MENTAL MODEL · M6710

MEREC

DecideHigh supportDecision Science

Included

Updated 2026-08-11

Loading revision record…

INTRODUCTION

English translation pending.

CORE DEFINITION

Method based on the Removal Effects of Criteria (MEREC) is an objective weighting technique. For each criterion in turn, the method recomputes the overall performance score of every alternative with that criterion removed, and measures the distance between the reduced evaluation and the original one. A large change indicates that the criterion strongly differentiates among the alternatives, so it receives a high weight; a small change means the criterion is nearly redundant. The weights are normalized from these distances. The method derives weights entirely from the data, so it removes subjective input and inherits any bias in the data.

SCAFFOLDING EFFECT

psychology

Reduce cognitive load

- Use removal test: delete one criterion at a time and measure how much the ranking of alternatives shifts. - Use differentiating power: read a large shift as evidence that the criterion separates alternatives well. - Use direction handling: classify criteria as benefit or cost before computing the removal effects.

anchor

Anchor fast decisions

A criterion earns its place in a decision by distinguishing among the alternatives. Removing it and recomputing the overall scores turns that intuition into a measurement: if the ranking barely moves, the criterion was carrying almost no information about differences. Iterating over all criteria produces a set of removal effects that can be normalized into weights without subjective input. The approach is fully data-driven, which is a strength with sound data and a weakness with biased data, since no external judgment enters to correct it.

MINIMUM ACTION

In progress 0/1

Practice this model in one real situation:

Build the raw decision matrix and normalize it. 2. Classify each criterion as benefit or cost type. 3. For each criterion, recompute the scores with that criterion removed. 4. Measure the distance between the reduced and original evaluations. 5. Normalize the distances into weights and rank the alternatives.

Check to track your progress (stored locally)

Learning progress0%

account_treeGenealogyexpand_more

menu_bookReferencesexpand_more

Source support: Explicit

link
doi.orghttps://doi.org/10.3390/sym13040525ZH · Explicit
verified

RELATED MODELS