Main effect¶
The marginal effect of one factor on a response averaged over the levels or distribution of the other factors in a factorial model.
Core Idea¶
Main effects depend on coding, balance, weighting and model scale and can be misleading in the presence of strong interactions; an omnibus test does not localize pairwise differences. Cell means or fitted predictions are averaged across other-factor levels under declared weights, and contrasts among the resulting marginal means quantify the focal factor. 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.
Scope of Application¶
Main effect belongs to experimental design and is useful where the analyst can specify the typed experimental design carrier, including objects, relations, parameters, conventions, evidence, boundaries, and comparison targets, then evaluate the response and experimental units, focal factor and levels, other factors, design balance and weights, model and link scale, marginal means or contrast, interaction terms, null hypothesis and uncertainty are explicit. The scope is broad within that domain but bounded by the need for the response and experimental units, focal factor and levels, other factors, design balance and weights, model and link scale, marginal means or contrast, interaction terms, null hypothesis and uncertainty are explicit.
Clarity¶
The abstraction clarifies a crowded vocabulary by making the response and experimental units, focal factor and levels, other factors, design balance and weights, model and link scale, marginal means or contrast, interaction terms, null hypothesis and uncertainty are explicit the center of the account. A claim should name the carrier, the governing operation or relation, the applicable assumptions, and the recognition test.
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 Main effect. Main effect 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 experimental design carrier, including objects, relations, parameters, conventions, evidence, boundaries, and comparison targets. Reject examples whose alleged carrier belongs to a different problem. 2. Lock the constitutive rule. Express the response and experimental units, focal factor and levels, other factors, design balance and weights, model and link scale, marginal means or contrast, interaction terms, null hypothesis and uncertainty are explicit independently of one notation or implementation.
Knowledge Transfer¶
Knowledge transfers strongly among subfields of experimental design because they reuse the typed experimental design carrier, including objects, relations, parameters, conventions, evidence, boundaries, and comparison targets, Cell means or fitted predictions are averaged across other-factor levels under declared weights, and contrasts among the resulting marginal means quantify the focal factor., and type the carrier, state every parameter and convention in the definition, test that the response and experimental units, focal factor and levels, other factors, design balance and weights, model and link scale, marginal means or contrast, interaction terms, null hypothesis and uncertainty are explicit, compare the nearest accepted identity, and report counterexamples, uncertainty, and limiting cases.
Relationships to Other Abstractions¶
Current abstraction Main effect Domain-specific
Parents (1) — more general patterns this builds on
-
Main effect is a kind of Aggregation Prime
The proposed strict upward parent is
prime:aggregation.
Hierarchy path (1) — routes to 1 parentless root
- Main effect → Aggregation → Micro Macro Linkage
Neighborhood in Abstraction Space¶
Main effect sits in a crowded region of the domain-specific corpus (27th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.
Family — Engineering Design & Requirements (47 abstractions)
Nearest neighbors
- Fractional factorial design — 0.93
- Statistical unit — 0.92
- Demand characteristics — 0.91
- Cultural probe — 0.90
- Engineering design process — 0.90
Computed from structural-signature embeddings · 2026-09-08