Table 9: Vollands et. al., Project A
Edinburgh reading comprehension
Neale reading comprehension
Standard deviations in parentheses
Thus, the AR group did better on one test, declined less on another, but did not gain as much on the follow-up. This study is not a comparison of AR versus recreational reading alone. It is a comparison of two programs in which students were held accountable for what they read, and the AR group had far more exposure to comprehensible text. In addition, the sample sizes are so small that test results are probably not reliable.
The comparison group in Project B engaged in two kinds of activities. They had 15 minutes per day of sustained silent reading time during which they could read whatever (novel) they wanted to. Vollands et. al. (1999) noted that "children would write their name on a publicly displayed chart when they had finished their book" (p. 54). The comparison group was thus also involved in a kind of incentive program. Comparison children also spent 20 to 60 minutes per week, depending on their reading level, reading from a selection of ten novels, with all reading done aloud by students, and answering comprehension questions in class or at home. This is hardly free voluntary reading. If we count only the genuine sustained silent reading time, comparison children spent a total of 30 hours in actual free reading (15 minutes per day * 120 sessions (six months) = 1800 minutes = 30 hours).
AR students in this study took the tests but received no rewards, bringing their treatment somewhat closer to genuine free voluntary reading. Points were, however, displayed in public. This study, thus, compared two versions of incentives: Points gained in AR tests displayed in public (the AR group) versus titles of books read displayed in public (comparison group).
As was the case in project A, the AR group suffered from a lack of access to books they could read and get credit for, which certainly affected their progress negatively. On the positive side, they were also read to, as were the AR readers in Project A, which most likely boosted their performance; the readaloud time, however, was counted as part of reading time (Vollands et. al., 1996, p. 78).
The AR group read for 15 minutes four times per week for the first three months of the project, and 20-30 minutes for the last three months. This amounts to 1920 minutes or 32 hours (for the first three months, or 12 weeks: 48 sessions * 15 minutes = 720 minutes; for the second three months, or 12 weeks, 48 sessions * 25 minutes = 1200 minutes, or a total of 1920 minutes).