While research has shown a positive correlation between instructor presence and student learning outcomes, the role of instructor synchronous activities in online courses remains underexplored. This review addresses that gap by critically evaluating the literature on synchronous instructor presence and its impact on student learning.
Originally intended as a meta-analysis, this review followed a systematic search process aligned with the US Department of Education (2020) and the PRISMA guidelines (Page et al., 2021). Due to a limited number of studies on synchronous instructor presence, inconsistent definitions, diverse study designs and variable outcome measures, meta-analysis was deemed unfeasible. The review proceeded as a critical synthesis of the existing research.
From the initial pool of 7,591 studies published between 2010 and 2022, only ten met the criteria for inclusion in the critical review: seven examined course grades and three examined grade distributions. None satisfied the stricter criteria for meta-analysis. Most studies lacked a standardized definition, valid effect size reporting and consistent methodological rigor.
This review highlights a significant gap in the current evidence and calls for more robust, clearly defined and methodologically sound research on synchronous instructor presence. It underscores the need for improved reporting practices to enable future meta-analyses and evidence-informed practice in online education.
Introduction
In the field of online teaching and learning, there has been a growing interest in the direct interactions between instructors and students in predominantly asynchronous online environments. Distance education research has consistently shown that well-designed online instruction results in learning outcomes equivalent to those achieved in traditional classrooms (Simonson, 1999, 2019; Simonson, Schlosser, & Orellana, 2011). In the past few years, coinciding with the proliferation of various forms of online instruction that were borne out of the COVID-19 pandemic, there has been a renewed interest in including synchronous components in online courses that have previously been predominantly asynchronous (Capranos, Dyers, & Magda, 2023). During that time, the distinction between remote teaching and intentionally designed distance education was especially emphasized (Simonson, 2021). In online education, there has been a continual interest in increasing the presence of instructors in order to foster connections between faculty and students. Introducing synchronous components in online courses is one strategy that may increase the presence of the instructor (Martin, Parker, & Deale, 2012). This heightened presence is not just about more frequent interactions; it is about fostering a sense of community and belonging among learners. Research has shown that when students feel a stronger connection with their instructors, it can lead to a more immersive and engaging learning experience (Caskurlu, Richardson, Maeda, & Kozan, 2021). This sense of connection is particularly crucial in an online learning environment, where physical distance can lead to a feelings of isolation. This lack of physical connection brings unique challenges in maintaining student engagement and motivation. In the Wiley's Voice of the Online Learner survey (Capranos et al., 2023), around 33% of student respondents preferred fully online and asynchronous programs, but a significant portion (approximately 79%) were open to including synchronous components. Within this group, about 50% would welcome synchronous sessions once a week, while around 25% would prefer them every second week or once a month. During these synchronous sessions, students appreciate the opportunity to ask real-time questions (58%) and receive clearer explanations from instructors (52%).
Empirical studies have begun to shed light on the relationship between the presence of the instructor and positive student outcomes, such as improved grades and higher retention rates (Rosser-Majors et al., 2022), reinforcing the view that instructional design quality and active instructor involvement are essential for success in distance education (Simonson, 2006, 2019). This perspective aligns with early research on media effects on learning. Specifically, Clark (1983) argued that media by themselves do not influence learning, while Kozma (1991, 1994) emphasized that the ways instructors design and facilitate learning through media can meaningfully shape students' cognitive engagement and outcomes. These findings suggest that the role of the instructor extends beyond the mere delivery of content; their presence and active engagement with students are critical components of effective learning. While existing research suggests a positive link between instructor presence and student satisfaction and perceived learning (Caskurlu et al., 2020, 2021; Richardson, Maeda, Caskurlu, Kozan, & Swan, 2024), there is a considerable gap in understanding the effect of synchronous instructor activities in online courses. The current literature often lacks specific insights into the role of different types of instructor presence, such as synchronous lectures, real-time discussions and individual consultations, and their distinct impacts on learning.
Despite their potential for immediate feedback and engagement, the effectiveness of these real-time interactions in improving student learning outcomes still needs to be explored. Most studies have focused on broader online learning aspects, overlooking the unique contributions of synchronous instructor activities. Thus, this review fills this gap by examining the literature on synchronous instructor presence to understand its impact on student learning.
Teaching presence in the Community of Inquiry framework
One of the distinctive differences between asynchronous and synchronous online courses is whether the instructor is directly engaged in the teaching and learning process. In synchronous online learning, instructors provide direct instruction, hold live question-and-answer sessions, or guide collaborative learning as part of the course design. In this context, the instructor's engagement is different from simple co-location in the online space or managing student learning behind the scenes (such as overseeing the online discussion board, grading and commenting, posting learning materials, etc.).
In defining instructor presence for this paper, we adopted the Community of Inquiry (CoI) framework, where social, teaching and cognitive presences contribute to creating meaningful learning experiences (Garrison, Anderson, & Archer, 2000). According to the CoI framework, teaching presence refers to “the design, facilitation, and direction of cognitive and social processes for the purpose of realizing personally meaningful and educationally worthwhile learning outcomes” (Anderson, Rourke, Garrison, & Archer, 2001, p. 5; Castellanos-Reyes, 2020). Teaching presence plays a critical role in facilitating student learning (Swan et al., 2008), as “teaching presence refers to the methods that an instructor utilizes to promote a quality online environment and facilitate an effective community of inquiry” (Bangert, 2008, p. 40).
Synchronous instructor presence
In this critical review, we conceptualized instructor presence as the synchronous and observable instructional actions that operationalize teaching presence within the CoI framework. The COI component of teaching presence encompasses design, facilitations and direction of learning processes (Anderson et al., 2001) across all modalities; whereas in this analysis, synchronous instructor presence refers specifically to the enacted behaviors of instructors that learners directly experienced during real-time interactions. While social and cognitive presence are related constructs within the CoI, they typically represent the outcomes of instructional decisions rather than the instructor behaviors themselves. Accordingly, we focused our analysis on synchronous instructional activities that demonstrate instructor facilitation and direction during live interactions, rather than on student-perceived indicators of social or cognitive engagement. Although teaching presence overlaps with cognitive presence - such as course design and organization of the content during the preparation phase and asynchronous feedback and grading—instructors’ synchronous engagement in facilitating and guiding student learning is critical (Anderson et al., 2001; Swan et al., 2008; Zhang, Lin, Zhan, & Ren, 2016). In online courses, teaching presence can be identified by instructional activities such as the instructor facilitating student learning, providing direct instruction, conducting assessments and giving timely feedback (Anderson et al., 2001), and these types of instructional activities are correlated with student satisfaction and outcomes (Richardson et al., 2024).
As teaching presence integrates intersections between social presence and cognitive experience, teaching presence addresses teaching elements such as instructional design and organization (ID & OR), facilitating discourse (FD) and direct instruction (DI) (Anderson et al., 2001). Among these elements of teaching presence, ID & OR can be prepared asynchronously or prior to the teaching of the course without direct interaction between the learner and the instructor. On the other hand, FD can be done as the instructor interacts either synchronously or asynchronously with the learner, and DI engages the instructor to present the course content or provide feedback, either synchronously or asynchronously, interacting with the learner. As we were interested in examining the effects of synchronous learning, we focused on the FD and DI perspectives where active instructor presence could have a considerable effect. In this context, we narrowed down our focus on the instructor's presence from the overarching concept of teaching presence. Additionally, teaching presence in the context of teacher education literature often refers to the physical presence of the teacher in face-to-face classroom teaching. Therefore, we have chosen to use instructor presence to refer to the online instructor's engagement as a part of teaching presence, in the context of higher education.
While instructor presence can be implemented both synchronously and asynchronously, we also wanted to focus on synchronous presence, in which learners are more aware of the instructor's active presence, in contrast to asynchronous presence, in which learners' perceived level of instructor presence varies by the course context. Also, as instructor presence overlaps with social and cognitive presence, the nature of interaction and content presentation has an effect on learners' perception of instructor presence. (Martin, Wang, & Sadaf, 2018; Watson et al., 2023). Therefore, to determine synchronous instructor presence, we have analyzed and coded the course settings described in the studies for this systematic review, instead of using the instructors' or learners' perceived level of instructor presence.
Literature focused on outcomes
Research studies using the CoI framework mainly used the instrument created by Arbaugh et al. (2008) to examine if one or more of the presences can be detected in a given learning environment and to validate the construct. Following the suggestion from Arbaugh et al., researchers used “CoI as a dependent measure in comparing courses and the implementation of emerging technologies within courses.” (p.136). In this context, the existing studies examined the relationship between the CoI presence and: (1) students’ sense of learning community (Shea, Li, & Pickett, 2006), (2) students’ level of interaction (Aslan, 2021) and (3) perceived learning and satisfaction (Akyol and Garrison, 2008, 2011, 2008; Richardson et al., 2024). On the other hand, some studies used various outcome measures such as text analysis ratings of online discussions in relationship with social or cognitive presence (Guo, Saab, Wu, & Admiraal, 2021), student satisfaction, perceived progress and final grades (converted into percentages of the maximum points possible) in a K-12 setting (Zhang & Lin, 2021). A recent systematic review on collaborative learning and learner engagement in relation to the CoI model, Das and Madhusudan (2023), lists studies based on students’ perceived level of collaborative learning and engagement.
Perceived learning, student satisfaction and academic outcomes, such as grades, serve as common metrics to evaluate the effectiveness of synchronous components of online courses. Within the CoI framework, both social and cognitive presence are associated with positive learning outcomes (Rockinson-Szapkiw, Wendt, Whighting, & Nisbet, 2016). However, the active presence of the instructor serves as the uniting force that brings the cognitive and social elements of an online learning environment together, intending to achieve personally meaningful and educationally valuable outcomes (Miller, Hahs-Vaughn, & Zygouris-Coe, 2014). Students highly value synchronous instructor presence due to its ability to provide real-time interaction and immediate feedback, and to foster a sense of community, thus substantially enhancing their perceived learning experience. These results reinforce that learning outcomes in online environments are shaped by the presence of synchronous sessions and the quality of their design and implementation (Simonson, 2019).
Meta-analyses and reviews
There is continued research interest in the CoI, and several meta-analyses or systematic reviews have been conducted to understand the impacts of the three presences. Martin, Wu, Wan and Xie (2022) conducted a meta-analysis on the effects of CoI presence on learning outcomes in online and blended learning environments. To examine the effects on learning outcomes, as most previous studies focused on learners' perceptions, Martin et al. examined effect sizes on “actual learning”, a construct suggested by Bacon (2016), which includes scores from tests, projects, presentations and performances, by the three CoI presences. They found that teaching presence was positively related to satisfaction (r = 0.51), actual learning (r = 0.35) and perceived learning (r = 0.39). These findings were similar to a 2020 meta-analysis by Caskurlu et al. (2020) that showed a positive relationship between teaching presence and student satisfaction (r = 0.59) and perceived learning (r = 0.60).
While these meta-analyses examined teaching presence, they did not examine the specific impacts of synchronous instructor presence. A review by Watts (2016) concluded that previous research on synchronous versus asynchronous methods used varied methodologies and produced mixed results, leading them to conclude that it was difficult to identify which was the best method for online instruction. Martin, Ahlgrim-Delzell and Budhrani (2017) reviewed research on synchronous online learning from 1995–2014. Their review discussed the relationship between synchronous interaction and social presence but did not address the impact of instructor presence.
Critical review
This critical review fills the gap by examining this question: How does instructor presence in the form of synchronous activities in online courses relate to student satisfaction and student outcomes (e.g. grades)? Initially planned as a meta-analysis, this review underwent a remarkable transformation in scope and objective. The following limitations in the available literature rendered a meta-analysis impractical and inappropriate for addressing this research question: (1) very few published studies examined instructor presence; (2) even fewer examined synchronous instructor presence; (3) in those that did, synchronous instructor presence was inconsistently defined; and (4) limited information was available about the synchronous interaction frequency, types and activities (e.g. Martin & Bolliger, 2018). These pervasive issues created a fragmented research landscape and made meaningful aggregation and comparison of studies problematic, thus compromising the feasibility of conducting a rigorous and reliable meta-analysis.
Thus, this paper presents a critical review highlighting the current state of the research on synchronous instructor presence. It lays the groundwork for future studies to build upon, ensuring that they are better equipped to contribute meaningful insights into the role of synchronous instructor presence in online learning.
Methods
This critical review was conducted using the five steps for a systematic review cited by the US Department of Education, Institute of Education Sciences, What Works Clearinghouse Procedures Handbook (https://ies.ed.gov/ncee/wwc/Docs/referenceresources/WWC-Procedures-Handbook-v4-1-508.pdf) and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (Page et al., 2021) checklist (http://www.prisma-statement.org/).
Search process
The search process included databases, searching reference lists and author searches. We began with a scoping review of these databases: ACM Digital Library, EBSCO Academic Search, ERIC, ProQuest, PSYCHIFNO, Web of Science and Wiley Online Library. From the initial review, the search was focused on EBSCO-Academic Search, and ERIC (via EBSCO), PSYCHINFO (via EBSCO) and Web of Science. Database searches were modified to include only full-text, peer-reviewed scholarly journals, available in the English language. The following terms were used in the search process: online learning, synchronous, instructor presence, communication with instructor, teaching presence, communication with students, outcomes and satisfaction. After an article was pre-reviewed for inclusion in the analysis, the reference list was searched regardless of the decision to include and code the study or not. In addition, in Web of Science, we reviewed articles that were suggested by articles that appeared in the initial search. Overall, the initial database searches identified 7,584 records. The additional reference lists and author searches identified four and three additional records, respectively, resulting in a total of 7,591 potential studies published from January 1, 2010, to December 31, 2022.
Inclusion criteria
Initial criteria
Studies had to meet all of the following inclusion criteria: (1) Study published in peer reviewed journals between 2010 and 2022, (2) available in English language, (3) an empirical study based in post-secondary settings, (4) included online courses with synchronous activities in which students had the opportunity to have interaction with the instructor in real-time, (5) reported a comparison between synchronous and asynchronous components (within or between courses) and (6) reported grades or grade distributions, or course satisfaction as outcomes in the study.
Criteria revised
Due to the lack of studies meeting all of the above criteria, we dropped criterion #5 and did not exclude studies that lacked a comparison between synchronous and asynchronous components. One study found in the reference list search was published in 2009 and thus did not meet criterion #1, but met all other criteria. This study was included in the critical review.
Application of inclusion criteria
Inclusion criteria were applied in three PRISMA-aligned stages, with exclusions at each stage representing studies that did not meet the inclusion criteria. At Stage 1, one researcher applied the inclusion criteria and removed the study if it clearly did not meet the criteria. At this stage, studies were excluded for the following reasons: (1) studies based on blended, flipped course designs, because in these modalities synchronous instruction typically occurs in face-to-face settings rather than in an online environment. (2) studies on MOOCs, (3) courses in vocational education and (4) studies on K-12 courses. This stage resulted in removal of 7,515 studies.
At Stage 2, the remaining 76 studies were reviewed by pairs of researchers who applied the inclusion criteria. As shown in Figure 1, 47 studies were excluded because they did not meet the inclusion criteria. At Stage 3, the remaining 29 studies were coded separately by two researchers and reviewed by a third researcher. Applying the inclusion criteria during this coding process resulted in excluding 19 studies, which resulted in a total of 10 studies included for the final analysis (see Figure 1). A few studies had several outcomes, as shown in the results section.
The vertical flow diagram made of rectangular boxes connected by directional arrows, with three labeled sections shown along the left side as tall vertical boxes: “Identification”, “Screening”, and “Included”. The process is titled“Identification of studies via databases and registers”.The flowchart begins in the “Identification” section with a large rectangular box lists sources under “Records identified from: Databases (n equals 7,584)” with “E R I C (n equals 308)”, “E B S C O (n equals 2,601)”, “PSYCHINFO (n equals 844)”, and “Web of Science (n equals 3,831)”, alongside “Reference List (n equals 4)” and “Author Search (n equals 3)”, concluding with “Total equals 7,591”. A downward arrow moves the flow into the “Screening” box, which spans the next stages, including “Stage 1: Records screened by one researcher (n equals 7,591)”, which connects via a right-pointing arrow to a box labeled “Records excluded (n equals 7,515)”. A downward arrow from Stage 1 leads to “Stage 2: Records screened by two researchers (n equals 76)”, which connects to a right-side box labeled “Records excluded: n equals 47”, listing reasons exactly as shown, including “No synchronous components equals 14”, “No appropriate outcome equals 10”, “No instructor presence equals 6”, “Qualitative study equals 6”, “Non-degree course equals 2”, “Blended course equals 2”, “Threats to validity equals 1”, “No online component equals 1”, “COVID hyphen 19 pandemic study equals 1”, “K 12 study equals 1”, “Unable to isolate instructor presence equals 1”, “Theoretical paper equals 1”, and “Both treatment and control were synchronous equals 1”. A downward arrow from Stage 2 leads to “Stage 3: Records screened by three researchers (n equals 29)”, which connects to another right-side box labeled “Records excluded: n equals 19”, listing “No synchronous components equals 2”, “No appropriate outcome equals 3”, “No instructor presence equals 5”, “Insufficient data equals 1”, “Threats to validity equals 1”, “Unable to isolate instructor presence equals 1”, “Flipped classroom equals 1”, “Methodological issues equals 1”, “No difference between treatment and control group equals 1”, “Poor reporting on methods and outcomes equals 1”, “Vocational course equals 1”, and “Duplicate equals 1”. A downward arrow leads into the “Included” box at the bottom, aligned with the rectangular box labeled “Studies included in review (n equals 10)”.PRISMA 2020 flow diagram for new critical review. From: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. doi: 10.1136/bmj.n71. For more information, visit http://prisma-statement.org/
The vertical flow diagram made of rectangular boxes connected by directional arrows, with three labeled sections shown along the left side as tall vertical boxes: “Identification”, “Screening”, and “Included”. The process is titled“Identification of studies via databases and registers”.The flowchart begins in the “Identification” section with a large rectangular box lists sources under “Records identified from: Databases (n equals 7,584)” with “E R I C (n equals 308)”, “E B S C O (n equals 2,601)”, “PSYCHINFO (n equals 844)”, and “Web of Science (n equals 3,831)”, alongside “Reference List (n equals 4)” and “Author Search (n equals 3)”, concluding with “Total equals 7,591”. A downward arrow moves the flow into the “Screening” box, which spans the next stages, including “Stage 1: Records screened by one researcher (n equals 7,591)”, which connects via a right-pointing arrow to a box labeled “Records excluded (n equals 7,515)”. A downward arrow from Stage 1 leads to “Stage 2: Records screened by two researchers (n equals 76)”, which connects to a right-side box labeled “Records excluded: n equals 47”, listing reasons exactly as shown, including “No synchronous components equals 14”, “No appropriate outcome equals 10”, “No instructor presence equals 6”, “Qualitative study equals 6”, “Non-degree course equals 2”, “Blended course equals 2”, “Threats to validity equals 1”, “No online component equals 1”, “COVID hyphen 19 pandemic study equals 1”, “K 12 study equals 1”, “Unable to isolate instructor presence equals 1”, “Theoretical paper equals 1”, and “Both treatment and control were synchronous equals 1”. A downward arrow from Stage 2 leads to “Stage 3: Records screened by three researchers (n equals 29)”, which connects to another right-side box labeled “Records excluded: n equals 19”, listing “No synchronous components equals 2”, “No appropriate outcome equals 3”, “No instructor presence equals 5”, “Insufficient data equals 1”, “Threats to validity equals 1”, “Unable to isolate instructor presence equals 1”, “Flipped classroom equals 1”, “Methodological issues equals 1”, “No difference between treatment and control group equals 1”, “Poor reporting on methods and outcomes equals 1”, “Vocational course equals 1”, and “Duplicate equals 1”. A downward arrow leads into the “Included” box at the bottom, aligned with the rectangular box labeled “Studies included in review (n equals 10)”.PRISMA 2020 flow diagram for new critical review. From: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. doi: 10.1136/bmj.n71. For more information, visit http://prisma-statement.org/
Coding process and framework
The codebook was developed and adapted during the coding process. Both deductive and inductive coding processes were used for the codes shown below. The studies were coded by the authors using a Google Sheet. Each study was assigned to rotating pairs of authors who coded independently. The pairs then discussed their coding results and came to a consensus. If consensus was not reached, a third coder reviewed the study until all three coders came to a consensus. When items were coded as open-ended, the codes were categorized based on frequency.
Articles were coded for the following characteristics: course delivery modes, synchronous and asynchronous course components, year of publication, institution name, characteristics of students (education level and background), characteristics of the courses involved in the studies (discipline and length), number of participants and attrition, study design, effect name, outcome for analysis, statistical significance, validity and reliability, risk of bias and reason for rejection. A summary is shown in Table 1.
Summary of coded elements
| Article information | Full reference; year of publication; institution name(s) |
|---|---|
| Course Delivery Modes in the study | Coded as online synchronous, online asynchronous, both online synchronous and asynchronous |
| Course Delivery Comparisons in the study | Open-ended coding of what course delivery modes were being compared |
| Course Components | Open-ended coding of synchronous and asynchronous course components |
| Course Characteristics | Open-ended coding of discipline and length of course(s); education level and background of students in the course(s) |
| Study Design | Coded as Experimental, quasi-experimental, non-experimental |
| Outcome for analysis | Coded as course grades, grade distribution, satisfaction, and other |
| Statistical information | Open-ended coding of statistical tests, group sizes, and descriptive statistics (M, SD) |
| Risk of Bias | Open-ended concerns related to the quality of the study and threats to the validity |
| Reason for Rejection | Coded reasons were: qualitative studies, insufficient data, no synchronous components, no instructor presence, no appropriate outcome, risk of bias, threats to validity, and other |
The following sections describe the coding process for key variables.
Course delivery modes
Course delivery modes were coded as (1) online synchronous, (2) online asynchronous and (3) both online synchronous and asynchronous, reflecting how the course(s) in the study were delivered. To find the effect of interest for this analysis, the study needed to have both online and synchronous components. In some studies, the two course delivery modes were compared between two different sections of a course (e.g. Griffith, Faulconer, & McMasters, 2021; Roberts, Griffith, Faulconer, Wood, & Acharyya, 2019). Other studies included both synchronous and asynchronous components within one course (e.g. Duncan, Kenworthy, & McNamara, 2012; Skylar, 2009).
Course components: synchronous
Our key variable of interest was the synchronous components of the courses in these studies, in which we could identify synchronous instructor presence. The following components were identified as synchronous components: (1) Live lectures, (2) One-on-one meeting(s) with instructor, (3) Live chat with instructor, (4) Live online discussions with instructor and other students, (5) Twitter chat, (6) Group texting, (7) Live discussion through online platforms (e.g. Discord, Jamboard and Teams) and (8) Collaborative brainstorming tools (e.g. Google docs, Jamboard and Perusall). In the coding process, we noted if these synchronous components were a required component of the course and if there was direct synchronous interaction between the instructor and student.
Course delivery comparisons
We coded studies for comparisons between the synchronous course delivery components and the absence of those components (a typical fully asynchronous online course). Many of the studies included a comparison of a section of a course with synchronous-only interaction to the same course with asynchronous-only interaction (e.g. Olson & McCracken, 2015; Bourdeau, Griffith, Griffith, & Griffith, 2018; Faulconer, Griffith, Wood, Acharyya, & Roberts, 2018). In some studies, there were no course delivery comparisons, but relevant correlations were reported. For example, Nieuwoudt (2020) reported the correlation between synchronous class attendance and grades.
Outcome for analysis
Outcomes were coded as: (1) Course Grades, (2) Grade Distribution, (3) Satisfaction and (4) Other. Course grades included summative evaluations of learning, final course grades and GPA. Grade distribution was a course-level outcome. Satisfaction was based on one or more survey questions completed by students.
Reason for rejection
We coded the reasons that studies did not meet the inclusion criteria during the coding process. Initially, these reasons were recorded as open-ended. Later in the coding process, the reasons for rejection were categorized by frequency by the authors. Coded reasons for rejection are shown in Figure 1.
Results
An overview of the ten studies examined for this critical review is shown in Table 2. None of the studies reporting satisfaction as an outcome met the inclusion criteria for the final analysis. The systematic search resulted in a total of 10 studies, including seven studies that examined course grades and three studies that examined grade distributions. Five studies reported on mean differences and three studies reported chi-square tests of course grade distributions. Table 2 shows the outcome for analysis, statistics, education level and course discipline for each of the ten studies.
Studies included in the systematic review
| Authors | Outcome for analysis | Statistics | Education level | Course discipline/Topic |
|---|---|---|---|---|
| Griffith et al. (2021) | Course grades | Means | Undergraduate | Statistics |
| Nieuwoudt (2020) | Course grades | Correlations | Undergraduate | Study skills & strategies |
| Roberts et al. (2019) | Course grade distributions | Chi-square | Undergraduate | Research methods |
| Bourdeau et al. (2018) | Course grade distributions | Chi-square | Undergraduate | English composition |
| Faulconer et al. (2018) | Course grades distributions | Chi-square | Undergraduate | Physics |
| Blau et al. (2017) | Course grades | Means | UG and Graduate | Psychology (lab study) |
| Olson and McCracken (2015) | Course grades | Means | UG and Graduate | Unknown |
| Strang (2012) | Course grades | Means | Undergraduate | Statistics |
| Duncan et al. (2012) | Course grades | R square | Graduate | Accounting |
| Sklyar (2009)* | Course grades | Means | Graduate | Education |
Note(s): *Study is outside the search parameter years; identified during the reference list search
Due to this small number of studies and the problems uncovered in the review and coding process, the research team was not able to draw conclusions to sufficiently address the original research question about how instructor presence in the form of synchronous activities in online courses relates to student satisfaction and student outcomes.
The following section reviews and summarizes how this small set of studies defined synchronous instructor presence and measured outcomes. We will also discuss methodological and reporting factors that impacted the critical review.
Course grades
Table 3 shows the analysis of the seven studies that examined course grades.
Studies examining course grades
| Authors | Course delivery comparisons reviewed for analysis | Synchronous components | N | Statistics | Significance | Effect size |
|---|---|---|---|---|---|---|
| Griffith et al. (2021) (Outcome 1) | Video synchronous classroom (VC) vs. Asynchronous Online | Video synchronous classrooms = students connected with the instructor via synchronous video from satellite locations | VC = 12 | VC: M = 88.21 (SD = 8.87) | NS | *d = 0.208 |
| Async online = 230 | Async online: M = 84.17 (SD = 19.76) | |||||
| Griffith et al. (2021) (Outcome 2) | Video synchronous Home (VH) vs. Asynchronous Online | Video synchronous Home = students connected with the instructor via synchronous video from home | VH = 43 | VH: M = 79.43 (SD = 25) | NS | *d = −0.229 |
| Async online = 230 | Async online: M = 84.17 (SD = 19.76) | |||||
| Nieuwoudt (2020) (Course 1) | No comparisons: correlation between synchronous class attendance and grade | One-on-one meetings, chat, shared whiteboard with instructor (once per week) | N = 114 | Spearman Rank order correlation (Rho) = 0.596 | p < 0.001 | Spearman's Rho = 0.596 |
| Nieuwoudt (2020) (Course 2) | No comparisons: correlation between synchronous class attendance and grade | One-on-one meetings, chat, shared whiteboard with instructor (once per week) | N = 45 | Spearman rank order correlation (Rho) = 0.416 | p < 0.01 | Spearman's Rho = 0.416 |
| Blau et al. (2017) | Two-way video conferencing (VC). vs One-way video conferencing | “Lecture” was delivered in a 2-way video conferencing both the instructor and the students were seen on video | Two-way = 26 One-way = 26 | Two-way M = 0.83 (SD = 0.19) | NS | *d = −0.366 |
| One-way M = 0.80 (SD = 0.18) | ||||||
| Olson and McCracken (2015) | Asynchronous vs. asynchronous + weekly synchronous lesson | Weekly session: lecture, real-time discussions and the instructor interacted with students via chat in Adobe Connect | Async = 10 | Async M = 88.8% (SD = 9.2) | NS | *d = −0.278 |
| Async + sync = 10 | Async + sync M = 84.1% (SD = 22.1) | |||||
| Strang (2012) | Asynchronous discussion forums vs. synchronous sessions | Chat, audio and document sharing via Skype | Async = 29 | Async M = 64.59 (SD = 5.84) | p < 0.01 | *d = 0.607 |
| Sync = 32 | Sync = M = 69.02 (SD = 8.42) | |||||
| Duncan et al. (2012) | Within subjects: students' engagement in both asynchronous and synchronous discussions (chat) | A weekly 2-h chat session with the instructor | N = 120 | Regression coefficient, sync engagement on Final Exam score = 3.820 | p < 0.05 | NA** |
| Sklyar (2009)* | Asynchronous online vs synchronous web conferencing (within subjects – all experienced both conditions) | Synchronous web conferencing lectures with two-way audio and video, breakout rooms, chat, polling, & whiteboard | N = 44 | Async M = 38.45, (SD = 6.114) | NS | *d = 0.321 |
| Synch M = 40.19, (SD = 4.618) |
Note(s): *Effect size calculated by authors
** Insufficient information to calculate effect size
This table shows only specific comparisons or analyses that were relevant to this review and includes the relevant statistical information and effect sizes calculated by the research team. Some studies included more than one relevant outcome, as noted in the authors column. The second and third columns show the wide variety of study designs and types of synchronous instructor presence in these studies. Some studies compared asynchronous and synchronous course sections (e.g. Griffith et al., 2021) while others used within-subjects designs (e.g. Duncan et al., 2012). Most of these studies were conducted within course settings where the course and research designs allowed for a comparison between a condition in which there was synchronous instructor presence in the course and a condition or course that lacked synchronous instructor presence, typically an asynchronous online course. However, Nieuwoudt (2020) did not have any course delivery comparisons but examined the correlation between attendance in synchronous classes and grades within courses. Unlike the other studies, Blau, Weiser and Eshet-Alkalai (2017) was based on a laboratory study that directly manipulated instructor presence by randomly assigning participants to one of three learning conditions: face-to-face, one-way video conferencing and two-way video conferencing via Zoom.
The Griffith et al. (2021) study included two relevant outcomes. This study compared two synchronous conditions: students connected with the instructor via synchronous video from satellite classrooms or from their homes to an asynchronous online condition. Little information was provided about the nature of the synchronous interactions, other than that they were regularly scheduled meeting times. Their analysis showed no statistically significant differences in these comparisons, but the average course grades were highest in the video synchronous classroom condition. However, this condition had a very small sample size (n = 12).
Both Olson and McCracken (2015) and Nieuwoudt (2020) examined the impact of weekly synchronous lessons. Olson and McCracken's (2015) study included ten students who attended an asynchronous course with the addition of a weekly synchronous lecture in which they participated in real-time discussion. The comparison group was another section of ten students who only participated asynchronously. While not statistically significant, the asynchronous-only students had higher average grades. In contrast, Nieuwoudt (2020) did not have a comparison group but examined two online courses with weekly live synchronous sessions where the students and instructor interacted. This study showed significant, moderate correlations between attendance in weekly synchronous class sessions and higher course grades in two different courses.
The laboratory study by Blau et al. (2017) delivered a 25-min lecture via two-way video conferencing or one-way video conferencing. The average course grades were slightly, but not significantly, higher for the two-way group. There are few details provided in this paper about the nature of the interaction between students and the instructor in one-way and two-way video conferencing groups.
Studies by Strang (2012) and Duncan et al. (2012) were focused on synchronous discussions. Strang's study (2012) included a comparison of asynchronous discussion forums versus synchronous sessions described as “problem-based learning tutorials” (p. 11). The results showed significantly higher average grades for the group who experienced the synchronous sessions, with a medium effect size. In contrast, Duncan et al.'s (2012) study was a within-subjects design where students were engaged in both an asynchronous discussion board and a 2-h weekly synchronous chat with the instructor. They concluded that while asynchronous discussion was important, engagement in weekly synchronous chat sessions had a greater impact on final exam grades.
Skylar's study (2009) was also a within-subjects comparison of asynchronous, text-based lectures and synchronous web conference lectures in two sections of a course. Using an alternating treatment design, students experienced both synchronous and asynchronous conditions. The study showed that average grades were slightly but not significantly higher for content that was delivered synchronously (d = 0.32).
Grade distributions
Table 4 shows the analysis of the three studies that examined course grade distributions. These three studies were conducted by researchers at one institution and used the same research methodology with courses in three different disciplines (see Table 1).
Studies examining grade distributions
| Authors | Course delivery comparisons reviewed for analysis | Synchronous components | N | Statistics | Significance | Effect size (calculated by authors) |
|---|---|---|---|---|---|---|
| Roberts et al. (2019) (Outcome 1) | Video synchronous home (VH) vs asynchronous online | students connected synchronously with the instructor from home | VH = 214 | Chi-square = 18.3546, df = 4; * | p < 0.01 | Cramer's V = 0.107 |
| Async = 1,393 | ||||||
| Roberts et al. (2019) (Outcome 2) | Video synchronous classroom (VC) vs asynchronous online | students connected via video synchronously with the instructor from satellite locations | VH = 289 | Chi-square = 23.22, df = 4; * | p < 0.001 | Cramer's V = 0.118 |
| Async = 1,393 | ||||||
| Bourdeau et al. (2018) (Outcome 1) | Video synchronous home (VH) vs asynchronous online | students connected synchronously with the instructor from home | VH = 193 | Chi-square post hoc = 19.95 df = 4, * | p < 0.001 | Cramer's V = 0.0899 |
| Async = 2,278 | ||||||
| Bourdeau et al. (2018) (Outcome 2) | Video synchronous classroom (VC) vs asynchronous online | students connected via video synchronously with the instructor from satellite locations | VC = 146 | Chi-square post hoc = 12.93, df = 4; * | p < 0.05 | Cramer's V = 0.0730 |
| Async = 2,278 | ||||||
| Faulconer et al. (2018) (Outcome 1) | Video synchronous home (VH) vs asynchronous online | students connected synchronously with the instructor from home | VH = 157 | Chi-square post hoc = 7.93, df = 4; * | NS | Cramer's V = 0.0734 |
| Async = 1,318 | ||||||
| Faulconer et al. (2018) (Outcome 2) | Video synchronous classroom (VC) vs asynchronous online | students connected via video synchronously with the instructor from satellite locations | VH = 323 | Chi-square post hoc = 10.42, DF = 4; * | p < 0.05 | Cramer's V = 0.0797 |
| Async = 1,318 |
Note(s): *Asynchronous grades higher
Table 4 shows two comparisons from each of the three studies that were relevant to this review. In these studies, there were two synchronous course delivery modalities. The video synchronous home section consisted of students connecting synchronously with the instructor via video conferencing from their home locations. The video synchronous classroom section consisted of students connecting synchronously with the instructor from remote satellite classrooms. The third modality that each of these was compared to was an asynchronous online section (see also Griffith et al. (2021) as discussed above).
Both the Roberts et al. (2019) and the Bourdeau et al. (2018) studies showed that the students in the asynchronous online sections of the course had significantly more “A”s and fewer “B”s and “D”s than the two video synchronous sections (home and classroom), although effect sizes were small. These findings were replicated by Bourdeau et al. (2018), in which online asynchronous students earned more “A”s than in the two video synchronous sections, again with very small effect sizes. The third study by Faulconer et al. (2018) showed mixed results. They found no significant differences in grade distributions between video synchronous home and asynchronous online sections. However, the grade distributions of video synchronous classroom and asynchronous online sections were statistically different, with a very small effect size.
Discussion
In this review, we planned to conduct a meta-analysis to examine the impact of a synchronous instructor's presence on student learning in the post-secondary online learning environment. Specifically, we were interested in empirical studies on courses with activities that included a structured instructional presence as an integral part of the course design. These activities represent predictable and planned forms of synchronous student-instructor engagement and include synchronous lectures, tutorials, discussions and Q&A sessions, web-based video conferencing and live chats. However, inconsistent definitions of the instructor's presence, unclear nature and frequency of synchronous interactions, variability of research designs, discrepancies in the outcome measures and the lack of data from diverse institutions in the reviewed studies posed a major threat to the validity of results and rendered a meta-analysis unfeasible. Thus, instead of attempting a meta-analysis of a thicket of incompatible studies, we shifted our focus to writing a critical review highlighting the problems in current research on this topic. Consequently, a goal of this critical review is to spark a much-needed discussion about research quality in online learning and serve as a call to action for researchers to adopt more standardized, transparent and rigorous research and reporting methods.
Challenges: threats to validity
Operational definitions
The first challenge we encountered when synthesizing our findings was a lack of a universally accepted operational definition of synchronous instructor presence. The operational definitions ranged from those activities typically associated with the traditional course setting, such as synchronous lectures (Griffith et al., 2021; Olson & McCracken, 2015), two-way conferencing (Blau et al., 2017; Sklyar, 2009) and group discussions (Olson & McCracken, 2015), to the activities associated with online communication outside the educational settings, such as online chats (Duncan et al., 2012; Nieuwoudt, 2020; Olson & McCracken, 2015; Sklyar, 2009; Strang, 2012) and collaborative work via online platforms (Sklyar, 2009). A number of studies compared students' synchronous lecture attendance from home and other satellite locations (Bourdeau et al., 2018; Faulconer et al., 2018; Roberts et al., 2019). Often, a single course incorporated different types of synchronous components (Nieuwoudt, 2020; Olson & McCracken, 2015; Strang, 2012; Sklyar, 2009).
Research design
In addition to the inconsistencies with operational definitions, the variability of study designs further confounded our ability to draw conclusions about the impact of synchronous instructor presence on student learning. While the majority of the studies compared synchronous and asynchronous course components using a between-subject design (Bourdeau et al., 2018; Falconer et al., 2018; Griffith et al., 2021; Roberts et al., 2019; Strang, 2012), a small number of studies used a within-subject design (Duncan et al., 2012; Sklyar, 2009). Furthermore, three studies used three different methodological approaches: Blau et al. (2017) compared two types of synchronous components, Olson and McCracken (2015) compared an asynchronous-only component with a component containing both asynchronous and synchronous interactions, and Nieuwoudt (2020) reported only a correlation between synchronous class attendance and grades.
It is important to emphasize that the various research designs used in the studies we reviewed reflect the complexity and richness of the online course design landscape. As such, each of these designs has its merits in research. However, the limited number of methodologically sound studies utilizing different methodological approaches that cannot be grouped and analyzed together hindered our ability to compare study results and draw meaningful conclusions.
Measurement of outcomes
Students' final course grades (Blau et al., 2017; Duncan et al., 2012; Griffith et al., 2021; Nieuwoudt, 2020; Olson & McCracken, 2015; Sklyar, 2009; Strang, 2012) and grade distributions (Bordeau et al., 2018; Faulconer et al., 2018; Roberts et al., 2019), reported as the outcome measures in the reviewed studies, provide valuable information about the impact of synchronous instructor presence on students' performance (Deschacht & Goeman, 2015; Lewin, 2021). However, these two groups of studies are incomparable because final grades and grade distributions are different measures of academic performance. Final course grades are a composite measure of students’ performance. On the other hand, the grade distributions highlight the frequency of different grades for multiple modalities within the course. Thus, the final course grade provides an aggregated measure of the overall course achievement, while a grade distribution presents a broader picture of the variance in students’ performance within the course. Thus, the final course grade and the grade distribution inherently provide different kinds of information, and findings based on one measure cannot be directly compared to the findings based on the other.
Single institution
Finally, several studies in this review were conducted within a single institution using the same research methodology (Bourdeau et al., 2018; Faulconer et al., 2018; Griffith et al., 2021; Roberts et al., 2019). These studies offer greater methodological consistency, allowing for a more reliable comparison of the outcomes across studies. However, these studies were all conducted with a specific population of students at an aeronautical university, which raises concerns about the generalizability of these findings. Without replication of these results across different institutions and student populations, it is impossible to conclude whether the insights gained from these studies accurately reflect the effects of synchronous instructor presence on learning in different educational contexts or represent a reflection of the homogeneous institutional culture.
Opportunities
The challenges we encountered in our research underscore the urgent need for a robust discussion about the quality of the research methodologies and data reporting in the field. These challenges are often rooted in the use of various technological solutions and platforms (Sato et al., 2024), different learning activities and types of interactions in online courses (Sato et al., 2024; Yount & Tandoh, 2016) and the diversity of course structures and teaching practices within different pedagogical models of online education (Picciano, 2017). However, the unexpectedly high number of publications on the topic that we excluded from the analysis, and our inability to synthesize results from a handful of studies we reviewed, indicate that the research in the field lacks methodological rigor, clarity and transparency in data analysis and reporting. Implementing high methodological and reporting standards will promote effective evidence-based teaching practices.
The first step towards improving the research methodology is the widespread adoption of unambiguous operational definitions of synchronous instructor presence and measurement procedures (Slife, Wright, & Yanchar, 2016). Without a clear definition of what synchronous instructor presence involves, with respect to variety, type, quality, frequency and duration of the student-instructor interaction, it is impossible to effectively compare its impact across different studies. Qualitative studies using thematic analysis (Belt and Patrick, 2023) suggest that instructors consistently agree that synchronous interaction benefits their students' learning and community building. They reported using synchronous communication technologies in various ways, but there was no unifying or dominant use of synchronous interaction. However, all instructors agreed that the success of these interactions depended on both situational factors and student demographics (see also Garrison & Arbaugh, 2007; Garrison, Anderson & Archer, 2010; Caskurlu, 2018).
It is important to acknowledge that controlling for every variable in the continuously evolving realm of online education (as in controlled experimental research) is not possible. However, the unique nature of the research in the field should not be viewed as a justification for eschewing high-quality research. On the contrary, it underscores the importance of pursuing scientific rigor and high publishing standards. That includes selecting the appropriate methodology, reporting sufficient details about it and correctly reporting the results (Lodge, Corrin, Huijser & Han, 2024). Thus, a large number of well-designed studies, with unambiguous operationalization of variables, outcome measures and replicable study designs would enable the kind of comparative analysis that was beyond reach in this review. This progression would deepen our understanding of the instructor's presence in online teaching and enhance our educational practices (Caskurlu et al., 2020; Richardson et al., 2024). The current lack of evidence restricts the ability of educational institutions to make informed decisions about integrating synchronous teaching methods into their online courses. It is the responsibility of all researchers to conduct more systematic and comparable studies to inform the most effective, evidence-based educational practices across various educational contexts.
