submissions are signed under local regulations and organisational procedures.
Agricultural and food conclusions affect formulation, fertilisation, feeding or safety judgements, and unverified extrapolation can cause real losses.
Control: Use an independent validation set and comparison with a baseline, define failure thresholds and stop conditions clearly, and keep the raw data and review records.
Research data may contain commercial formulations, undisclosed variety information or farmer information, and leakage would harm the client and individuals.
Control: Manage access and retention by data classification, use de-identified or synthetic data for demonstrations, and obtain authorisation before any external sharing.
If a conclusion cannot be traced back to its inputs, versions and parameters, neither review nor reproduction is possible and responsibility becomes hard to define.
Control: Record the data version, tool or model identifier, parameters and evaluation results, and keep the change history.
Unit conversion, period basis or denominator errors can directly change the direction of a conclusion, especially when comparing across regions.
Control: Verify units, periods and statistical bases item by item, and recalculate important values with a program and leave an audit trail.
When it is not clear who reviews and who approves, a wrong implementation recommendation may end up with no one responsible.
Control: Write the reviewer, approval authority and escalation path into the process, and keep the decision records.
Practise experimental design, statistics and one data-handling tool until you can complete a small project independently; build portfolio work with public or synthetic data, and do not collect private or commercial confidential material.
Fill in the biology and process knowledge of the target direction, and use one small experiment to show you can handle field variation and record-keeping requirements.
Practise moving from execution to design and interpretation: write hypotheses, controls and exclusion conditions clearly, and learn to express conclusions with statistics and uncertainty.
Establish boundaries for AI use and record standards, specify reviewers and approval authority, and manage tool output separately from formal conclusions.
Insufficient evidence: this page obtained no study of AI adoption rates, efficiency gains or overall substitution for this occupation. Sampling, method validation and approval responsibility follow actual authorisation, and an aggregate growth projection cannot be used to promise that the occupation will not be substituted.
Source: BLS Occupational Outlook Handbook (2025 edition)Original source ↗
Check the statistics, spreadsheet, scripting or modelling requirements against the target position; the O*NET technology list does not prove that programming is a threshold for all positions or that its importance is rising. Domain knowledge and experimental design still have to be checked independently.
Source: O*NET 19-1011.00 technology list and task materialOriginal source ↗
Input-organising and candidate-generating tasks such as literature search and summarising, record and log organisation, code and chart drafts, and image and sensor first-pass screening; the degree of impact depends on data quality and on whether the organisation arranges review.
Source: O*NET 19-1011.00 tasks and technology list (editorial synthesis)Original source ↗
The BLS combined group is projected to move from 37,400 in 2025—2035 to 39,800 people, a growth of 6% and a net increase of about 2,400; the original chapter also lists animal 6%, food 6% and soil and plant 7% separately. The combined group's growth cannot be applied to each direction, and it does not separate out AI's contribution.
Source: BLS Occupational Outlook Handbook: agricultural and food scientistsOriginal source ↗
No. Test conclusions and decisions such as release and recall require sign-off by personnel with the relevant qualifications and authorisation, and tool output can only serve as a draft or supporting evidence. The specific signing authority and procedure are set by the employer's rules and local regulations; for food-related requirements in mainland China, see the Food Safety Law of the People's Republic of China in the sources.
Source: O*NET 19-1011.00 tasks and work activities and organisational authorisation boundaries (editorial note)Original source ↗
This page provides no year of disappearance. Task exposure research measures the degree to which a task can be handled technologically, and does not equal an actual adoption rate or a share of positions reduced; treating exposure as a disappearance date has no supporting evidence.
Source: This page's evidence cards and source policy (editorial note)Original source ↗
Suggested order: experimental design and statistical foundations, biological and process knowledge in the target direction, and data management with traceable records; only then learn data cleaning and modelling tools. Check specific courses and certificates against the target position's requirements, and do not pile up certificates to pad a CV.
Source: O*NET 19-1011.00 education and task material (editorial suggestion)Original source ↗
The material above separates technical capability, actual adoption and job outcomes; the next section verifies the wage and employment data basis for this occupation.
Insufficient evidence: no wage statistics for this occupation broken down by unit type (government or public institution, state-owned, foreign-invested, private) were obtained.
Scenario comparison requires the city, employer, duties and pay basis to be consistent; a recruitment or project sample cannot represent the national distribution.
| Educational background | Common job directions | Notes and verification advice |
|---|---|---|
| Bachelor's degree (new graduate) | Testing and quality control, and field and laboratory technical positions | The scenario is an exploration direction proposed by the editors from the tasks; this page obtained no mainland China bachelor's-degree recruitment or starting-pay sample, and it is not claimed to come from verified recruitment material. |
| Master's degree | Product development, process and formulation research | The scenario is an editorial synthesis; specific requirements are governed by the target employer's original job posting. |
| Doctorate | Independent research, and university and research institute positions | Insufficient evidence: this page obtained no sample of contract types or pay for doctorate-level positions in mainland China, and it cannot be generalised that these are mainly project-based or permanent-establishment positions. |
| Career change or in-work further study | Moving from a testing or production role into research and development | The editors suggest adding statistics and target-process coursework; whether a move into research is possible depends on verified education, actual experience and recruitment requirements, and the frequency of this path has not been checked. |
When checking your own quote, first confirm the work location, unit type (government or public institution, state-owned enterprise, private), the job duties and whether bonuses are included; do not annualise a single project's or a single employer's quote into a national level.
Check the basic wage, bonuses, project pay and renewal arrangements in the contract; this page did not measure differences in this occupation's wages and stability across unit types.
Check real postings in the target region, the industry affiliation and the cost of living; this page obtained no data linking industrial clusters to the number of positions in this occupation.
Check education, training and authorisation against the actual research and signing duties; this page did not measure the contribution of education, project funding or results to wages in this occupation.
Editors may trial literature, data and image drafts, but there is still no measurement of AI adoption, efficiency gains or wage causation for this occupation. Industry averages do not give the main reasons for individual wage differences; the specific contract, duties and pay terms must be checked.
| Median $81,970/year; the medians for the three sub-occupations are $68,940/$88,720/$78,850 respectively, and the different bases must not be mixed. |
| Highest 10% | Above $142,300/year | Official annual wage, not an individual's actual pay | The highest 10% earn above $142,300/year; this is a position in the combined group's wage distribution, and this page did not verify the causes of income differences or attribute them to research and development or management duties. |
|---|
Pay · 10th percentile：Below $48,730/year；The lowest 10% of workers earn an annual wage below this value； this is the national combined occupation group basis, not a newcomer's starting pay.；BLS Occupational Outlook Handbook: agricultural and food scientists (2025 edition)
Pay · 25th percentile：Insufficient evidence: this page obtained no 25th percentile annual wage for the 19-1010 combined group；O*NET provides the 25th percentile only for the separately listed occupations (19-1011/19-1012/19-1013)； this page does not fill the combined group with a single-occupation basis.；BLS Occupational Outlook Handbook: agricultural and food scientists (2025 edition)
Pay · median：$81,970/year；Half of the covered workers are above and half below this value； it does not cover the self-employed or some agricultural workers.；BLS Occupational Outlook Handbook: agricultural and food scientists (2025 edition)
Pay · 75th percentile：Insufficient evidence: this page obtained no 75th percentile annual wage for the 19-1010 combined group；As above, a separately listed occupation's percentile is not treated as the combined group distribution.；BLS Occupational Outlook Handbook: agricultural and food scientists (2025 edition)
Pay · 90th percentile：Above $142,300/year；The highest 10% of workers earn an annual wage above this value； this is not an automatic ceiling reached after a certain number of years.；BLS Occupational Outlook Handbook: agricultural and food scientists (2025 edition)
No. A median means half of the covered workers are above that value; starting pay is affected by region, specialism, education and employer type.
A statistical hourly wage multiplied by an individual's actual hours must not be treated as that person's annual salary; the OEWS hourly and annual wages are the same statistical conversion and do not represent an individual's paid hours or take-home income.
From the BLS Occupational Employment and Wage Statistics (OEWS) for 2025, month 5 (May) and the BLS Occupational Outlook Handbook 2025 edition; both are published on the national combined occupation group 19-1010 basis.
Specialism (animal / food / soil and plant), industry, region, education, whether independent research and signing responsibility is held, and employer type all affect the actual quote.
| Industry | Median wage for this occupation in this industry | How to understand this value |
|---|---|---|
| Research and development (physical, engineering and life sciences research) | $93,950/year | Median wage for this occupation in this industry (BLS, 2025, month 5 (May)); the industry basis is not equal to the national median. |
| Food manufacturing | $86,020/year | Median wage for this occupation in this industry (BLS, 2025, month 5 (May)); the industry basis is not equal to the national median. |
| Government (excluding state and local education and hospitals) | $80,690/year | Median wage for this occupation in this industry (BLS, 2025, month 5 (May)); the basis excludes state and local education and hospitals. |
| State universities and professional schools | $66,100/year | Median wage for the combined occupation in BLS state universities and professional schools, $66,100/year; this page did not check the share of teaching or differences by employment status. |
OEWS publishes both annual and hourly wages; 2,080 hours is a statistical conversion basis, not each person's actual paid hours, and it cannot be used to derive an individual's take-home monthly income.
$81,970 is the median for the 19-1010 combined occupation group; in 2025, month 5 (May), the medians for the three separately listed occupations are $68,940 for animal scientists, $88,720 for food scientists and technologists, and $78,850 for soil and plant scientists.
Median wages for the same occupation differ markedly across industries (research and development $93,950, food manufacturing $86,020, government $80,690, state universities and professional schools $66,100); job duties and responsibility for results also affect the quote.
The BLS combined group had national employment of about 37,400 people in 2025, and the projection scope includes the self-employed； table 1.2 gives the self-employed share as 4.4%. The OEWS wage scope, which excludes the self-employed, must not be applied here.
The national combined group had a net increase of about 2,400 people； the OOH also lists animal 6%, food 6% and soil and plant 7% separately, so it cannot be said that no split by specialism exists or that the net change can be recomputed from rounded headcount differences.
Average annual openings include replacement demand from transfers and exits and are not equal to net new positions.；BLS Occupational Outlook Handbook: agricultural and food scientists (2025 edition)
Which Holland interest type is closer to agricultural and food scientists?
The Holland interest code is IR (Investigative · Realistic, O*NET 19-1011.00). Investigative corresponds to research and analysis work and Realistic to outdoor, hands-on and equipment-operation elements; this describes an interest direction only and is not a hiring standard.
Do not use it this way: Caution is needed for people who like purely theoretical reasoning and are unwilling to sample hands-on or verify records.
Test my Holland vocational interestsWhich Big Five traits are more relevant to this job?
A Big Five self-assessment can help you reflect on experiment planning and verification habits; this page obtained no validation study in which scores predict performance in this occupation, and similarity between conscientiousness or openness and the tasks is not treated as an empirical association.
Do not use it this way: Treating one assessment result as a career conclusion ignores real task experience.
View my Big Five personality traitsCan MBTI be used to judge whether I am suited to being an agricultural and food scientist?
MBTI is not a career-matching tool, and there is no evidence supporting 'a certain type is best suited to agricultural and food scientists'; this page uses it only as language for self-observation, not as a basis for choosing a career.
Do not use it this way: Replacing task experience with a type label makes it easy to miss real fit information.
Understand my MBTI preferencesCan the Enneagram tell me whether I am suited to this occupation?
The Enneagram is likewise a self-exploration tool and cannot judge whether you are suited to this occupation; a more reliable approach is to record your own real feelings during sampling, analysis and interpretation tasks.
Do not use it this way: Explaining failure with a type label reduces the chance of changing specific practices.
Explore my Enneagram patternCan I be an agricultural and food scientist if I am not good at mathematics?
IQ and numerical reasoning: this occupation requires statistical and data-handling ability, but there is no uniform IQ threshold; a weak mathematical foundation can be built up step by step with coursework and practice.
Do not use it this way: Treating 'bad at maths' as a fixed label underestimates trainability.
Understand my intelligence and reasoning performanceDo agricultural and food scientists need high emotional intelligence?
Emotional intelligence: research work also needs cross-team communication and conflict handling, but emotional intelligence is not a hard hiring criterion; what matters more is whether you can state boundaries clearly when a conclusion is uncertain.
Do not use it this way: Avoiding communication or treating disagreement as a personal conflict slows projects down.
Understand my emotional intelligence abilitiesFits better when: Suits people who build up sampling, testing and recording experience first and then decide whether to pursue further study.
Watch for: Requires accepting an execution role and established methods, with limited authority to interpret.
Explore related career: Agricultural and food science techniciansFits better when: Suits people who prefer equipment, facilities and engineering systems.
Watch for: Requires additional knowledge of engineering design and physical systems.
Explore related career: Agricultural engineersFits better when: Suits people interested in livestock nutrition, breeding and husbandry sites.
Watch for: Requires accepting animal facilities, ethics and biosafety conditions.
Explore related career: Animal scientistsFits better when: Suits people interested in formulation, processing and food safety regulations.
Watch for: Requires working close to the production line and taking on regulatory and labelling responsibility.
Explore related career: Food scientists and technologistsFits better when: Suits people who care about crops, soil and resources and can accept seasonal outdoor work.
Watch for: Field conditions cannot be controlled, and sampling and spatial variation are the difficulties.
Explore related career: Soil and plant scientistsFits better when: Suits people who would rather focus on modelling and engineering.
Watch for: Lacks biological and regulatory background, so domain knowledge must be added.
Explore related career: Data scientistsThere is no 'suited' personality. It is more reliable to look at behaviour: whether you can accept long experiments, repeated verification and expressing uncertainty.
An introvert label does not determine suitability. Do a written verification and a mock explanation separately, and record whether you state the data and the limits and whether you are willing to repeat; a single performance does not infer long-term competence.
An extrovert label does not prove the ability to drive things across departments. Use the same case to verify data, hear objections and explain limits, record omissions and revisions, and draw no conclusion about personality advantage or hiring.
No. MBTI is not a career-matching tool, and there is no evidence supporting any one type being best suited to this occupation.
The O*NET 19-1011.00 interest code is IR (Investigative · Realistic) and 19-1012.00 is IRC. An interest code describes work-content tendencies and is not a hiring standard.
The correlation between personality scores and job performance is limited and cannot predict individual performance; use actual tasks and outputs to judge.
Mathematics is not a fixed threshold, but statistics and data handling are core abilities in this occupation; they can be built up through coursework and practice.
Emotional intelligence helps but is not a hard criterion; what matters more is stating boundaries when a conclusion is uncertain and managing collaboration well.
No. The Enneagram is a self-exploration tool and does not constitute evidence of occupational fit.
Prepare a small output for each direction: a feeding-data analysis note for the animal direction, process parameters and a safety assessment for the food direction, and field sampling and analysis for the soil and plant direction; compare how the three outputs actually feel.
Possible consequence：Samples cover only a few batches, locations or varieties but are written up as a general rule, which can mislead farmers or production decisions.
How to reduce risk：Define the target population and exclusion conditions first, use random or stratified sampling and record missing samples; state the range of generalisation in the report, and do not use a larger sample size to conceal a wrong sampling frame.
Evidence：O*NET OnLine 19-1011.00 tasks and work activities · O*NET OnLine 19-1012.00 tasks and work activities
Typical scenario：A batch of samples had no temperature recorded during transport and the test results deviate markedly from historical levels, yet a report was still demanded directly.
Roles most affected：Laboratory testing and quality control positions
Possible consequence：Loss of control over contamination, temperature, the sample chain or instrument calibration makes results unreviewable and can even lead to a wrong release or rejection.
How to reduce risk：Use blanks, controls, replicates, blind samples and calibration records, and keep timestamps and handover records; isolate and retest anomalous results first, then let authorised personnel decide whether to escalate.
Evidence：O*NET OnLine 19-1011.00 tasks and work activities · O*NET OnLine 19-1011.00 Work Context
Typical scenario：A model performs well in its training region but is required to be used directly for prediction in another region with a different climate and varieties.
Roles most affected：Research positions doing data modelling and prediction
Possible consequence：Treating correlation as causation, or extrapolating a regional model to other varieties and climate conditions, can produce wrong formulation, fertilisation or risk conclusions.
How to reduce risk：Use an independent validation set and comparison with a baseline, run sensitivity analysis and register versions; decisions affecting food safety, animal health or the environment are carried out under local regulations and organisational procedures.
Evidence：O*NET OnLine 19-1011.00 Work Context · O*NET OnLine 19-1012.00 tasks and work activities
Typical scenario：An external partner asks for raw data containing formulations and farmer information, and no data-use agreement has been signed yet.
Roles most affected：Research positions involving formulations, undisclosed varieties and farmer information
Possible consequence：Research data may contain commercial secrets and personal information, and sharing it externally or using it improperly harms the client and other people's rights.
How to reduce risk：Manage access and retention by data classification, use de-identified or synthetic data for demonstrations, and obtain authorisation before any external sharing.
Evidence：O*NET OnLine 19-1011.00 Work Context · O*NET OnLine 19-1012.00 tasks and work activities
Typical scenario：Key experimental records exist only on a personal computer, and method versions and parameters are not retained centrally.
Roles most affected：All positions that keep experimental and testing records
Possible consequence：When records are missing or leave no trace of revisions, conclusions cannot be reproduced, and both review and tracing of responsibility are blocked.
How to reduce risk：Keep the raw data, method versions, parameters and evaluation metrics, and retain the change history and failure records.
Evidence：O*NET OnLine 19-1011.00 tasks and work activities · BLS entry and education requirements
Typical scenario：The project report presents only significant results, and failed batches and excluded samples do not appear in it.
Roles most affected：Positions involved in publication and project reporting
Possible consequence：Keeping only successful experiments and hiding failure and exclusion records damages scientific integrity and misleads subsequent research.
How to reduce risk：State the primary metrics in the research plan in advance, keep failure and exclusion records, separate exploratory from confirmatory analysis, and have key conclusions independently reviewed.
Evidence：O*NET OnLine 19-1012.00 tasks and work activities · BLS entry and education requirements
| Entry directions | Early tasks | Evidence employers recognise as proof of ability | Preparation evidence you can present without experience | How it differs from adjacent positions |
|---|---|---|---|---|
| Animal science direction | Carries out the sampling, testing and recording of experiments on feed, nutrition, breeding and disease control in the laboratory and at livestock farms. | Sampling and testing completed to standard methods, raw data and versions traceable, and able to restate the method version and calibration status. | Use public data to produce a controlled-experiment plan for a feed nutrition or breeding metric (question, hypothesis, control and metrics). | Unlike technicians in the animal science field, this occupation takes responsibility for research design and interpretation of results; data scientists provide general methods but do not take responsibility for animal experiments. |
| Food science direction | Carries out microbiological, chemical and process-related testing, trial formulation and data recording at processing facilities and in the laboratory. | Testing carried out to standard procedure, a complete sample chain and calibration record, and able to separate the known from the unknown and state the range of applicability. | Use synthetic data to complete a controlled analysis of a food process parameter (including baseline, replicates and uncertainty). | Unlike food quality management positions, this occupation is responsible for research design and interpretation; release and recall are decided by authorised positions under regulations. |
| Soil and plant science direction | Carries out crop, soil and spatial sampling, testing and field-trial recording in the field and in the laboratory. | Field trials designed to standard, complete sampling and testing records, and conclusions that state the applicable variety and region. | Use public weather and soil data to complete an analysis and limitations statement for a small-scale field comparison experiment. | Unlike agricultural engineers, this occupation answers whether and why; equipment and engineering systems are the engineer's responsibility. |
| Experimental data and records | Build a data dictionary and versions, record missing values, outliers and exclusion reasons, and keep the raw data and analysis scripts. | A complete sample chain, calibration records, evaluation metrics and reasons for decisions, so conclusions can be independently reviewed. | An analysis record template containing the data dictionary, cleaning steps and exclusion reasons. | Unlike a research assistant, this role takes responsibility for method choice and conclusions, and does not merely execute set steps. |
| Laboratory and compliance operations | Carries out sample management, safety training and record keeping to laboratory safety and food regulation requirements. | Complete safety and compliance training records, operations within the scope of authorisation, and a trace left for signing and handling. | A self-check list of key points for laboratory biological and chemical safety and sample management. | Unlike a regulatory body, this occupation provides the experimental basis; release, recall and regulatory handling are decided by authorised positions in accordance with the law. |
The typical U.S. BLS basis is a relevant science bachelor's degree, and some positions require a master's or doctorate; the animal ONet43/22/13 are education feedback, not the combined group distribution.
Insufficient evidence: this page obtained no precise recruitment sample for this occupation in China; a bachelor's degree or above in a relevant field can only serve as exploratory preparation, and must not be passed off as a verified general employer requirement.
On-site work is managed by the employer through job duties, experimental facilities and safety training, and any biological or chemical safety work requires the corresponding training; contract and identity conditions follow the Labour Contract Law of the People's Republic of China and the employer's rules.
On the mainland China side, laboratory qualifications, inspection and testing body accreditation and food safety management requirements follow the regulations in force and the competent authority's rules (Food Safety Law of the People's Republic of China, amended in 2021); a laboratory conclusion does not automatically amount to a basis for release, and signing and handling must take place within the scope of authorisation.
Animal scientists 19-1011 feedback: 48% consider accuracy extremely important and 55% are outdoors in all weather at least once a week; this cannot be applied to the soil and plant direction.
The main workplaces are the laboratory, the office, processing facilities, farms and the field, in different proportions by direction; these are group descriptions of the occupation and are not used to judge whether an individual is suited.
Task statement: a fictional greenhouse uses 6 independent trial pots of the same batch and variety of seedlings, randomly assigned to two nutrient regimes A/B with 3 pots per group and all other growing conditions identical; after 14 days the above-ground fresh weight of each pot is measured in grams, with no missing values.
Group and measurement record: group A 10, 12 and 11 grams; group B 13, 14 and 15 grams. The material is for paper analysis only and is not actual cultivation or fertilisation advice.
The task statement provides no price, yield, significance or approval information; it must not be added by the candidate, and no population, causal or commercial-promotion conclusion may be inferred from it.
List the groups and the measurement record for each pot, calculate the means of groups A and B and the absolute and relative difference of B against A, keeping the working and the units.
Write a conclusion with its limitations, and state what further evidence is needed before generalising; the conclusion must give the scope of this observation and the sources of uncertainty.
First check against the A/B grouping that the 6 measured values have no missing entries, then calculate step by step: mean A 11 grams, mean B 14 grams, difference 3 grams, relative difference 3/11 ≈ 27.3% (keeping the working and the units).
The conclusion describes only this observation of these 6 pots, and does not establish population significance, adequately proven causation or effective commercial promotion; the small sample, the single batch and site, and measurement error still limit the conclusion.
One requirement per row (education and field, experimental and safety training, method and statistical skills, and any qualification relating to inspection and testing or food safety), listing the verification source and status.
Map each requirement to the rule of the competent authority, the educational institution or the employer, noting the date checked; record the items not yet met or needing strengthening, and the plan.
Check this exercise deliverable item by item: whether the groups and measurements match the task statement, whether the mean and difference calculations are reproducible, and whether the conclusion states its limitations explicitly.
The verification material uses only fictional data and adds no price, yield, significance or approval information absent from the task statement; all reference material comes from official channels and is dated.
Compile the same fictional case into a one-page deliverable: the group and measurement record, the means and difference for A/B (mean A 11 grams, mean B 14 grams, difference 3 grams, relative difference about 27.3%), a conclusion with its limitations, and the follow-up verification plan before generalising.
Preparation records should demonstrate an understanding of method and record requirements and your own analytical foundation, not internal or business data; do not put online rumours or unofficial information into preparation records.
| Selection and training stage | Verifiable checkpoints | Observable evidence | Boundaries and warnings |
|---|---|---|---|
| Education and professional training | Whether systematic training in agriculture, food, biology or a related field has been completed. | Can provide education and coursework records, and has completed basic laboratory or field work. | An academic qualification by itself does not prove the ability to design reviewable experiments, and needs project evidence to support it. |
| Experimental and statistical skills | Whether experimental design, sampling and testing, and statistical analysis methods are mastered. | Able to explain the control, replicates, metrics and exclusion conditions, and to explain the analysis script. |
Understand the A/B grouping and measurement conditions of the 6 pots in the fictional greenhouse (same batch and variety, random assignment, all other conditions identical), and write out the grouping table.
Transcribe the above-ground fresh weight after 14 days (group A 10, 12 and 11; group B 13, 14 and 15 grams), check that there are no missing values and note the unit.
Calculate the means of groups A and B separately and keep the working: mean A 11 grams, mean B 14 grams.
Calculate the absolute and relative difference of B against A: difference 3 grams, relative difference 3/11 ≈ 27.3%, noting the denominator and the unit.
Describe only this observation of these 6 pots, and do not conclude significance, causation or commercial promotion; the small sample and measurement error limit the conclusion.
| Path or preparation item | Who it applies to | Effect on entering this occupation | Whether it is a general threshold | Conditions of application | Investment and duration | Limitations and risks |
|---|---|---|---|---|---|---|
| A bachelor's degree or above in a relevant field | People intending to enter | Whether systematic training in experimental design, sampling and testing, statistical analysis and the professional direction can be obtained | Common path | To the curriculum rules of the institution and programme | The length of undergraduate or postgraduate education |
Typical scope：Carries out feed nutritive value measurement and the organisation of feeding data, and assists with writing technical reports.
Readiness evidence：Able to complete a feed composition measurement independently and explain the sources of error.
Credential boundary：Editorial preparation suggestion: add relevant science coursework and laboratory training; the typical U.S. basis is a bachelor's degree, and the jurisdiction and the qualification for specific duties must be checked separately, so it cannot be stated broadly that there is no mandatory qualification.
Next moves：Fill in animal nutrition and experimental design, and take independent responsibility for a feeding comparison trial.
Typical scope：Takes part in selection and mating plans and reproduction data management, and writes up interim analyses.
Readiness evidence：Able to explain the assumptions and limitations of genetic parameter estimation.
Credential boundary：Coursework in animal genetics and breeding; some positions require a master's degree.
Next moves：Learn genetic evaluation methods and statistical models, and take on the validation of one improvement plan.
Typical scope：Turns research conclusions into feeding and health management plans and tracks the implementation results.
Readiness evidence：Able to produce a one-page recommendation with implementation preconditions and monitoring metrics.
Credential boundary：Training in animal science or veterinary medicine (specific requirements are governed by the employer).
Next moves：Takes on cross-department implementation plans and designs effect monitoring.
Typical scope：Responsible for research objectives, resource allocation and results compliance, and supervises team members.
Readiness evidence：Able to take responsibility for the boundaries of conclusions and the ownership of results.
Credential boundary：Usually requires a doctorate or long research experience; it is not a uniform grade.
Next moves：Extends into external collaboration and research portfolio management.
Typical scope：Completes ingredient specification testing and formulation data organisation, and outputs analysis records.
Readiness evidence：Able to carry out standard testing methods and record anomalies.
Credential boundary：Training in food science or chemistry; laboratory safety training.
Next moves：Learn formulation design and shelf-life testing, and complete one small formulation validation independently.
Typical scope：Designs processing and preservation trials and determines process parameters and shelf life.
Readiness evidence：Able to explain how trial conditions affect the range of applicability of a conclusion.
Credential boundary：Coursework in food engineering or a related field; some positions require a master's.
Next moves：Works with the production team to complete pilot-scale trials and designs monitoring metrics.
Typical scope：Assesses microbiological, additive and labelling requirements against standards and regulations, and forms a compliance conclusion.
Able to list the regulatory basis and the situations where it does not apply.
Typical scope：Completes field sampling and soil physical and chemical testing to the plan, and organises the raw data.
Readiness evidence：Able to explain the relationship between sampling point layout and spatial variation.
Credential boundary：Training in agricultural resources and environment; field safety protection training.
Next moves：Learn spatial statistics and experimental design, and design one field trial independently.
Typical scope：Designs cultivation and fertilisation comparison trials and produces an executable plan.
Readiness evidence：Able to produce trial records and conclusions with controls and replicates.
Credential boundary：Coursework in crop science or soil science; some positions require a master's.
Next moves：Works with extension and production teams to verify the conditions under which a plan applies.
Typical scope：Assesses soil degradation or contamination and proposes improvement and monitoring plans.
Able to distinguish the conditions under which survey results and remediation recommendations apply.
Sampling, testing and recording completed to standard methods, stopping and reporting when anomalies occur; able to restate the method version and calibration status.
Writes an experimental plan with the target population, control, replicates and metrics, and can compare treatment with control using statistical methods.
Turns conclusions into process, formulation or quality standards, helps production and quality teams implement them, and designs monitoring metrics.
Responsible for research objectives, resource allocation, review arrangements and external reporting, and manages data and results compliance.
Does not establish：The employment projection includes the self-employed, while the OEWS wage does not; a national projection cannot prove that a particular state, specialism or employer is expanding hiring.
United States national (combined occupation group 19-1010) · Combined occupation group
6%
Employment growth rate
BLS projects employment growth of 6%, from about 37,400 people to about 39,800 people, a net increase of about 2,400; the same release and the same occupational scope.
Does not establish：This cannot be interpreted as a guarantee of individual employment, nor can each specialism be assumed to grow at the same rate.
United States national (combined occupation group 19-1010) · Combined occupation group
About 2,700 per year
Average annual job openings
BLS projects about 2,700 job openings per year on average; the openings include replacement demand from transfers and exits and are not equal to net new positions.
Does not establish：Annual openings are not the number of positions an individual can obtain; region, specialism, degree and research facilities all change the opportunities.
These are capability relationships, not automatic promotions or credential waivers.
Transfers：Laboratory, sampling and recording tasks overlap and can serve as preparation material; this does not guarantee hiring or direct mutual recognition of position competence.
Build next：Editorial transition check: technicians assist with research, sampling and testing, analysing results and compiling records; check the target method, equipment, quality control and the actual recruitment requirements.
View career profileTransfers：Understanding of the field and the process helps with engineering collaboration.
Build next：Requires additional knowledge of engineering design and physical systems.
View career profileTransfers：Shares experimental design and statistical ability.
Build next：Requires adding animal nutrition, breeding and biosafety conditions.
View career profileTransfers：Shares experimental design and data analysis ability.
Build next：Requires adding food processing, regulation and labelling responsibility.
View career profileTransfers：Shares statistical and field trial ability.
Build next：Requires adding knowledge of crops, soil and spatial variation.
View career profileTransfers：Statistical and data-handling ability transfers directly.
Build next：Requires adding engineering modelling and business delivery responsibility.
View career profileTransfers：Understanding of standards and testing processes is helpful.
Build next：Requires adding knowledge of regulatory enforcement and enforcement procedures.
View career profileTransfers：Laboratory analysis and recording tasks overlap; whether the target position's requirements are met depends on the specific test object and method.
Build next：Editorial check: the target analytical method, testing specifications, deviation investigation and record requirements; an analyst role is not expanded into a regulatory enforcement position.
View career profileThere is no evidence supporting wholesale replacement. AI will change tasks such as literature organisation, data cleaning and image first-pass screening, but sampling design, method validation, anomaly handling and release sign-off remain human responsibilities; technical capability, adoption rates and the number of positions must be judged separately.
No precise national wage distribution for this occupation was obtained. The National Bureau of Statistics publishes annual average wages by industry sector and unit type, which can serve only as a broad reference and cannot replace an individual's starting pay, median or percentiles.
No. The median annual wage in 2025, month 5 (May), is $81,970, with the lowest 10% below $48,730 and the highest 10% above $142,300; these are positions in a statistical distribution, not an individual quote or a guaranteed starting pay for a new graduate.
The typical U.S. entry point is a relevant science bachelor's degree, and some independent research positions require a master's or doctorate; the portfolio should demonstrate sampling, raw data, analysis code, verification, deviation and executable recommendations. A degree does not amount to reviewable experimental ability.
Mainly sampling bias and over-extrapolation, loss of control over the sample chain and testing, model extrapolation and treating correlation as causation, and leakage of data and commercial information. Controls, replicates, calibration, blind samples, version records and human review are the basic controls.
BLS projects employment growth of 6% for this occupation group over 2025—2035, a net increase of about 2,400 people (from about 37,400 to 39,800) and about 2,700 openings per year; this is a national combined occupation group projection and does not distinguish specialisms.
Switches between the laboratory, the office, processing facilities and the field. O*NET 19-1011.00 shows that 78% of respondents work more than 40 hours per week, 48% consider precision and accuracy extremely important, and 55% have face-to-face discussions every day.
First look at the research object (animal / food ingredients and processing / crops and soil), then at the form of output and who is responsible, and finally check the coursework, laboratory, internship and the target employer's actual requirements.
FermatMind careers content editorial team
2026-10-08 (nine-section revision limited to this round)
O*NET 19-1011.00/19-1012.00/19-1013.00 occupation profiles; BLS Occupational Outlook Handbook 2025 edition and OEWS 2025, month 5 (May); National Bureau of Statistics 2025 urban unit wage statistics; Food Safety Law of the People's Republic of China (amended in 2025).
Limitations：Used for occupational identity, tasks and work context; a U.S. single-occupation basis.
Limitations：Used for the U.S. median wage, employment size and the 2025—2035 projection; national combined occupation group 19-1010.
Limitations：Used for the occupational identity and task boundaries of the food direction.
Limitations：Used for animal scientists wage percentiles; a single-occupation basis.
Limitations：Used for mainland China annual average wages by industry sector and unit type; it contains no separate wage for this occupation.
Limitations：Used for the occupational identity and task boundaries of the soil and plant direction.
Limitations：Used for work context and working hours respondent shares.
Limitations：Used for the software and technology types common in job postings; it does not indicate adoption rates or technical effectiveness.
Limitations：Used for optional certification information; it does not constitute a mandatory qualification for this occupation.
Limitations：Used for optional training information.
Limitations：Used to explain that the total wage bill is on a before-tax basis and what it comprises.
Limitations：Used for general information on labour contracts and employment conditions in mainland China.
Limitations：Used for the responsibilities of mainland China food producers and traders and for food safety management requirements.
Limitations：Used for occupational tasks, work activities and record requirements.
Limitations：Used for food scientists wage percentiles.
Limitations：Used for soil and plant scientists wage percentiles.
Limitations：Used for detailed task material in the food direction.
Limitations：Used as the industry distribution reference for this occupation.
Limitations：Used for detailed task material in the soil and plant direction.
Limitations：Used for the occupation overview and related-occupation entries.
Limitations：Used as the industry distribution reference for the food direction.
Limitations：Used for respondent shares on education requirements for new hires.
Limitations：Used for entry education requirements and the description of the working environment.
Limitations：Used for wages, employment and projection data.
Limitations：Used for reference on tasks and interest codes.
Limitations：Statistical product, jurisdiction and actual duties are checked separately, and this does not prove an AI effect or an individual's authorisation.
Limitations：Statistical product, jurisdiction and actual duties are checked separately, and this does not prove an AI effect or an individual's authorisation.
Limitations：Used for occupational identity, tasks and work context; a U.S. single-occupation basis.
Limitations：Used for the U.S. median wage, employment size and the 2025—2035 projection; national combined occupation group 19-1010.
Limitations：Used for the occupational identity and task boundaries of the food direction.
Limitations：Used for animal scientists wage percentiles; a single-occupation basis.
Limitations：Used for mainland China annual average wages by industry sector and unit type; it contains no separate wage for this occupation.
Limitations：Used for the occupational identity and task boundaries of the soil and plant direction.
Limitations：Used for work context and working hours respondent shares.
Limitations：Used for the software and technology types common in job postings; it does not indicate adoption rates or technical effectiveness.
Limitations：Used for optional certification information; it does not constitute a mandatory qualification for this occupation.
Limitations：Used for optional training information.
Limitations：Used to explain that the total wage bill is on a before-tax basis and what it comprises.
Limitations：Used for general information on labour contracts and employment conditions in mainland China.
Limitations：Applies only to the subjects, tasks and jurisdictions specified in the Food Safety Law, and is not a uniform confidentiality rule for all research information.
Limitations：Used for occupational tasks, work activities and record requirements.
Limitations：Used for food scientists wage percentiles.
Limitations：Used for soil and plant scientists wage percentiles.
Limitations：Used for detailed task material in the food direction.
Limitations：Used as the industry distribution reference for this occupation.
Limitations：Used for detailed task material in the soil and plant direction.
Limitations：Used for the occupation overview and related-occupation entries.
Limitations：Used as the industry distribution reference for the food direction.
Limitations：Used for respondent shares on education requirements for new hires.
Limitations：Used for entry education requirements and the description of the working environment.
Limitations：Used for wages, employment and projection data.
Limitations：Used for reference on tasks and interest codes.
Limitations：Statistical product, jurisdiction and actual duties are checked separately, and this does not prove an AI effect or an individual's authorisation.
Limitations：Statistical product, jurisdiction and actual duties are checked separately, and this does not prove an AI effect or an individual's authorisation.
The U.S. OEWS wage uses an employee basis, while the employment projection includes the self-employed; the combined group and the three sub-occupations are used separately. China's NBS industry and position categories are a broad reference, not converted into individual take-home income or extrapolated across jurisdictions.
Big Five personality (English) · MBTI (Chinese) · Enneagram (Chinese) · Intelligence and reasoning (Chinese)
The labour contract and the employer's rules prevail (Labour Contract Law of the People's Republic of China).
When food safety conclusions are involved, check the Food Safety Law and the organisation's procedures.
Where student and employee results are involved, confirm against institutional rules and agreements.
The O*NET 19-1011.00 technology list and certification directory can be checked, and list the equipment, technologies and certification entries commonly used in this occupation; the actual scope of use depends on the direction (animal, food, soil and plant), the experimental facilities and the employer type.
Preparation should build a solid foundation in experimental design, sampling and testing, statistical analysis and safety compliance from public material, and should find out about qualification and training requirements through official channels.
The portfolio only demonstrates the level of analytical preparation; it does not constitute actual cultivation or fertilisation advice, a skills test or a hiring test, and does not represent any qualification or authorisation; the open part may propose different reasonable follow-up verification plans (such as replicates across more batches and sites, or a pre-registered analysis plan), but must not add price, yield, significance or approval information that the task statement does not contain.
| Presenting only final charts without being able to explain the code and the deviations cannot be relied on. |
| Safety and compliance training | Whether training in laboratory biological and chemical safety, sample management and the relevant regulations has been completed. | Complete training records, with operations within the scope of authorisation and a trace left. | Signing or handling conclusions beyond the scope of authorisation creates compliance risk. |
|---|
| Qualification and authorisation (involving testing or food release) | Whether the corresponding qualification or authorisation has been obtained under the regulations in force in mainland China and the competent authority's rules. | Complete qualification and authorisation records, with conclusions issued within the granted scope. | Issuing testing conclusions externally without qualification or standing in for release responsibility is handled in accordance with the law. |
|---|
List the evidence needed before generalising (replicates across more batches and sites, a pre-registered analysis plan); do not add information beyond the task statement.
Debrief the week: are you willing to work to the record–calculate–limited-conclusion process and record uncertainty truthfully?
| An academic qualification by itself does not prove the ability to design reviewable experiments |
| Specialised experimental and statistical skills | Research assistant and laboratory positions | Whether experimental design, sampling and testing, and statistical analysis can be completed independently | Common requirement in positions | To the rules of the research group and the employer | The duration of training and on-the-job accumulation | Conclusions must not be issued independently without mastery of method selection and validation |
|---|
| Laboratory biological and chemical safety training | Laboratory positions | Whether experiments and sample management can be carried out safely and compliantly | Common requirement in positions | To the employer's safety management rules | The training duration | The relevant experimental areas must not be entered without training |
|---|
| Inspection and testing body qualification accreditation (involving testing) | Applies to bodies that issue inspection and testing data with evidentiary effect to the public; personnel competence and signing are checked separately. | Body accreditation relates to statutory evidentiary effect and business scope; it is not required for all research output and is not a personal licence. | When data or results with evidentiary effect are issued to the public in accordance with the law, they are checked against the applicable regulations and exceptions. | To the regulations in force in mainland China and the competent authority's rules | The accreditation cycle | Issuing testing conclusions externally without accreditation is handled in accordance with the law |
|---|
| Food safety management and release authorisation (involving food) | Food enterprises and quality positions | Determines whether responsibility for food safety conclusions and release can be taken | Essential threshold (involving release) | Under the Food Safety Law of the People's Republic of China and organisational procedures | The organisation's authorisation and training | A laboratory conclusion does not automatically amount to a basis for release |
|---|
| A doctorate or research-direction degree (for comparison) | Independent research positions | Affects whether an independent research direction can be led and projects applied for | Common requirement in research positions | To the rules of the educational institution and research body | The length of doctoral education | Research positions often end when project funding ends; the contract term and renewal conditions must be asked about clearly |
|---|
Credential boundary：Requires familiarity with the Food Safety Law and related standards; authorisation follows organisational rules.
Next moves：Takes on communication with regulators and the preparation of risk assessment documents.
Typical scope：Responsible for product development objectives and resources, and manages data and results compliance.
Readiness evidence：Able to take responsibility for the boundaries of responsibility for release-related conclusions.
Credential boundary：Usually requires a higher degree and project experience; it is not a uniform grade.
Next moves：Extends into product portfolio and technology roadmap planning.
Credential boundary：Training in environment and soil; where remediation is involved, local regulations apply.
Next moves：Takes on long-term monitoring and effect review of remediation plans.
Typical scope：Responsible for the research direction and resources, guides the team and manages results and compliance.
Readiness evidence：Able to take responsibility for the continuity of data from long-term fixed-site trials.
Credential boundary：Usually requires a doctorate or long research experience; it is not a uniform grade.
Next moves：Extends into regional collaboration and long-term monitoring networks.
Limitations：Editorial ratings use the existing occupational responsibilities and AI task analysis on this page. They are not adoption rates, measured effects, job-disappearance or personal-unemployment probabilities, or official scores from Stanford or another external institution. No measured time weights or future robot deployment are assumed.
