[b-hebrew] standardized testing

Randall Buth randallbuth at gmail.com
Sat Oct 17 15:38:10 EDT 2009

>> Now if we could get comprehensive, standardized testing we could even make
>> this truly objective. This does not exist for biblical Hebrew, and I've been
>> lamenting this for over ten years in other forums.
> How do you expect to get standardized testing in a language that, based on
> discussions here on this list and elsewhere, is only imperfectly understood?
> Whose standards would be used?

Tests should be machine readable, 'objective'

the place to start is Masoretic Biblical Hebrew. It has a recognized
(If someone didn't study or use 'masoretic' Hebrew, then the test would not be
applicable, of course. Other tests could be made, but I would start
with one based
on the MT.)

Tests should be language neutral. It should not 'prejudice' Korean or
English or . . .
other speakers.
Optimally, questions and answers should be in Biblical Hebrew. However, there
would need to be some compromises, where intro/question/answer sheets for
different language backgrounds.

Tests should be theory neutral, TOEFL and modern language exams can serve
as a model here.
that is, they could test comprehension of the language,
and correct/standard forms of the language,
but they would not test metalanguage or "explanations" of how or why the
language works.

A TOEFL English test does not ask what word order is basic or why one
order or another occurs. But it may propose different examples in a context
where only one is acceptable.

Tests require 'calibration' as to hardness of questions.

basic meanings, vocabulary collocations, fields of synonyms, 'cloze'
fill-in, and
appropriate morphology are relatively unambiguous questions to write and can
distinguish many skill levels.

The biggest problem are the reading comprehension types of questions. Good
questions can be written, but the Bible is such a small corpus that
there is always
a likelihood that someone will answer a question based on background knowledge
rather than reading and understanding the text and the test.

One solution to such problems is to rewrite biblical
accounts in such a way that the character names, places, etc., do not give
away the correct answer.
Testees would be warned that details in the test may
be different or may contradict texts in the Masoretic Bible.
Of course, even in a Masoretic Hebrew-based test,
texts could be provided unvocalized as well as vocalized.

Since this test doesn't exist, I don't know how well it will
differentiate higher levels.
But we can probably get a pretty good reading up to about a 4000-5000
word vocabulary
level. After that stage, uncertainty in the biblical texts and
language may prevent
more precision. Stephen Krashen has reported that vocabulary level serves as a
fairly good indicator of language level and acquistion, so even if it
were the main
category in place, it would be a step forward.


Randall Buth, PhD
randallbuth at gmail.com
Biblical Language Center
Learn Easily - Progress Further - Remember for Life

More information about the b-hebrew mailing list