Recent discussions among LMI Directors have raised questions regarding the continued requirement that state LMI Offices populate the Workforce Information Database (WID). These questions include:
- Why maintain a full WID relational database when alternative structures are often preferred for more efficient delivery of data on web sites?
- What is the continued purpose of the WID in light of new data portal technologies?
- Are there technologies that states can use to share data instead of having to store each other’s data in the WID?
- Should we broaden the discussion around future LMI data needs and their implications for the WID as we strive to keep it relevant to changing technologies?
The Analyst Resource Center (ARC), through its grant from the Employment & Training Administration, is responsible for maintaining the WID and supporting WID related activities. At our regular meetings, the ARC and its WID Structure Subcommittee discuss WID needs, including necessary structure improvements, how to integrate the data structures with data portals, data analytics and presentation graphics. The ARC also has conducted seminars on these topics in the LMI and WID DBA forums.
What are the advantages to having a standardized database structure like the WID?
Defining Critical Components of LMI
The WID structure identifies and describes the important elements of LMI data. These elements, described in the structure as tables with data fields, define what LMI is and how it is used.
Enables Sharable Application Development
WID plays an important role in many Web and other applications. It enables applications to be developed around a standard. Not only does it provide a repository of data for the application, but also a common LMI data model applications can use.
Allows Third-party Development
This common model and repository allows the construction of third party applications for multi-state projects (e.g., Geographic Solutions, LMInformer) and the sharing of data between states that would be difficult without using a common structure. This common structure, provided by the WID data dictionary, ties all state LMI data efforts together, while providing some flexibility for state specific data.
Facilitates Technical Support to States
Without a common platform, states would be on their own in solving technical database issues as they arise. Having a common platform provides expertise across the states that others can turn to when staff turnover or new requirements precludes an in-house solution.
Data issues vs. structure issues
All data has issues: breaks in code, breaks in collection, other data cleaning. No structure can fix this; one does need to know the data to use it. We do think some sort of documentation of code changes, geography changes, etc., could be useful. To that end, we intend to come up with a metadata table or tables for the WID that will accomplish this.
Data structure vs display structures
Any structure used for storing data in a logical and efficient manner, such as the WID, is not always the most user-friendly way to display it. One needs to do some work on any data extract to make it usable for a specific purpose, a specific tool. This will be true regardless of the structure
We are in the process of creating a “sandbox” comprised of sample applications, code snippets and other tools that WID administrators and users may find useful (e.g., data loaders, web service and display tools.)
For the past 20+ years, the Workforce Information Database (WID), and its predecessor America’s Labor Market Information Database (ALMIS), has played a catalytic role in improving and modernizing the way states use and distribute Labor Market Information. Before the advent of these common national structures, each state had its own structure and limited distribution mechanisms. For the reasons identified here, we believe the requirement that all states adhere to a common structure is of significant benefit to the LMI Community at large.