In Ethiopia in 1968 we were faced with the prob-
lem of trying to find out how much English had
been learned by high school students who had been
taught by American Peace Corps volunteers. The
usual way of doing this there, as elsewhere, is to
give the student a "fill in the blanks," multiple-
choice objective test to see whether the student
knows the meaning of words, understands correct
grammatical forms, etc. We felt that this left out
the most important part of the criterion behavior:
the ability to use English to communicate. So we
asked students to write brief stories which we then
coded objectively, not for grammatical or spelling
correctness, but for complexity of thought which
the student was able to express correctly in the
time allotted. This gave a measure of English
fluency that predictably did correlate with occupa-
tional success among Ethiopian adults and also
with school success, although curiously enough it
was significantly negatively related to a word-game
skill (English antonyms) that more nearly ap-
proximates the usual test of English competence
(Bergthold, 1969).