Abstract
Lecturing has been criticized for fostering a passive learning environment, emphasizing a one-way flow of information, and not adequately engaging students. In contrast, active-learning approaches, such as team-based learning (TBL), prioritize student interaction and engagement and create multidirectional flows of information. This paper presents an exploratory analysis of whether lecturing or TBL was better for teaching content; developing skills, such as critical thinking; and creating an enjoyable learning environment in a sociology course. Results showed few differences in these outcomes between groups taught with the two different methods, although TBL students thought they improved their oral communication and creative thinking skills more than the lecture students, and they reported getting to know their professor and classmates more than did the lecture students.
The changing college fiscal environment has put pressure on schools to increase enrollment while simultaneously cutting pedagogical costs. While large lecture classes may seem like a simple solution to diminishing resources, this method of teaching has been criticized as resulting in poor attendance, high student apathy, and passive and impersonal learning environments (Fertig 2010; Vevea and Harris 2011). In contrast, active-learning techniques, such as team-based learning (TBL), utilize the “flipped classroom,” wherein content delivery is pushed out of class time in favor of concept engagement. These methods have been gaining attention in the higher education literature as alternative teaching methods that can combat some of the problems of lecturing (e.g., Horn 2013; Rosenberg 2013; Tucker 2012). Little empirical research, however, has been conducted to test whether or not these newer methods produce better learning outcomes compared to lecturing. This paper presents an exploratory analysis of whether students perform better in a sociology course taught by lecturing versus the same course taught with an active-learning method called TBL.
Lecturing in college classes can take many forms, but it is generally characterized as a didactic approach to teaching that relies on an expert (lecturer) imparting knowledge to a listener (student). Its primary benefits are (1) efficiency, because it can be easily implemented in very large classes; (2) immediacy, as the instructor’s performance can impress students in the moment; and (3) control over the content, flow, and class environment (Bauerlein 2011; Jones 2007). Skeptics, however, criticize the one-way flow of information from lecturer to student and the perceived lack of engagement with students due to either poor attendance or fleeting attention during class. To counter these limitations, some instructors have incorporated more visual aids, such as videos or PowerPoints, personal response systems, Twitter feeds, or think-pair-share activities, to make lectures more engaging and interactive. Empirical evaluations of the effectiveness of such enhancements, however, have found mixed results (Fies and Marshall 2006; Filer 2010; Hill et al. 2012; Stowell and Nelson 2007; Svinicki and McKeachie 2010).
In contrast, active-learning approaches prioritize student interaction and engagement to create multidirectional flows of information. Learning core content is still important, but it is either transferred outside of the classroom in the flipped model or delivered primarily through activities rather than lecturing. TBL is one such technique that uses the flipped-classroom model and active-learning philosophy to promote learning through frequent interactions. Students learn basic course content on their own through readings and/or videos or other media; then most of the class time is spent working on activities in stable, small groups. The instructor’s role focuses less on delivering content and more on designing effective team activities and facilitating discussions within teams, between teams, and with the class as a whole. The benefits of this approach are (1) fostering independent learning and personal accountability, (2) high levels of interaction and frequent feedback for both students and instructors, (3) actively practicing critical and creative thinking skills, and (4) advancing soft skills, such as communication, deliberation, and decision making (Michaelsen, Knight, and Fink 2004; Michaelsen and Sweet 2008).
Evaluations of TBL to date have been sparse, but there is documented evidence of both instructor and student satisfaction with the method (e.g., Harde 2011; Jacobson 2011; Kühne-Eversmann, Eversmann, and Fischer 2008; Restand 2012; Stamatel, Bushway, and Roberson 2013). At the same time, there are start-up costs for instructors (who have to learn the new method) and there can be student resistance to the active-learning environment (Stamatel, Bushway, and Roberson 2013). Thus, the question for most instructors who are considering changing teaching methods is whether or not active-learning approaches, such as TBL, produce better learning outcomes than lecturing.
There have been a few empirical tests of outcomes in TBL classes versus more traditional methods in the health and medical sciences, but to our knowledge, none to date have been performed in the social sciences. These studies have shown better results for the TBL classes than their comparison groups, but there were some clear methodological limitations in each of these studies (Carmichael 2009; Levine et al. 2004; Sisk 2011). Given the bureaucratic challenges of course scheduling, teaching assignments, and student enrollments, it is very difficult to implement randomized, controlled experiments of teaching methods and learning outcomes in higher education. We were not immune to these structural barriers when designing this study. This exploratory comparison of learning outcomes and student satisfaction from a sociology course taught by lecturing versus TBL is an attempt to collect empirical evidence of this kind for social sciences courses.
Hypotheses
Each of the authors of this study had been teaching different sections of a 300-level sociology course called Introduction to Crime, Law, and Deviance at a large, state university in the East South Central region for several semesters prior to conducting this evaluation, with one instructor using the lecture model and the other using TBL. In order to develop hypotheses for this evaluation, we first identified three common learning objectives that we wanted students to achieve in this course regardless of teaching method:
Understand the meaning of core concepts of a sociological student of crime, law, and deviance. For example, what is the difference between deviance and crime? What is social stratification?
Draw informed conclusions about patterns of crime and victimization. For example, what do we know about the gender composition of criminal offenders? Why is that the case? What does it mean that young people are “overrepresented” in violent crime statistics?
Compare, contrast, and apply theories of criminal behavior. For example, how would the 1999 shootings at Columbine High School be explained by a neoclassical theorist versus a social process theorist?
In order to achieve these learning goals, students in each section of the course were assigned the same textbook readings, but how content was delivered and how skills were developed varied by teaching method. For example, the section on theoretical explanations of criminal behavior starts with classical and neoclassical perspectives. In the lecture sections, the instructor used a PowerPoint-enhanced lecture to cover the assumptions of classical and neoclassical theories, the history of societal responses to crime, and foundational thinkers, followed by a more detailed exploration of two popular neoclassical theories (rational choice and routine activities theories). While there were some interactions (questions asked and answered by the instructor or students), the bulk of the class consisted of a one-way flow of information. Students listened, took notes, and were expected to read the associated textbook chapter outside of class.
In the TBL sections, students were first assigned the reading from the textbook to complete outside of class, and then they took a Readiness Assessment Test (RAT) on the date the reading was due, prior to any lecturing by the instructor. The RAT was a multiple-choice quiz that was first taken by students individually, and then the same quiz was repeated as teams (Michaelsen et al. 2004). The RAT was designed to test basic comprehension, hold students accountable for reading on their own, generate discussion about course content within teams, and provide feedback to the instructor regarding difficult concepts that need further elaboration (Sibley and Ostafichuk 2014). Next, the instructor gave a 10-minute lecture providing an organizational framework for the unit on crime theories, discussing major differences across bodies of theories, and then situating the classical and neoclassical theories within this framework. Students were given an opportunity to ask questions, and then they were assigned a case study to analyze within their teams. They had to choose one of three neoclassical theories that best explained the case and provide a rationale for why their choice was best. When the teams finished, the class compared answers across teams and had a discussion about the strengths and limitations of the three theories presented in the assignment.
As this example shows, both approaches require students to learn new content, but they differ in how it is delivered. Because lecturing offers more control over how the content is delivered and provides more overlap between the written content and the oral presentation, we expected students in the lecture section to have a slight advantage over TBL students in terms of mastering the first two learning objectives (understanding core concepts and drawing informed conclusions about patterns of criminal behavior). In the TBL course, lectures are brief and do not cover all of the content in the textbook, so students who do not complete all of the reading on their own are at a distinct disadvantage compared to students in the lecture section who do not do the required reading but attend the lectures. As such we hypothesized the following:
In contrast, we expected students in the TBL sections to better accomplish the third learning objective compared to the lecture sections because they frequently practice comparing, contrasting, and applying through regular activities, whereas these skills are typically demonstrated in the lecture environment but less frequently exercised.
In addition to knowledge acquisition, we wanted to see how students fared with skill development in the two learning environments. TBL proponents argue that one of the strengths of the method is an emphasis on skill development, such as writing, oral communication, critical thinking and creative thinking. This is argued to be more pronounced as compared to more passive lecture environments. Similar to the second hypothesis, we predicted that this would also be the case with our students, given more frequent opportunities to practice these skills in an active-learning environment compared to the lecture classes.
In addition to comparing learning outcomes, we also wanted to gauge student satisfaction with their respective learning environments. In this study, students were expected to perform the same tasks outside of the classroom, essentially reading the textbook, but the TBL class had more frequent measures of accountability, which might make students feel like they are working harder than they do in other types of courses. They also have to do more than listen during class time; therefore, we hypothesized the following:
On the more positive side, TBL instructors often receive favorable comments on evaluations stating that the course is fun due to the high levels of interactivity and collaborative atmosphere and that they enjoy working with other students (Stamatel, Bushway, and Roberson 2013). Instructors who lecture also get both positive and negative feedback on evaluations, but our experience has been that the positive comments have been more enthusiastic in the TBL course evaluations. Therefore, we wanted to test this difference more systematically by asking students in both courses to rate their experiences.
Methods
We collected data from undergraduate students enrolled in a 300-level sociology course called Introduction to Crime, Law, and Deviance at a large, state university in the fall of 2012 and the spring of 2013. During each semester, two sections of the course were taught: one using lectures and the other using TBL. The lecture sections were taught by one instructor, and the TBL sections were taught by another.
We were not able to randomize assignments to the sections, so students self-selected based on scheduling and word-of-mouth knowledge of the instructors and the course expectations, as is typical for college students. Students were not provided with information about teaching methods in the course catalogue, but this is introduced on the first day of classes before the drop/add period ends so students have the opportunity to drop the class or change sections if they are unhappy with any aspect of the course, as is the case for any other college course. This course is one of the more popular ones in our department, so it is possible that the reputations of the instructors, including their teaching styles, were tied to student preferences about which section to select.
There were 117 students enrolled in the two sections taught with lecturing and 127 students enrolled in the two sections taught with TBL. In all sections, few students had declared a sociology major or minor, and there were slightly more women than men. There were twice as many first- and second-year students in the TBL sections than in the lectures, but they were still less than a quarter of the sample. Self-reported grade point averages were similar across the sections.
There were other structural differences between the two sections and across semesters that we could not control. In particular, the scheduling of the sections was determined at the administrative level. The lecture class was taught once a week for 150 minutes in the fall and twice a week for 75 minutes each in the spring. The TBL sections were taught in shorter, 50-minute blocks, three times a week, both semesters.
Following our institutional review board (IRB) protocol, students were asked to voluntarily participate in this study by allowing us to compare their test scores and volunteering to participate in a brief survey at the end of the semester. There were 244 students enrolled in the four study sections, and 71.6 percent agreed to allow us to analyze their test scores for this study, but only 49 percent of the students completed the voluntary survey. This was due to both absences on the day the survey was administered and nonresponse.
The first hypothesis was tested using a short-answer exam about basic criminological knowledge, such as types of crimes, sources of crime data, and patterns of crime. The second hypothesis was tested with an essay exam requiring students to compare, contrast, and apply criminological theories. All of the material on both exams was covered in the textbook. Students in all sections of the course also received the same study guide at the same point in the semester prior to each exam in order to standardize expectations about what material would be covered.
Per our IRB protocol, we hired a PhD in sociology who had previously taught this same course at our university to grade both exams in order to eliminate potential bias in the grading and to ensure anonymity of the study participants. All students were required to take the tests as part of their course grades, but they could choose whether or not to have their grades included in this study. A graduate student recruited the undergraduates from our classes to participate in the study, removed identifying information from the students’ tests prior to grading, and maintained an anonymized data set for analysis. The instructors were not involved in the actual data collection, nor could they identify students who participated in the study or not. The grader was given an answer key for the short-answer exam and a rubric for the essay exam to standardize grading expectations between instructors and across sections.
In addition to comparing test scores, we also gathered data from a short survey of students’ perceptions of their learning experiences. The survey was administered at the end of the semester on a different day than the required university course evaluations. Students were asked to rate their mastery of the course material. The question asked, “By taking this class, I am better able to explain: patterns of crime, theories of crime, types of crime, criminal justice issues, what the field of criminology is, where our perceptions of crime come from, and structural inequalities (e.g., race, gender, and class).” They answered each item on a Likert scale of 1 (strongly disagree) to 5 (strongly agree). This question allowed us to test hypothesis 1 with subjective assessments of learning to supplement the exam scores.
Additionally, the survey asked students to rate how their course improved their skill development. “By taking this class, I have improved my ability to write, express myself verbally, synthesize course material, analyze course material, apply course material, think critically, think creatively.” Students answered on the same 5-point scale of 1 (strongly disagree) to 5 (strongly agree). Some of these items overlap with the skills tested for hypothesis 2 using the essay exam, again providing subjective data to complement the test scores and they were used to test hypothesis 3, looking more broadly at skill development.
Finally, to gauge student satisfaction with their class experiences, we asked two other sets of questions. To test hypothesis 4, we asked, “How did the workload for this course compare to your other courses this semester, other sociology courses, and other 300/400-level courses?” Students answered these questions on a scale of 1 (significantly less) to 5 (significantly more). To test hypothesis 5, we asked students to rate the following four statements on a scale of 1 (strongly disagree) to 5 (strongly agree): (1) “I feel that I got to know my professor during this course.” (2) “I feel that I got to know my classmates during this course.” (3) “I enjoyed the content of this course.” (4) “I enjoyed the way this course was taught.”
Findings
T tests were used to compare means between lecture classes and TBL sections across different measures. Hypothesis 1 predicted that students in the lecture sessions would have an advantage over TBL students in acquiring and comprehending knowledge. There was no statistically significant difference between the two groups. The mean for the lecture sections was 81.49 percent (n = 74, SD = 13.12) and it was 83.61 percent (n = 101, SD = 10.89) for the TBL students. Additionally, the student survey asked students to assess how much they thought they learned about the content of the course. The results presented in Table 1 also show no statistically significant differences between the two groups. Students in both groups performed reasonably well on the test, earning a low B on average, and felt reasonably confident in what they had learned.
Students’ Perceptions of Content Mastery and Skill Development.
Note: 1 = strongly disagree, 5 = strongly agree.
t tests statistically significant at p ≤ .05.
Hypothesis 2 predicted that students in the lecture sections would perform less well than TBL students at comparing, contrasting, and applying knowledge, as measured by an essay exam. Once again, there was no statistically significant difference between the two groups on the second exam. The mean for the lecture students was 80.78 percent (n = 74, SD = 20.91), and the mean for the TBL students was 81.78 percent (n = 101, SD = 15.29). Interestingly, the standard deviations for the TBL classes were smaller than the lecture sections, perhaps indicating that the team experience may assist students who might otherwise struggle on their own. This cannot be directly measured with our data, but it is a point worthy of further consideration.
Hypothesis 3 predicted that students in lecture classes would be less adept at certain skills than the TBL students. Table 1 shows this to be the case only for two of the seven skills measured. TBL students reported more improvement with respect to verbal expression and creative thinking than students in the lecture sections. The mean for the lecture students on verbal skill improvement was a 2.77 (n = 48, SD = 1.04) on a 5-point scale, whereas TBL students had a mean of 3.4 (n = 73, SD = 1.08). The mean for the lecture students on creative thinking improvement was 3.23 (n = 48, SD = 0.97), whereas TBL students had a mean of 3.63 (n = 73, SD = 1.15).
Hypothesis 4 predicted that students in TBL classes believed they worked harder than students in lecture classes. Table 2 shows support for this hypothesis from the student survey data when students compared their workloads for the classes we were studying to other courses they were taking that semester. Lecture students had a mean of 2.54 (n = 48, SD = 0.87) on a 5-point scale, whereas TBL students had a mean of 3.04 (n = 73, SD = 0.70). The hypothesis was also supported when comparing the study courses to other 300- and 400-level courses. Lecture students had a mean of 2.5 (n = 48, SD = 0.80) and TBL students had a mean of 2.79 (n = 73, SD = 0.71). There was no difference comparing the study courses with other sociology courses.
Students’ Perceptions of Workload and Class Environment.
1 = significantly less, 5 = significantly more.
1 = strongly disagree, 5 = strongly agree.
t tests statistically significant at p ≤ .05.
Finally, the last hypothesis predicted that students in the TBL classes would enjoy the class atmosphere more than students in the lecture sections. This was true to the extent that TBL students were more likely to agree that they got to know their professors during the course. Lecture students had a mean of 3.21 (n = 48, SD = 0.99), and TBL students had a mean of 3.55 (n = 73, SD = 1.09) on a 5-point scale. There was also a statistically significant difference in ratings of how well students got to know their classmates during the course. Lecture students had a mean of 2.23 (n = 48, SD = 1.04), and TBL students had a mean of 4.48 (n = 73, SD = 0.71) on this measure. There were not statistically significant differences on the measures of how much students enjoyed the course content or the teaching method.
Discussion and Conclusions
With increasing pressures for college instructors to “do more with less” in conjunction with demands for more evidence of effective teaching, some instructors are turning to teaching methods other than lecturing to improve learning in the context of growing class sizes and diminishing resources. TBL is one such approach that incorporates several recent trends in higher education, including backward engineering, active learning, and the flipped classroom; however, there have been very few empirical tests of whether TBL produces better learning achievements than lecturing. This exploratory study sought to address this gap in the literature by systematically comparing learning achievements and students’ perceptions of their learning experiences between two groups of students in a 300-level sociology course, one group taught with lecturing and the other with TBL.
What we found from this study is that one method is not necessarily better than the other for many of the items that we compared. Most importantly, both objective assessments of students’ mastery of core content and their own assessments of what they learned showed very few statistically significant differences between teaching methods. In fact, the test scores and subjective assessments showed that both approaches were successful for teaching content and developing skills, like synthesis, analysis, and application.
We did find a few differences in students’ perceptions of their learning and their satisfaction with their courses. TBL students perceived more improvements in verbal expression and creative thinking relative to students in the lecture classes, which is likely due to the active-learning style that fosters high levels of participation and the fact that application activities often involve problem-solving tasks that require both critical and creative thinking. This is exactly the kind of skill improvement that proponents of TBL cite as unlikely to occur in the environment of the traditional lecture class. Because the communication tends to be instructor to student, even if content mastery is the same, lectures do not provide the opportunity for students to practice the ability to verbally express themselves. In a sense, lectures “silence” students, not allowing them to develop a “voice.” Additionally, TBL students reported stronger relationships with the instructor and other students, reflecting the high levels of interactivity and continual feedback that are central to the TBL method. Importantly, TBL students also believed that their workload was higher than the students in the lecture classes. This could be an important barrier to creative teaching methods, as they often require student buy-in to be successful. College students are exposed to lecturing frequently, so the expectations are usually understood. With other teaching methods, particularly those utilizing the flipped classroom and active learning, instructors have to prepare students for different expectations associated with new learning environments.
These results need to be interpreted cautiously as there were some confounding factors for which we could not control in this research design due to structural limitations regarding course assignments and student enrollments. For example, both instructors have strong teaching records so that weaknesses in a teaching method could be compensated with a highly effective instructor. Second, because students self-selected into the course sections, they could have chosen the teaching environment that better suited their learning style. Third, the exams we administered may not have been refined enough to capture significant differences in long-term learning goals, and intense studying just prior to the exam could account for similar test scores as opposed to how the material was taught by the instructors. Finally, the response rate for the survey was low. We administered the survey on only one day at the end of the semester, so low attendance negatively affected our data collection.
While the findings of this exploratory analysis are interesting, we acknowledge that we need more controlled comparisons of teaching methods to get a better sense of the conditions under which each might work more effectively. We also need to think more carefully and creatively about the outcomes we wish to maximize and how to best measure them. Traditional measures of teaching effectiveness, like test scores, may not be the best way to measure the nuances between the different teaching methods.
We should also consider the possibility that there were no statistically significant differences in many of our comparisons because both methods could be effective under the right circumstances. For example, some of the criticisms of the passive nature of lecturing could have been overcome by the instructor’s ability to engage students. Additionally, the flipped-classroom model of TBL may make it harder for some students to learn course content than with lecturing. We conclude that there is not a “one-size-fits-all” approach to effective teaching, and instructors should match their methods to their learning objectives and teaching strengths.
Footnotes
Acknowledgements
The authors would like to thank Bill Roberson for guidance on the project, Kaitlyne Motl and Cynthia Glass for research assistance, and the editor and reviewers for their comments.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The American Sociological Association’s Carla B. Howery Teaching Enhancement Grants Program provided support for this project.
Editor’s Note
Reviewers for this manuscript were, in alphabetical order, Angel Hoekstra, Karl Kunkel, and Esther Wilder.