Item-total correlation¶

The correlation between one scored assessment item and a total or rest score, used to evaluate whether the item aligns with the construct measured by the scale.

Version: v2 · 2026-09-19 · History
Domain-specific #: 5129
Domain group: Formal Sciences
Origin domain: Experimental Design & Statistics
Subdomain: Psychometrics → Experimental Design & Statistics

Core Idea¶

Item–total correlation quantifies association between item responses and an aggregate test score, with corrected item–total correlation excluding the focal item to avoid part–whole inflation. Responses are scored across persons, a total is constructed, and covariance is normalized; item difficulty, dimensionality, reliability, range restriction, and overlapping content affect interpretation. The abstraction is therefore identified by a declared carrier, a transformation or constraint over that carrier, and an invariant that tells an analyst whether the named structure is genuinely present.

How would you explain it like I'm…

Does This Question Fit?

Look at one question on a quiz. Do the kids who got it right also usually get a high score on the whole quiz? If yes, that question goes along well with the rest of the quiz. Item-total correlation is a number that says how strongly they go together.

One Question vs Whole Test

Item-total correlation is a number that shows how well one test question lines up with the total score. If people who do well on that question also tend to score high overall, the number is high. There's a catch: the question is itself part of the total, which can make the match look better than it really is. So a 'corrected' version compares the question with the total of all the other questions instead. Things like how hard the question is can also change the number.

Corrected Item-Total Correlation

Item-total correlation measures how strongly responses to a single test item go along with an aggregate test score across the people who took the test. It is computed by scoring each person's answers, building a total score, and normalizing the covariance between item and total. Because the item is included in the total, the plain version is inflated: part of the item is being correlated with itself. The corrected item-total correlation removes the focal item from the total to avoid this. Interpretation also depends on item difficulty, whether the test measures one thing or several, reliability, a narrow range of scores, and overlapping content. A correlation shows association, not that the item causes or defines the total.

Item-total correlation quantifies association between responses to a single item and an aggregate test score across respondents: responses are scored, a total is formed, and their covariance is normalized into a correlation coefficient. The uncorrected version includes the focal item in the total, producing part-whole inflation; the corrected item-total correlation excludes it. The statistic is used to flag items that fail to align with the rest of the scale. Its interpretation depends on item difficulty, dimensionality, reliability, range restriction, and content overlap among items. A reported value is meaningful only once scoring, sample, total-score construction, inclusion or exclusion of the focal item, the correlation estimator, missing-data handling, and dimensional assumptions are declared.

Scope of Application¶

Item-total correlation belongs to psychometrics and is useful where the analyst can specify the typed psychometrics carrier, defining objects and relations, parameters, conventions, evidence, boundary cases, and comparison targets, then evaluate scoring, sample, total-score construction, inclusion or exclusion of the focal item, correlation estimator, missingness, and dimensional assumptions are declared. The scope is broad within that domain but bounded by the need for scoring, sample, total-score construction, inclusion or exclusion of the focal item, correlation estimator, missingness, and dimensional assumptions are declared. The entry records a descriptive analytical identity; practical use requires the governing domain's evidence, standards, and safety obligations.

Clarity¶

The abstraction clarifies a crowded vocabulary by making scoring, sample, total-score construction, inclusion or exclusion of the focal item, correlation estimator, missingness, and dimensional assumptions are declared the center of the account. A claim should name the carrier, the governing operation or relation, the applicable assumptions, and the recognition test. A bare label is insufficient because the name Item-total correlation can be used for a formal identity, an implementation, or a neighboring result unless carrier and convention are stated.

Manages Complexity¶

Without the abstraction, an analyst must reason directly over many local details: the carrier roles, admissibility assumptions, competing conventions, derived invariants, boundary cases, and proof or validation obligations specific to Item-total correlation. Item-total correlation compresses them into the roles in the structural signature. That compression permits comparison across instances without erasing the variables that determine validity. It also exposes which details may be varied safely and which are constitutive.

Abstract Reasoning¶

Identify the carrier. State what the elements, states, objects, or observations are: the typed psychometrics carrier, defining objects and relations, parameters, conventions, evidence, boundary cases, and comparison targets. Reject examples whose alleged carrier belongs to a different problem.
Lock the constitutive rule. Express scoring, sample, total-score construction, inclusion or exclusion of the focal item, correlation estimator, missingness, and dimensional assumptions are declared independently of one notation or implementation.

Knowledge Transfer¶

Knowledge transfers strongly among subfields of psychometrics because they reuse the typed psychometrics carrier, defining objects and relations, parameters, conventions, evidence, boundary cases, and comparison targets, Responses are scored across persons, a total is constructed, and covariance is normalized; item difficulty, dimensionality, reliability, range restriction, and overlapping content affect interpretation., and type the carrier, state every parameter and convention in the definition, test that scoring, sample, total-score construction, inclusion or exclusion of the focal item, correlation estimator, missingness, and dimensional assumptions are declared, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases.

Relationships to Other Abstractions¶

Current abstraction Item-total correlation Domain-specific

Parents (1) — outgoing typed relations

Item-total correlation is a kind of Correlation Prime

What makes it its own entry: the domain-specific identity determined by scoring, sample, total-score construction, inclusion or exclusion of the focal item, correlation estimator, missingness, and dimensional assumptions are declared.

Hierarchy path (1) — routes to 1 parentless root

Item-total correlation → is a kind of Correlation

Neighborhood in Abstraction Space¶

Item-total correlation sits in a crowded region of the domain-specific corpus (6^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Psychometric Measurement & Judgment Biases (29 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-25