Table 2 Literature Review: Summary of...

Table 2

Literature Review: Summary of Research Articles on Class Size in Online Courses

Authors/Source	Class Size Recommendation	Notes
Andersen and Avery (2008)	No size recommendation. Found that actual teaching time in hours per credit hour for web-based courses = 46.1 hours (SD 16.7); actual teaching time in hours per credit hour for face-to-face (F2F) courses = 39.4 hours (SD 13.2).	Descriptive study utilized time records from 11 web-based and five face-to-face graduate-level nursing courses in one large U.S. Midwestern university. Study focused on “direct teaching time” only; omitted inclusion of course development and preparation time during the semester.
Arbaugh and Benbunan-Fich (2005)	A study reported in this chapter was that class section sizes (of 30 or more students) were negatively associated with student learning.	Different class sizes create different group dynamics. Large classes are more impersonal and less individualized. Additional research needed to identify optimal student-instructor ratios.
Arbaugh (2005)	Study found no support for greater student satisfaction or self-perceived student learning with class sizes of 30 or less.	Characteristics of web-based instruction were predicted to impact two dependent variables: student satisfaction with the delivery medium, and self-perceived student learning. Sample: 40 web-based MBA classes over 11 semesters, taught by 16 different faculty, in a single Midwestern university, 1998-2001; enrollment range was 9-31 students/course.
Ascough (2002)	The author found that the workload for online course delivery is 50% to three times more work to design and run than teaching in the classroom.	This is an informational article on suggestions for designing and managing an online course, including considerations of faculty workload.
Berry (2008)	Asynchronous discussion group size should be limited to 4-9 students.	Conference presentation. Recommendations are based on author's literature review and experiences himself and with other faculty at the American Public University System.
Blood-Siegfried et al. (2008)	Indicated discussion groups should be limited to 10-12 students/group. Authors' university limits all online courses to a maximum of 25 students, perceived as necessary for effective student/faculty communication.	A group of six faculty members at a single university received funding to determine best practices in online courses. The group developed an evaluation rubric to measure quality in the graduate online curriculum, then applied the rubric to the core courses required of all graduate students. Concepts deemed of high importance to online education: learner-centeredness faculty-student interaction student-student interaction
Brook and Oliver (2003)	Asynchronous environments: limit group size to 25. Synchronous environments: limit group size to 10.	The number of participants will influence class community development strategies in online courses.
Buckingham (2003)	For online discussions, recommends keeping the size of discussion groups to 6-10 students, with one faculty assigned to a group.	Article focused on key elements of quality online education: course design, research design, class size, levels of computer skills, technological support, and timeliness of feedback.
Burruss, Billings, Brownrigg, Skiba, and Connors (2009)	Findings complex and mixed. For both graduate and undergraduate students: Student-faculty interaction-more exists in smaller classes; Peer interaction-more exists in larger classes; better online than F2F; Diverse learning-less opportunity in larger classes. Graduate students: Professionalism [defined as the extent to which students “believe the content and concepts of the course prepare them for professional practice and for acquiring the values of the [nursing] profession” (p. 39)—better in smaller graduate classes (< 20 students). Satisfaction: satisfaction decreased as class size grew; found the asynchronous conversations in larger classes unwieldy. Connectedness: less in very large classes (> 40 students). Social presence: higher in smaller graduate classes (< 20 students). Undergraduate students: Satisfaction: size increase from small to medium produced higher satisfaction (> 20 students). Social presence: class size increase from small to medium produced higher social presence (> 20 students).	This exploratory descriptive study examined class size in relation to the use of technology and to particular educational practices and outcomes. The sample consisted of undergraduate (n = 265) and graduate (n = 863) students enrolled in fully web-based courses. Class sizes were defined as very small (1 to 10 students), small (11 to 20 students), medium (21 to 30 students), large (31 to 40 students), and very large (41 students and above). Study reported more negative outcomes with very small or very large class sizes. Small, medium and large classes registered some differences in student satisfaction. Need fewer students enrolled in courses using active learning strategies. There were big differences between undergraduate and graduate course student outcomes (see middle column).
DiBiase and Rademacher (2005)	No specific class sizes recommended. Author asserts that economies of scale are possible (to some extent) with increasing enrollments in distance courses. This study reported a 12% gain in efficiency when moved from 18 to 49 students. Most time-consuming instructor activities: 40-55% of instructor effort: communications with students by threaded discussion and email, a consistently large share of effort, 34-36% of instructor effort: Student assessment and feedback.	Research explores scalability in distance learning. The authors studied time spent teaching a course in geography eight times over a 3.5 year period. When faculty increased the average class size by a factor of 2.7 (from 18 to 49 students), their course-related workloads increased by a factor of about 2.5 (from 47.5 hours to 116.7 hours total), and average faculty time spent per student dropped from 3.2 to 2.4 hrs. Student satisfaction with the course was high overall and suffered no significant decline as a result of larger classes and increased instructional efficiency. Key variables exerting the most influence on teaching efficiency and student satisfaction: stability of the subject matter, instructor experience and knowledge, pedagogical approach, level of institutional support, and student maturity. Concludes that courses requiring the least instructor effort [objectivist teaching strategies] are the most scalable and efficient but may not be effective for satisfying students.
Drago and Peltier (2004)	Authors indicate that as the number of students increases in online classes, it becomes more time-consuming for the instructor to deal with issues as they come up. However, results from this study found that larger class sizes did not predict studentperceived course effectiveness, value of the course content, instructor support, better course structure or interactions; for two dependent variables, large class size predicted positive student perceptions. Authors conclude that online education quality factors important to students do not depend on class size.	Studied MBA courses, including 31 taught online, during an academic year at a large, regional university. Used class size (range of 22-83 students) as the independent variable; dependent variables representing elements of online effectiveness included: course content, instructor support, course structure, student-to-student interaction, instructor-to-student interaction, and global course effectiveness. Results indicated no significant relationship between class size and global course effectiveness. Class size showed some significance in predicting instructor support and course structure, but unexpectedly the direction of this association was positive.
Dykman and Davis (2008)	The quality fulcrum is around 20 students. For a new course or a new instructor, 15-20 students in an online class are ideal. For an established online class with an experienced online teacher, 25-30 students might be a workable number. Online classes may be too large (> 30) or too small < 10-12). Large courses can be taught online, but they should call for less student contact with the instructor. Large classes demand that the fullness of the pedagogy be limited: fewer deliverables, much less feedback for students, much more of the “sink or swim” mentality for students.	Author comments: “[T]he size of the class determines the learning objectives and course design. An online course designed for 15-20 students is necessarily a different kind of course from one designed for 35 or 50. It is unwise to take a class designed for 20 students and enroll 40-50 students in it. When it comes to online education, one size does not fit all!” “Nothing is more destructive to online student motivation than a faculty member who is not interacting with them.”
Keeton (2004)	Optimal class size is 20 students.	Article reports on part of a larger study about best practices in online instruction; ultimate goal of research is to help faculty improve the quality of their teaching. Studied eight faculty who have each taught an average of 16 online courses. Measures of effective online teaching practice used in study were based on 20 years of best practices research in F2F instruction.
Kenny (2002)	Recommends small group sizes with similar skill level to enhance learning.	Author conducted a small qualitative study to explore the experiences of Australian students with online learning. Used individual interviews and focus group interviews of 21 students enrolled in a health informatics course.
May and Short (2003)	Reports on studies showing larger online classes (e.g., 50 students) are negatively associated with students' perceived learning and satisfaction (overcrowding).	Report on class sizes is from authors' own literature review.
Orellana (2006)	Actual class size of the surveyed 131 faculty was 22.8; a class size of 18.9 was perceived as optimal to better achieve the course's intended level of interaction; and a class size of 15.9 was perceived as optimal to achieve the highest level of interaction.	Article presents findings of a study conducted to determine instructors' perceptions of optimal class sizes for online courses with different levels of interaction. A web-based survey method was employed. Online courses studied were those taught sometime in the last 5 years by a single instructor in undergraduate or graduate programs from U.S. colleges. Instructors described the level of interactive qualities in their most recently taught online course. Used a version of Roblyer and Wiencke's (2004) rubric for assessing interactive qualities in distance courses.
Rovai (2002)	Eight to 10 students are a reasonable estimate for the minimum critical mass needed to promote good interactions. At the opposite end of the continuum, 20-30 students are the most learners that a single online instructor can reasonably handle in a single class—if it contains active discussions.	Class size in online environments strongly influences learning activities. Too few members generate little interaction, and too many members generate a sense of being overwhelmed. Large classes (> 30 students) can be managed by team teaching or use of a teaching assistant.
Schellens and Valcke (2006)	Reviews literature, indicates that most researchers suggest a group size of 10-12 participants. Large group sizes (> 12) cause participants to deal with too high a number of messages and tend to invoke extraneous cognitive load. Larger groups require well-designed learning-oriented task structures. Study found large groups invoke more communication about nonsense topics and technical issues.	Study of 300 students working over 6 months in 38 electronic discussion groups. Purpose was to analyze knowledge construction and discourse in collaborative learning. The results clearly indicated that group size affects the types, structure and phase of knowledge construction in asynchronous discussion groups. Smaller- and average-sized groups performed at a quantitatively and qualitatively higher level.
Tomei (2006)	Online teaching demanded a minimum of 14% more time than traditional instruction. The calculation of ideal class size for the online format was 12 students.	This study examined the impact of substituting traditional F2F courses with distance-based education. Had a small sample of two classes, each 11 students: one taught online, the other F2F. Compared the time demands for each, analyzing the impact of distance learning demands on faculty teaching loads, and computed the ideal class size for an online course.
Visser (2000)	Found that the total work hours for developing and delivering a graduate-level course in public administration were twice as high for distance education as for traditional courses.	A case study of author's own work expenditure developing and delivering an online course. Conjectures that the accumulation of instructor experience in distance learning, the level of institutional support, and the support of a technically knowledgeable graduate assistant may help decrease the faculty's work expenditure.

Authors/Source	Class Size Recommendation	Notes
Andersen and Avery (2008)	No size recommendation. Found that actual teaching time in hours per credit hour for web-based courses = 46.1 hours (SD 16.7); actual teaching time in hours per credit hour for face-to-face (F2F) courses = 39.4 hours (SD 13.2).	Descriptive study utilized time records from 11 web-based and five face-to-face graduate-level nursing courses in one large U.S. Midwestern university. Study focused on “direct teaching time” only; omitted inclusion of course development and preparation time during the semester.
Arbaugh and Benbunan-Fich (2005)	A study reported in this chapter was that class section sizes (of 30 or more students) were negatively associated with student learning.	Different class sizes create different group dynamics. Large classes are more impersonal and less individualized. Additional research needed to identify optimal student-instructor ratios.
Arbaugh (2005)	Study found no support for greater student satisfaction or self-perceived student learning with class sizes of 30 or less.	Characteristics of web-based instruction were predicted to impact two dependent variables: student satisfaction with the delivery medium, and self-perceived student learning. Sample: 40 web-based MBA classes over 11 semesters, taught by 16 different faculty, in a single Midwestern university, 1998-2001; enrollment range was 9-31 students/course.
Ascough (2002)	The author found that the workload for online course delivery is 50% to three times more work to design and run than teaching in the classroom.	This is an informational article on suggestions for designing and managing an online course, including considerations of faculty workload.
Berry (2008)	Asynchronous discussion group size should be limited to 4-9 students.	Conference presentation. Recommendations are based on author's literature review and experiences himself and with other faculty at the American Public University System.
Blood-Siegfried et al. (2008)	Indicated discussion groups should be limited to 10-12 students/group. Authors' university limits all online courses to a maximum of 25 students, perceived as necessary for effective student/faculty communication.	A group of six faculty members at a single university received funding to determine best practices in online courses. The group developed an evaluation rubric to measure quality in the graduate online curriculum, then applied the rubric to the core courses required of all graduate students. Concepts deemed of high importance to online education: learner-centeredness faculty-student interaction student-student interaction
Brook and Oliver (2003)	Asynchronous environments: limit group size to 25. Synchronous environments: limit group size to 10.	The number of participants will influence class community development strategies in online courses.
Buckingham (2003)	For online discussions, recommends keeping the size of discussion groups to 6-10 students, with one faculty assigned to a group.	Article focused on key elements of quality online education: course design, research design, class size, levels of computer skills, technological support, and timeliness of feedback.
Burruss, Billings, Brownrigg, Skiba, and Connors (2009)	Findings complex and mixed. For both graduate and undergraduate students:Student-faculty interaction-more exists in smaller classes; Peer interaction-more exists in larger classes; better online than F2F; Diverse learning-less opportunity in larger classes. Graduate students: Professionalism [defined as the extent to which students “believe the content and concepts of the course prepare them for professional practice and for acquiring the values of the [nursing] profession” (p. 39)—better in smaller graduate classes (< 20 students). Satisfaction: satisfaction decreased as class size grew; found the asynchronous conversations in larger classes unwieldy. Connectedness: less in very large classes (> 40 students). Social presence: higher in smaller graduate classes (< 20 students). Undergraduate students: Satisfaction: size increase from small to medium produced higher satisfaction (> 20 students). Social presence: class size increase from small to medium produced higher social presence (> 20 students).	This exploratory descriptive study examined class size in relation to the use of technology and to particular educational practices and outcomes. The sample consisted of undergraduate (n = 265) and graduate (n = 863) students enrolled in fully web-based courses.Class sizes were defined as very small (1 to 10 students), small (11 to 20 students), medium (21 to 30 students), large (31 to 40 students), and very large (41 students and above).Study reported more negative outcomes with very small or very large class sizes. Small, medium and large classes registered some differences in student satisfaction. Need fewer students enrolled in courses using active learning strategies. There were big differences between undergraduate and graduate course student outcomes (see middle column).
DiBiase and Rademacher (2005)	No specific class sizes recommended. Author asserts that economies of scale are possible (to some extent) with increasing enrollments in distance courses. This study reported a 12% gain in efficiency when moved from 18 to 49 students. Most time-consuming instructor activities: 40-55% of instructor effort: communications with students by threaded discussion and email, a consistently large share of effort, 34-36% of instructor effort: Student assessment and feedback.	Research explores scalability in distance learning. The authors studied time spent teaching a course in geography eight times over a 3.5 year period. When faculty increased the average class size by a factor of 2.7 (from 18 to 49 students), their course-related workloads increased by a factor of about 2.5 (from 47.5 hours to 116.7 hours total), and average faculty time spent per student dropped from 3.2 to 2.4 hrs. Student satisfaction with the course was high overall and suffered no significant decline as a result of larger classes and increased instructional efficiency. Key variables exerting the most influence on teaching efficiency and student satisfaction: stability of the subject matter, instructor experience and knowledge, pedagogical approach, level of institutional support, and student maturity.Concludes that courses requiring the least instructor effort [objectivist teaching strategies] are the most scalable and efficient but may not be effective for satisfying students.
Drago and Peltier (2004)	Authors indicate that as the number of students increases in online classes, it becomes more time-consuming for the instructor to deal with issues as they come up. However, results from this study found that larger class sizes did not predict studentperceived course effectiveness, value of the course content, instructor support, better course structure or interactions; for two dependent variables, large class size predicted positive student perceptions. Authors conclude that online education quality factors important to students do not depend on class size.	Studied MBA courses, including 31 taught online, during an academic year at a large, regional university.Used class size (range of 22-83 students) as the independent variable; dependent variables representing elements of online effectiveness included: course content, instructor support, course structure, student-to-student interaction, instructor-to-student interaction, and global course effectiveness. Results indicated no significant relationship between class size and global course effectiveness. Class size showed some significance in predicting instructor support and course structure, but unexpectedly the direction of this association was positive.
Dykman and Davis (2008)	The quality fulcrum is around 20 students. For a new course or a new instructor, 15-20 students in an online class are ideal. For an established online class with an experienced online teacher, 25-30 students might be a workable number.Online classes may be too large (> 30) or too small < 10-12). Large courses can be taught online, but they should call for less student contact with the instructor. Large classes demand that the fullness of the pedagogy be limited: fewer deliverables, much less feedback for students, much more of the “sink or swim” mentality for students.	Author comments: “[T]he size of the class determines the learning objectives and course design. An online course designed for 15-20 students is necessarily a different kind of course from one designed for 35 or 50. It is unwise to take a class designed for 20 students and enroll 40-50 students in it. When it comes to online education, one size does not fit all!”“Nothing is more destructive to online student motivation than a faculty member who is not interacting with them.”
Keeton (2004)	Optimal class size is 20 students.	Article reports on part of a larger study about best practices in online instruction; ultimate goal of research is to help faculty improve the quality of their teaching. Studied eight faculty who have each taught an average of 16 online courses. Measures of effective online teaching practice used in study were based on 20 years of best practices research in F2F instruction.
Kenny (2002)	Recommends small group sizes with similar skill level to enhance learning.	Author conducted a small qualitative study to explore the experiences of Australian students with online learning. Used individual interviews and focus group interviews of 21 students enrolled in a health informatics course.
May and Short (2003)	Reports on studies showing larger online classes (e.g., 50 students) are negatively associated with students' perceived learning and satisfaction (overcrowding).	Report on class sizes is from authors' own literature review.
Orellana (2006)	Actual class size of the surveyed 131 faculty was 22.8; a class size of 18.9 was perceived as optimal to better achieve the course's intended level of interaction; and a class size of 15.9 was perceived as optimal to achieve the highest level of interaction.	Article presents findings of a study conducted to determine instructors' perceptions of optimal class sizes for online courses with different levels of interaction. A web-based survey method was employed. Online courses studied were those taught sometime in the last 5 years by a single instructor in undergraduate or graduate programs from U.S. colleges. Instructors described the level of interactive qualities in their most recently taught online course. Used a version of Roblyer and Wiencke's (2004) rubric for assessing interactive qualities in distance courses.
Rovai (2002)	Eight to 10 students are a reasonable estimate for the minimum critical mass needed to promote good interactions. At the opposite end of the continuum, 20-30 students are the most learners that a single online instructor can reasonably handle in a single class—if it contains active discussions.	Class size in online environments strongly influences learning activities. Too few members generate little interaction, and too many members generate a sense of being overwhelmed.Large classes (> 30 students) can be managed by team teaching or use of a teaching assistant.
Schellens and Valcke (2006)	Reviews literature, indicates that most researchers suggest a group size of 10-12 participants. Large group sizes (> 12) cause participants to deal with too high a number of messages and tend to invoke extraneous cognitive load.Larger groups require well-designed learning-oriented task structures. Study found large groups invoke more communication about nonsense topics and technical issues.	Study of 300 students working over 6 months in 38 electronic discussion groups. Purpose was to analyze knowledge construction and discourse in collaborative learning. The results clearly indicated that group size affects the types, structure and phase of knowledge construction in asynchronous discussion groups.Smaller- and average-sized groups performed at a quantitatively and qualitatively higher level.
Tomei (2006)	Online teaching demanded a minimum of 14% more time than traditional instruction. The calculation of ideal class size for the online format was 12 students.	This study examined the impact of substituting traditional F2F courses with distance-based education. Had a small sample of two classes, each 11 students: one taught online, the other F2F. Compared the time demands for each, analyzing the impact of distance learning demands on faculty teaching loads, and computed the ideal class size for an online course.
Visser (2000)	Found that the total work hours for developing and delivering a graduate-level course in public administration were twice as high for distance education as for traditional courses.	A case study of author's own work expenditure developing and delivering an online course. Conjectures that the accumulation of instructor experience in distance learning, the level of institutional support, and the support of a technically knowledgeable graduate assistant may help decrease the faculty's work expenditure.

[ViewLarge]