Froyd et al (2017) From Dissemination to Propagation: A New Paradigm for Education Developers PDF

Change: The Magazine of Higher Learning ISSN: 0009-1383 (Print) 1939-9146 (Online) Journal homepage: http://naspa.tandfonline.com/loi/vchn20 From Dissemination to Propagation: A New Paradigm for Education Developers Jeffrey E. Froyd, Charles Henderson, Renée S. Cole, Debra Friedrichsen, Rai...

Change: The Magazine of Higher Learning ISSN: 0009-1383 (Print) 1939-9146 (Online) Journal homepage: http://naspa.tandfonline.com/loi/vchn20 From Dissemination to Propagation: A New Paradigm for Education Developers Jeffrey E. Froyd, Charles Henderson, Renée S. Cole, Debra Friedrichsen, Raina Khatri & Courtney Stanford To cite this article: Jeffrey E. Froyd, Charles Henderson, Renée S. Cole, Debra Friedrichsen, Raina Khatri & Courtney Stanford (2017) From Dissemination to Propagation: A New Paradigm for Education Developers, Change: The Magazine of Higher Learning, 49:4, 35-42, DOI: 10.1080/00091383.2017.1357098 To link to this article: https://doi.org/10.1080/00091383.2017.1357098 Published online: 29 Sep 2017. Submit your article to this journal Article views: 259 View Crossmark data Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at http://naspa.tandfonline.com/action/journalInformation?journalCode=vchn20  FROM DISSEMINATION TO PROPAGATION: A New Paradigm for Education Developers By Jeffrey E. Froyd, Charles Henderson, Jeffrey E. Froyd is a Research Professor at Texas A&M Uni- versity. His research interests include sustained adoption of Renée S. Cole, Debra Friedrichsen, improvements in learning and teaching, systematic reviews in Raina Khatri, Courtney Stanford engineering education, evaluation of faculty campus climates, and assessment of complex learning outcomes. Charles Henderson is a Professor at Western Michigan University with a joint appointment between the Department of Physics and the Mallinson Institute for Science Education. His research focuses on promoting productive changes in In Short higher education at the individual, department, institution, and   Scholarly studies and national reports national levels. document failure of current efforts to Renée S. Cole is an Associate Professor of Chemistry at the achieve broad, sustained adoption of University of Iowa. She teaches introductory and advanced research-based instructional practices, courses in chemistry and runs a research program focused despite compelling bodies of evidence on the design and assessment of instructional materials and supporting efficacy of many of these teaching strategies and faculty professional development. practices. Debra Friedrichsen founded and operates a small consulting   A dissemination paradigm characterizes company. She received her Ph.D. from Oregon State Univer- patterns of these current, failing efforts. sity in Chemical Engineering with a dissertation focused on Change agents, working within the engineering education. Debra also has a MBA, a MS, and a dissemination paradigm, try to convince patent for sensor development. adopters that their innovations can help their students. Raina Khatri is a doctoral student in the Mallinson Institute   Alternatively, change agents, working for Science Education at Western Michigan University. Her within the propagation paradigm, research focus is understanding how education developers in engage with adopters early and often to STEM can promote sustained adoption of their innovations. understand their instructional systems Courtney Stanford is a postdoctoral researcher in the Depart- and interactively develop a strong ment of Chemistry at Virginia Commonwealth University. Her product adaptable to specific contexts. current research focuses on the identification, development, and assessment of process skills (also known as professional skills) in active learning, undergraduate STEM classrooms. www.changemag.org 35 Introduction on to point out the deficiencies of this approach (Seymour, A developer of educational innovations laments failed 2001, p. 92). Along with Seymour, we argue that efforts efforts to achieve sustained adoption of an evidence-based made within the confines of the dissemination paradigm will teaching practice, “I know that developing a strong product not lead to large-scale change of teaching practices across and telling people about it does not lead to sustained adop- undergraduate STEM education. tion; I just do not know what else to do.” Poor results from working within the dissemination The quote illustrates a national challenge. Numerous paradigm are increasingly recognized by organizations that researchers, working in many contexts and disciplines and fund STEM educational development. In response, they supported by a wide variety of funding agencies, have built are emphasizing expectations for propagation of educa- an extensive set of evidence-based practices and demon- tional innovations rather than simply expecting projects strated their effectiveness with respect to science, technol- to “disseminate their work.” For example, the most recent ogy, engineering, and mathematics (STEM) learning. How- solicitation for the National Science Foundation (NSF) ever, this knowledge is not being translated into practice by Improving Undergraduate STEM Education (IUSE) pro- large numbers of STEM teachers (Committee on Undergrad- gram states, “transferability and propagation are critical uate Physics Education Research and Implementation, 2013; aspects for IUSE-supported efforts and should be addressed Graham, 2012; Prince & Felder, 2006). throughout a project’s lifetime….” Similarly, the Alfred While many papers have examined various reasons for P. Sloan Foundation, in reference to higher education in the lack of systemic adoption (e.g., Foertsch, Millar, Squire, STEM, states, “[s]uccessful proposals are expected to be… & Gunter, 1997; Fox & Hackerman, 2003; Henderson & concerned with the … portability of results to other institu- Dancy, 2007; Seymour, 2001), we argue many change tions.” These shifts suggest that funders of educational agents who are working to promote systemic adoption of reform agree with our assertion that change efforts within these teaching practices use approaches that are character- the dissemination paradigm have not led, and are unlikely ized by what we call the dissemination paradigm. Efforts to lead, to systemic adoption of evidence-based instruc- within the dissemination paradigm have not led, and are tional practices. unlikely to lead, to systemic adoption. We offer the propaga- We have worked for some time to better understand the tion paradigm as an alternative with the potential to produce extent to which evidence-based instructional strategies improved results. have been incorporated into undergraduate teaching prac- Education reformer Elaine Seymour describes the dissem- tice (Khatri et al., 2017), as well as identifying barriers to ination paradigm as the belief that “good ideas, supported adoption. As part of our work we have analyzed typical by convincing evidence of efficacy, will spread ‘naturally’— dissemination practice, specifically 71 education develop- that, on learning about the success of particular initiatives, ment proposals funded by NSF in 2009 (Stanford et al., others will become convinced enough to try them.” That is, 2017). With input from experts in disciplinary-based educa- awareness and compelling evidence are necessary and suffi- tion research, we have also identified 44 well-propagated cient conditions to support systemic adoption. Seymour goes innovations in undergraduate STEM and have conducted in-depth case studies of three (Khatri et al., 2017). We have also reviewed the relevant literature from a wide variety of disciplines (Henderson, Beach, & Finkelstein, 2011). Based on this research, we have developed resources (www. increasetheimpact.com) to address developer concerns about Our work has led us to how to propagate educational innovations. Our work and that of others leads to the conclusion that continuing efforts within the dissemination paradigm will articulate a propagation not lead to widespread adoption of evidence-based in- structional strategies (e.g., Borrego & Henderson, 2014; paradigm that may help D’Avanzo, 2013; Graham, 2012; Henderson, Cole, Froyd, & Khatri, 2012; Khatri et al., 2016; Stanford et al., 2016). achieve the goal of systemic Therefore, a new paradigm is needed. Our work has led us to articulate a propagation paradigm that may help achieve the adoption of effective teaching goal of systemic adoption of effective teaching and learning strategies for undergraduate STEM education. and learning strategies Overview Table 1 compares and contrasts the two paradigms across for undergraduate STEM a set of dimensions related to the goal of systemic adoption of the educational innovation. education. Four key terms in Table 1 may need some explanation: efficacy, fit, affordances, and barriers. We use the Diffusion 36 Change July/August 2017  Table 1. Characteristics of the dissemination and propagation paradigms Dissemination Paradigm Propagation Paradigm Goal Systemic adoption of the educational Systemic adoption of the educational innovation innovation Focus of the paradigm Evidence. Telling potential adopters about the Fit. Interacting with potential adopters innovation and its efficacy. throughout the development and dissemination process. What innovation Innovations should be optimized for outcomes. Innovations should be optimized for usability. characteristics should It is important for the developer to collect It is important for the developer to work with the developer prioritize? strong data regarding efficacy, preferably using potential users in multiple settings to develop a an experimental design. product with wide applicability. What is the purpose of Raise Awareness. Focus on getting the word Support Adoption. Focus on learning through developer interactions out about the innovation and evidence of its engaging with potential adopters to promote with potential adopters? efficacy. productive implementation, including customization. How does the Instructional systems are largely a source of Instructional systems provide both affordances instructional system barriers to adoption, increasing the need for and barriers, and are key to successful adoption. influence adoption? more convincing evidence. Changing the Customization is necessary to adapt to the innovation to fit local situations is ignored or local instructional system and, when done viewed as undesirable and likely to degrade appropriately, will likely improve efficacy. efficacy. to understand and apply; (4) trialability—perceived ease of Innovations framework (Rogers, 2003) to help define of giving it a try; and (5) observability to people within the these terms. According to this framework, the attributes of social system. an innovation that promote adoption are: (1) relative advan- For most educational innovations, relative advantage is tage—perceived benefits compared to existing practice; (2) communicated in terms of evidence supporting the efficacy compatibility—perceived consistency with individual needs of an educational innovation in achieving the intended and values as well as consistency with practice in the local outcomes—usually improvements in student learning. The instructional system; (3) complexity—perceived difficulty second term, fit, can be viewed as a composite of compatibility, complex- ity, and trialability. Another way to Figure 1. Relationships among fit, efficacy, and adoption, think of fit is as the degree to which based on the diffusion of innovations framework. an educational innovation is compati- ble with the instructional system and/ or the pedagogical beliefs of potential adopters. Mediating factors in addi- tion to these characteristics include affordances and barriers that are part of the instructional system. Affor- dances are aspects of the instructional system that support or can be used to support changes to teaching prac- tice while barriers prevent or deter change. The dissemination paradigm falls in the lower, right-hand quadrant (high efficacy and low fit) where systemic adoption is unlikely (see Figure 1). In contrast, the propagation paradigm, emphasizing both fit and efficacy, falls in the upper, right-hand quadrant, where systemic adoption is likely. www.changemag.org 37 Illustrative Scenarios developers and the instructors involved in the first round of Two scenarios describe and contrast the two paradigms, testing are now able to go out and spread the word about illustrating the elements in Table 1. In the dissemination the instructional innovation as well as to support additional paradigm, a change agent/educational developer identifies adopters of the innovation. If the innovation ultimately a local instructional problem and has an idea to fix it. The becomes widely popular, by that time there may be a com- change agent, perhaps with a few collaborators in his or her munity of adopters associated with it. Their propagation department, writes and receives a grant based on this idea strategies focus on fit—interacting with potential adopters and then works hard to develop the innovation for specific working in different instructional contexts throughout the needs within the specific instructional context. The devel- development and dissemination process. oper and colleagues use the new instructional innovation for a few semesters in their classes and collect evidence of The key difference between the paradigms is that improved student performance. change agents working within the dissemination para- After development and implementation efforts are nearly digm tell and try to convince adopters that an innovation complete, they carefully compile and document curricular is effective in achieving the desired outcomes. Change resources related to the instructional innovation and then agents working within the propagation paradigm en- start to share this information with potential adopters. They gage with adopters early and often to understand their create a web site for the materials, give presentations at instructional systems and interactively develop a strong professional conferences, and write journal articles. In many product adaptable to these contexts. cases, they also give partial-day workshops at professional As the scenarios illustrate, the underlying assumption of conferences. Their dissemination strategies focus on rais- the dissemination paradigm is that a potential adopter who ing awareness, telling potential adopters about the innova- becomes aware of an innovation that solves an instructional tion, offering compelling evidence within their instructional problem and sees that it “works” (that is, there is sufficient context, and convincing potential adopters their innovation evidence for efficacy of the innovation) will decide to try the will work. innovation. Awareness and compelling evidence viewed as In the propagation paradigm, a change agent/educational are necessary and sufficient conditions to support systemic developer identifies a local instructional problem and has adoption. an idea to fix it. The change agent talks with other instruc- Once a “strong” product is developed, if the product is not tors from his or her discipline at a professional conference used, one or both of the following explanations are offered and finds that they also experience the same problem and for the lack of adoption: (1) potential users do not yet know think that the idea, with modifications, could also work in about the product, or (2) potential users are not convinced their institutions. The change agent incorporates these sug- by the existing evidence. Thus, if systemic adoption has not gestions and insights into a grant proposal, asking some of yet occurred, change agents, working within this paradigm, these colleagues to serve in advisory roles or as beta testers. should focus on (a) doing a better job of getting the word out The change agent might also coordinate a collective effort and/or (b) developing increasingly strong studies to demon- to write and receive a multi-institutional collaborative grant strate effectiveness. based on this idea. If the innovation is not adopted and the change agent Once the grant is received, the developers refine the in- decides that the reason is lack of awareness, then the change structional innovation a bit through use in their own classes agent should make more presentations, publish more journal and quickly develop a set of core ideas. Before solidifying papers, and improve the website. If the innovation is not their ideas, they talk to a dozen or so instructors from a adopted and the developer (i.e., change agent) decides that variety of institutions who are teaching the same target class. the reason is insufficient compelling evidence, then the Through these conversations, the change agents are able to change agent should invest more resources in more rigorous identify ways to make their ideas fit in a variety of contexts. comparisons (e.g., greater numbers of students in treatment They then continue development of the instructional innova- and comparison groups, random assignment to treatment and tion at their institutions. Once an early version of the instruc- comparison groups, more efforts to reduce effects of con- tional innovation is ready, the change agents again solicit the founding factors, etc.). help of other instructors in testing the innovation. Alternatively, lack of adoption may be characterized as The goal of these tests is to document efficacy as well as resistance. Instructors are labeled as “resistant” to change to identify productive modifications and implementation when they cite features of the instructional system around difficulties. The developers carefully compile and document them that serve as barriers to change (e.g., lack of time or curricular resources related to the new instructional innova- a reward system that does not value teaching). Within the tion. These resources include recommended variations in use dissemination paradigm, it is thought that this resistance can and expected difficulties that will need to be overcome. They be overcome by stronger evidence of efficacy. Developers also create a web site for materials and give presentations, therefore focus on convincing instructors, who are assumed workshops, and write journal articles. to make individual decisions about instructional approaches, Their most important source of potential users, how- that their current (traditional) teaching practices should ever, are developed through word of mouth. The multiple 38 Change July/August 2017  Figure 2. Contrasting Dissemination and Propagation Paradigms be exchanged for an evidence-supported innovation that teaching, large class sizes, necessity to cover so much mate- improves student learning. The expectation is that potential rial) hindered their adoption of evidence-based instructional adopters will want to use “proven” materials that will im- practices. Despite the importance of the instructional system prove student learning. on adoption decisions, these potential adopters are labeled In the dissemination paradigm, the instructional system within the dissemination paradigm as resistant to change. therefore plays, at most, a minor role. However, significant Significant work within the dissemination paradigm research has found that many potential adopters hold beliefs has been done, and strong bodies of evidence support the and values about teaching and learning that are more aligned efficacy of many types of innovative instructional strate- with the evidence-supported innovation than their current gies. The work to develop these instructional strategies and teaching practices, but often these views have not resulted in demonstrate their effectiveness has been invaluable. How- changed practice (Henderson & Dancy, 2004; Yerushalmi, ever, only limited numbers of instructors have adopted these Henderson, Heller, Heller, & Kuo, 2007). Instead, potential instructional strategies. Continued work within the dissemi- adopters make their teaching decisions within the context of nation paradigm alone will not lead to systemic adoption. a larger system that is not addressed effectively in the dis- In contrast to the dissemination paradigm, the essence semination paradigm. of the propagation paradigm is that a potential adopter will Evidence collected from interviews with “physics faculty adopt an instructional innovation that solves an instructional about their instructional practices, conceptions about teaching problem or improves some aspect of student learning or and learning, and experiences with educational innovation” course delivery (efficacy) and fits well within their local (Henderson & Dancy, 2007, p. 020102-1) indicated “that instructional system. Fit, i.e., alignment with both individual these faculty agreed with PER [physics education research] instructor needs, preferences, and beliefs, as well as the researchers on many of the problems with traditional instruc- larger instructional system, critically influences adoption. tion and were all aware of a variety of research-based alterna- Therefore, developers should focus many more of their tives” (Henderson & Dancy, 2007, p. 020102-1). However, efforts on understanding the instructional systems of poten- these faculty indicated that aspects of the instructional system tial adopters; identifying characteristics of the innovation (e.g., institution does not reward time spent focused on that address concerns; working with potential adopters to www.changemag.org 39 design appropriate instructional innovations; and supporting Therefore, providing users support following initial adoption adopters in adapting these innovations to their instructional will be just as important as engagement to promote initial system. adoption. Implementing an educational innovation likely requires In the propagation paradigm, lack of adoption is viewed that some aspects of the instructional system be changed. as typically being a result of an innovation not matching suf- These aspects can be envisioned in terms of four structural ficiently with the local environment or insufficient support levels: individual, departmental, institutional, and extra-in- being provided to sustain adoptions. Barriers to adoption are stitutional (Henderson, Finkelstein, & Beach, 2010; Stanford seen as legitimate reasons for concern instead of undesirable et al., 2016). Within the propagation paradigm, developers sources of resistance. Therefore, developers explicitly iden- understand what will need to change, what elements of the tify (a) who makes adoption decisions, (b) characteristics system will aid adoption (affordances), and what elements of these potential adopters, and (c) affordances and barriers of the system will hinder adoption (barriers). They then arising from the instructional environments in which adopt- apply this knowledge in the development of the innovation, ers reside. determining what factors should be considered in the evalu- Instead of being unaware of barriers, ignoring barriers, or ation of the innovation (e.g., gathering evidence on content promoting adoption despite barriers, developers modify their coverage) and determining supports for implementation. In innovations or offer workarounds. They identify potential in- general, the greater the change in the instructional system compatibilities early in the development process and modify required by adoption, the more difficult it will be to make or redesign innovations to reduce incompatibilities when the change. Propagation strategies need to adjust depending possible. When not possible, developers construct strategies on the degree of change required. that acknowledge and address barriers. In addition, develop- Unlike the dissemination paradigm, developers working ers are aware of and take advantage of affordances. within the propagation paradigm expect individual instruc- Within the propagation paradigm, developers focus on tors will almost always modify an instructional innova- both product development and potential adopter develop- tion for better fit with the local instructional system. Thus, ment (Blank & Dorf, 2012; York & Danes, 2014). That change agents work with potential users from the very is, developers must understand factors that positively and beginning to develop a product that emphasizes compat- negatively influence adoption and address these factors. This ibility with instructional systems of individual users as involves understanding the multiplicity of instructional con- well as effectiveness in addressing intended learning goals. texts and making the innovation easily adaptable and provid- Evidence is still important, but it is embedded in the context ing opportunities for potential adopters to participate in the of fit. Further, at least one-third of instructors who try an development process. Developing archetypes of specific educational innovation stop using it (Henderson, Dancy, & adopters and the instructional systems in which they work Niewiadomska-Bugaj, 2012); we call this “adopt and drop.” allows change agents to better identify specific individuals representing these archetypes who can provide feedback. Engaging potential adopters early in the process pro- vides insight into what drives the decision to adopt, what A recent correspondence received by one of the the major barriers to adoption are, and how changes in the authors exemplifies the difference between the two innovation can minimize those barriers without sacrificing paradigms in engagement with potential adopters. the innovation’s effectiveness. Engaging several different types of individuals from different contexts can inform a An instructor (and educational developer) has been developer about how features of an innovation fit in different developing new instructional materials and has given contexts with different people. After an innovation is mostly presentations about their development. These presenta- developed, continuing this interaction through initial imple- tions often result in requests for access to the materi- mentation by additional adopters (beta testers) provides a als, which the author has declined. Her reasoning was developer with information about the issues instructors have that the materials were not ready to share with others when implementation is less controlled and what actions and yet—she wanted them to be “finished” before she let structures will support adoption both in initial implementa- others see them. tion and through to sustained adoption. After hearing about our work and reflecting on the Keeping the communication channels open also provides implications, the educational developer is rethinking insight into additional support that may be needed as an her assumptions about when she should begin sharing adopter becomes more familiar and comfortable with an in- the materials. This change in mindset has come from a novation as well as how to empower adopters to share their realization that early interaction with potential adopt- experience and enthusiasm in a way that increases interest ers allows the developer to determine affordances and in the innovation. Identifying and interacting with potential barriers for adoption early in the development process, adopters to better understand the affordances and barriers of allowing for the optimization of the former and mini- the instructional system that need to be addressed to encour- mization of the latter. age and support implementation is as important as develop- ing a strong product. 40 Change July/August 2017  convincing evidence to support their efficacy. However, ef- Systemic adoption of Peer Instruction illustrates forts within this paradigm have not led to widespread use of the importance of fit of an educational innovation these innovations. within the instructional system. We believe shifting to the propagation paradigm has the potential to achieve desired widespread adoption. The propa- Peer Instruction is well propagated and has wide ap- gation paradigm is characterized by a focus on potential peal because its developer, Eric Mazur, understood adopters and the fit of innovations. This includes involv- instructional systems in physics. Mazur, a physics ing adopters in development of innovations and supporting professor at Harvard, knew that his lectures were not adopters in customizing these innovations for their instruc- achieving the outcomes he wanted. First, he thought tional system. about teaching students in smaller groups—quickly Shifting paradigms is never easy and typically takes a nixed. While small-group instruction would likely long time; however, we have reasons for hope. Although we have been effective, and he might have been able to are not aware of anyone explicitly articulating the propaga- pull it off with Harvard’s resources, he wanted to have tion paradigm, we are aware of movement in this direction. an impact on college physics teaching as a whole. Any For example, as noted earlier, both funders and education systemic change would need to be inexpensive and not developers have expressed dissatisfaction with the results require significant extra time on the part of instructors of working within the dissemination paradigm. Similarly, or TAs. Peer Instruction was successful in part because funding agencies have begun to move away from programs Mazur understood the instructional systems in physics that work within the dissemination paradigm and begun to and what was likely to be adopted widely. develop program elements more consistent with the propa- gation paradigm. We believe contrasting these two paradigms can provide language and support conversations that speed transition Conclusion towards the propagation paradigm. Over 20 years ago, Barr We contend widespread adoption of new educational and Tagg contrasted the instruction paradigm and the learn- strategies and teaching materials is hindered because efforts ing paradigm (Barr & Tagg, 1995). This framed and facili- to develop and support systemic adoption of these strate- tated constructive conversations that have slowly created a gies have occurred and continue to occur within the dis- shift to the learning paradigm. Similarly, we hope articulat- semination paradigm. The dissemination paradigm focuses ing the propagation paradigm will help catalyze conversa- primarily on the innovation. This focus has been productive tions to shift how education developers think about and in developing many high-quality innovations and amassing engage in educational change. C Resources Barr, R. B., & Tagg, J. (1995). From Teaching to Learning: A New Paradigm for Undergraduate Education. Change: The Magazine of Higher Learning, 1383(March), 12–26. doi: 10.2307/40165284 Blank, S. G., & Dorf, B. (2012). The Start-Up Owner’s Manual: The step-by-step guide for building a great com- pany. K&S Ranch, Inc. Borrego, M., & Henderson, C. (2014). Increasing the use of evidence-based teaching in STEM higher education: A comparison of eight change strategies. Journal of Engineering Education, 103(2), 220–252. doi: 10.1002/jee.20040 Committee on Undergraduate Physics Education Research and Implementation. (2013). Adapting to a Changing World—Challenges and Opportunities in Undergraduate Physics Education. Washington, D.C.: The National Acad- emies Press. D’Avanzo, C. (2013). Post-vision and change: do we know how to change? CBE Life Sciences Education, 12(3), 373–82. doi: 10.1187/cbe.13-01-0010 Foertsch, J., Millar, S. B., Squire, L., & Gunter, R. (1997). Persuading Professors: A Study of the Dissemination of Educational Reform in Research Institutions. Madison, WI. (continued) www.changemag.org 41  Resources (cont’d) Fox, M. A., & Hackerman, N. (Eds.). (2003). Evaluating and Improving Undergraduate Teaching in Science, Tech- nology, Engineering, and Mathematics. Washington, DC: National Academics Press. Graham, R. (2012). Achieving excellence in engineering education: The ingredients of successful change. London. Henderson, C., Beach, A., & Finkelstein, N. (2011). Facilitating change in undergraduate STEM instructional practices: An analytic review of the literature. Journal of Research in Science Teaching, 48(8), 952–984. doi:10.1002/ tea.20439 Henderson, C., Cole, R., Froyd, J., & Khatri, R. (2012). Five Claims about Effective Propagation, 1–4. Henderson, C., & Dancy, M. (2007). Barriers to the use of research-based instructional strategies: The influence of both individual and situational characteristics. Physical Review Special Topics—Physics Education Research, 3(2), 20102. doi: 10.1103/PhysRevSTPER.3.020102 Henderson, C., & Dancy, M. H. (2004). Teaching, Learning and Physics Education Research: Views of Mainstream Physics Professors. In Proceedings of the Physics Education Research Conference. Henderson, C., Dancy, M., & Niewiadomska-Bugaj, M. (2012). Use of research-based instructional strategies in introductory physics: Where do faculty leave the innovation-decision process? Physical Review Special Topics— Physics Education Research, 8(2), 20104. http://doi.org/10.1103/PhysRevSTPER.8.020104 Henderson, C., Finkelstein, N., & Beach, A. (2010). Beyond Dissemination in College Science Teaching: An Intro- duction to Four Core Change Strategies. Journal of College Science Teaching, 39(5), 18–25. Khatri, R., Henderson, C., Cole, R. S., Froyd, J. E., Friedrichsen, D., & Stanford, C. (2017). Characteristics of well-propagated teaching innovations in undergraduate STEM. International Journal of STEM Education, 4(2), 1–10. doi: 10.1186/s40594-017-0056-5 Khatri, R., Henderson, C., Cole, R., Froyd, J. E., Friedrichsen, D., & Stanford, C. (2016). Designing for sustained adoption: A model of developing educational innovations for successful propagation. Physical Review Physics Educa- tion Research, 12(1), 10112. doi: 10.1103/PhysRevPhysEducRes.12.010112 Prince, M. J., & Felder, R. M. (2006). Inductive Teaching and Learning Methods: Definitions, Comparisons, and Research Bases. Journal of Engineering Education, 95(2), 123–138. doi: 10.1002/j.2168-9830.2006.tb00884.x Rogers, E. (2003). Diffusion of innovations (Fifth ed). New York, NY: Free Press. Retrieved from http://books. google.com/books?hl=en&lr=&id=v1ii4QsB7jIC&oi=fnd&pg=PR15&dq=Diffusion+of+Innovations&ots=DJ_ vqNZtcX&sig=NAZCoco-0KFyM2iMFh4vzZtDDMI Seymour, E. (2001). Tracking the processes of change in US undergraduate education in science, mathematics, engineering, and technology. Science Education, 86(1), 79–105. doi: 10.1002/sce.1044 Stanford, C., Cole, R., Froyd, J. E., Friedrichsen, D., Khatri, R., & Henderson, C. (2016). Supporting sustained adoption of education innovations: The Designing for Sustained Adoption Assessment Instrument. International Jour- nal of STEM Education, 1–13. doi: 10.1186/s40594-016-0034-3 Stanford, C., Cole, R., Froyd, J. E., Henderson, C., Friedrichsen, D., & Khatri, R. (2017). Analysis of propagation plans in NSF-funded education development projects. Journal of Science Education and Technology, 26(4), 418–437. Yerushalmi, E., Henderson, C., Heller, K., Heller, P., & Kuo, V. (2007). Physics faculty beliefs and values about the teaching and learning of problem solving. I. Mapping the common core. Physical Review Special Topics—Physics Education Research, 3(2), 1–31. doi:10.1103/PhysRevSTPER.3.020109 York, J. L., & Danes, J. E. (2014). Customer Development, Innovation, and Decision-Making Biases in the Lean Startup. Journal of Small Business Strategy, 24(2), 21–39. doi:10.1111/j.1748-5827.2011.01219.x 42 Change July/August 2017