Abridged Big-Five Circumplex (AB5C)

This is the full IPIP AB5C collection: 45 scales and 485 public-domain items measuring the facets of the Abridged Big Five Circumplex (Hofstee, de Raad, & Goldberg, 1992), in which each facet is a blend of its two strongest Big Five components. The circumplex approach captures trait shadings — for example, sociability colored by assertiveness versus sociability colored by warmth — that a five-score summary flattens out.

By the early 1990s two geometries competed for describing trait space. Simple structure assigns each trait to exactly one factor — tidy, but false to the many terms that plainly mix factors. Circumplex models arrange traits continuously around factor pairs — faithful to blends, but hard to extend past two dimensions. Hofstee, de Raad, and Goldberg (1992) proposed the Abridged Big Five Circumplex as the integration: keep the Big Five as reference axes (Goldberg, 1990), and characterize every trait by its two strongest factor loadings, primary and secondary, each with sign.

That bookkeeping yields 45 facet cells — five 'pure' factor facets plus forty two-factor blends — and the IPIP assembled a public-domain scale for each cell (Goldberg, 1999). The payoff is interpretive precision at the level where everyday trait words actually live.

These 45 scales are the AB5C model's native public-domain operationalization — original IPIP items keyed to the circumplex cells, not a version of some proprietary inventory. Scale membership and reverse-keying are taken verbatim from Goldberg's IPIP multi-scale map (ScalesFor967items), with construct names as catalogued at ipip.ori.org.

All 45 facets are reported against Eugene–Springfield Community Sample reference values in the norms section. Its length makes it a research and deep-self-study instrument rather than a quick profile.

“Accept people as they are.”

Illustrative only. During administration items are presented one screen-set at a time; response-key direction is never shown to respondents.

Item responses are summed within each scale (reverse-keyed items recoded first) and expressed as a population percentile against the cited reference sample, with a reliability-based confidence range where α is published.

Norms computed from the Eugene-Springfield Community Sample (ESCS) — Lewis R. Goldberg’s adult community panel from Eugene and Springfield, Oregon (Harvard Dataverse, doi:10.7910/DVN/UF52WY). This is a community sample and is NOT nationally representative. The mean, standard deviation, reference N and Cronbach’s α were computed directly from the raw IPIP item-level responses on THIS scale’s exact item set and reverse-keying (complete cases); reference N varies by scale and is shown with each scale below. The confidence range uses the standard error of measurement (SEM = SD·√(1−α)) from that computed α.

Public-domain items from the International Personality Item Pool (Goldberg, L. R., 1999; ipip.ori.org). Scale membership and reverse-keying taken verbatim from Goldberg's IPIP multi-scale map (ScalesFor967items). Construct names from ipip.ori.org.