Introduction
"Is this program viable?" is the question every program review committee eventually asks, and it is secretly two questions. Can we keep supplying it, and can the labor market keep absorbing its graduates? Higher ed counts the supply side with one federal code system and the demand side with a completely different one, and neither NCES nor BLS reconciles them for you. A memo that answers only one half is the standard failure: completion trends with no market, or labor market data with no connection to the actual program. Here is the full pipeline, in the order that works.
The two codes, side by side
| CIP | SOC | |
|---|---|---|
| Counts | What the institution produces: completions by program | What the labor market absorbs: employment, wages, openings |
| Maintained by | NCES, U.S. Department of Education | BLS, U.S. Department of Labor |
| Current edition | 2020 CIP | 2018 SOC |
| Data on top of it | IPEDS Completions; Scorecard and PPD field-of-study files | OEWS; Employment Projections; O*NET |
| Where you are in it | You report it: your keyholder assigns the code | You only read it: employers are surveyed, not colleges |
The vocabularies do not overlap by design, which is why the crosswalk between them is its own federal product. Both sides are covered from the ground up in what is a CIP code and what is a SOC code; this post is about the connection.
The five-step workflow
Fix the CIP code and the credential level
Everything downstream inherits this decision. Confirm the program's 6-digit code is right (how to find the CIP code for any program), then decide which credential levels count as supply. A program that awards both associate and bachelor's credentials is two different supplies, and step 4 decides which one the memo means.
Count supply at three scales
Your completions at the CIP and credential level you fixed; your peers' completions from IPEDS, same code, same level; and the national total for context. The peer number is the one committees actually react to, because it shows whether the program is unusually large or unusually small for its sector.
Bridge with the crosswalk, and pick an anchor
Pull the CIP-to-SOC mappings for your code. Watch the shape: one CIP often maps to many SOC codes, and residual "Other" CIP codes map to nothing at all. If one CIP code dominates the feeder pool for the occupation, that is your anchor, and the ratio you compute is meaningful. If the feeders are spread thin with no anchor, no honest single ratio exists; say that in the memo instead of averaging it away. The crosswalk explained covers the rules.
Pull demand by SOC, labeled by what it actually is
OEWS employment and wages for your market area are an employer survey: they count jobs where they are located, and wages across all experience levels, not entry-level graduate pay. National projected openings combine growth with replacement needs, so a flat occupation can still show thousands of openings a year. There is no official state-level openings number; anyone who produces one built it themselves, and their allocation method belongs in your footnote. What is a SOC code covers the reading rules.
Write the memo with sources on every number
Every figure carries its file and vintage: IPEDS Completions 2024-25, OEWS May 2025, projections 2024-34. The crosswalk is labeled as a theoretical bridge maintained by NCES and BLS, not an outcome measurement. Where actual graduate outcomes exist, PSEO for earnings by field of study, cite them as the empirical check on the crosswalk's theory.
The credential-level trap
The crosswalk maps programs to occupations; it does not encode the credential level the occupation requires. Ignore that and the supply number is quietly wrong. The canonical case is registered nursing: completions under CIP 51.3801 include associate degrees, bachelor's degrees, and RN-to-BSN completions, and the RN-to-BSN population is already licensed nurses being counted a second time. In our graduates per opening analysis, the national ratio looked stable at 1.39 graduates per projected opening, but rebuilding the numerator at the associate-only level, the credential that actually matches the licence pathway for most new entrants, materially changed the state rankings underneath. The rule: rebuild the numerator at the award level that matches the occupational licence before quoting any ratio, and name the largest producer when the ratio is at the state level, because one institution can hold a third of a state's pipeline.
A memo skeleton that survives scrutiny
- The program, its CIP code, the credential levels in scope, and the date the code was last verified.
- Supply: your completions, peers, and national, at the fixed credential level, from IPEDS with the collection year named.
- The bridge: the anchor CIP-to-SOC mapping and its shape, or the explicit statement that no anchor exists.
- Demand: OEWS employment and wages for the market area, and national openings with the growth-versus-replacements split stated.
- The empirical check: PSEO earnings and sector destinations for the field, where the file publishes them, with the coverage caveat that most program cells are suppressed (see why your program has no earnings number).
- What the memo does not claim: no causal link between the program and the occupation, no state openings number presented as official, no entry-level wage passed off as graduate pay.
Where to go next
The pipeline in tool form: browse CIP-4 programs with their completions and earnings data, the CIP-to-SOC crosswalk from the program side, the SOC-side crosswalk from the occupation side, and the occupations directory ranking fields of study by graduates per year. For the supply-demand method with the full worked example, read graduates per opening, and for the terms the earnings-accountability regime attaches to these same fields, start with STATS vs FVT/GE vs PPD.
Run a program review end to end
Supply from IPEDS, demand from BLS, the crosswalk between, sources and vintages on every number, in one conversation.
Book a demo