Summer Session Organizational Models at Canadian Universities

The issue of summer session organizational models continues to be of interest to summer session deans/directors and university administrators. The University of Victoria surveyed Canadian universities on this issue in 1994. Based on a similar survey done in 2009, this paper updates the status of Canadian university summer session organizational models, and looks at changes that have occurred over the last decade and a half. It appears that the predominance of more centralized models of summer session administration at Canadian universities has continued. Nevertheless, closer examination suggests that the predominant model is actually a hybrid model wherein responsibilities are vested in a centralized summer session unit with selective responsibilities devolved to faculties and/or departments.


Organizational Models
In 1994, the University of Victoria surveyed Canadian universities on the issue of how those institutions organized their summer sessions. The resulting report (University of Victoria, 1994), explored the extent of centralization or decentralization of summer sessions. Similar to the definitions in the University of Victoria study (1994), the concepts of centralization and decentralization are typically framed in terms of decision making control: • Centralized-a model in which the control for decisions related to the planning and administration of summer session, including budget (and payroll for instructors and assistants), advertising (including calendar production), course coordination (planning, scheduling, and room bookings) reside with a specific person or department (dean/director of summer session) for the entire university. Typically, in this model faculties/departments provide or approve instructors, deal with academic matters, and participate in course planning.
• Decentralized-a model where the control of decision making resides in individual faculties or departments of the university with no person designated as dean or director of summer session. Typically, this means a transfer of decision-making authority and responsibility for decisions related to summer session to the faculties/departments; often referred to as devolution.
Ultimately, the degree of centralization/decentralization provides an organization with certain benefits. A more centralized structure will result in greater control over decision making, greater unity of purpose and consistency of action, potentially increased cost efficiencies, maximization of expertise, and reduction of inequities. On the other hand, a more decentralized structure provides greater discretion and autonomy to units, increased responsiveness and flexibility, and customization to "local" needs (Coggburn, 2005;Fleurke & Hulst, 2006;Goddard & Mannion, 2006;Heikel, 2000;Ho, 2006;Hutchcroft, 2001;Iwe, 2006;Rickards, 2007). Centralization, according to Richards (2007), "facilitates an enterprise approach to the provision of . . . infrastructure, systems, and services rather than a fragmented, localized approach. It can deliver benefits not only in terms of technology, people, and processes but also in terms of services to students and staff" (p. 4). In fact, Heikel (2000) found that centralized summer session models were more financially successful and more successful in meeting student needs than summer sessions that operated under decentralized models. This occurred, in part, because with a centralized model, the summer session program had a "big picture" focus that tended to serve a wider range of students across the university. In Heikel's view, decentralized models run the risk of the units "becoming overly parochial by focusing on the needs of their own immediate students and not offering courses and programs of interest or need to students in other majors or programs" (p. 40).
In the literature discussing centralization/decentralization generally, rather than summer sessions in particular, Peckham, Exworthy, Powell, and Greener (2008) summarized decentralization as "the process of throwing off the shackles and constraints that are perceived to be part of working within the public sector and achieving greater responsiveness, with managers taking greater responsibility. It is argued that decentralization, with devolved power, provides local agencies [units] and managers with the autonomy to act and manage [with resulting improvement in service]." (p. 560) Challenges for each model are summarized by Coggburn (2005) as follows, "Centralized models can become rigid, complex, and unresponsive. Decentralized models can produce a loss of consistency in approach, a loss of equity, and loss of control, along with limitations in the expertise available within individual units" (p. 425).
Several authors, including Mintzberg (1979), warned against taking too simplistic a view when examining the values and challenges of centralized versus decentralized organizational models. One approach, proposed by Acemoglu, Aghion, Lelarge, Reenen, and Zilibotti (2007) to come to terms with the issue, is to think of a trade-off between the consequences of too much control (centralization) and the negative effects of delegated decision making (decentralization). Assuming that polar extremes do not exist allows the concept to be thought of in terms of a continuum, where centralization and decentralization are relative and an organization is either more or less centralized or more or less decentralized (Hutchcroft, 2001). The key in determining where an organization fits on the continuum appears to be avoiding a one size fits all approach and considering instead the outcomes that the organization wants to achieve (Goddard & Mannion, 2006;Hutchcroft, 2001;Richardson, Vandenberg, Blum, & Roman, 2002;Waggener, 2007), and the context of the organization (Coggburn, 2005;Richardson et al, 2002).
With respect to summer sessions, what does a university want to achieve in providing students with opportunities to meet their learning needs? Posing this question is in keeping with management historian Alfred Chandler's (1962) claim that structure follows strategy. Furthermore, what is the organizational context-both the internal and external-of the university? Internal factors including levels of expertise, available resources, interests, and motivations are critical for an institution to achieve its desired outcomes, and external factors such as diversification of services/products, organizational size, and the predictability/stability of the organization's environment impact the appropriateness of organizational structure (Acemoglu et al, 2007). For example, in a study of British and French companies, Acemoglu et al (2007) identified contextual factors that affected the degree of centralization/decentralization. Greater decentralization occurred in firms closer to the technological frontier (dealing with new technologies), firms working in more heterogeneous environments (varied experience), and young firms (limited histories). Making the "right" decision about the degree of centralization/decentralization requires consideration of this plethora of contextual factors and desired outcomes. Fleurke and Hulst (2006) discuss this issue in terms of a contingency approach in which consideration is given to situational factors rather than assuming a general approach to centralization/decentralization. In other words, consideration needs to be given to organizational context (internal and external) when making decisions about the appropriate structure. As well, if the organizational model is to function effectively, consideration needs to be given to the relationship between the centralized unit and the other units within the organization to which authority has been delegated to ensure that role expectations are clear, and the effort-to-reward ratio is perceived as fair (Eisenhardt, 1989). Katz (2007) suggested that it is not a question of centralization or decentralization but, rather, a question of centralization and decentralization (p. 19). He went on to say, "the glue (in our world) that will make it possible for us to create rich hybrid services that balance the institution's need to innovate with its need to economize and account for outcomes will be . . . governance" (p. 19). Others, such as Iwe (2006), agreed that organizations need to determine the best way to centralize and decentralize services, functions, and administrative tasks to gain maximum benefit to the organization and its clients. According to Piper (1996), there is no correct organizational structure for administering summer sessions in universities, which, in his opinion, is why the discussion about the appropriateness of a centralized versus a decentralized administrative model continues. He claims that in practice there are no fully centralized or decentralized models. Young and McDougall (1991) reported that organizational structures for summer sessions "range from a total high degree of centralization, a highly centralized structure for part of the programs and activities surrounded by a host of decentralized parts, to a loosely  Heikel (2000) recommended that summer sessions should be administratively centralized and programmatically decentralized in order to maximize the benefits of both organizational forms (p. 39).

A Hybrid Approach
The idea of hybrid organizational structures (an appropriate blend of centralization and decentralization) for the administration of summer session makes sense: the institution then needs to determine what functions need to be centralized and what functions need to be decentralized in order to create a model that best achieves desired outcomes. Mintzberg (1979) talked about this as selective decentralization where the power for different decisions rests in different places in the organization. He added that, along with selective decentralization, efforts need to be made to deal with the interdependencies of the units in terms of coordination and control, which Mintzberg claimed, are achieved by the use of liaison devices (p. 187) such as liaison positions, task forces, committees, and matrix structures (p. 175). The use and effectiveness of such devices depends on the specialization, complexity, and differentiation of the work being done, as well as the level at which the work is done in the organization (p. 178, 179).
The majority of Canadian universities appear to be using hybrid organizational models for summer sessions, described in survey data (in both 1994 and 2009) as models of shared responsibilities. Even in universities reporting a centralized model, decision making is shared with faculties and departments. Administrative decisions, including program planning, scheduling, promotion, and budgeting are centralized, with academic decisions decentralized: coordination and control are provided by the summer session units headed by a dean or director. These units operate from a university-wide perspective and act as a liaison device.

The Pendulum Swing-Factors Influencing Decisions to Change Organizational Models
Regardless of the popularity of and support for one model over the other, from time to time decisions are made to change summer session administration models. In other words, the pendulum swings from centralization to decentralization and back again (Evaristo, Desouza & Hollister, 2005;Axelsson, 2000). Each change is made to remedy the problems of the existing structure, but often only creates a set of challenges that are a consequence of the new structure (Coggburn, 2005).
The rationale for the pendulum swing is sometimes unclear. Axelsson (2000) suggested the forces behind the oscillation between centralization and decentralization for over a century in the Swedish healthcare system were technological advancements, economic conditions, new management approaches, consumer/public pressure, and changes in political climate. Evaristo, Desouza, and Hollister (2005) also addressed the notion of pendulum swing by looking at the trend by organizations to recentralize IT hardware architecture. They identified the primary factors that impacted changes in centralization/decentralization over the last 50 years, including economics (telecommunications and hardware costs), new technology, pressure to improve responsiveness and flexibility, a need for data integration and simplification, security concerns, and the e-commerce movement. Kops (1998) looked at factors that influenced decisions to change summer session organizational models, based on case studies of six universities that made such changes. The factors were categorized as external to the university, internal to the university, and specific to summer session. The key factors that were drivers in virtually all cases were actions by provincial/state governments on funding (and subsequent budget decisions by the university), changes in the university's senior administration (coupled with a review of strategic plans and mission), and several factors particular to summer session. Specifically, the relationship of summer session to the university's mission and goals, the performance/style of the summer session dean/director in terms of their ability to relate to faculty deans and department heads, and issues of sharing summer session revenues were drivers that affected organizational change.
An important theme in making organizational change is to be aware of the drivers of the change, as well as the expected outcomes of a new organizational model. Axelsson (2000) advocates assessing and evaluating models before they are implemented so as to counteract the pressure to make change for the sake of change.
What is the situation with respect to summer session organizational models at Canadian universities, and have models changed? The University of Victoria's 1994 study on the question of centralization revealed most summer session operated under a centralized or hybrid model. The purpose of this study is to determine whether the situation has changed over the last decade and a half, and, if significant changes have occurred, what factors have influenced decisions for organizational change.

Survey of Canadian Universities
Centralized models are the most common organizational form for summer sessions, which is evidenced by the University of Victoria survey (1994) that found most (87%) summer sessions at Canadian universities operated under a centralized or combined centralized model. Essentially the same picture emerged in the 2009 study with almost 80% of the summer session programs structured under a centralized or alternative/centralized model. Interestingly, using similar definitions, a study done by Heikel (2000) of 144 public institutions in the United States found that 84% were classified as centralized or hybrid models and 16% as decentralized.

Methodology of the 2009 Survey
Survey methodology was used to collect the data. The survey instrument, including definitions, replicated the survey used in the 1994 University of Victoria study in order to collect data that would allow comparison of the status of summer session organizational models between 1994 and 2009. The research project and survey instrument were approved by the Education and Nursing Research Ethics Board of the University of Manitoba. The survey was sent to all universities that were members of the Canadian Association of University Continuing Education in 2009, and included all but one of the universities in the University of Victoria study (total of 40 universities). Specifically, the survey was directed to individuals responsible for summer sessions or degree studies within the universities' continuing education units. The survey was distributed in both electronic and print form, including an introductory letter and consent form. Reminders were sent to participants by email at intervals following the initial distribution of the survey. In total, 28 universities responded for a response rate of 70%. Of the 22 universities that responded to the 1994 survey (and were included in the 2009 survey), 16 responded. Twenty-one respondents completed the survey in full, while seven provided only partial responses: where there was no answer to a question, a "no response" was recorded. The data were collected and analyzed to present the following descriptive analysis.

Summer Session Models
In this study, 18 institutions reported a centralized or an alternative model with key functions centralized (an alternative model was not identified as centralized or decentralized as defined earlier, but did have a central unit responsible for several key responsibilities). In terms of those with a summer session program (not including trimester), this represented 79% of the study's participants. Five institutions reported a decentralized model or alternative model with key functions decentralized: this represented 22% of institutions with a summer program. Two institutions reported operating on a trimester, and three institutions reported offering no credit summer session programming.
Compared to the 1994 University of Victoria survey where 20 universities reported centralized summer session models (87% of the total respondents), two reported decentralized models (9%), and one operated on a trimester, in 2009, the proportion of Canadian universities with a centralized summer session remained at about 80%.
Jurisdiction refers to the reporting relationship of the summer session program-to what unit or person does the summer session program report. Based on the 2009 survey results, 14 of the 18 centralized units reported through continuing education or an equivalent unit, three reported to the Registrar's Office, and one reported to the Dean's Office. Centralized summer session programs reporting to continuing education units declined slightly from 85% in 1994, to 78% in 2009. In the case of three universities that maintained centralized models (Alberta, Saskatchewan, and St. Mary's), the reporting relationship changed from continuing education to the Registrar's Office, and at Mount Saint Vincent University, the shift was from the Office of Academic Advising to the Dean's Office.

Longevity of Models
While in 1994 almost all universities (95%) reported that the current model of summer session administration had been in place for quite some time, it appears that changes have occurred more frequently since then with 78% of responding institutions in 2009 indicating a longstanding model. Based on the 2009 data (not including trimester), summer session organizational models at 14 universities have been in place for 8 years or more (with almost 80% in place for over 15 years), and at four institutions the model has been in place for less than 8 years. Five institutions did not respond to the question.

Revenue Sharing
Revenue sharing between the summer session program and participating faculties has become more popular in recent years. The purpose of revenue sharing is to create a financial incentive for participation of faculties in summer session. In the University of Victoria study (1994), only three universities indicated that revenue was shared-in most cases the revenue went to the university (central administration). In the 2009 survey, revenue was shared at five of the institutions with a summer session program (not including trimester). Thirteen respondents indicated that all revenue went to a single unit-five that revenue was retained by the summer session program, six that all revenue went to the university (central administration), and two that all revenue went to the faculties and/or departments. Five institutions did not answer the question.

Responsibilities for Summer Session
As discussed earlier, there are no polar extremes, which suggests that centralization and decentralization are relative-i.e., an organization is either more or less centralized or more or less decentralized. This appears to the case for summer sessions at universities generally (Young and McDougall, 1991), and specifically in Canadian universities where the majority of summer sessions are more centralized. The centralized models identified in the 2009 survey are closer to what is described as a hybrid model where selected functions and responsibilities are decentralized. As outlined in Tables 2 and 3, the responsibilities that are typically devolved to the faculties/departments are directly related to academic matters, including instructor selection and approval, course approval, and matters such as student academic appeals. A range of other functions are carried out centrally by the summer session program. The survey did not poll universities on the basis of selective delegation, but since most universities have had the current summer session model in place for some time, it could be assumed that delegation of selective responsibilities was done advisedly in order to maximize desired outcomes for students and the university. The result is that academic units deal with academic matters, and attention is paid to organization-wide issues and programs by specialized staff in a centralized summer session unit. In addition to organization-wide issues such as budgeting, marketing, special programming (e.g., travel/study programs, summer institutes), enrolment management, and comprehensive program planning, the summer session units serve as a liaison device providing coordination and control of what happens across the institution and on behalf of the institution, including monitoring performance measures such as enrolment and registration, retention, and revenues and expenses.

Frequency of Mention
Instructor selection/approval/appointment/evaluation 15 Course selection/approval/cancellation 11 Academic matters, including academic appeals 4 Other-ordering texts, course expense decisions 1

Benefits to the University
According to the organizational literature, the advantages of a centralized model are greater control over decision making, greater unity of purpose and consistency of action, potential increased cost efficiencies, maximization of expertise, and reduction of inequities. For the most part, these are manifested in the advantages mentioned by respondents to the 2009 survey as outlined in Table 4. On the other hand, a decentralized approach provides more discretion and autonomy to units, greater responsiveness and flexibility, and customization to "local" needs. The respondents identified benefits that were more specific to operating efficiencies and curriculum planning (see Table 5). Again, the predominant centralized/hybrid model leverages many of the benefits of the two approaches by allowing faculties and departments discretion in academic decisions of course selection, instructor recruitment and selection, and specific academic matters, while a centralized summer session unit provides the advantages of unity of purpose for summer session across the institution, attention to comprehensive program development, utilization of expertise, cost efficiencies, consistency and equity of image and policy application, and a single point of contact and representation for summer session. Tensions may exist in this arrangement, but are less likely if the roles of the faculties/departments and the centralized summer session unit are clearly understood, and the outcomes of service to students and shared revenue are accepted.

Frequency of Mention
Increased efficiencies/cost savings 2 Regularize summer as part of year-round curriculum 1 Increased utilization of faculty in summer 1 Final Comments It appears that the predominance of more centralized models of summer session administration at Canadian universities has continued over the last decade and a half. However, closer examination suggests that the predominant model is actually a hybrid model that has selective responsibilities devolved to faculties and/or departments. This is appropriate because these devolved responsibilities are academic in nature, where the faculties and departments have both the mandate and expertise to make decisions on programs and staff. A range of other responsibilities and functions, including budgeting, marketing, special programming, enrolment management, comprehensive program planning, and performance-monitoring measures are handled by a centralized summer session unit that offers expertise, economies of scale, and an institution-wide perspective. A hybrid structure with selective delegation provides an opportunity to maximize benefits of both centralized and decentralized approaches, and to play on the strengths of both a centralized summer session unit and the academic faculties and departments. Making the "right" decision about the degree of centralization and decentralization requires consideration of a multitude of contextual factors and desired outcomes. One of the important considerations in making decisions about organizational structure is to determine the desired outcomes of the organization (Waggener, 2007;Goddard and Mannion, 2006;Richardson et al, 2002;Hutchcroft, 2001). What are the desired outcomes of Canadian universities in terms of summer session? If the desired outcomes are a robust, comprehensive, university-wide program to serve a varied range of students, a more centralized structure that provides greater unity of purpose and consistency of action, potential cost efficiencies, maximization of expertise, greater control over decision making, and reduction of inequities may be the right one. However, a desired set of outcomes that focuses on year-round programming of faculties and departments to serve their own students may be better achieved by a more decentralized structure that provides more discretion and autonomy to units, and customization to local faculty/student requirements. Needless to say, an examination of desired outcomes is an important step in any decision to structure (or restructure).
Another key consideration in making decisions about structuring organizations is organizational context (Coggburn, 2005;Richardson et al, 2002). Katz (2007) stated that universities deal with two opposing pressures-the pressure to innovate and the pressure to economize (p. 18). Further, he suggested that the accepted notion is that centralization is best where efficiency (i.e., economies of scale) is important and decentralization is most sensible where innovation is key.
What is the context of Canadian universities, specifically with respect to summer sessions? One expects, as Katz (2007) suggested, that Canadian universities operate in an environment where both innovation and efficiency are realities. There is a need to be conscious of both when creating organizational structures because the question of what is right is complex and simple solutions will not be successful (Katz, 2007;Hutchcroft, 2001;Richardson et al, 2002). Consideration needs to be given to a complex of factors, including internal factors of expertise, available resources, interests and motivations, and external factors such as diversification of services, organizational size, and predictability/stability of the organization's environment. All are critical for the institution to achieve its desired outcomes.
Whether by design or as a result of successful experience over time, more centralized organizational models continue to be the most popular approach for summer session administration at Canadian universities. Changes in structure seem to be a bit more frequent in recent years as indicated by the relatively short time that some summer session models have been in place. This suggests a pendulum swing from more centralized to more decentralized structures or from more decentralized to more centralized structures. As a pendulum swing gains momentum, it is important to be aware of the drivers pressuring for change, and careful consideration needs to given to the complex task of formulating organizational change. Katz (2007) stated that, "the job is to keep making the point that we need to do both [strive for efficiency and innovation] and keep building not only the infrastructure but also the trusting environment that will allow us to create the best blend of centralization and decentralization" (p. 20). For example, a perception often exists that there is a direct connection between organizational models and distribution of revenue or revenue sharing. Specifically, the belief is that a centralized model would have all revenues retained by a central unit and a decentralized model would result in all revenues distributed to individual units (faculties/departments). One of the drivers of organizational change is pressure from deans for a greater share of summer session revenue (Kops, 1998). Hutchcroft (2001) addresses this type of situation by stating that, "where local bosses exert a great deal of coercive and socioeconomic power . . . strategies of devolution need to be approached with extreme caution" (p. 46). There is no reason to believe that there is a direct relationship between summer session organizational structures and revenue sharing arrangements. Accepting such disconnect reduces the impact of this driver to change summer session administrative models, and it also reduces tension that may develop between faculties/departments and the centralized summer session unit. This does not imply that revenue sharing arrangements cannot be changed or modified, but simply that these decisions need to be made separately from decisions about organizational structure. As with other drivers for organizational change, it is advisable to ensure that the change is based on a careful examination of multiple factors and desired outcomes in order to ensure that the revised organizational structure results in more benefits than challenges to the organization (Axelsson, 2000).
Another driver of change to summer session organizational models (and likely other organizational arrangements) at universities is actions by provincial governments on funding and subsequent budget decisions by the university. One can assume that the pressure for change is likely to continue in the future as governments and universities struggle with decreasing revenues. It will be interesting to update the status of summer session organizational models in another decade to determine whether changes have occurred and, if they have occurred, what have been the drivers that compelled the organizational changes.