Why do we teach? If you ask 20 university faculty, you will likely get 100 different answers. Hopefully, most answers will be affirming: to train students how to think critically, to prepare students for life, and to motivate students to know how to know. Knowing why we teach is important. It dictates who, what, when, and how we teach; the why generates the curriculum for our courses. Developing curricula and teaching in ways that prioritize students’ participation in learning can be exciting. Furthermore, research demonstrates that using innovative and engaging strategies is a successful and sustainable model for student learning (Froyd, 2007; Kuh, 2008). Therefore, learning how to teach using an active learning approach, an approach where students “read, write, discuss, or be engaged in solving problems” (p. iii) can increase life-long learning, ultimately benefiting students academically, socially, and professionally (Bonwell & Eison, 1991).

Just as there are many ways to teach students using active learning pedagogy (cf. Bonwell & Eison, 1991; Wolff, Wagner, Poznanski, Schiller, & Santen, 2015), there are many ways to teach teachers how to teach using this same framework. Workshops are a primary method of increasing instructor use of a variety of student-centered teaching strategies (Hayward, Kogan, & Laursen, 2015). Other strategies include mini-grants for faculty instructional development (Feldhaus et al., 2015) and faculty learning communities (FLCs; Cox, 2016).

Over the last 15 years, there have been several large-scale initiatives to train science, technology, engineering, and math (STEM) faculty and instructional staff to improve classroom teaching. These initiatives predominantly grew from a National Research Council (2003) report that deemed the pedagogical development of STEM faculty essential in reforming undergraduate education. Evidence, while limited, suggests that pedagogical professional development opportunities for faculty should provide long-term support and model classroom instruction (Ebert-May et al., 2015).

In 2014, a diverse group of academic personnel from North Dakota State University (NDSU), including faculty, instructional staff, and educational data analysts, came together to discuss how to create institutional change by shifting the university’s traditional, large-lecture model of teaching in STEM courses to an active learning–based model. At the time, this transition seemed like an impossible task because most of the STEM faculty in our large science and engineering programs had little to no training in pedagogy, and nearly all the faculty used a traditional, lecture-based format. This structure was especially true for “gateway” courses, those entry-level courses with large enrollment and high DFW rates (i.e., students receiving a “D” or an “F” as a final grade or who withdrew for the semester; Gateways to Completion [G2C], 2018; ).

Funding provided by the National Science Foundation for the Gateways-ND (Gateways–North Dakota; ) faculty training initiative at NDSU enabled the leadership team to create an instructional model that can be implemented with minimal costs at any postsecondary institution. Our assessment data show that Gateways-ND improves faculty teaching and fosters campus-wide social change. With administrative buy-in, minor internal funding, and commitment from pedagogical experts on campus, we believe institutions can move from a lecture-based instructional “norm” to an active learning–based norm without external funding.

In 2014, NDSU shared many of the same problems that define postsecondary education at state universities and land grant colleges throughout the United States. An average of 53% of NDSU’s first-time, full-time students graduated from the campus within 6 years for cohort years 2002–2007 versus 63% of first-time, full-time students at Integrated Postsecondary Education Data System peer institutions. These graduation rates compared even less favorably for four-year graduations (25.4% at NDSU vs. 46.7% at peer institutions). Departments within the NDSU College of Science and Mathematics mirrored these institutional graduation rates, with average six-year graduation rates of 51.4% for biological sciences, 50.0% for chemistry and biochemistry, 48.0% for computer science, and 43.7% for psychology.

Numerical comparisons showed that poor graduation outcomes correlated with student performance in several key gateway STEM courses at NDSU. The average DFW or incomplete (I) rates from spring 2009 to fall 2013 were 35% for General Chemistry I, 41% for Calculus I, 36% for Engineering Mechanics, and 41% for Human Anatomy & Physiology. Data also showed that students who received a D, F, W, or I in gateway STEM courses were less likely to return to the university for the next fall semester. For example, 58% of the students who earned a D, F, W, or I in the Human Anatomy & Physiology course did not return to NDSU the subsequent fall semester. Of the returning students, 28% received a D, F, W, or I on their second attempt at the course. This pattern holds true for General Chemistry I (55% vs. 25%), Intro to Psychology (49% vs. 17%), and a host of other gateway STEM courses. Although pinpointing attrition to one grade in a single course is nearly impossible, the data suggest that there is an observable relationship between student retention and academic success.

Several surveys informed our understanding of lagging student outcomes. From 2007 to 2013, NDSU students reported lower levels of engagement than students at peer institutions on four administrations of the National Survey of Student Engagement (NSSE; 2018). In 2013, both first-year and senior-level NDSU students reported less satisfaction with classroom teaching practices, and student behavioral patterns did not indicate that high-quality, student–faculty interactions were common (NSSE, 2018). Overall, lower satisfaction with instruction and minimal student–faculty interaction seemed to be affecting the tone for learning across campus.

In the years leading up to 2014, the educational landscape at NDSU changed in five key ways opportune for gaining momentum in active learning-based STEM education: (a) introduction of a campus-wide teaching assistant (TA) training program; (b) creation of an Office of Teaching and Learning; (c) construction of a new classroom building, designed specifically for interactive STEM education, including large SCALE-UP (Student-Centered Active Learning Environment with Upside-down Pedagogies; ) interactive classrooms; (d) turnover of the upper administrative leadership to one with an undergraduate education–focused agenda; and (e) institution-wide investment in the John N. Gardner Institute–led G2C (2018) initiative to improve select first-year courses on the NDSU campus. Each of these changes was useful, but not necessary, to support instructional change on the campus, though the G2C set the tone for Gateways-ND.

NDSU participated in the G2C program from 2013 until 2016. The John N. Gardner Institute aids postsecondary institutions in improving student learning and increasing the retention of “low-income, first-generation, and historically underrepresented students” (). Four NDSU departments participated in the G2C program, including Biological Sciences, Chemistry, History, and Psychology. As an outcome of the program, changes in participating first-year courses included: (a) realigning course goals and learning outcomes; (b) adding classroom learning assistants to aid in peer mentoring; (c) incorporating classroom response systems (e.g., clickers) to engage students; (d) implementing earlier and more frequent opportunities for assessment and feedback; (e) identifying and supporting struggling students early in the semester; (f) adding recitation and small-group discussion sections; and (g) improving classroom coordination with campus-based student support services. These changes led to lower DFW rates in some of the departmental gateways courses, shown in Figure 1 for the fall 2011–2016 offerings of gateway courses.

To build on the G2C program, a two-year professional development opportunity for STEM faculty was developed titled Gateways-ND. Three goals and nine objectives guide all Gateways-ND programming, which are listed in Table 1.

Programming was developed in response to student needs extracted from the 2018 NSSE data on student engagement (NSSE, 2018). Student feedback was consistent with the results of studies summarized in Froyd’s (2007) review of the efficacy of active learning pedagogies. Table 2 outlines the set of needs and corresponding group of Gateways-ND interventions.

The Gateways-ND program is designed to establish the relationship between educational research and teaching practice. Over the course of 2 years, each cohort of STEM faculty completes five professional development workshops. The five workshop themes are: (a) beginnings and introductions, (b) building instructional strength, (c) building active engagement, (d) supporting student growth and each other, and (e) being a campus educational leader.

In addition to the workshops, cohort members participate in two additional aspects of the program: COPUS (Classroom Observation Protocol for Undergraduate STEM) and FLCs.

The COPUS is an observational tool that assesses the presence of instructor and student behaviors in the classroom (Smith, Jones, Gilbert, & Wieman, 2013). The COPUS has 12 “instructor is doing” and 13 “students are doing” codes that reflect instructional practices and student behaviors that can be considered passive and active. To elaborate, the codes range from more passive behaviors like instructor lecture and student listening to more active learning behaviors such as the instructor conducting a demonstration and students making predictions regarding the demonstration. A code is marked on the observation protocol if one (or more) of the instructor and student behaviors listed on the COPUS (Smith et al., 2013) occurs within a 2-minute interval.

The COPUS has been noted as an adequate protocol for assessing STEM classroom behaviors (Lund et al., 2015; Smith et al., 2013; Smith, Vinson, Smith, Lewin, & Stetzer, 2014). Compared with other observation protocols (e.g., Reformed Teaching Observation Protocol, Sawada et al., 2002; Teaching Practices Inventory, Wieman, & Gilbert, 2014; Teaching Dimensions Observation Protocol, Hora, Oleson, & Ferrare, 2013), the COPUS has several advantages including reasonable training for use of the observation protocol as well as a straightforward design that limits a number of biases that could influence the protocols’ use (Smith et al., 2013). Additionally, the COPUS was designed specifically for STEM classrooms and includes instructor and student behaviors important to active learning.

Instructors involved in the Gateways-ND project receive two COPUS observations per semester. We found that classes with instructors just beginning the program tended to have more passive instructor and student behaviors (i.e., instructor lecture and student listening). As instructors progress through the Gateways-ND program, instructors display a decrease in passive instructor behavior and an increase in active instructor behavior. Specifically, instructors from the first two cohorts exhibit a year-over-year reduction in lecturing as determined by the percentage of lecturing in COPUS code intervals (M = 66.02%, SD = 22.3% vs. M = 59.31%, SD = 21.1%; t(45) =—1.79, p = .08.) Moreover, instructors demonstrate a twofold year-over-year increase in the use of group work (M = 8.7%, SD = 14.5% vs. M = 16.3%, SD = 19.6%; t(45) = 2.47, p =.02), and a 2.5-fold increase in the use of clicker questions inside the classroom (M = 2.0%, SD = 4.7% vs. M = 5.5%, SD = 11.5%; t(45) = 2.30, p = .03) as indicated by the presence of these activities in COPUS code intervals.

An FLC is “a cross-disciplinary group of 10 or so teachers who engage in an extended (typically yearlong) planned program to enhance teaching and learning and which incorporates frequent activities to facilitate learning, development, and community building” (Cox, 1999, p. 40). FLCs provide a framework for sharing inquiry that benefits individuals and groups, and both learning and nurturing a community are essential outcomes of FLCs (Ortquist-Ahrens & Torosyan, 2009). FLCs are most likely to succeed if they are personally meaningful, voluntary, and characterized by a sense of shared responsibility, a nonthreatening and engaging atmosphere, and genuine inquiry.

FLCs are an important tool in improving teacher attributes, such as commitment to students and self-efficacy (Park, 2005) and are expected to both provide and sustain environments for faculty to collaborate and share professional norms and values (Dooner, Mandzuk, & Clifton, 2008; Vescio, Ross, & Adams, 2008). FLCs are characterized by shared values and vision, collective responsibility, reflective professional inquiry, promotion of individual and group learning, and collaboration (Stoll, Bolam, McMahon, Wallace, & Thomas, 2006).

Taylor (1997) suggested five features that underlie the creation of FLCs: (a) time for intensive learning, (b) probing of belief systems and evaluation of belief systems, (c) safe “zones” to explore new ideas and practice, (d) opportunities for people on different belief sides to discuss and collaborate, and (e) institutional support. This support could take the form of funding from the university, recognition by colleges, and/or recognition of efforts by faculty in processes such as promotion and tenure. If FLCs receive long-term sustained support, they can impact institutional culture (Cox, 1999).

An important feature of FLCs is linking participation to a common desire for a specific outcome or outcomes. For Gateways-ND, the FLCs are cohort-focused and designed to address and meet the needs of members in areas of teaching, learning, and development. During the program, Gateways-ND participants participate in five FLCs. The curriculum of the FLCs is partially informed by the content of the workshops; therefore, FLCs are used to reinforce material covered in the workshops and give time for participants to plan and implement new ideas. Examples of FLC topics include: (a) writing syllabi, (b) developing course goals and objectives, (c) discussing active learning strategies, (d) writing effective assessments, (e) conducting peer observations, (f) providing feedback, (g) sustaining instructional change, and (h) taking leadership on campus.

As we begin Year 4 of the project, surveys, workshop evaluations, and journal reflections aid in improving the program model. Although workshop speakers or methods may change, the program goals and objectives ensure consistency in the participant learning experience. On the basis of reflection and participant feedback, the following changes were made to improve participant learning: more time to discuss and work, understanding direction and program alignment, and continued support for a community of teachers.

Because the fall and spring professional development workshops occur a week before classes start, faculty typically have little time to implement what is addressed in each session. We regularly receive participant feedback about the timing of these workshops, but based on faculty contracts and the varied time commitments of faculty, the week before the semester yields the least number of scheduling conflicts.

Within the workshops, opportunities have been added for attendees to discuss and apply ideas and techniques presented by our expert speakers, which periodically include national STEM education leaders. As one Cohort 2 attendee wrote, “Sometimes allowing time for discussion provides the chance to really think about the big picture and why a specific technique might matter.” Many through Cohort 1 and 2 echoed this sentiment. By providing guided work time, participants are able to make changes to their course documents and take detailed notes on how they might make changes to content delivery, formative assessment, etc., once in the classroom. All of the workshops are scaffolded to deepen understanding. Providing time to work and scaffolding the material is a strategy used to help eliminate faculty feeling overwhelmed.

Frequently, Cohort 2 wanted to skip key building blocks of the program. One participant wrote, “Documenting lesson plans and syllabi are not why I joined Gateways. I want to experience a variety of learning activities that I may then adapt in some way to my classroom should they be applicable.” Although we were excited that participants wanted to learn about active learning, we wanted participants to understand the importance of writing measurable objectives and creating documents such as syllabi and lesson plans that align with those objectives.

During the first workshop for Cohort 3, we dedicated additional time to discuss the programmatic goals and outcomes of Gateways-ND. Although these goals were discussed with Cohort 1 and 2, feedback suggested this could be improved. Like students, participants benefited from explicitly hearing why they were completing a task and how research supported it. Because we wanted participants to employ the backward design–based curricular model in their own teaching, we decided to demonstrate how each workshop aligns with the overall programmatic goals and objectives. This instructional method was an effective way to introduce and model backward design.

During our last workshop of Cohort 2, we invited participants to share what they learned in interactive sessions with their peers. This idea was provided by a Cohort 1 member: “I’d love to see a panel comprised of Gateways-ND participants, where each participant gets 5–10 minutes to talk about something they’ve done in their classrooms. It could be a short description (with goals), how it went, and how it would be changed for a future use in a different class.” In addition to this feedback, we also received feedback asking us to facilitate continued discussions regarding teaching and learning discussion. As a result, optional FLCs are offered to support faculty who have completed the Gateways-ND program.

A program like Gateways-ND can be developed and implemented on a relatively modest budget. Table 3 lists low-cost strategies for implementing a long-term teacher training program similar to the one outlined in this article. Especially in the areas related to data analysis, course release time, and priority room scheduling for participants, administrative buy-in is vital. The importance of priority room scheduling cannot be overstated; at NDSU, active learning classroom space is at a premium, and priority scheduling was cited as a motivating factor by many participants.

Currently, three cohorts have completed the initial two-year commitment of workshops, and Cohort 4 is partway finished. We have selected the participants for Cohort 5. The final workshop for all cohorts focuses on leadership, specifically how faculty might extend the Gateways-ND program and the educational outcomes to the broader campus and beyond. We challenge each cohort to define what leadership means and identify how they can help the campus and wider academic community take pride in their commitment to teaching and learning. Table 4 lists several examples of faculty-based educational leadership that emerged from the Gateways-ND program.

Throughout implementing the Gateways-ND program, we learned that institutional change is possible. As we approach the start of Cohort 5, the program continues to be popular. To date, only 15% of the cohort members have left the program early for jobs at other universities. Two of the roughly 140 members left because of dissatisfaction with the program. Despite only offering a modest personal stipend and institutional incentives (such as highest priority in our active learning classroom building), we continue to have engaged cohorts and faculty who are excited about the program. Overall, Gateways-ND demonstrates that a community of active learning–based educators is possible, and ongoing support, via FLCs and other aspects of the project, is useful for the teacher and the institution. As we look forward, we will continue to evaluate the program based on our goals and objectives.