Exploring the dynamics of organizational learning: identifying the decision chains science and math faculty use to plan and teach undergraduate courses

Exploring factors influencing faculty decision-making

Higher education researchers have long been interested in delineating the specific factors that shape faculty work, particularly in regard to curriculum design and instructional practice. In the 1970s and 1980s, researchers sought to develop explanatory models of course planning to help designers create and improve degree programs (e.g., Mayhew and Ford [1971]), though these early efforts were critiqued for positing an overly linear and simplistic account of faculty practice that likely did not exist in real-world settings (Conrad and Pratt [1983]). Later research attempted to account for a more complex array of factors including environmental inputs (e.g., professions and community), the nature of the material, instructional traditions, and resource allocations, to name but a few (Conrad and Pratt [1983]; Lattuca and Stark [2009]).

A shift to more multifaceted and systemic account of faculty practice was also the hallmark of the research program of Joan Stark (Stark [2000]). Given the autonomy that many faculty enjoy in their work, Stark hypothesized that personal characteristics (e.g., beliefs about teaching and learning) would largely drive their decisions about the structure and content of a course, but these individual-level factors were influenced or ‘filtered’ by features of the organizational context that ultimately determined how a course was designed and taught (Stark [2000]). A key finding from this line of inquiry was that much of postsecondary instructors’ planning involves minor alterations to existing materials on account of the fact that instructors often teach the same course(s) year to year—what Stark ([2000]), p. 420 called ‘routine maintenance’. However, as this work was primarily survey based, it did not capture or describe the specific features of faculty practice as it unfolded in real-world situations.

More recently, research on STEM education has begun exploring similar territory, largely based on the widespread recognition that the pace of pedagogical change has not been what policymakers or educational leaders anticipated or desired. As previously noted, a growing body of research is exploring which individual, cultural, and organizational factors appear to be most influential in supporting or inhibiting faculty adoption of interactive teaching modalities. For example, Henderson and Dancy ([2007]) found that factors including student attitudes towards school, departmental expectations of content coverage, lack of time, and class size each acted as barriers to the adoption of new teaching methods. Building on the promise represented by such analyses, researchers have subsequently called for additional work to identify ‘the factors at work within the multiple contexts that affect faculty work’ , so that educational leaders can target these ‘multiple change levers’ when designing and implementing reform initiatives (Austin [2011], p. 3)^c. However, a theoretical framework that is able to model decision-making within complex systems, as well as research methods to empirically identify them, is needed to help the field of STEM education meet these goals.

Organizational learning in postsecondary institutions

Organizational learning has a long history in organization and administration science, but it is important to distinguish between the empirical research program of organizational learning on the one hand and the managerial theory of the learning organization on the other hand (Kezar [2005a]). The empirical research program is generally focused on the investigation of the processes of organizational functions and change. In contrast, the learning organization approach is a more of a proscriptive approach to management that describes or advocates an idealized state of affairs. While the latter is grounded in the former, some consider the learning organization approach to be a popular yet fleeting management theory similar to total quality management (TQM) (Birnbaum [2000]; Kezar [2005a]). While the concepts are certainly related, the long-standing research program of organizational learning is the main concern of this paper.

At its core, research on organizational learning focuses on the underlying processes of how organizations learn, change, and adapt (Levitt and March [1988]). Much of the later research in this area was strongly influenced by advances in cognitive psychology. Concepts such as cognitive ‘scripts’ and ‘schema’ , which refer to mental representations related to particular information stored as neural pathways in the brain, became widely used in the organizational learning literature (e.g., Gioia and Poole [1984]). For example, information processing theory focuses on the development of organizational systems to process information as a key function of organizational success in uncertain environments (Galbraith [1977]). This approach emphasizes the role of information pertaining to organizational experience (e.g., established routines and effective strategies) in organizational functioning and improvement. This stored information is important for understanding organizational behavior because, once retrieved, it may act as ‘decisional stimuli’ that elicit particular responses and actions (Walsh and Ungson [1991] p. 61).

In any case, one of the most appealing features of organizational learning theory for education researchers is the prospect that an organization can in fact grow, evolve, and change in a positive manner, and that these changes can be empirically studied to provide evidence for organizational improvement. When considering this application, it is useful to understand the long-standing debate in the field regarding what constitutes evidence of learning and change. For some theorists, evidence of organizational learning is only apparent by increased performance or observable changes in behavior, whereas others argue that changes resulting from learning may not be visible or immediately evident. For example, Huber ([1991]) argues that ‘an entity learns if, through its processing of information, the range of potential behaviors is changed’ (p. 89). This focus on alterations to the mechanisms that underlay the potentialities of human behavior is similar to Argyris and Schön ([1974]) well-known distinction between single- and double-loop learning. In this view, single-loop learning occurs when an organization changes superficial behaviors or protocols but leaves internal norms and practices intact. In contrast, double-loop learning focuses on the internal mechanisms of the group or organization and involves enacting changes to deeply held traditions, assumptions, and behaviors (Argyris and Schön [1974]). Interestingly, just as an organization may learn by altering entrenched behaviors, an ‘unlearning’ process can be useful when organizations discard old and unproductive information or practices (Nystrom and Starbuck [1984]).

In delving more deeply into the specific processes of organizational learning, Huber ([1991]) argued that researchers should think in terms of information acquisition, distribution, processes of interpretation, and memory functions. In particular, organizational memory, or what Levitt and March ([1988]) call the recording and conservation of experience, is perhaps the most critical part of the learning process. This is due in part to the fact that most organizations experience personnel turnover and regular upheavals (e.g., course rotation procedures) that make the storage of information critical because newcomers will draw upon these storehouses of information to guide their behavior.

From these different repositories, information is retrieved during preparation for a task or during the task itself. This retrieval process can be thought of as either a deliberate, controlled process or an automatic, habituated process where decision makers are not consciously aware of how they retrieve information or from where. Once retrieved, the information enters into an active decision-making process and is thus subject to a variety of influences including sociocultural and physical characteristics of the immediate task environment and the preexisting cognitive attributes of the decision maker. That is, the prior experiences, knowledge, and beliefs of individuals will shape how the new information is interpreted and then used (Daft and Weick [1984]; Hora [2014]). This process of interpretation highlights the fact that the organizational information, once retrieved, is not simply enacted into practice but instead undergoes a translation between the retention structure and task performance.

Educational applications

What constitutes learning is particularly salient in the field of postsecondary education, where the concept of organizational learning is widely viewed as a productive way to think about improving education. As noted, a distinction must be made between the empirical research program on organizational learning and the managerial theory of the learning organization. This distinction is particularly significant in higher education, given that many scholars use the terms interchangeably and/or to refer to a collective of managerial strategies (e.g., total quality management and management by objectives) (Kezar [2005a]). While there are empirical studies about the nature of organizational learning in postsecondary institutions, most researchers advocate for colleges and universities to become learning organizations (e.g., Boyce [2003]; Petrides [2002]; Kezar [2005b]). A promising line of inquiry that uses the organizational learning framework focuses on issues of equity and student learning. Bensimon ([2005]) analyzes individual faculty and their ‘cognitive frames’ , which are seen as instrumental in shaping the ways in which they teach and think about student learning, all of which can ‘create the problem of unequal outcomes’ (p. 101). A core idea motivating this work is that organizational learning is not about adopting new programs or best practices per se but instead is dependent upon institutional actors changing how they think about students and learning.

In this study, we adopt an organizational learning framework to examine an activity that is common to all postsecondary institutions: course planning. The focus on course planning is important because of the centrality of curricular design to the instructional process (Lattuca and Stark [2009]; Hora and Ferrare [2013]). Course planning practices, such as creating course syllabi and selecting instructional materials and teaching activities, constitute the primary means whereby faculty design learning environments for their students. This activity is particularly amenable to analyses using the organizational learning framework in general and the construct of organizational memory in particular. This is because, during the planning process, faculty retrieve information about the curriculum, course, and topic from various sources (i.e., the organizational memory).

Besides documenting the nature of these processes, we investigate whether or not a particular project influenced the organizational memory functions among the faculty participants in our study. As noted, in situations where new information or policies are introduced, the processes of routinized practice (i.e., memory functions) can be brought into bold relief. Further, since the organizational learning framework can be used as a way to detect deep-seated changes to an organization, studying these processes in light of the USE project represents an opportunity to field test this framework as one approach to capturing project effects on a specific set of departmental functions. Thus, detecting whether or not instructional reforms such as the USE project altered these memory functions of storage and retrieval would indicate the degree to which the organization has ‘learned’ or changed in response to the initiative. Finally, insights into these issues can shed light on the processes of organizational change in general and how faculty respond to innovations and new policy initiatives in particular.

The Undergraduate Science Education (USE) project

The USE project was a department-driven change effort undertaken at the study site. The 5-year project was funded by a large private donation and led by science faculty members in collaboration with several departments. The project’s goals were to improve student learning outcomes by improving faculty members’ understanding of student learning and to encourage the use of inquiry-based teaching practices and high-quality formative and summative assessments of learning. In addition, an overarching goal of the project was to ‘change the culture’ of how research-oriented faculty were thought about and engaged in their teaching practice. To achieve these goals, the USE project collaborated directly with departments, which had committed to participate in the change effort through successful submissions as part of a competitive proposal process. The main charge to participating departments was to identify a sequence of courses to be revised, generally starting with high-enrollment, lower-division undergraduate courses, and then proceeding to upper-division courses. The revision process involved identifying and agreeing upon quantifiable learning goals for each course based on what the department wished students to be able to do at the course’s end. Central to this process was the hiring of several postdoctoral researchers who were located within each participating department. The researchers worked with faculty for two or three semesters to guide and support efforts to revise the selected courses, providing both intellectual and practical assistance. It is important to note that neither authors were directly involved in this initiative.

Sampling procedures

Data collection

All data were collected in spring 2012. The first phase involved three researchers, led by the first author. The second phase, conducted by the second author, included interviews with the four additional participants in the USE project who were identified at a later date. All researchers underwent the same training procedures, which entailed in-depth discussions about research instruments and their use in the field. For this study, we followed the critical-decision method approach by asking a focused question about a recently performed task and then asking follow-up questions to delve more deeply into the decision-making process of each respondent (see Feldon [2010]). We used a semi-structured interview protocol that elicited an extensive account of the respondent’s most recent planning activities. Following this question were probes that examined the role of the course syllabus, course type, prior experience, and knowledge of educational research in shaping the respondent’s planning process (see Appendix A). These probes were based on evidence from the literature regarding the primary factors that influence course planning (e.g., Stark [2000]). While we did not explicitly elicit information about organizational memory functions, the question resulted in rich data about respondents’ retrieval processes and the sources from which they drew while planning their classes. Then, to explore if and how teaching-related professional development (e.g., the USE project) influenced respondents’ planning procedures, we asked: ‘How, if at all, has your participation in professional development (or the USE project) influenced how you plan and teach?’ Not all respondents had participated in such projects and thus not all answered this question.

Data analysis

The research team transcribed and entered all interviews into NVivo® qualitative analysis software. Our analysis drew on a structured approach to grounded theory, which involves a combination of inductive analyses of textual data and consideration of theory that is external to the data (Corbin and Strauss [2007]). We began by segmenting the complex transcripts into more manageable units. To do this, we used an inductive, open-coding process in which new codes were created from terms, phrases, or ideas in the text. To develop the initial codes, the analysts reviewed five randomly selected transcripts and independently created a list of codes that described higher-order categories that aligned with the interview questions (e.g., steps and processes in planning and effects of professional development). The first author applied the coding scheme to all 24 transcripts; then, the analysts focused on the data coded as ‘steps and processes in planning’ and data coded as ‘effects of professional development’. This involved an open-coding process whereby the analysts independently developed new codes from approximately 10% of the coded text. Each analyst then compared each successive instance of the code to previous instances in order to confirm or alter the code and its definition (i.e., the constant comparative method) (Glaser and Strauss [1967]). After coding the initial selection of data, the analysts met to discuss their new codes and collaboratively developed a new code list. Then, the analysts coded another 10% of the data and repeated this process of code revision until they developed a final coding scheme. Examples of codes included ‘personal experiences of the course’ , ‘fine-tuning lecture notes,’ and ‘inheriting course materials’. The results of an inter-rater reliability test showed that the proportion of instances where both analysts applied the same code was 89%. The first author then applied the coding scheme to all transcripts, which resulted in an extensive NVivo® library of coded text.

Identifying characteristics of organizational memory and retrieval

At this stage of the analysis, we began to consider how organizational learning theory could inform the interpretation of the data and at which point constructs such as organizational memory and retrieval entered into the analytic process. That is, the codes (and thus, the organizational memory and retrieval functions as reported in this paper) were not originally derived in light of this theoretical framework but were identified through an inductive process. We later used the theoretical framework to interpret these findings. Using this theory as a guiding framework, we re-analyzed the themes identified as core components of planning in order to identify the primary components of the retention structure according to the framework set forth by Walsh and Ungson ([1991]). Thus, we mapped the codes onto these bins: individual, cultural, procedural, structural, artifactual, and archival. For example, we mapped the theme of ‘personal experiences of the course’ as a key factor shaping course planning onto the memory bin of ‘individual memory’ and so on, until the results indicating the nature of the organizational memory structure were identified. Then, to identify the nature of information retrieval functions, we examined specific references to the individual retrieving information from these sources in order to prepare for the course. It is important to note that we report themes that were reported by a total of four or more respondents only. Thus, with the exception of the detailed decision chain models of two instructors, more idiosyncratic findings limited to an individual or small groups are not reported in this manuscript due to space limitations.

Identifying temporal steps in planning routines

Next, we sought to identify how organizational memory and retrieval functions operated in practice. To do this, we used the verbal analysis technique of Chi ([1997]), which is a variation of protocol analysis that uses discourse-based data to identify structural features of human cognition. This is similar to causal network analysis, which is an approach for identifying relationships between concepts in a graphic and time-ordered fashion (Miles and Huberman [1994]). For this analysis, we revisited the data coded as ‘steps and processes’ and identified clearly demarcated steps in how respondents planned their courses. Not all respondents provided sufficiently detailed data, which resulted in a more limited dataset (n = 18). These data represented decision chains for each individual, comprising specific components of his or her planning activities. Overall, we identified 75 unique decision components (e.g., use lecture notes) across the entire set of 18 transcripts.

With these data in hand, we created two graphics. The first represents aggregated decision chains across seven individuals who exhibited similarities in their use of the organizational memory. In developing this graphic, we identified ‘use lecture notes’ and ‘revise lecture notes’ as the most common components reported by multiple respondents. The decision chains that included one or both of these codes were visually depicted in order to examine the structure and temporal nature of course decision-making. This approach involves graphically depicting the decision steps as a network map (see Chi and Koeske [1983] for a seminal example of this method). Importantly, the resulting displays represent the accounts of a relatively small number of respondents from the study and thus should not be extrapolated to entire departments or institutions within the study sample or viewed as definitive accounts of action and behavior within these administrative units. In addition, we included in this graphic only those behaviors clearly described by the respondent. Thus, no analyst inferences were made regarding contextual influences or other factors that may have impinged upon their behaviors.

Next, we selected three individuals from this analysis in order to delve more deeply into the sociocultural and contextual factors that influenced their planning. To do this, we focused on the three participants in the USE project who were included in the previous analysis in order to begin exploring the effects of the project on their behaviors. For this analysis, we returned to the full interview transcript and in places where the respondent discussed their planning, which often took place throughout the interview, we identified other factors that influenced their planning. These sociocultural and contextual factors were then graphically overlaid on the decision chains identified in the previous analysis.

Identifying the effects of the USE project

Finally, we analyzed the sections of the interviews coded as ‘effects of professional development’ to identify whether or not the USE project or other similar initiatives had any demonstrable impact on organizational memory or retrieval functions at this institution. Using the inductive analytic techniques described earlier (e.g., open-coding and constant comparative method), the analysts first identified instances where respondents clearly attributed a change in their course planning activities to the USE project. Thus, the analysts did not infer effects but instead relied on respondent accounts. These statements of attribution were mapped onto the bins of the retention structure used in previous analyses (e.g., individual memory and curricular artifacts).

Limitations

Limitations to the study include a small, self-selected sample and the reliance on self-reported data to analyze course planning procedures and the effects of the USE project. The issue of self-selection is difficult to avoid in field-based research such as this, and we attempted to address the possibility of collecting data only from faculty engaged in teaching-related activities by creating a split sample, though we recognize the limitations of this sampling approach. In addition, ideally planning procedures could be observed ‘live’ , but practical considerations precluded this possibility for faculty participating in the study. In addition, the sample represents only a small portion of the disciplines represented in the ‘STEM’ acronym, and thus results should not be generalized to broader populations of STEM faculty. Similarly, because this study was conducted in a large, research institution, the study population reflects a unique organizational setting. Finally, the reliance on respondent self-reports of the effects of the USE project was unavoidable, given the lack of a pre- and post-project evaluation established at the inception of the project.

The information repositories faculty drew upon when planning

First, we report the types of repositories where information pertaining to curricular and instructional issues were stored. The locations and the frequency with which they were referenced are depicted in Figure 1, below.

These data demonstrate the diversified forms in which teaching-related information are stored and that these repositories contain information about what to teach, how to teach, what works in the classroom, and what resources or materials to use.

The processes whereby faculty actually prepare for their classes

Next, we report the different processes whereby faculty actually prepare for undergraduate classes, which centered on whether respondents ‘fine-tuned’ existing materials or prepared new ones (see Figure 2, below).

It is important to note that in reporting these data, we focus on the two major actions that the faculty reported in regard to their planning and not the discrete steps taken to fulfill these tasks or the factors that influenced them. In the next section where we report faculty decision chains, we delve into these nuances of faculty practice.

Specific decision chains that faculty use when preparing for their courses

While it is important to document the nature of the memory retention structure and related retrieval processes, it is by tracing the actual flow of decision-making and information retrieval that we can arrive at an accurate portrayal of organizational memory functions in action. In this section, we present results from two different decision chain analyses of the steps that respondents went through while planning and if/how aspects of organizational memory were engaged. The first graphic reflects an attempt to model decision chains across multiple individuals (n = 7) at a more coarsely grained level, and the second graphic is a more detailed analysis of three individuals where aspects of the sociocultural and organizational context are included.

Decision chains in the aggregate

Eighteen respondents provided clear indications of the temporal process of their planning, and the most common ‘chain’ of planning steps involved two themes: looking through old lecture notes and updating these notes. These themes were reported by a total of seven respondents, either as a direct link between the two themes (4 respondents), an indirect link between the two themes (2 respondents), or only as a reference to a single theme (1 respondent). Examining these seven respondents’ planning procedures, it is possible to see the aggregate patterns in planning and the degree to which different repositories and/or retrieval processes were involved (see Figure 3).

The figure depicts the processes whereby individuals retrieved curricular information from the retention structure while planning their classes. Each decision chain (i.e., sequence of boxes) represents this process for an individual instructor as depicted by his or her code in the study sample. When the decision chain component could be clearly linked to one of the information repositories previously identified, it is outlined in thick, black lines. Within each box is one of the 75 decision steps identified in the analysis of the entire sample.

For example, respondent C01 (a biologist teaching a genetics course) reported the following steps in her planning: ensuring materials were in place, looking through old notes, and then making sure class activities were ready. Respondent C11 (a chemist teaching a general chemistry course) reported a more complex sequence of steps but shared the common theme of looking through old notes. Indeed, each of the seven respondents depicted retrieved curricular information that was stored in the curricular artifacts repository—lecture notes. These artifacts were either the result of prior experience teaching the course or were inherited from prior instructors. As a result, this process implicates two other aspects of the retention structure: individual memory (personal experience of the course) and social networks (experienced colleagues)^f.

These results highlight the importance of preexisting curricular artifacts in shaping how some faculty in the study sample plan their courses, but that this process of artifact adoption is often not one of simply ‘plugging and playing’ old notes with no adaptations. As one biology faculty described this process:

“Since this is my first semester teaching I took many colleagues’ advice, which is to start with somebody else’s notes. So I start with those, and then I take a look at the homework assignment, look at the book and see what the students will be asked to do, and compare that to my notes. At that point I revise, maybe introduce new examples or replace examples. And then it becomes a question of how am I going to present that content? And to personalize the teaching for myself, hoping that the way I understand it offers an alternative way for them to see it as well.”

This quote illustrates how a single instructor fine-tunes existing notes by retrieving information from existing notes to update them while also drawing upon other resources, such as the textbook and beliefs about student learning. Importantly, as discussed above, all seven respondents depicted in Figure 3 referenced the core process of looking through notes to fine-tune them as a sort of habituated or automatic practice. This behavior was described by many, especially those who had taught for several years, as part of their regular routine for preparing a course or individual classes. Finally, it is worth noting that for two respondents (C16 and C18), the final stage of their planning process included altering their lecture notes immediately after the class and adding new observations about what worked or did not work, which reflects an effort to update these artifacts based on reflection about the efficacy of their own teaching practices.

Individual-level decision chains that include contextual factors

Next, we provide a more in-depth analysis of three individual’s decision chains (see Figure 4).

In this graphic an assortment of personal, sociocultural, and contextual factors are included to provide some background to each decision chain. For example, the relatively simple decision-making process of respondent C01 becomes more nuanced when her other considerations are included (e.g., time constraints and availability of local professional development). While the three individuals depicted here are certainly not representative of their departments or institution, it is notable that each reported past experiences teaching the course, local professional development programs, and cultural norms of peers as salient influences. When referring to cultural norms, each of the respondents identified two groups: one group of faculty supportive of pedagogical reform and one group that were either unsupportive or uninvolved in such activities. Each respondent noted that the presence of a supportive group was an important factor in their decision-making by acting as a source of information as well as a support system.

It is important to note that each individual underscored the fact that no single factor influenced or determined their planning behaviors; rather, a complex web of factors influenced their behaviors. This fact highlights the nonlinear and complex nature of decision-making processes, as well as challenges inherent in discerning meaningful implications from such data. Finally, it is also notable that no respondent described organizational procedures or policies that governed the existence of continuous improvement mechanisms that would require faculty to reflect upon their own teaching. The existence of such systems is a key facet of facilitating organizational change and learning (Walsh and Ungson [1991], Kezar [2005a]). The only evidence of such mechanisms in place were at the micro-level, where individual faculty took it upon themselves to engage in reflective practice at the conclusion of each class period, making notes on their curricular artifacts about what went wrong and what worked.

Effects of the USE project on faculty engagement with organizational memory

Finally, we consider the influences of the USE project on the elements of the retention structure and subsequent information retrieval processes. By examining organizational memory functions in places where reforms are being attempted, we can illuminate local practices and procedures that influence the success of these initiatives, and we can track changes in the organizational memory as a way to assess the depth and sustainability of reform efforts. Of course, definitive claims about the effects of the initiative on faculty behaviors cannot be made here, given the nature of the data. That said, we suggest that effects based on respondent self-reports are useful in highlighting whether and how STEM education reforms are targeting organizational memory functions (see Figure 5 below).

An approach for understanding instructional decision-making in postsecondary settings

The approach described in this paper for studying faculty practice moves beyond a singular focus on individual decision-making and the role of isolated factors that shape practice to a more comprehensive accounting of organizational functions and the role of information storage and retrieval. A particular contribution of this study is the method of identifying interconnected factors that comprise decision-making in real-world settings, or what we call ‘decision chains’. It is through these chains that it becomes possible to account for the influence of multiple levels of organizational functions (i.e., micro, meso, and macro) without privileging any single level in describing how organizations ‘learn’ and change. That is, by focusing on individual-level decision-making as it occurs within the sociocultural and organizational contexts of postsecondary institutions, a more realistic and nuanced account of faculty work that accounts for the complex intersection of disciplinary, professional, and organizational influences becomes possible (Austin [1996]; Umbach [2007]). Future research in this area should explore the intersections between this perspective on organizational learning and integrative theories of practice that are widely recognized in fields such as the learning sciences, such as activity theory (Leontiev [1978]), situated and embodied cognition theory (Lave [1988]; Wilson [2002]), and distributed approaches to educational practice (Spillane et al. [2001]).

Insights into organizational learning and change processes among science and math faculty

One of the most important findings from this study is that curricular information is not stored in a single location or format. Instead, information about what and how to teach is stored in formalized systems and artifactual or physical forms (i.e., the ‘hard’ memory) as well as in informal systems and individual and/or sociocultural forms (i.e., the ‘soft’ memory). While planning for a course or a class, individuals will likely draw upon both sources of information. For example, a chemist in our study (respondent C11) drew upon both soft (i.e., memories of previous class topics and successes) and hard (i.e., old notes and topics in the syllabus) repositories of information when planning a class.

As a result, an exclusive focus on hard memory sources, such as lecture notes or course syllabi, which is possible given the ease with which hard memory sources can be identified, will overlook a significant portion of the information available to faculty as they plan their courses. That is, less tangible soft memory sources, such as cultural norms, also play a considerable role in faculty decision-making. Even entrenched beliefs and cultural norms must interact with the technical and structural features of departments and institutions, such that the hard and the soft features of organizational life and information co-exist in an interactive and even recursive dynamic.

While a variety of information repositories are accessed by faculty in practice, it is important to highlight the finding regarding the centrality of curricular artifacts in the information retrieval process. That is, the regular pattern of retrieving old lecture notes and subjecting them to fine-tuning or ‘tweaking’ is a dominant mode of planning practice. This highlights two critical features of organizational learning: the importance of artifacts and the habituated nature of many organizational routines.

First, curricular artifacts play a considerable role in dictating how faculty approach course planning (Hora and Ferrare [2013]). This suggests that once an instructor creates lecture notes, PowerPoint slides, and clicker questions, he or she is highly likely to return to them time and again. Retrieval of preexisting artifacts is understandable given the limited time that the faculty members have to prepare courses. While it is beneficial for well-designed materials to have an enduring impact on an individual’s teaching and subsequent experiences for students, poorly created materials will have a similarly long-lasting, though this time negative, impact. The role of artifacts in mediating action within complex social environments is also an area that future researchers should examine (Wertsch [1991]).

Second, the tacit and habituated nature in which curricular information is retrieved is an important factor to consider in relation to organizational learning. While the data do not provide evidence that the retrieval of information from lecture notes was automatic, the respondents did describe it as recurring and habitual. One respondent characterized the retrieval and refinement of lecture notes as ‘just something that I’ve always done’ , which shows that this common practice is a deeply ingrained and non-problematic behavior for some individuals. This finding highlights the role of habit, which can supersede or override controlled or conscious decision-making processes (Bargh and Williams [2006]). Over time, these can become internalized as unconscious decision rules or heuristics, such that, when particular situations are encountered (e.g., being unexpectedly assigned a new course), habit may dictate individual behavior (Greeno [1998]). This may be particularly true in academic settings, where time constraints on faculty press individuals to constantly seek ways to make their work more efficient. In research universities, this is particularly the case with teaching, which generally takes a backseat to research. As one respondent stated, having course materials ‘in the can’ that could be easily retrieved each semester made her teaching much easier.

Finally, we suggest that the results hint at how culture operates in postsecondary settings. In doing so, we take a broad view of culture that encompasses a variety of phenomena including physical artifacts, recurrent practices and behaviors, and tacit beliefs or understandings. Importantly, these manifestations of culture do not operate in a homogenous fashion, influencing all faculty within a department in the same manner. Instead, different configurations of these elements can have different results for different faculty. That said, the centrality of curricular artifacts in cultural practice in general, and among the faculty in this study in particular, underscores their importance and influence. In any case, we suggest that by tracing the decision chains of faculty as they interact with organizational memory functions, it is possible to illuminate certain aspects of local cultural practice.

Implications for educational improvement at the undergraduate level

Besides contributing to the theory of organizational learning in higher education, the approach described in this paper also has implications for educational improvement efforts. Specifically, we suggest that this approach can be used to complement existing metrics for program evaluation and to diagnose leverage points for new change efforts.

Intro script

Thank you very much for taking the time to talk with me today and for letting me observe your class. As the email mentioned, we are studying course planning and classroom teaching at the undergraduate level in your field, and the interview will focus on how you went about planning the class that I’ll be observing this week.

Before we start I have some consent forms for you to sign—one for the interview and the other for the observation. Thanks. And is it OK if I record this conversation?

Grand tour/warm-Up

Q1. Can you please briefly tell me about your background and current position?

Planning freelist

Q2. Please list all of the things that come to mind that influence instructors as they plan a specific lecture or class at the undergraduate level, using only single words or phrases. For example, the things that influence how a tomato grows could be sunlight, good soil, fertile seeds, and so on.

Planning for the observed class—critical decision method

Q3. Now I have a few questions about your planning process for the class I’ll be observing on [fill in day]. I’d like you to think about how you went about preparing for this class, starting as far back as you can in the planning process, and walk me through the exact steps that you went through, paying particular attention to specific points at which you made important decisions about things like course content and/or teaching strategies-activities.

(If not volunteered, probe for when exactly the planning took place).

[Take detailed notes of the process being laid out, and identify 1–2 decision point that addresses the selection of a particular teaching method or strategy. These points will be the foundation for subsequent probes. If respondent answers probes below for the 1st teaching strategy-activity rapidly, then ask the same probes for the 2nd teaching strategy-activity].

Now I’d like to ask a few follow-up questions about specific aspects of your planning process.

Q4. Did considerations of departmental or colleagues expectations for your work factor into this decision, and if so, how?

Q5. Did considerations of the type of students in the course factor into the decision?

Q6. Did considerations of the course syllabi factor into the decision?

Q7. When you decided to [name teaching strategy-activity] were you drawing on any prior experiences when making this decision?

Q8. When you decided to [name teaching strategy-activity] were you drawing on any particular knowledge base about teaching or learning when making this decision?

Q9. When you decided to [name teaching strategy-activity] did you have any specific goals for students at the time, and if so, what were they?

How do you know if you’ve achieved these goals?

[Repeat for 1 more teaching strategy-activity if time; e.g., if only 13–15 minutes have elapsed].

Focus on cultural forces

Q10. How, if at all, does the culture of your department and/or institution influence how you plan for this class—especially what you teach and how in this class?

[IF interview is moving quickly, probe for any interesting responses here].

Q11. How, if at all, does your particular position (as lecturer, tenure-track faculty, etc.) influence how you plan for this class?

Plan format

Q12. What is the physical format of your plan? (e.g., paper, PowerPoints, etc.). [at this point, if plan is in hard-copy format, ask if you can take a picture or otherwise get a visual/copy of the artifact; if slides, then ask for copy of slides to be emailed to you].

Professional development

Q13. Have you participated in any teacher training or professional development in the past 2 years? If so, why? If not, why not?

Q14. How, if at all, has your participation in professional development (or the USE project) influenced how you plan and teach?

Q15. What, if any, aspects of your department or institution have supported (or discouraged) changes to your teaching?

Q16. What, if any, aspects of your department or institution have supported (or discouraged) changes to teaching in general in your department?

Do you have any additional points you’d like to make about any of the questions I’ve asked?

Thank you very much for your time.