Data Collection
Triangulating the Data for Reliable Results
We incorporate a mixed-methods approach using qualitative and quantitative methods to triangulate data, increase the validity of findings, and ensure the results are useful for continuous quality improvement.
Qualitative
TEG uses qualitative methods, such as key informant interviews, open-ended survey questions for online and paper surveys, meeting minutes and administrative document review, observations, and focus groups (with parents, teachers, students, participants, etc.). Qualitative data analysis is guided by code development informed by scholarly literature, stakeholder panels, and member checking.
Quantitative
TEG develops customized, quantitatively-based instruments, including fidelity indices, surveys, checklists, and observation protocols. Quantitative data is analyzed using descriptive statistics (i.e., means, standard deviations, frequencies, and percentages); parametric (t tests, ANOVA, regression, etc.) and non-parametric (chi square, Wilcox Signed Rank test, etc.) inferential statistics; and effect sizes disaggregated by subgroups.
Fidelity Indices
TEG is a leader in the development of fidelity indices that produce: 1) a core component-level fidelity score, specific to each core component, and 2) an overall-level fidelity score that combines the data across all core components. The fidelity score is a composite index derived from four weighted measures: quality - clear and correct service delivery; dosage - the amount being delivered; reach - the extent to which participants received the component; and reactions - the extent to which to the component stimulates interest and participants are satisfied.