A Comparison of Computer-Assisted Instruction<br>and the Traditional Method of Teaching Basic<br>Statistics

Carmelita Ragasa
Journal of statistics education

The objective of the study is to determine if there is a significant difference in the effects of the treatment<br>and control groups on achievement as well as on attitude as measured by the posttest. A class of 38<br>sophomore college students in the basic statistics taught with the use of computer-assisted instruction and<br>another class of 15 students with the use of the traditional method from the University of the East, Manila<br>(SY 2003-2004) were the focus of this study. The research method used was the quasi-experimental, nonequivalent<br>control group design. The statistical tool was the Multiple Analysis of Covariance. The researcher<br>made use of the CD-ROM prepared by Math Advantage (1997) to serve as the teaching medium for the<br>experimental group. The following summarizes the findings of the study. The achievement posttest of the<br>treatment group has higher estimated marginal means than the control group and it is reversed in the attitude<br>posttest. Using Hotelling's Trace for the multivariate test, the achievement pretest, attitude pretest, and the<br>two groups have a significant effect on the dependent variables, achievement posttest and attitude posttest.<br>Using covariates to control for the effects of additional variables that might affect performance the attitude<br>pretest accounts for about 56% of the variability in the two groups while achievement pretest about 15%.<br>Levene's test shows that the homogeneity of variances assumption between the two groups is met for<br>achievement posttest but not for attitude posttest. The univariate effects for achievement posttest that are<br>significant are achievement pretest, college entrance test overall score, and groups. The univariate effects<br>that are significant for attitude posttest are attitude pretest and high school general weighted average.

The CAUSE Research Group is supported in part by a member initiative grant from the American Statistical Association’s Section on Statistics and Data Science Education