Date of Award


Document Type

Thesis and Dissertation

Degree Name

Doctor of Philosophy (PhD)


Department of Psychology: School Psychology

First Advisor

Gary Cates


This dissertation reports the results of a research project that examined the effects of varying rates of reinforcement on students’ math performance and assignment preference. The relationship between students’ instructional level and the rate of reinforcement threshold for influencing assignment preference was also explored. Participants were fourth and fifth grade elementary students who were receiving remedial math services through the school’s Response to Intervention program.

Twenty elementary students (12 females, 8 males) completed all parts of the study and were included in the analyses. All participants completed a series of control-experimental assignment pairs with varying rates of interspersing at each student’s instructional and frustration level. After completing each assignment pair, participants were asked to rate their perceptions of assignment time, difficulty, and effort to complete on a dichotomous and continuous rating scale. Further, participants assignment preference was assessed by asking students to choose an assignment they would like to take home for homework. Results were analyzed using visual analysis, planned comparisons, and ANOVAs. There was evidence that participants relative problem completion rates were lower on 1:1 and 2:1 instructional level assignment pairs than frustration assignment pairs at the same rate of interspersing. Moreover, a visual analysis indicated that relative problem completion rates increased as the rate of interspersing became denser on instructional level assignments. Total problem completion rates increased on experimental assignments as the rate of interspersing became denser; however, there was minimal to no visible difference on target problem completion rates for control and experimental assignments at the same instructional level. Total problem accuracy increased as the rate of interspersing became denser on experimental assignments at frustration level only. There was a visible difference for target problem accuracy between instructional and frustration level assignments, but there was minimal to no visible difference between experimental and control assignments at the same instructional level despite the rate of interspersing. Instructional level assignments were perceived as requiring less time, effort, and as being less difficult overall than frustration level assignments; however, participants tended to perceive control assignments at instructional level as more favorable than experimental assignments on continuous and forced-choice rating scales. With regard to assignment preference, participants were more likely to choose experimental assignments for homework than paired controls across rates of interspersing.


Imported from ProQuest Sullivan_ilstu_0092E_11505.pdf


Page Count