About the Data

The NLx Research Hub is your one-stop destination for comprehensive labor demand data

Watch the video to learn how the NLx data is being utilized.

The NLx Difference

It all begins with the data. The strength of the NLx Research Hub comes from the unique process of the NLx job feed.

Data Formatted for Every Application

The NLx Research Hub API is created to integrate easily into visualizations and data tools. Refreshed every 24 hours, the NLx Research Hub API is created to help you easily integrate the NLx data into visualizations and data tools. Visit our Data Resources page for more specifications.
Looking to create your own custom dataset for your research project? Our API allows you to tailor your queries of the entire NLx database based on geography, occupation, data range, employers, and more. Visit our Data Resources page for more specifications.
In addition to populating all API endpoints, the entire Job Table is available in monthly files via the Research Hub’s secure FTP Server; and represents a catalog of the aggregated job listings. When a job listing is updated in the NLx job feed, the details of that job listing are updated in the Job table. In other words, the records in the Job table show only the most recent version of each job listing. This format is perfect for identifying emerging occupations, identifying in-demand professional certifications, and gap analyses between in-demand skills and education curriculums.
Available only in monthly files via the Research Hub’s secure FTP server, the Jobhistory Table is an “append only” table containing a row for every entry that has ever appeared in the NLx job feed. As each new day’s job feed is received into the NLx Research Hub data warehouse, it is appended to this table. This provides a longitudinal dataset of all NLx job postings, and due to the NLx job feed “kill and fill” process, is exclusively available via the NLx Research Hub. All fields from the Job table are contained in the JobHistory table, with the exception of the description field due to storage and functionality limitations.

Strengths of the NLx Data

Comprehensive

Job postings from all 50 states, D.C., Guam, Puerto Rico, and the U.S. Virgin Islands are included in the data. Because of its sourcing methods, the NLx captures jobs from small employers that are manually posted through state job banks and are otherwise not available online.

High-Quality

NLx jobs come from only original sources: (1) corporate career websites and applicant tracking systems, (2) state job banks, and (3) USAjobs.gov. Data is evaluated by experts to ensure there is no duplicate or fraudulent job postings and that no fraudulent employers are included.

Accessible

From cost, to access methods, to data formats, the Research Hub strives to eliminate as many barriers as possible between the labor market information (LMI) community and this rich data source. This is done to ensure that the NLx data remains a public good for the workforce and education ecosystem.

Transparent

Information on how data is gathered, processed, and distributed is shared publicly to ensure ethical research and data collection practices. The Research Hub also works with users to ensure that any tools, methods, and products developed with the NLx data are done so following open-source principals whenever possible.

Responsive

The NLx job feed - the source data for the NLx Research Hub - is refreshed every 24 hours. This ensures that the data is up-to-date and reflective of labor market trends as they occur. Additionally, the Research Hub is designed to quickly address any known issues in the data, and work to develop products, services, and technical enhancements to meet the evolving needs of its users and stakeholders.

Trusted

NASWA and DirectEmployers have long-standing relationships with states and employers around the country. Each business is evaluated before being added to the NLx Research Hub. It is these foundational relationships – and the goal to maintain trust – that strengthen the NLx Research Hub

Known Limitations of the NLx Research Hub

Not every job opening is posted online
While many employers and jobseekers now use online tools throughout job search and hiring processes, a significant portion of hiring still happens through traditional and informal methods.
Job postings are advertisements
The goals, preferences, and biases of HR departments and hiring managers may predominate over actual indications of skill demand in the marketplace. Depending on the use case, significant effort may be required to clean up grammar, format job descriptions, and/or insert neutral placeholders for phrases that reflect biased terminology. Aside from mitigating bias, sophisticated parsers are needed to extract usable data from job descriptions, which can sometimes be long and poorly worded.
One job posting might not equal one job vacancy
For certain industries and occupations, one job posting may correlate with precisely one job vacancy. In other cases, a single job posting may represent multiple vacancies (e.g., an HR manager might post a single job ad for a fast-food cashier position and hire 50 cashiers from the candidate pool across several locations).
Remote work and location-flexible positions distort aggregate job numbers
It can be difficult to tell whether a job posting represents a virtual/remote opportunity or if it represents multiple vacancies in several different locations. For compliance reasons, companies sometimes list jobs in every state where they will consider applicants; if a company lists a single job ad in 10 different metro areas, there will be 10 data records for that job in the NLx.
Data field completeness varies
In the NLx and other sources of job posting data, some job ads are more structured than others. Fields like required experience, education, etc. are often left blank, though a good text parser can glean this information from the job description when included.
The NLx and similar data sources have matured over time
The total number of NLx job postings has grown year over year, which reflects broader HR trends and the privatization of job posting technology rather than significant shifts in the labor market. Moreover, it is difficult for any source of job advertisements to understand the extent to which they fully represent every sector of the economy.
Some jobs are evergreen
“Evergreen” jobs are positions that are always open, either due to high turnover, high demand, or specific skills required. These jobs are extremely common in datasets of online job ads, and they make it difficult to evaluate the lifecycle of a job. Users may want to consider setting a manual expiration date for evergreen jobs or utilizing statistics such as the median or other percentiles of posting time, to reduce variance in measurements for how long a job is active.
NLx data only include job postings
Resumes, hiring data, and other indicators that may be considered along with job postings for certain analyses are not included in the NLx dataset.
NLx data is provided as advertised
By providing job postings directly from original sources, the data may be prone to certain inconsistencies. For instance, there may be a slight time lag in employers removing job postings after they have been filled.