Issue link: https://takingitglobal.uberflip.com/i/1542824
R. Adeboye, C. Flewelling,V. Ogbole, E. O'Sullivan 15 Coding, Analysis, and Interpretation Interviews Interviews were coded using NVivo. A hybrid coding approach was applied, whereby the PIAs and Outcomes were the principal codes, with subsidiary codes added to reflect emergent subthemes. Sentiment analysis was also applied, distinguishing positive from negative comments. Responses from principals, Connected North staff, and Content Providers were not quantified due to the small sample sizes. Rather, rare or divergent perspectives were noted as such. In contrast, the teacher interviews were sufficient in number to support basic counts of recurring themes, which provided additional clarity on the relative salience of different views while maintaining a qualitative orientation. The following shorthand is used in the report, with each term corresponding to a numerical range: "one" = 1; "a few" = 2–3; "some" = 4–6; "many" = 7– 10; and "the majority" = 11–16. Univariate Quantitative Analysis of Rating-based Survey Questions As noted above, the teacher, Content Provider, and student surveys each included questions that were answered using rating scales, typically a 1-5 scale representing low (1) to high (5) levels of agreement with various statements or ideas. To utilize these responses to address the evaluation questions, each survey question was reviewed and classified as having no 5 , mild, moderate, or strong alignment with the PIAs or Outcomes. These alignments with their justifications are included in the aforementioned technical methodology report; but as an example, teacher agreement with the statement ""I feel that Connected North sessions help to engage my students, contributing to their attendance in class and motivation for learning" was classified as moderately aligned with the short-term outcome "increased student motivation." Although the statement directly references motivation, it is only one element of a multifaceted question 6 , hence the ranking as being moderately rather than strongly aligned. 5 Only items with at least mild alignment were considered and discussed as part of the evidence base. 6 Although analyses of the program's data collection instruments was beyond the scope of this evaluation, in some cases alignment with the Program Theory was less than optimal because the questions themselves did not fully align In the simplest terms, the evaluation questions set out to assess the credibility of the claim that Connected North is contributing to increased graduation rates: if the evaluation finds that Connected North is appropriately targeting its intended outcomes (Q1); and that both the program design (Q2) and the hypothesized outcomes chain (Q3) are supported by educational research and practice; and that the program is being implemented as intended; then it is reasonable to conclude that Connected North is contributing to increased graduation rates—particularly if this inference is reinforced by evidence that the expected outcomes are being observed (Q5).