A quasi-experimental multiple time series design was used to analyze and compare the impact of two types of instructional coaching, face-to-face and virtual covert audio provided with Bluetooth technology, on teacher transfer of knowledge learned in professional development into classroom practice. Teacher transfer across baseline, intervention, and maintenance phases was analyzed. The study was conducted at a public elementary school in a Florida suburban school district with approximately 750 students. Twelve teachers were randomly selected from teachers who volunteered to attend professional development. Six teachers (one from each grade level K-5) in the treatment group received virtual covert audio coaching. Six teachers (one from each grade level K-5) in the control group received face-to-face coaching. Professional development was on RallyCoach™, a Kagan cooperative learning structure, which allows students to interact and practice procedural learning such as calculating math algorithms, defending a point of view, or editing writing. This structure was chosen to provide teachers with an instructional tool to teach and provide students practice for the speaking and listening strand of the Common Core State Standards. RallyCoach™ was also chosen to increase student engagement. Data analysis included descriptive statistics and visual analysis methods. Both the control and treatment groups increased the mean (level) percentages of RallyCoachTM components implemented across time from baseline to intervention and from intervention to maintenance. There was an increasing trend line for implementation of RallyCoachTM iv components across phases for both study groups. The decreasing standard deviation across phases represented a decreasing variability of data and can be considered to show a treatment affect for both types of coaching. Teachers who received both types of coaching continued increased implementation into the maintenance phase when the coaching intervention was removed. Data analysis revealed an increasing percentage of student pairs providing positive student-to-student interaction with an increasing trend line and a decreasing standard deviation (reduced variability) across time over phases. Face-to-face and virtual coaching had a positive impact on student-to-student positive interaction.