For 20 years we've ignored
the significant differences in the meanings of three critical
words in the quality lexicon: registration, certification
and accreditation. We tend to use them interchangeably.
As a result, there's tremendous confusion in the marketplace.
This seems strange, considering that professed quality practitioners
(people who should understand the pitfalls of ambiguous
definitions) use these three words constantly.
Let's consider how they relate to the quality of the output
of two organizations that may soon merge--the International
Accreditation Forum and the International Laboratory Accreditation
Registration is also called certification, especially
in Europe, which adds to the confusion. Here we're talking
about quality system registration (or certification). To
receive it, an organization must have a quality system that's
been assessed as compliant with the requirements of a standard
like ISO 9001. Any organization can be examined for this
compliance, but this only means that its quality system
meets requirements. It doesn't attest to the organization's
output--except, perhaps, by inference. But no one is checking
the output; the output's quality can vary depending on the
design specifications. Thus, a product's user doesn't really
know how one producer's output compares with another.
When carried out properly, registration indicates that
a user will receive what the producer has designed. But
neither the product nor the production process is examined
to ensure that, in fact, you get a specified product.
Product certification goes one step further than registration.
It attests to the fact that the assembly process is adequate
to produce the product you intended. Certification makes
no judgment about how similar products compare. ISO Guide
65 offers guidelines for product certification, but it doesn't
verify the production process through inspection or assessment.
ISO defines about five different ways to provide product
certification. One way is 100-
percent testing, wherein all products from the assembly
line are tested to see if they meet the requirements. Another
way is to build in a sampling procedure as well as other
controls, a useful alternative when the test is destructive.
Based on certification, one can't tell which product is
best. Using ISO Guide 65, one can only judge--through documentation
and review--if the production process is adequate to meet
the producer's design goals. Most product certifiers insist
that the organization producing the product build a quality
system into the process, and some require quality system
registration, or at least ISO 9001compliance.
The term accreditation is also used ambiguously, a fact
that isn't recognized by the quality community and those
hoping to combine ILAC and IAF. Many assume accrediting
laboratories is the same process as accrediting quality
system certifiers, but the two are quite different.
On one hand, IAF accreditation bodies accredit certifiers
by evaluating them to see that they meet the ISO standard
for certifiers and that the quality systems of those certified
meet ISO 9001 requirements. IAF can't accredit product certifiers
as laboratories are accredited because there's no standard
that states how certification and assessment should take
On the other hand, ILAC accreditation bodies assess a
lab's quality system, examine its production process, and
then judge the lab's competence to produce the intended
product based on reproducible test methods. Laboratory accreditation
requires that the product from different labs is the same--within
the stated uncertainties--and is accepted as meeting the
test method requirements. Further, ISO/IEC 17025 requires
the lab to ensure that the testing requirements are understood
and that the lab uses test methods that meet the client's
needs. By contrast, ISO 9001 speaks only to the client's
The goal of laboratory accreditation is to attain an acceptable
test report everywhere in the world so that products don't
have to undergo multiple and duplicate testing with attenuating
time delays and increased costs. In other words, test data--a
testing lab's product--can be directly compared with test
data produced by all other accredited laboratories.
Laboratory accreditation not only makes judgments about
an organization's quality system and production process,
but also about the product's quality. Accreditation, in
this sense, isn't the same as accrediting certifiers. Incorporating
the IAF and ILAC into one organization may diminish the
significance of the findings of a laboratory accreditation
system because the two approaches are only loosely related.
Two conclusions seem eminently clear:
A standard for the conduct of product certifications, one
similar in scope to ISO/IEC 17025, is needed to make an
objective statement about a product's quality.
AF and ILAC differ in the depth to which they perform assessments,
and even though they both use the word "accreditation,"
the term means different things for each body.
If they do merge, there will continue to be existing confusion.
John W. Locke is a former president of the American Association
for Laboratory Accreditation and manager of the National
Voluntary Laboratory Accreditation Program. Letters to the
editor regarding this column can be sent to firstname.lastname@example.org.