HOME-STUDY DRIVER EDUCATION
instruction methods not requiring a computer are compared in the “No Computer” sub- analysis.
Because the PEN course was added fairly late in the study, there was the possibility that comparisons of PEN students to those in the other instruction methods, some of whom had been assigned before PEN was in the study, would be biased by some sort of calendar effect. Supplementary analyses comparing correctly-assigned non-driver training PEN students to analogous students in the other courses assigned only after PEN was added were also conducted.
A supplementary analysis was also conducted for each outcome measure to determine if the pattern of differences/non-differences among the instructional methods varied as a function of the driver education provider school the students attended. Only students who were correctly randomly assigned and who did not enroll in driver training were included in these analyses. Further, only five provider schools had high enough volumes of these students to make a within-provider analysis meaningful. These schools were Sacramento Driving School, Roadrunner Driving School, Baldy View Driving School, Pegasus Driving School, and American/El Cajon Driving School.
Correlation of Outcome Measures
For descriptive purposes, study exit exam knowledge and attitude (CPI) scores and DMV written and drive test outcomes were correlated using a series of Pearson correlation coefficients. An alpha level of .05 was used to determine the statistical significance of each correlation. The results are shown in Table 3. For purposes of these analyses, DMV written and drive test pass rates were coded so that a higher value indicates passing the test.
Correlations of Study Outcome Measures
DMV written test
DMV written test
Exit exam knowledge
DMV drive test
Note. Higher DMV written and drive scores indicate a passing score. *p < .05.
As can be seen from the table, the only statistically significant correlation among the outcome measures was a positive correlation between exit exam knowledge scores and DMV written test pass rate (p < .05). The meaning of this correlation is that students who had higher exit test knowledge scores also tended to pass the DMV written test.