1.1 Administrative roles in higher education

Institutions of higher education within the United States may at first glance seem to be built structurally the same; however, ‘in fact no two have been found to be exactly alike’ (Powers & Schloss, 2017, p. 27). Despite this fact, the administrative roles within U.S. higher education remain consistent across all institutions though they may vary depending on institutional size, scope and history (Johnson Bowles, 2022). However, most U.S. higher education administrative structures or chains of command can be categorised into five distinct levels (1) the governing authority (i.e., state governor and system board of trustees and/or regents); (2) system senior administration (i.e., system president or chancellor and system vice presidents or chancellors); (3) campus governing authority (i.e., institutional governing board of trustees/regents/visitors or college council); (4) campus senior administration/upper management (i.e., institutional president or chancellor, divisional leaders and chief officers); and (5) campus administration/middle management or mid-level administrators (i.e., associate or assistant vice presidents/chancellors, deans, assistant or associate deans and department chairs and directors) (Johnson Bowles, 2022).

The campus senior administrative and administrative level roles hold the largest amount of power related to decision making and creating policies and procedures within their institutions (Powers & Schloss, 2017). Their power lies in the ability to make executive decisions that create change. Middle management or mid-level administrators are empowered by campus senior administrators to implement policy and practice to enact change. They are also responsible for the oversight of their subordinates—faculty and staff—(Johnsrud et al., 2000; Rosser, 2000) and have the power to make decisions related to day-to-day departmental needs. Rosser (2004) noted that mid-level administrators play a pivotal role within institutions, but that ‘by virtue of their “middleness” within academic organizations, mid-level leaders must find balance between superiors’ directions and the needs of those who require their support and service’ (p. 319). Notably, their power lies in their ability to communicate needs between their subordinates and senior administrators.

1.2 Defining online programmes and needed support

Defining online programmes can be complex due to the various definitions and terminology found in the literature. In some cases, the terms conflict with one another (Singh & Thurman, 2019) or there is inconsistent use of terminology (Moore et al., 2011). For this reason, we define online programmes as those in which the majority of coursework is provided in an asynchronous format allowing students, who so desire, to complete their degree without the requirement of ever meeting in a physical classroom. Therefore, online programmes require a great deal of student and faculty support (Earnshaw & Bodine Al-Sharif, 2023).

Offering online student support is challenging from a U.S. perspective due to the differing structures and governance of academic divisions versus student support services. For some institutions, online programmes and student support services are housed within academic departments and some are under a global or world online campus (Lieberman, 2017). At some institutions, student support services are centralised, while online programming is decentralised and vice versa. This matters because the misalignment between student support services and online programmes can be a roadblock to collaboration and understanding of online student and programme needs.

Stewart et al. (2013) identified several areas at the university level to support students: ‘admissions and registration, advising, orientation [to the university and to online learning], academic support, scholarships and awards, library resources, computing and technology resources, articulation with other institutions, career placement and communication’ (p. 290). However, most online student support is at the course level (Rotar, 2022). At the course level, institutions require technical infrastructure for courses to run effectively and efficiently. This includes having privacy and data security (Hirumi et al., 2021), updated software and hardware for design, development and delivery (Dousay & Hall, 2021; Martin & Kumar, 2021), as well as the appropriate staff to support both faculty and students.

In addition, faculty support should align in the areas of training, financial resources, needed equipment and software packages to meet their teaching, learning, research and assessment needs (Shelton & Pederson, 2017). Providing online faculty support is more than addressing the needs of the classroom such as learner-centred approaches to teaching (Weimer, 2013), but also includes supporting faculty as professionals by developing communities of practice (Lave & Wenger, 1998), creating work-life balance (Denson & Szelényi, 2022), providing non−/tenured faculty mentoring programmes (Ng et al., 2021) and addressing issues of inclusivity (Xu, 2012) for the professoriate. This echoes the work of prior research (Gregory & Salmon, 2013; Vaill & Testori, 2012) that has also identified this as a weakness in online faculty support.

1.3 Current assessment for online programmes

Administrators currently assess online programming through a variety of methods, including, but not limited to, student evaluations and quality assurance frameworks (Piña, 2021; Piña & Bohn, 2014; Tobin et al., 2015). This is often driven by legislation and the need to meet the requirements of accrediting agencies (Higher Education Act of 1965, 2022; U.S. Department of Education, 2023). Student evaluations of the instructor (Piña & Bohn, 2014) are the most commonly used assessment tool, but they may not be customised for teaching online (Martin & Kumar, 2021; Miller & Digges-Elliott, 2021) and faculty may receive lower ratings for online courses than face-to-face ones (Marzano & Allen, 2016). Quality assurance frameworks such as Quality Matters (2018) or the OLC OSCQR Course Design Review Scorecard (Online Learning Consortium, 2018) focus heavily on the instructional design of the course (Tobin et al., 2015) and not on the instructor. Since some instructors may not even be responsible for the design and development of their courses (Piña & Bohn, 2014), these methods do not provide adequate information to make informed decisions about both the quality of online faculty (Piña & Bohn, 2014) and the online course development process (Piña, 2021).

1.4 Theoretical framework

For this work, we have selected two theoretical frameworks: role identity theory (Sluss et al., 2011) and psychological empowerment (Quinn & Spreitzer, 1997; Spreitzer, 1995). Role identity theory examines the ways in which people define themselves within designated social positions (roles) and then how they act within those roles based on differing social contexts (Sluss et al., 2011). Therefore, role identity theory can aid in understanding how individuals define themselves within the roles that they occupy, as well as provide ‘an understanding of how individuals respond to the demands placed on them when facing multiple sets of role expectations’ (Anglin et al., 2022, p. 1474). Sluss et al. (2011) suggested that role identity is more than just the structural position, but also includes ‘the goals, values, beliefs, norms, interaction styles, and time horizons that are typically associated with a role’ (Ashforth, 2001, p. 6). Thus, role identity is heavily influenced by the meanings and expectations associated with one's role.

In addition, psychological empowerment was also selected as a theoretical framework (Quinn & Spreitzer, 1997; Spreitzer, 1995). Based on Spreitzer's (1995) early work with psychological empowerment in the workplace, Quinn and Spreitzer (1997) identified four characteristics of empowerment which include (1) self-efficacy—an individual's internal belief that they have the ability to fulfil a task; (2) meaning—how one personally cares about their own work; (3) sense of competence—one's sense of confidence in their ability to do their work; and (4) sense of impact—one's belief that they influence their work environment or that ‘others listen to their ideas’ (p. 41). In addition, Quinn and Spreitzer (1997) noted, that there must be trust which is related to the strength and reliability in the relationship between the employee(s) and leadership.

Quinn and Spreitzer (1997) stated that ‘many of us implicitly discourage empowerment by reinforcing organizational structure and control systems that either intentionally or unintentionally send the message that we really do not trust people’ (p. 42). This messaging is embedded in bureaucratic culture, multi-level conflict and personal time constraints (Quinn & Spreitzer, 1997). Bureaucratic culture maintains the status quo, stalls change and function through top-down directives. Multi-level conflict is created when divisions exist across areas that intentionally or unintentionally pit individuals against each other in competition for resources and advancement and ‘creates a hostile working environment in which people worry about protecting themselves rather than doing what is right for the organization’ (Quinn & Spreitzer, 1997, p. 43). Personal time constraints occur when downsizing and layoffs force individuals to perform the work tasks of multiple individuals, inducing a stressful working environment that stifles innovation and new initiatives.

Together, role identity theory (Sluss et al., 2011) and psychological empowerment (Quinn & Spreitzer, 1997; Spreitzer, 1995) provide an appropriate lens of analysis for this work. Role identity theory aids in understanding how our participants perceive their roles as administrators over online programmes while psychological empowerment aids in understanding how participants enact their ability or power to support and assess online programmes.

2.1 Research team

Our research team consisted of three members all of whom are serving in the southeastern Carnegie region (CCIHE, n.d.) as assistant professors and have been or are currently programmes coordinators of master's level online programmes in different fields of study, as well as have had experience teaching online at an R1 institution. The first researcher has worked in higher education for 15 years as a professional practitioner, administrator and faculty. Her former work experience provides insight and understanding of the administrative processes and challenges in supporting online programmes. The second researcher has over 20 years of experience in online learning as a student, course designer and faculty. Her academic and professional practitioner background as an instructional designer/developer gives insight into the infrastructure and pedagogical needs of online programmes. The third researcher's professional and academic background in psychology allows her to view online learning as a complex phenomenon that involves cognitive, social, developmental, neurological and other processes. She strives to improve the psychological factors that underlie online learning so that it can move from technology-centred to human-centred.

2.2 Trustworthiness

To ensure trustworthiness, a research team was created who collected data from multiple sources and institutional types, was consistent with data analysis and agreed to meet on a regular basis to discuss the research and triangulate findings. However, it is noted that CDA researchers often ‘reject a neutral, objective stance in research’ (Mullet, 2018, p. 120). Therefore, the researchers of this work have elected to ‘remain aware of the social, political, and economic motives that drive their own work and acknowledge that they do not occupy a superior position’ (Mullet, 2018, p. 120).

2.3 Participants

Purposive sampling was utilised for this research (Robinson, 2014). Purposive sampling is the intentional selection of participants based on their knowledge and experience with the subject matter being researched. For this research, participants are in administrative roles such as chairs, deans, provosts and vice presidents of academic affairs and are between the ages of 18 to 89. They are all supervisors of online programmes at their institutions.

In total, 11 participants volunteered to be interviewed. The interviews were semi-structured, recorded and transcribed verbatim (Schwandt, 2007). Respondents were from seven Carnegie-assigned regions of the United States (see Table 1 for specific regions) and from nine public and two private for-profit institutions (Carnegie Classification of Institutions of Higher Education [CCIHE], n.d.). Six of the participants were female, five were male. All but one participant identified solely as white/European American, one participant identified as Latino/Hispanic and one participant identified as white/European American and other (not specified). Pseudonyms were assigned to protect the participants' identities.

2.4 Data collection

Upon appropriate IRB approval from The UAB Institutional Review Board for Human Use under protocol number IRB-300008594, the researchers sent a recruitment email to the American College Personnel Association (ACPA) listserv to recruit potential participants for the research study. The email included information about the purpose of the study, qualifications to participate and a Qualtrics link they could click to provide initial demographic information and consent to participate in the study. The researchers conducted the interviews using 18 semi-structured interview questions (Schwandt, 2007) that focused on infrastructure, faculty development and the marketing and recruitment for their online programmes. Interviews lasted no longer than 2 hours and were transcribed using transcription software.

2.5 Data analysis

Intertextuality was our method of data analysis. Intertextuality looks at how discourse intersects across differing aspects of language (Kristeva, 1980). At the points of intersection, intertexts are formed and can expose structures of power, social problems and societal meanings (Fairclough, 1992). This is achieved by examining text from two axes—horizontal connecting the researchers to the research and vertically connecting individual discourses to one another (Kristeva, 1980). Intertexts are the ‘weaving of voices’ (Barthes, 1970/1974, p. 20). Barthes (1970/1974) and Kristeva (1980) both noted that no text stands alone; all texts are interwoven and gain their meaning from prior texts. According to Barthes (1970/1974), intertexts form their meaning through five distinct types of voices found within texts and each voice creates a different code. The five voices and their related codes include: empiric voices which produce proairetic codes that determine cause and effect; person voices which produce semic codes that determine connotations and help to define discourse; science voices which produce cultural codes that determine cultural order/truths that are considered common knowledge; truth voices that determine hermeneutic codes which help to reveal mysteries; and symbol voices that determine symbolic codes that reinforce dominant culture and expose conflicts (Barthes, 1970/1974).

Figure 1 demonstrates the consistent need for the researchers to be connected to the text as meaning is developed through intertextual analysis. Researchers look at each text individually and then in relation to how it is interwoven within the other texts – moving back and forth between the vertical and horizontal axes.

The intertexts revealed were analysed in relation to Barthes' (1970/1974) five voices and their associated codes and placed in a spreadsheet. Themes evolved only where the collective voices of all participants were found. As the codes were examined, four themes emerged in relation to administrative support of online programmes: (1) the role of mid-level administrators as gatekeepers, (2) administrators do not view online faculty holistically, (3) assessment as a check-the-box culture and (4) student engagement as a secondary thought. Each of these is further expounded on in the findings section below.

3.1 Mid-level administrators are gatekeepers

Our data analysis revealed that mid-level administrators are the main gatekeepers to providing support for online programmes. Their role is socially constructed and influenced by their power relationship with campus senior administrators (Patel, 2015). They serve as the initial access point to advocate for needed resources and support for online programmes and faculty. However, due to their self-identified middleness, they are caught in a top-down bureaucratic reporting structure giving the greatest social power for decision making and change (French & Raven, 1959) to senior administrators while they utilise their power to enact campus senior administrative directives toward their subordinates. This power relationship (van Dijk, 1995) reveals itself through their discourse about their role expectations and responsibilities.

Blake, a department chair and mid-level administrator, did not feel that he had the ability to impact or influence (Quinn & Spreitzer, 1997) individuals who may be in charge of marketing and recruiting for the online programmes that he oversees, because he believed that was not part of his mid-level administrative role expectations (Anglin et al., 2022). Blake stated, ‘I don't seem to be privy to a lot of those conversations, which I would jokingly say… it's above my paygrade’. For faculty that may be recruiting for their own programmes and need support, this discourse is problematic because Blake is the middle link in the chain of command between faculty and campus senior administrators.

CDA asserts that those in elite positions maintain a higher level of social power and therefore, have more control than ordinary people (van Dijk, 1995). Though Drew and Blake do not have the same level of institutional power found in campus senior administrative roles, they are in a pivotal mid-level administrative role where their awareness along with inaction asserts negative power over their subordinates whose needs are not being communicated to those who could create avenues for change and support. This could lead to competition for resources by their subordinates and in turn create a hostile working environment where individuals are consistently having to protect their own interests (Quinn & Spreitzer, 1997).

Frankie, like Drew and Blake, completely missed an opportunity to use the power of her position to advocate for assessment processes that could look past the typical accrediting agency requirements related to learning outcomes (U.S. Department of Education, 2023). Through the lens of CDA, these mid-level administrators all seem to misunderstand the messages that they are sending to their subordinates about how they are valued (Mogashoa, 2014). Likewise, they are showing that in their roles and the social contexts (Sluss et al., 2011) of their institutions of higher education, it is acceptable to maintain the status quo even if it negatively impacts the ability to promote growth and innovation.

3.2 Administrators do not view online faculty holistically

As administrators shared their perspectives, it became very clear that online faculty were not viewed holistically. Our administrative participants often assessed online faculty as if their roles were identical to face-to-face faculty regarding workload responsibilities, professional development and assessment. They had mixed ideas about the online faculty role and its non−/similarities to the face-to-face faculty role, despite the research that has been done in this area that suggests otherwise (Martin et al., 2022; Martin & Kumar, 2021; Piña & Bohn, 2016).

CDA often informs where advocacy is needed, but it is up to the individuals, in this case Jackie, to act (French & Raven, 1959; van Dijk, 1995). While she understood the power limitations within her role to enact change (Anglin et al., 2022; Sluss et al., 2011), she also has not utilised the power that she does have to convey the need up her chain of command and act by requesting assistance.

Both Alex, an associate provost, and Parker, a vice provost, noted the different ways in which they have provided technological support for their online faculty as campus senior administrators. Alex stated, ‘Our instructional technologists work on a completely flexible schedule. They don't have set hours; they are given complete flexibility and they work remotely’. Through the lens of CDA (French & Raven, 1959; van Dijk, 1995) and role identity theory (Sluss et al., 2011), we can see that while Alex celebrates the flexibility of the role of the instructional technologist, he fails to acknowledge the role of the online faculty who are also working non-traditional hours without that same flexibility and support.

Although instructional designers provide pedagogical assistance, there are additional aspects of the faculty role including the need for schedule flexibility and support outside of course development (i.e., research and service) that is needed.

Jordan, an associate provost, seemed to be the only campus senior administrator to have a greater awareness of the need for a holistic approach to supporting online faculty and programmes. She stated, ‘The biggest challenge we're facing… [is that] we have to make things easier, both for students [and] for faculty, but also for administrators trying to enact change’. She pointed out the need to push back against the bureaucratic culture (Quinn & Spreitzer, 1997) that maintains the status quo and prevents change at all levels. She had a strong understanding of her role and the expectations associated with it (Ashforth, 2001).

3.3 Assessment has become a check-the-box culture

Overall, our findings show that the assessment of online programmes and faculty has taken on a minimalist approach that checks-the-box to say the task of assessment, as defined by accrediting bodies and legislation (Higher Education Act of 1965, 2022; U.S. Department of Education, 2023), as well as U.S. institutional teaching quality standards (Goos & Salomons, 2017), has been completed through student evaluations, hiring practices and informal mentoring, but does not serve the deeper need for the development of these programmes nor the faculty that teach within them.

While her advocacy within her role is admirable, she does not recognise all of the responsibilities or needs for assessment related to online faculty or online programmes. Her sense of impact is high, but she is still utilising her power to reinforce ‘organizational structures and control systems’ (Quinn & Spreitzer, 1997, p. 42) that promote the status quo.

When questioned about how online faculty are assessed, Blake, a department chair and mid-level administrator, stated he assesses online faculty at the point of hire. He noted, ‘We hire really intentionally… It's in our job ads that, experience teaching online is a minimum qualification, … minimally, it's they've taught online before’. He expounded, ‘to teach our students, … I think it is really personality driven. I think there's always been good people. Just good souls, you know, in that area’. Mogashoa (2014) spoke about the ways in which CDA can be utilised as a tool to help members within a profession to understand what messages they are sending. Here, we see Blake telling us that in his role as a mid-level administrator, he is tasked with hiring for his department and his way of assessing qualifications and needs for online faculty is to hire ‘good souls’ who have experience teaching online, even if only minimal experience.

CDA often exposes the relationship between power and action or in this case, lack thereof (van Dijk, 1995). Jamie is in a position as a mid-level administrator that allows him the power to create a formal assessment plan, but instead, he has created an informal mentorship programme that lacks a data-driven approach to onboarding new online faculty.

3.4 Student engagement is a secondary thought

Taylor, an assistant dean and mid-level administrator, spoke about student engagement based on enrollment management services (Shelton & Pederson, 2017) but not how students engaged within online programmes. She stated, ‘Then, it's really the delivery. Again, when it comes to all those services [marketing, admissions, records, and advisors], … we can partner. So, we've done all that for our students’. The reinforcement of enrollment management processes that push quantity over quality contributes to bureaucratic cultures that maintain the status quo and stall positive systemic change (Quinn & Spreitzer, 1997). Both Kennedy and Taylor are reinforcing organisational structures that centre on the processing of students to meet institutional enrollment goals, but not on student engagement within their online programmes. Here CDA reveals how the reproduction of dominant institutional power structures controls the student engagement experience (van Dijk, 1995).

Kelly, an associate dean and mid-level administrator, noted that it is difficult to provide equitable engagement and support of students in online programmes because ‘there's a lot of mindset that campus services are available to everyone, but they're not always accessible from [a] distance’. Kelly understands that one of the expectations of her role is to provide opportunities for student engagement (Anglin et al., 2022), but lacked a sense of self-efficacy (Quinn & Spreitzer, 1997) in her ability to do so. Parker, a vice provost and campus senior administrator, also noted that in his role, he had not ‘done a good job of that. We have been much more, you know, offices are open from nine to five, and hopefully, you can make it to campus’.

While he is displaying a sense of competence through the confidence in his actions (Quinn & Spreitzer, 1997), he failed in his role to recognise the need to assess his students' experiences to see if this is what they needed.

4.1 Limitations

A limitation of our study is related to the timing of data collection. As many institutions began to resume face-to-face coursework, the hyperfocus on online learning and services during the COVID-19 pandemic revealed the weaknesses in online/technological infrastructures at many institutions (Garrett et al., 2020) and may have influenced some of our participants' perceptions of the importance of these infrastructures.

4.2 Implications for practice

We provide the following four suggestions to administrators in higher education. First, mid-level administrators who directly oversee online programmes need additional training in working within the online environment. Second, training should specifically include how to better assess all aspects of online programming beyond technological needs and current student evaluations. Assessment tools are needed that focus specifically on teaching quality and student engagement like that introduced by Piña and Bohn (2014). Third, specific focus must be placed on the holistic support of online faculty regarding onboarding, training in how to teach online and their own professional growth. Fourth, administrators must view student engagement from an online programme perspective looking beyond the recruitment and enrollment processes.

4.3 Implications for research

We strongly suggest that more research be conducted on the specific differences between online and face-to-face faculty workloads based on faculty classification, programme responsibilities and career trajectories. We also strongly suggest that further research be conducted to address the online student experience and how their engagement and/or lack thereof impact retention and graduation rates.