A survey of the educational system of the Philippine islands by the Board of educational survey, created under acts 3162 and 3196 of the Philippine Legislature.

MEASUREMENT OF INSTRUCTION

119

fundamental conclusions and recommendations which have been drawn
would not have been altered.
Table 2 shows that a very large number of group tests were given.
From Grade III through the first year of high school, in no grade were
there given less than 10,000 subject-tests. In Grades IV and VII each
the number was practically 40,000, in the first-year high school 50,000,
and in the fourth-year high school 26,000. Eight thousand tests were
given in the University. In order to provide conclusive evidence concerning the work of the school on four distinct levels, a very large
sample was taken in these four grades; the fourth, seventh, first, and
fourth-year high-school grades. Thus we have adequate measures of
achievement at the close of the primary school, at the close of the
intermediate school, and at the beginning and the end of the high school.
In addition, a sufficient number of tests was given in Grades III, V,
and VI to tie together the records of achievement in the primary and
intermediate schools. It is possible to report, therefore, a quantitative
measure of the development of the Philippine school system from Grade
III through the first year of the University.
The findings of this study of the work and ability of the pupils, have
great statistical validity. Care was taken in selecting the pupils of a
given grade in a municipality either to test all of the pupils of a school
(generally the central school) or to take a thoroughly random sample.
For example, in municipalities in which pupils had been sectioned in
ability groups, samples were taken from the best section, the poorest
section and two or more middle sections. In no case was the selection of pupils left to officials of the Bureau of Education. The decision
as to which pupil would be tested was always made by the Commission. In many municipalities several times as many pupils were tested
in a given grade as would have been necessary to satisfy statistical
criteria. The number of tests per school system and per grade is
so large that if the measurements were duplicated, average achievements
of municipalities would not be altered one point per test.
TABLE 2.-TOTAL NUMBER OF TESTS GIVEN IN EACH GRADE
GRADE I
Word Pronunciation Tests (individual) -........-....-.........................-....... 571
Oral Reading Tests (individual)................................................ 402
Total................................................................................................. 973
GRADE II
W ord Pronunciation Tests (individual).............................................................. 320
Oral Reading Tests (individual)......................-.. --- —..... —. 126
Total......................................... ---- - - ----- ------------—. --- —- - 446