My article, “The Effect of Engaging Prior Learning on Student Attitudes toward Creationism and Evolution” ( BioScience 55: 996–1003), included errors in the data reported and in some statistical comparisons. I take full responsibility for these mistakes, which would have been impossible for reviewers or editors to find. Although regrettable, the errors do not change the basic conclusion of my article, which was that students in sections A and B (comparative pedagogy) experienced a greater degree of attitude change with respect to creationism and evolution than did students in sections C and D (evolution-only pedagogy).

I have reexamined the original survey sheets and note here the most important corrections; full details and corrected figures are posted on my Web site at  www.cwu.edu/verheys/. Table 2 and figure 3 contain the primary data regarding changes in attitude. Table 2 included errors relating to sections C and D (evolution-only pedagogy). The correct information for section C is as follows: 12 students did not change attitudes, 5 students changed by less than one CL–AE unit, and 2 students changed by one or more CL–AE units. The correct total number of section C students is thus 19. This is the correct information for section D: 10 students did not change attitudes and 1 student changed by less than one CL–AE unit. The correct total number of section D students is 11, and the total for sections C and D combined is 30. Using the corrected data, one-way between-subjects ANOVA comparing the amount of change in each of the four sections indicates that the differences between the groups is significant (F = 3.656, p = 0.0169). The published p value for this test was 0.028.

Counting only change in the rationalist direction, as suggested by Professor Black, it is appropriate to omit the students in sections AB and CD whose initial views precluded further change in that direction. Thus, for sections AB (comparative pedagogy), 9 students changed toward more rationalist views, and 26 students could have changed in that direction but did not; for sections CD, 2 students changed toward more rationalist views, and 25 students could have changed in that direction but did not. Fisher's exact test yields p = 0.059 (one-tailed), p = 0.094 (two-tailed).

Because figure 2 was constructed using data from table 2, it is inaccurate as published. The difference between sections AB and CD for changes of less than one unit is not significant for the corrected data (p = 0.5716, ² = 0.32, df = 1). Other comparisons based on figure 2 data yield p values that are little changed.

As published, figure 3 omitted four students from pooled section AB and one student from pooled section CD. All of these students reported theistic evolution as their before- and after-class attitude. The article states,“None of the 5 students in the other sections whose direction of change could be determined moved toward the nonrationalist end of the spectrum.” The correct number is 2, not 5. A minor error appears in the text on page 1001, where I incorrectly wrote that 8 of 15 preclass CL and YE creationists were in sections A and B. The correct total of CL and YE creationists is 16.