Group Math Invervention
135
Dependent Variables
The number of digits correct per minute (DCPM) was used as the main dependent
variable. On each assessment sheet there were
10
possible correct digits: Each quotient
consisted of
1
digit, and there were
10
division problems to a sheet. After scoring an
assessment sheet for the number
of
correct digits, the total was then multiplied by
60
and divided by the number of seconds it took to complete the worksheet, resulting in
DCPM (Skinner et al.,
1989).
In addition, the components of DCPM (speed and ac-
curacy) were examined independently, resulting in two additional dependent measures.
Design
The DCPM score obtained by each student on assessment sheets was tracked using
a multiple baseline design across the three sets of division facts, replicated across subjects.
“The baselines (from untreated sets) served as within subject comparisons for evaluating
treatment effects” (Skinner et al.,
1989,
p.
414).
In other words, if rate and accuracy
of responding increase for division facts within the set to which the intervention is ap-
plied and remain stable with untreated sets, a functional relationship is demonstrated
between the intervention and the dependent variable@).
Procedure
The students remained in their regularly assigned classroom, and the normal seating
arrangement was maintained. The intervention was conducted at the same time each
day, two to three times a week. A packet containing three assessment sheets (A,
B,
and
C), three training sheets for the appropriate set, an index card, and a stopwatch was
distributed to each student at the beginning of the period. Students timed themselves
and recorded the number of seconds it took to complete assessment sheets A,
B,
and C.
During assessments the classroom teacher and the experimenter spot-checked the
accuracy of students’ self-timing and recording by periodically following individual
students’ times with other stopwatches. If errors were found while spot-checking the
accuracy of students’ self-timing, the timing procedure was reviewed and the recorded
time was corrected. In the case
of
observed errors in recording the obtained time, the
student was corrected with the statement “the time you recorded does not match the
time on your stopwatch” and was required to correct the recorded time. Although a
couple
of
students’ times had to be corrected occasionally according to the above pro-
cedures, spot checks of accuracy did not reveal major problems with students’ self-timing.
During all sessions, each student completed three assessment sheets (A,
B,
C) and
three training sheets (either Set A,
B,
or
C) where the student had not yet achieved
mastery. Students completed assessment sheets prior to the training sheets. Students
began with training sheets for Set A until they met the criterion of
40
DCPM and
90%
accuracy on three consecutive assessment sheets for Set A; at this point, the training
process was applied to Set
B
until mastery, and then finally to Set C. The continuous
administration of assessment sheets for each of the three sets provided follow-up data
for mastered sets.
During the training phase, students worked their way down each training sheet by
(a) silently reading the first problem and its solution on the left side of the paper, (b)
covering that problem and solution with an index card, (c) writing the problem and solu-
tion from memory in the blank to the right side of the covered problem, (d) uncovering
the problem and solution on the left side to check their written response, and (e) repeating
the procedure with each
of
the
10
facts on three training sheets.