Plausibility of comparisons supports relevance hypothesis

When each known successful variant differs from at least one other acceptable variant less than critical, this confirms the underlying relevance hypothesis. Would there be no other subcritical different reference example, this would mean (under the assumption of a complete example base), that there differing dimensions were more important than thought.

As soon as an inacceptable example is discovered that differs from the known acceptable variants less than critical, the relevance hypothesis is shattered. From this moment on the importance of the differing factors must be considered as being higher.

The relevance hypothesis is devaluated as well if successful examples are discovered that differ from all other known successes in more than critical dimensions. Intrinsically, a failure should be expected applying the current relevance hypothesis. The underlying relevance expectation can not persist.

The expectations concerning the relevance are disrupted whenever variants perform better than expected or worse than expected. Successful variants that should be failures or acceptable variants that are expected to fail devaluate the relevance hypothesis.

A motor company so far known for its luxory cars launches a station wagon for the first time. The management so far avoided this type of car as it feared the sales in its core market would plunge and the company image would suffer a damage. Despite all doubts, this new station car is a market success and sales in the core business is not hurt. The image of the brand can be enlarged to such a utility vehicle.

Based on inferences from a multitude of largely identical successes, the FACTORFINDER-Procedure calculates the relevance of all factors. The results build on the assumption, that the absence of specific values is not tolerable in important dimensions. Otherwise, the example base can not produce stable expectations concerning relevance. By discovering new ideal values all prior expectations are devaluated. The same happens if an inacceptable example is discovered that differs only subcritical from all examples that are considered to be ideal.

Contradictions between unexpectedly successful/inacceptable examples and a relevance hypothesis can be solved in two ways: