education-data-source-ipeds

What it does

Explains IPEDS as a central data system for ~6,500 Title IV-participating U.S. postsecondary institutions (1980-present) across 12+ survey components, with emphasis on where to find data (Education Data Portal mirrors) and primary identifier UNITID. It instructs analysts to use IPEDS data for enrollment, completions, graduation rates, finances, aid, admissions, and HR analyses, noting graduation rates track first-time full-time students and that cross-sector finance comparisons require care due to GASB vs FASB accounting.

How it works

Outlines a reference of file structure and components (survey-components.md, graduation-rates.md, enrollment-data.md, finance-data.md, financial-aid.md, institution-identifiers.md, completions-data.md, data-quality.md). Provides decision trees to guide data selection (e.g., graduation rates, enrollment counts, finances, aid, degrees, institutional info, HR, linking data). Describes variable encoding and the need to verify Portal codes against codebooks, with examples showing how to map NCES raw names to Portal names, and how to filter data (e.g., sex=99 for totals). Includes guidance on data access via mirrors, with paths and codebooks discussed, and notes known Portal gaps (e.g., missing distance education vars, open-admissions flag, finance data lag). Includes Python snippets for filtering and calculation of admit_rate, and a clarification on data availability lags and a critical cutoff for Finance data (2017) in Portal mirrors.

When to use it

Use when analyzing postsecondary enrollment, degree completions by CIP code, institutional finances, or admissions data, and when needing guidance on data encoding, codebooks, and known limitations (e.g., graduation cohorts, sector comparisons, and lag times).

What it can touch

The skill references data access code patterns and mirrors, including Python snippets that interact with Parquet data (e.g., filtering admissions totals to sex=99 and computing admit_rate). It mentions datasets and codebooks located in IPEDS mirrors and codebook paths such as ipeds/codebook_colleges_ipeds_directory and similar endpoints.

Caveats

Notes significant data lag for IPEDS datasets, with Finance data limited to 2017 as of Jan 2026 in Portal mirrors, affecting endowment values and financial ratios. Advises limiting analysis to available years, using NCCS 990 data as an alternative for private institutions, or forward-filling with caveats. Emphasizes relying on codebooks over documentation when discrepancies arise and that some variables are not mirrored in Portal datasets.