Dave Dennie

Date of Award





Doctor of Education


Counseling, Foundations & Leadership


A number of states throughout the United States, including Georgia, are implementing a relatively new metric, student growth percentiles, as part of teacher and leader evaluations. Student growth plays a tremendous role in evaluations, accounting for up to 50% of a teacher or leader evaluations, yet there is little to no peer reviewed research on classroom factors that influence student growth percentiles. This quantitative study examined the extent that teacher-student relationships influenced basic psychological needs, engagement, and student growth using the Self-systems Process Model as a framework using structural equation modeling. Based on prior research, it was hypothesized that context (teacher-student relationship) influenced self (basic psychological needs), which influenced action (engagement), and consequently, influenced outcome (outcome).

At the end of the 2015-2016 school year, data was collected from seventh and eighth grade students in a medium to large school district in southwest Georgia that was 73.4% white, 16.6% African-American, 3.4% Hispanic, and 5.1% multiracial, with 29.7% of the students receiving free lunch and 6.3% of the students receiving reduced-price lunch. The 512 student responses were representative of the school population.

Student responses to the modified Network of Relationships Inventory, Basic Psychological Needs Inventory, and Classroom Engagement Inventory showed that students perceived the following: that there was a positive teacher-student relationship, that their basic psychological needs were satisfied in the classroom, and that they were actively engaged. Student responses and their outcomes of Georgia Milestones standardized assessment norm-referenced scores, scale scores, student growth percentiles, and class GPA were used to complete a structural equation modeling analysis.

The findings of the study supported prior research that a positive teacher-student relationship positively influenced levels of engagement in the classroom and, consequently, student outcomes as measured by classroom GPA and standardized assessment results. Using an identical methodological setup that substituted student growth percentiles for scale scores, it was determined that teacher-student relationships, basic psychological need satisfaction, and level of engagement do not influence student growth percentiles across socioeconomic levels and race.