The high-dosage tutoring (HDT) field lacks agreed-upon definitions of tutoring dosage. This definitional ambiguity limits comparability across program models and leads to inconsistency and a lack of reliability when reporting program dosage as part of applied research. In practice, districts and schools procure a certain amount of dosage from tutoring providers, which may or may not align with program design or the amount of tutoring dosage that students ultimately receive in authentic school settings. The amount of dosage that providers commit to offer as part of a district procurement informs both the provider and district/school partners on the tutor capacity necessary to hire and the space and time in the school’s schedule necessary to secure. However, the amount of dosage that providers commit to offer will generally differ from the total dosage students actually receive, due to student (and tutor) absences, unforeseen scheduling conflicts, school closures, and other factors. The existence of multiple measures of dosage means that, unless the measures are well defined and distinguished from each other, many analyses that quantify the relationship between tutoring impacts, dosage, and costs will produce results that are not comparable across studies. Thus, ensuring that both providers and schools are operating under the same understanding of tutoring dosage available to students is essential.
In this brief, we aim to standardize tutoring dosage definitions to inject greater clarity into the HDT field and to improve comparability across programs. Below, we propose formal definitions for the following four dimensions of tutoring dosage: (i) program dosage; (ii) scheduled dosage; (iii) adequate dosage; and (iv) actual dosage. We then apply these definitions of tutoring dosage to the calculation of a tutoring program’s return on time (tutoring efficiency) and resources (cost-effectiveness) in terms of improvements in student learning.
Program Dosage
The total amount of tutoring a student would receive under ideal conditions, assuming full attendance and fidelity to program design.
This is the maximum hours of tutoring as defined by the program model, independent of a particular implementation setting, and defines the capacity needed for full program implementation. Program Dosage assumes that the program is implemented for the full duration of the program (i.e., the program commences on time), the implementation of tutoring is not limited by school scheduling, and there are sufficient personnel and monetary resources to fully implement the program as designed. Program Dosage is valuable for establishing clear parameters for program design activities and for describing the program to external partners. For example, Program Dosage is essential for ensuring there is alignment between the materials and the time students will have to engage with the materials. For decision makers, Program Dosage provides a starting point for identifying the time and space required to be allocated to fully implement a program.
Scheduled Dosage
The amount of tutoring an individual student would receive if they attended all tutoring sessions scheduled by a provider and its district or school partner for a given program implementation.
Scheduled dosage is the maximum amount of tutoring a student could receive based on the scheduled sessions as part of a particular tutoring implementation in a given set of schools and districts. Scheduled Dosage may differ from Program Dosage due to scheduling, budgetary, or other constraints that require deviations from Program Dosage. Scheduled Dosage is most typically calculated by multiplying session length (minutes/session), frequency (sessions/week), and the duration of the program (weeks). Scheduled Dosage is valuable for planning, procurement and resource allocation purposes. Scheduled Dosage informs cost-efficiency by estimating the expected tutoring time and required implementation resources. For providers, comparing Program Dosage and Scheduled Dosage can inform the extent to which program implementations in the field match the ideal design of the program.
Adequate Dosage
A provider-defined dosage threshold above which providers believe a student would realize substantively meaningful improvements in learning.
Adequate Dosage is the minimum hours of tutoring a program deems necessary to improve student learning. It is intended to help providers answer the question: “how much dosage is necessary for students to receive (or are on track to receive) in order to realize improved learning outcomes?” Adequate Dosage is not used to calculate tutoring efficiency or cost efficiency. Instead, Adequate Dosage provides insight into program implementation to assess, for example, the extent to which tutored students receive sufficient tutoring dosage (e.g., 80% of tutored students received adequate dosage). In practice, tutoring providers and school districts may over-schedule tutoring sessions in an attempt to ensure that students receive adequate dosage, so as to account for session cancellations and student/tutor absences. As a result, Adequate Dosage may often be less than Scheduled Dosage, while Program Dosage may often exceed Adequate Dosage because providers expect that additional tutoring beyond the adequate amount will continue to yield additional benefits for students and their program design reflects that expectation. Any implementation where Scheduled Dosage is less than Adequate Dosage is unlikely to be successful in improving student outcomes.
Actual Dosage
The amount of tutoring participating students receive (on average) across the implementation period.
This is calculated by averaging the amount of time (in minutes or hours) of tutoring received across all students who attended at least one tutoring session. The total amount of tutoring time students actually receive can be based on the number of sessions attended multiplied by the length of each session. Notably, the shorter each session is planned to be, the more important it is to directly measure the dosage received in each session. Actual Dosage is valuable for assessing program implementation, calculating the time efficiency of tutoring during program implementation (i.e., Tutoring Efficiency), and understanding the relationship between tutoring dosage received and student learning outcomes. Actual Dosage is used to calculate Tutoring Efficiency to measure the amount of additional learning a student realizes based on the actual amount of tutoring received during a particular program implementation.
| Dosage Type | Definition | Calculation | Applications |
|---|---|---|---|
| Program Dosage | Tutoring dosage under ideal conditions as designed by the program | Session Length (minutes/session) x Frequency (sessions/week) x Duration (weeks) | Program design; sales & marketing, provider staffing decisions |
| Scheduled Dosage | Tutoring dosage planned for in a given implementation setting | Session Length (minutes/session) x Frequency (sessions/week) x Duration (weeks) | Cost Efficiency; school-based planning |
| Adequate Dosage | Minimum dosage necessary (i.e. threshold) for student learning gains | Provider-defined | Monitoring; implementation & outcome analysis; sales & marketing |
| Actual Dosage | Tutoring dosage received | Sessions attended (count of sessions) x Session Length (minutes/session) | Tutoring Efficiency; implementation analysis |
Applying Scheduled and Actual Dosage: Tutoring Efficiency and Cost-Effectiveness
In May 2024, Accelerate introduced Tutoring Efficiency, a measure of the return on tutoring dosage in terms of student learning. Tutoring Efficiency, which we define as the hours of tutoring to improve student learning by one month, enables not only an assessment of the return on tutoring time across program providers, but it is also an essential component for calculating a program’s return on investment (i.e., cost-effectiveness). Together with Tutoring Efficiency, a program’s cost efficiency (which we define as the hours of tutoring dosage that can be purchased per $1,000 per pupil) is necessary to calculate a program’s cost-effectiveness, which we define as the additional months of learning gained by investing $1,000 per pupil (or, the amount of student learning a school can purchase for $1,000 per pupil).1
We leverage multiple measures of tutoring dosage to arrive at a program’s cost-effectiveness. First, Tutoring Efficiency relies on Actual Dosage, which measures the amount of tutoring students actually received. Tutoring Efficiency is calculated as the ratio of actual dosage to an estimate of the program’s impact on student learning (from the same program implementation setting) to generate a measure of the return on time in terms of student learning gains (see “Revisiting Tutoring Efficiency” in QRN 2.1 for a sample of Tutoring Efficiency estimates from Accelerate-funded program evaluations of tutoring providers). Second, cost efficiency relies on Scheduled Dosage, which accounts for the resources necessary to schedule and implement the program independent of the actual dosage of tutoring students ultimately receive, including the allocation of personnel to deliver tutoring, the time and space necessary to schedule tutoring, and the amount of materials or licenses necessary to support tutoring as planned in a particular program implementation. Finally, cost-effectiveness is calculated as the ratio of cost efficiency to Tutoring Efficiency, and measures the additional learning that can be generated for a given price per pupil. A program can become more cost-effective by maximizing tutoring attendance and tutoring dosage by, for example, reducing the difference between Scheduled and Actual Dosage.2
1 In Accelerate’s companion February 2025 report, we introduced our cost analysis tool for collecting program-specific cost data necessary for calculating a program’s cost-effectiveness.
2 A program’s cost-effectiveness may also be improved by lowering program costs (and thus increasing a program’s cost efficiency) and/or by increasing the amount of learning that occurs while students are being tutored (i.e., increasing the return on tutoring dosage and thus maximizing Tutoring Efficiency).
Accelerate thanks Robin Jacob and Catherine Asher (Youth Policy Lab at the University of Michigan), Gregory Chojnacki (Mathematica), and Steve Ross and Amanda Neitzel (Center for Research and Reform in Education at Johns Hopkins University) for helpful comments and feedback on this brief.
