|Abstract or Summary
- Statement of the problem:
The study sought to determine whether certain
measurable differences existed between students who had
withdrawn after acceptance into the accelerated mathematics
classes in the Portland schools and students who
had persisted after acceptance into the program. Examined
were: scores on the Iowa Tests of Basic Skills, the Iowa
Algebra Aptitude Test, the Seattle Algebra Test, the
Lankton First Year Algebra Test, the Iowa State Test of
Educational Development, and the School and College
Abilities Test; eighth grade and high school mathematics
grades and total grade point averages; attendance; age;
extra-curricular school activities and community
activities; part-time jobs during the school year; sex;
family composition; and occupations of parents.
Through the use of a random sampling technique, 160
subjects were selected from those 3,068 pupils who
commenced the accelerated mathematics program in the
Portland public schools in 1956, 1959, and 1960.
Data were collected from the files of the Mathematics
Department, School District Number One, Portland,
Oregon and the attendance records, transcripts, cumulative
folders and counselors' personnel files in eleven high
schools in the district.
The parametric statistic, Fisher's t, was employed to
test the difference between means for the twenty variables
which sufficiently satisfied the assumptions for its use.
The distributions of observations for nine variables not
satisfying these assumptions were examined by three nonparametric
tests: the Kolmogorov-Smirnov test, Fisher
exact probability test, and the x² The test selected
for each of the nine variables depended on the
associated statistical model and level of measurement.
A statistical analysis was not made for the variable of
occupations of rents.
Over all, except for the variable of sex, there were
no statistically significant differences (at the .05
level of significance) for the variables tested between students who were accepted into and who withdrew from the
accelerated Mathematics program in the Portland schools
and those who were still in the program at the time of
the study. This suggests that either the significant
variables were not selected for testing; that patterns of
multiple variables determine differences, rather than
individual variables; or that the sample size was too
small for the tests used.
Four of six comparisons made for the variable of sex
showed significance at the .05 level. The pattern set
was that of more girls in the groups who dropped and more
boys in the groups who persisted. The two groups that
failed to follow the pattern were the two samples from the