|
Frequently Asked Questions About Licensing Exams |
Multi-language exams
CLEAR Exam Review
(Summer 1992)
Eric Werner, M.A.
Question: I recently attended a meeting at which a state official announced that having English as a second, nonprimary language constituted a disability. Therefore, he said, state licensing boards must accommodate this condition (with interpreters and translated exams, for example) as part of their compliance with the ADA. Is this correct?
Answer: No. I've asked numerous lawyers who are familiar with the ADA and with the 1973 Rehabilitation Act whether ESL status is a protected disability. All have said that ESL status is not, by itself, a physical or mental impairment that substantially limits one or more of the major life activities (such as caring for one's self, performing manual tasks, walking, seeing , hearing, speaking, etc.). Therefore, it is not covered by the ADA.
CLEAR Exam Review
(Winter
1993)
Eric Werner, M.A.
Question: Some candidates, and officials of the schools that train them, have asked our licensing board to translate its written test into languages other than English. Persons making this request feel that translation would benefit test takers who have weak English skills by allowing them to be tested in their native tongue. What guidance or opinions have you in this area?
Answer: I suggest that before
deciding what to do, you examine the multiple dimensions of the
sensitive issue before you. These include public protection,
psychometrics, costs and benefits, legal/ governmental mandates,
and political/cultural realities. Dual-language testing is a
complex matter and different persons will reach different
conclusions about whether it should be done in a particular
circumstance. Therefore, you should not assume that my opinions
in this area will necessarily be shared by all other testing
professional, let alone by all other credentialing officials.
Licensing boards are receiving an
increasing number of requests for exam translations and
interpretations. Not many are providing translations, but some
are allowing the use of interpreters. There are many reasons why
most boards do not translate or interpret their tests. I think
that the reasons are generally sound, although each situation
should be evaluated independently. I recently looked into this
area and can offer you the following summary observations on your
question. Be sure to look at the Summer, 1990 CLEAR Exam Review
for two relevant articles.
Psychometric standards neither encourage
nor discourage dual-language testing. They emphasize that if it
is to be done, it must be done properly rather than expediently.
Proper procedures for translating written
tests exist. However, applying them to a sizeable item bank can
be extremely expensive and still may leave legitimate concerns
about the equivalence of English test forms and their
translations.
Many professional translating services
have little experience translating high-stakes multiple-choice
tests. They tend to be unaware of the technical and legal
difficulties that distinguish such translation tasks from the
translation of other forms of text. Consequently, they are likely
to underestimate the final costs of translation and to
overestimate its likelihood of success.
Very few licensing tests are translated,
and those that are translated are rarely accepted for licensure
outside one or two state jurisdictions. The few instances in
which licensing tests are translated tend to involve low-skill
occupations that call for little reading comprehension. Even in
these cases, specially developed basic English reading tests will
be likely to supplement the translated tests of occupational
knowledge.
Translating occupational and professional
licensing tests of multiple-choice format is different in
important ways from translating such assessments as driver's
license tests. For example, the latter often are quite short, use
either true-false or simple two-choice format, do not involve
technical or conceptual terms, and may be not be subject to
significant security conditions. The fact that many states
translate driver's tests does not provide a good basis for
arguing that occupational and professional licensing tests should
be translated also.
Many commercial test developers are
skeptical about proposals to translate or (especially) interpret
licensing tests because of the significant chance of lessening
test validity, reliability, and fairness. Fairness concerns not
only simple candidate equity issues, but also more serious
problems resulting from efforts to make testing more affordable.
The meaning and difficulty of translated
test questions may be different from English versions of the same
questions. This is because language structure often does not
translate with the original meaning, resulting in equity problems
between candidates who take different language versions of the
"same" test. Further, a test that has been translated
into a particular language often will not satisfy the needs of
all candidates who speak that language, because of dialect
variation among the candidates.
The number of candidates who would choose
to take a given translated form of a state test is likely to be
small. Therefore, it will usually be impractical to make the kind
of statistical item and test analyses that are necessary to
ensure that a translated test behaves the same as its English
counterpart.
Probably the majority of licensing
officials believe that some minimal level of English
comprehension is essential to the safe, effective practice of
most occupations and almost all professions.
Some candidates seeking to enter educationally less-demanding occupations are limited in English reading skills and not fully literate in their native tongue. These persons may do little, if any, better on translated tests than they do on English tests. Multiple modes of test presentation are likely to be necessary to ensure that these candidates understand the translated questions.
Back to
index
© 2002
Council
on Licensure, Enforcement and Regulation