In Search of Greater Understanding: The Impact of Mastery Learning on Social Science Education

Abstract

Mastery learning approaches were designed to improve student learning and elevate the level of understanding across a broader swath of students. These approaches operate under the belief that all students are capable of learning if given enough time. Little research has examined the utility or applicability of a mastery learning approach for social sciences outside of research methods courses. This study provides a review of the relevant literature on mastery learning, a discussion of the applicability of this approach to the teaching and learning of social sciences, and a review of the process and results of the conversion of more traditionally organized and taught courses to a mastery learning approach. Overall, our evaluation provides evidence that a mastery learning approach can make a significant impact on student learning.

Keywords

mastery learning knowledge acquisition knowledge retention

In the past several decades, undergraduate instructors have begun to reject the traditional “sage-on-stage” method of teaching to explore more alternative instructional techniques (Gilmer 2010; Haak et al. 2011; Fisher and Hanze 2019; Linneman 2019). The sage-on-stage model many of us aspire to move away from emphasizes lecture, prioritizes memorization of key details, and wrongly assumes that knowledge transfers directly (and easily) from teacher to student. Application-based learning, student-centered learning, and active learning strategies are all lauded as exceptional alternatives to the lecture-heavy curricula many of us are all too familiar with. Unfortunately, the fact remains that the sage-on-stage method is often the most accessible method as we must also juggle research and/or service in addition to teaching. Moreover, although studies have found the sage-on-stage method to be largely ineffective, students become accustomed to it in primary school and can be resistant to new ways of instruction (Roberts and Roberts 2008; Albers 2009; Linneman 2019). Thus, we as instructors are faced with the challenge of simultaneously navigating and managing our course load, meeting students’ expectations (about what the course should cover as well as how it should be covered), and ensuring student learning. While the sage-on-stage method may help us manage our course load and meet some student expectations, it often falls short of ensuring student learning. We believe that a better solution to the challenge faced by instructors today may come in the form of mastery learning.

With the preceding line of reasoning, two questions emerge: (1) What is the problem with the traditional, sage-on-stage method? and (2) Why might mastery be a better solution? In the sage-on-stage approach, instructors (not students) become the focus of the classroom. Mastery learning, on the other hand, seeks to shift the focus onto the students. Consider a traditional course structure: Students are assigned a reading and then are asked to attend a lecture that covers the material discussed in the reading the following day. The goal of the instructor is to deliver the content rather than ensure deep, conceptual understanding of the material (Handelsman et al. 2004). Under this approach, students are rarely asked to analyze the conceptual ideas underlying the material. Their ability to memorize what the professor lectures on becomes the goal (Mostrom and Blumberg 2012). While rote memorization is certainly a useful skill, extant literature suggests that this passive approach to learning produces little in the way of learning (Tagg 2003; Lee, Wrigley, and Dreby 2008; Addy, LePrevost, and Stevenson 2014). In addition, because students in a traditional classroom setting typically do not receive feedback until the unit or semester is over (at which point it is too late for students to modify their learning), there are serious concerns regarding the efficacy of the traditional teaching styles many of us are trained to imitate.

Conversely, mastery prioritizes deep, conceptual understanding and regular feedback (Guskey 2010). The ability to recite objective facts becomes less important than one’s ability to apply a concept or solve a problem. Mastery of material is achieved through active learning techniques and other strategies that emphasize application, analysis, and creation, with students receiving feedback as they progress through the material. Rather than cover each and every vocabulary word and fact included in a textbook, mastery instructors prioritize the most important, foundational concepts of the course. For example, in an Introduction to Sociology course taught using mastery, instead of asking students to memorize the difference between polygamy and polygyny, one might ask students to apply the concept of cultural relativism to norms and policies that exclude nonmonogamous family groupings. Students then learn about differences in family structures by virtue of learning how to apply one of the foundational concepts of the discipline. The value of limiting the number of concepts taught is that it (1) allows instructors to “dive deeper” into the most important ideas, (2) creates numerous opportunities for feedback, and (3) allows students to practice applying the concepts in a way that will help them better understand the world around them. The core assumptions of mastery are quite simple then: Students will learn better if merely given the necessary tools, namely, adequate time to learn the material, the opportunity to apply the material learned, and regular feedback from their instructor (Bloom 1968; Keller 1968; Bloom 1974; Guskey 2010).

Early iterations of mastery learning (see Bloom 1968; Keller 1968) encouraged instructors to think of themselves almost as tutors and to simulate one-on-one instruction in the classroom—something that is very difficult for present-day college professors. Mastery learning as it was originally proposed also demanded a significant amount of work on the part of the instructor. As previously discussed, under mastery learning, students should individually be provided with several opportunities and mediums to practice applying concepts; they should receive quick, regular feedback throughout each unit; and, finally, they should be provided with unlimited opportunities to retake assessments used to determine whether mastery has been achieved (Bloom 1968; Guskey 2010). The amount of work this requires on behalf of the instructor is simply unmanageable for many of us. Importantly, existing mastery models also lack the structure necessary for students who are new to college. Under one mastery model, for instance, class time is treated as more of an optional “study hall,” and students are expected to move through the curriculum at their own pace (see Keller 1968). A different approach requires the instructor to refrain from moving on to subsequent units until everyone in the class has achieved mastery (see Bloom 1968). In a 60-plus person lecture hall where each student is coming in at a different level and with a different set of skills, this is simply unfeasible. Clearly, mastery as it was envisioned by its founders is simply unrealistic for a modern college classroom. At the same time, its merits are undeniable. As such, we set out to determine whether we could update the mastery learning approach so that it is both accessible for busy instructors and effective for undergraduate learners.

Review of the Relevant Literature

Mastery learning is one of the most powerful research-supported strategies for improving student learning (Guskey 2010). Over the past 40 years, mastery learning has been continually implemented and evaluated across disciplines. Additionally, mastery learning’s core elements have formed the foundation for numerous teaching innovations that are implemented and praised in the classroom today (Guskey 2010).

There are two schools of mastery learning that have emerged since the concept was first theorized: Bloom’s (1968) Learning for Mastery (LFM) approach and Keller’s (1968) Personalized System for Instruction (PSI) approach. The LFM approach is group based and teacher paced. Students learn cooperatively and the teacher controls when students move onto further units. The majority of this approach was informed by Carroll’s (1963) work on school learning. Carroll posited that a student’s success in learning material was largely a product of time, operating under the idea that if given an appropriate amount of time, all students are capable of reaching any level of learning. Time needed would vary from student to student based upon a variety of factors, including a student’s aptitude for the specific content, the quality of instruction received, and a student’s ability to understand the instruction (Block and Burns 1977; Guskey 2007). Building upon Carroll’s work, Bloom (1968) believed that instructors should utilize a variety of instruction methods. Thus, at the heart of Bloom’s LFM approach are assessment of small units of learning, individualized feedback, collaborative learning, and providing students with the time needed for them to learn (for a full review of Bloom’s model, see Block and Burns 1977; Guskey 2007).

Contrary to Bloom’s (1968) LFM approach, Keller’s (1968) PSI approach is characterized as an individually based and student-paced course. Under the PSI approach, an individual student is responsible for managing the amount of time he or she spends on any one unit with minimal lecturing. The theoretical background for this approach originates from the programmed instruction movement that took place during the 1960s (for a full review, see Hartley 1974). Key to Keller’s PSI approach are the following: student-paced course structure that requires reaching mastery before moving to new units, lectures as rewards for course content mastery, and the use of a proctored testing system that provides immediate scoring and tutoring and allows for repeated testing (Keller 1968). Within both the LFM and PSI approaches, students who fail to demonstrate mastery are provided with correctives, while those who achieve mastery are assigned enrichments. Correctives are designed to signal areas where the students should devote more time, while enrichments are designed to bolster and build upon a student’s knowledge base.

Many studies have been developed to assess the effects of the mastery learning approach, specifically Bloom’s group-based model (Guskey and Gates 1986). Studies typically assess program effects in four key areas: student achievement, student learning retention, time variables (including time on task and time spent), and student affect variables. Most studies assess and find positive effects in regard to student achievement, most often measured as student scores on unit and course exams or letter grades attained by students (Guskey and Pigot 1988; Ozden 2008).

Mastery learning incorporates instructional strategies that seek to increase student achievement regardless of initial aptitude. Well-implemented mastery learning has consistently produced higher levels of achievement among students as well as increased their confidence in their ability to learn (Guskey and Pigot 1988; Kulik, Kulik, and Bangert-Drowns 1990). Core elements of mastery learning can be found in many recent instructional models and interventions, such as response-to-intervention models and understanding by design (Guskey 2010). Mastery learning emphasizes the importance of the use of multiple instructional approaches, including group-based learning, to engage students and tie concepts to their interests and experiences, much in line with current intervention models. Additionally, in recent years, scholars have stressed the necessity of active learning techniques to stimulate deep learning among students (Pedersen 2010). Deep learning is defined as “the process of preparing and empowering students to master essential academic content, think critically and solve complex problems, work collaboratively, communicate effectively, have an academic mindset, and be self-directed in their education” (Martinez and McGrath 2014:3). Within mastery learning and other current interventions, formative assessments to monitor student progress, along with descriptive feedback, are also important. These assessments reinforce course concepts and measure learning objectives in an effort to measure student and intervention success.

A New Approach to Mastery Learning

On the basis of previous literature, we were motivated to design a mastery learning approach that was cognizant of the various constraints faced in today’s higher education climate. Under the approach utilized for this study, components from Bloom’s (1968) LFM and Keller’s (1968) PSI approaches were selected and combined with the goal of increasing student achievement across seven undergraduate social science classrooms ranging in difficulty from introductory to upper-level courses. Reduced lecture time, increased use of alternative student learning activities and active learning strategies, a teacher-paced schedule, and opportunities for students to reattempt mastery achievement on unit quizzes characterize the new mastery model.

All of the courses included in the study met twice a week for 14 weeks with class durations between 75 and 100 minutes. Consistent with our approach, instructors lectured for roughly half of the class period and set aside the remainder of class time for “workshop” periods that focused on applying concepts and content learned in the first half of the class. Workshop periods differed from course to course but took several forms, including short video clips, hands-on application-based activities, and group discussions. In many courses, instructors utilized a greater number of short video clips to better illustrate key concepts and spark discussion among the students. In all courses, instructors designed hands-on, application-based activities that would have students apply or explore a key concept from that day’s lecture. These activities took a variety of forms and varied from course to course, such as finding examples of theoretical concepts in popular culture (to provide another point of reference or example for them to relate the content to), using maps of crime locations and examining why certain areas had greater amounts of crime, developing a self-report survey to administer to the class, and exploring abortion legislation in states deemed “hostile” by the Guttmacher Institute, just to name a few. Finally, most courses used the reduced lecture time to provide an opportunity for more engaged and focused (both small- and large-) group discussion on some of the more controversial or complex issues covered during the lecture period.

Consistent with Bloom’s (1968) LFM approach, the updated mastery model used a teacher-paced, rather than student-based, schedule. Instructors divided the semester into several learning units (ranging between five and seven units) designed to highlight the most important material for each class. Units lasted between one and three weeks with the goal of covering less material but going more in depth to increase student understanding. With the teacher-paced schedule, the class moved on to subsequent units regardless of whether mastery was achieved by all students on their first attempts following the reinforcement of concepts, which proved too difficult (more than two-thirds of students failed to demonstrate understanding).

At the end of each unit, students had the opportunity to demonstrate mastery of material by scoring 90 percent or higher on a short (10 questions maximum) unit quiz. Quiz questions were designed to assess knowledge, comprehension, and application and consisted of a variety of question types, including multiple choice, true/false, short answer, and problem solving. The use of 90 percent as our mastery cutoff is consistent with previous applications of mastery learning (Garver 1998). Following each unit quiz, students were provided with feedback on which concepts they needed to review more and were encouraged to attend office hours or meet with their instructor if they were not able to reach mastery after two attempts.

Students had the opportunity to attempt unit quizzes three times to achieve mastery. While concepts assessed remained consistent across all three attempts, the questions varied. In other words, students did not see the same questions on multiple retakes. Questions were varied to better ensure that our approach was testing mastery of the material rather than the mastery of quiz questions. While LFM and PSI allowed unlimited opportunities for reassessment, our method was designed with the practical constraints faced by higher education instructors today in mind. Additionally, Nation and Roop (1975) observed that capping reattempts could motivate students to devote more time and energy into the learning of course material, further supporting our decision to limit reattempt opportunities. To test the impact of our newly devised approach to mastery on undergraduate learning, we assessed the following: student knowledge growth (did students perform better on the final exam than they did on the pretest?), student achievement (did students earn high grades?), student performance (did students perform well on the unit assessments?), and finally, perceived value of the approach (how did the students interpret the mastery learning strategies we implemented?).

Methodology

Data and Sample

This study examines the impact of the implementation of a mastery learning approach to seven social science courses: SOC2000 (Principles of Sociology), GWS2000 (Introduction to Gender and Women’s Studies), SOC2830 (Methods of Data Analysis), SOC3140 (Ethnic Relations), SOC3620 (Criminology), SOC3650 (Punishment and Society), and SOC4540 (Juvenile Delinquency and Justice). Courses used were a product of teaching assignments of the research team and represent a convenience sample of courses. All courses were delivered at a university located in the Midwest with an undergraduate student population of approximately 18,000, had a course cap of 65 students (with the exception of SOC2830, which was capped at 35), and were core courses. Data were obtained from student in-course performance (unit quiz averages, mastery averages, final exam averages, pretest averages, and grade earned), student registration data (gender, year in school, major, semester enrolled, grade point average [GPA], and course taken), anonymous student evaluations (taken near the end of the semester), and open coding of student responses to open-ended evaluation questions. With institutional review board approval from the authors’ institution, data were collected from students who were enrolled in any of the authors’ courses during the fall and spring semesters of the 2018–2019 academic year and then deidentified before being added to the group’s data set. Additionally, a deidentified control sample was drawn from two instructors’ previous semesters for comparative purposes.¹ Our experimental sample included a total enrollment of 514 students, while our control sample included a total enrollment of 165 students. For analyses involving comparisons between the experimental and control groups, we reduced our experimental sample to include only those students who took courses from instructors able to draw a control sample. During these analyses, the experimental sample included a total enrollment of 282 students. Students in both the experimental and control samples possessed similar mean GPAs, gender distributions, and year in school distributions.

Student In-Course Performance

For students in our experimental sample, in-course performance data were collected through a variety of measures. Students in the experimental sample were given a pretest within the first week of their course beginning.² These pretests varied in length across courses, with the number of questions asked ranging from 10 to 17. The pretests were designed to assess student knowledge of key topics within the course at the onset to establish their baseline familiarity with the subject matter. To make the data from these pretests more comparable across courses, we measured the percentage of questions correctly answered rather than an absolute count. Students in the experimental group also had data collected on their overall percentage earned across all unit quizzes within a given course, the percentage of units mastered, and their average on the cumulative final exam. Students in both samples also had data collected on the final grade earned in the course (using the following scale: A = 90–100 percent, B+ = 86–89.99 percent, B = 80–85.99 percent, C+ = 76–79.99 percent, C = 70–75.99 percent, D+ = 66–69.99 percent, D = 60–65.99 percent, E = 59.99 percent and below). For each course, final grades were based on the following weighting system: attendance, 5 percent; participation/assignments, 20 percent; unit quizzes, 40 percent; demonstration of mastery, 10 percent; and final exam, 25 percent. The mastery-of-unit quizzes combined accounted for a total of 50 percent of a student’s grade. While the ability to reattempt quizzes can provide students with an opportunity to raise their overall grade, a portion of their grade (10 percent) is based entirely on their ability to reach mastery of the units. We expected students to reach mastery after three attempts, and if a student fell short of mastery, their overall final course grade was reduced by between 1.43 and 2 percent (depending upon the number of units in a course) for each unit not mastered.

Additionally, a variable was created for the experimental and corresponding control samples that measured how each student performed on equivalent questions (the same questions asked during each semester). Instructors who had previously taught the courses, and were able to draw a control sample, were asked to identify a selection of questions from assessments used prior to instituting a mastery learning approach that they felt met the goals of the approach and were relevant to the more limited scope of focus. Questions were considered to be equivalent if they tapped into the same course concepts. Within the mastery learning courses, these questions appeared in a similar but altered form (e.g., a multiple-choice question would become a short answer, or a true/false question would become a multiple choice, or the question wording and response options may change). There was variation in equivalent questions from course to course. The following percentages represent the percentage of final exam questions in our mastery learning courses that were deemed equivalent: SOC2830, 41 percent; SOC3620, 71 percent; SOC3650, 70 percent; and SOC4540, 44 percent.

Student Registration Data

Data were collected from digital registration information on all students (experimental and control). Data available based on this registration information included gender, class rank at the time of the course, which semester they were enrolled in the course, their cumulative GPA entering the course, and the course(s) in which they were enrolled.

Anonymous Student Evaluations

Anonymous student evaluations of the mastery learning model were completed at the beginning of a class period after 75 percent of a course was completed via paper-and-pencil surveys. In total, 411 out of 514 students (or an 80 percent response rate) in the experimental group completed the anonymous student evaluations. In the survey, students were asked to respond to 12 seven-point Likert scale items and five open-ended questions. In the analyses we present next, we focus on two of the 12 Likert scale items and two open-ended questions.³

The two Likert scale items allowed student responses to range from exceptionally low (1) to exceptionally high (7) on overall quality of the mastery learning approach and overall quality of the course. Following the Likert scale items, students were asked to respond to a series of open-ended questions; however, we focus on the following: “What about the mastery learning approach did you find to be most helpful in learning the course material? Why do you think this was the case?” and “This mastery approach focused on reduced lecture time, increased applied work time, increased group work, and regular assessment and feedback. What would you change about how the course was delivered? Why do you feel this way?”

Analysis Plan

Our analysis plan involved the use of mixed methodology to assess our approach both objectively and subjectively. To determine whether or not the mastery learning approach made an objective impact on student learning and understanding of the material, we first ran descriptive analyses on the full experimental sample to examine student knowledge growth (comparing pretest performance to the comparable final exam performance) and student achievement (grades earned). We then used an ordinary least squares (OLS) regression to compare students’ performance (measured by equivalent final percentage) between our experimental and control samples.

To assess the subjective impact of the mastery learning approach, we used student responses to our variety of Likert scale items as well as engaged in a qualitative content analysis approach to analyze the open-ended responses. The qualitative content analysis approach was used to determine common themes and elements present in student responses. We used an open coding approach outlined by Creswell (2009) in an attempt to expose the meaning, ideas, and thoughts contained within the text of our sample’s responses. To identify patterns in the qualitative data, we read through all of the student responses several times, taking notes of recurring themes. Over the course of our analysis, about 11 themes emerged. These 11 themes were later consolidated into three overarching categories that would serve as the final codes. The following codes were assigned a color and used to analyze the qualitative data: (1) “Mastery learning forced me to actually learn the material,” (2) “Having multiple attempts was helpful,” and (3) “The course structure (i.e., reduced lecture, emphasis on applied/group work, decrease in number of concepts taught, regular feedback) was useful/motivating.”

When students discussed more than one category within a single response, we coded for all of the categories present. Once the data were coded, we counted the frequency of each category within the anonymous responses to determine what percentage of students discussed the aforementioned categories. If a student mentioned any category multiple times throughout their responses, the theme was counted only once.

Results

Student Knowledge Growth

In Table 1 we present a comparison of student performance on the pretest (given within the first week of the semester) and comparable questions on the final exam (given at the end of the semester). To examine how the mastery learning approach impacts learners with different baseline knowledge, we examined student performance disaggregated by GPA. From the onset of each course, high- and low-achieving students held comparable levels of average subject knowledge (8.42 percent separated the A-level GPA students from those at the D+, D, or E level). When it came to student performance on the comparable final exam questions, a greater separation in performance scores was observed (as A-level GPA students scored 24.76 percent better than those at the D+, D, or E level). However, it is worth noting that there were students at all academic ability levels able to achieve a perfect score on the comparable final exam questions. In total, 24 A-level students (21.24 percent), 19 B+ students (18.10 percent), 14 B students (11.29 percent), 9 C+ students (12.16 percent), 3 C students (6.12 percent), and 1 D+, D, or E student (7.14 percent) were able to achieve perfect scores on the comparable final exam questions. In regard to the percentage difference from pretest to the comparable final exam questions, students on average scored 220.41 percent higher. Out of 437 students who completed both the pretest and final exam, only 28 students (6.41 percent) performed worse or no different on the final exam as compared to the pretest. These relatively few students who failed to demonstrate knowledge gain from the course represented a variety of academic ability levels (four A, seven B+, six B, six C+, and five C students). Overall, students of all academic abilities (as indicated by GPA) showed significant subject knowledge growth from pretest to final exam under our mastery learning approach.

Table 1.

Comparison of Student Performance on Pretest and Comparable Final Exam Questions.

	Student Performance
Grade Point Average	M	SD	Min.	Max.	n
Pretest average
A (3.7–4.0)	38.42	23.99	5.88	100	111
B+ (3.3–3.69)	36.61	24.17	0.00	100	101
B (2.7–3.29)	31.14	22.24	0.00	100	112
C+ (2.3–2.69)	35.12	23.13	0.00	90	68
C (1.7–2.29)	30.89	20.42	0.00	80	45
D+, D, or E (below 1.7)	30.00	12.56	6.67	50	11
Comparable final exam average
A (3.7–4.0)	85.46	12.96	46.67	100	113
B+ (3.3–3.69)	78.80	17.21	39.29	100	105
B (2.7–3.29)	75.37	17.07	35.71	100	124
C+ (2.3–2.69)	71.88	20.93	17.86	100	74
C (1.7–2.29)	66.78	20.43	25.00	100	49
D+, D, or E (below 1.7)	60.70	15.93	36.67	100	14
Percentage change pretest to final
A (3.7–4.0)	233.54%	245.55%	−12.50%	1300%	111
B+ (3.3–3.69)	205.04%	202.69%	−20.00%	1000%	99
B (2.7–3.29)	244.83%	226.25%	0.00%	1100%	107
C+ (2.3–2.69)	172.35%	166.10%	−16.67%	600%	65
C (1.7–2.29)	253.27%	376.84%	−20.00%	2257%	44
D+, D, or E (below 1.7)	141.04%	158.21%	25.00%	596%	11

Student Achievement

In Table 2, we present student course achievement (as measured by final grade earned) in relation to academic ability level (as indicated by GPA). At each academic ability level, a majority of students earned grades greater than or equal to their GPA. Of most import to our study was examining achievement that was higher than a student’s GPA would have predicted. Approximately 50 percent of B+ students, 41 percent of B students, 45 percent of C+ students, 38 percent of C students, and 13 percent of D+, D, or E students earned grades in our courses in excess of what would have been predicted given their GPA. In supplementary analyses, we examined these same relational comparisons and found the mastery learning approach to be particularly effective for B, C+, and C students.

Table 2.

Student Achievement in Relation to Grade Point Average (GPA) Level.

	Grade Earned in Relation to GPA Level
GPA	Higher	Equal	Lower
A (3.7–4.0)	~	69.57	30.43
B+ (3.3–3.69)	50.00	18.87	31.13
B (2.7–3.29)	41.41	21.88	36.72
C+ (2.3–2.69)	45.45	14.29	40.26
C (1.7–2.29)	38.00	26.00	36.00
D+, D, or E (below 1.7)	13.33	86.67	~

Student Performance

In Table 3, we present the results of our OLS regression predicting performance on equivalent exam questions using a subsample of our mastery course participants and a control group. Students in the mastery learning group scored approximately 5 percent higher overall on the equivalent exam questions than did students in the control group while controlling for gender, year in school, GPA level, and major. Outside of the effect of the mastery learning approach, significant effects were observed for GPA level and major. As would be expected, those with higher GPAs performed better on the exam than those with lower GPAs. Other majors (those outside of the home departments of our courses) also performed better than did criminal justice majors (reference group) but equivalent to sociology majors. Neither gender nor year in school produced significant effects on exam score results. In supplementary analyses, we examined mastery performance at the course level and observed the effect of mastery to vary. We observed the following coefficients for mastery learning for each course: SOC2830 = 16.82, SOC3620 = 12.82, SOC3650 = −12.24, and SOC4540 = 9.06. The coefficient observed in our full subsample would seem to be dragged down a bit by the student performance in one course (SOC3650). Overall, the mastery learning approach still had a significant impact in improving exam performance.

Table 3.

Ordinary Least Squares Regression Predicting Equivalent Exam Average (N = 406).

Variable	b	SE
Mastery learning	5.03**	1.74
Gender (reference female)
Male	1.65	1.75
Year in school (reference senior)
Freshman	−6.62†	3.68
Sophomore	−2.56	2.40
Junior	−1.26	2.18
GPA level (reference A)
B+	−10.83***	2.76
B	−13.72***	2.59
C+	−17.76***	2.84
C	−17.47***	3.32
D+, D, or E	−27.71**	8.79
Major (reference criminal justice)
Sociology	2.35	3.29
Other	5.56**	2.07
Constant	78.94***	3.06

Note: R² = .16.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Perceived Value

According to student responses to our two key Likert scale items (measuring quality of the approach and course), students on average felt that the mastery learning approach was of slightly higher-than-average quality and that the courses themselves were of high quality.

In addition to the Likert scale items, perceived value was also assessed by analyzing anonymous open-ended student evaluation responses. Approximately 37 percent of students in the experimental group felt that learning came more naturally to them in their mastery course. They felt that the combined aspects of mastery learning “forced” them to learn the concepts in a more meaningful way when compared to their other courses. For instance, Student A stated, “It forces you to understand concepts in functional ways to better retain the information.” Several students also explicitly noted this same perception of value compared to nonmastery courses with the undertone of coercion of learning. Whereas many of their courses allowed them to simply memorize material, students felt that deeper learning occurred for them as a result of our mastery learning approach. For example, Student B noted, “The material I studied throughout this class wasn’t just for memorization for a test, but I actually feel I was able to apply what I was learning after I learned the material.” These student perceptions illustrate the impact that our approach can have on student learning within social science courses.

Outside of commenting on their learning under our approach, about 36 percent of students mentioned how helpful it was for them to be able to try mastering the material in a unit more than once. Students who discussed the positive impact of multiple attempts pointed to test anxiety, stress from outside the classroom, pressure to perform, and the ability to learn from one’s mistakes when discussing their preference for the component of our mastery model that allows up to three retakes on unit quizzes. These ideas were illustrated in Student C’s response, “It puts less stress on studying correct answers for a test and allowed us to study in a more long-term way where you don’t have to worry so much about passing the first time,” as well as Student D’s response, “The ability to retake quizzes. Hands down this is the most beneficial part of the mastery approach. Allowing the student to fix their mistakes and prove that they are both trying to grow and studying.”

While the improved-learning and multiple-attempts themes were by far the most common, students frequently pointed to the “nontraditional” course structure when explaining what they liked most about the mastery learning approach. Nearly half (46 percent) of students felt that the emphasis on applied work, motivation to achieve mastery, reduced lecture time, regular feedback, and/or the decrease in the number of concepts taught made coursework more enjoyable and manageable. Student E stated, “This learning approach made me feel more accomplished after each section. I feel I know . . . a lot about each topic rather than a little about a lot of topics like in my other classes.” In reference to the emphasis on applied learning, Student F wrote, “We learn the material more hands on compared to other classes. I hate sitting in a lecture for a whole class so I like how this is more interactive.” Both Students E and F compared the active, in-depth learning in their mastery courses to the passive, surface-level learning in their nonmastery courses. In total, a majority of student responses (approximately 52 percent) indicated that the mastery learning approach employed in our courses had an overall positive impact on their learning.

However, there were students who expressed unfavorable opinions of the mastery model. Negative perceptions of the mastery learning approach can be categorized into three sentiments: (1) desire for more lecture, (2) dislike of group work, and (3) frustration with standard of mastery. Students who expressed negative perceptions of the mastery approach because of the reduced lecture time often stated that they “learned better” in a more traditional, lecture-style classroom. This explanation was also common among students who did not like the mastery approach because of the increase in time spent on group work. Students who were frustrated by our standard for mastery felt that mastery should be set at 80 percent rather than 90 percent. Importantly, however, this is the only negative sentiment specific to the mastery approach. The view that class time should be set aside for active learning techniques (e.g., group work), rather than lecture, is not unique to mastery. Instructors who “flip the classroom,” for instance, also structure class time similarly (Song and Kapur 2017). Moreover, while some students were frustrated by the high standard of mastery, it was more common for students to discuss the high standard as a motivating factor rather than a source of frustration.

Discussion and Conclusion

The mastery approach presented combines elements from both Keller’s (1968) and Bloom’s (1968) models and attempts to transform undergraduate learning from passive to active, surface level to deep, and teacher centered to learning centered. The current study was designed to test the effectiveness and perceived impact of this newly devised model for mastery through a series of quantitative and qualitative analyses.

Broadly, our findings provide support for the notion that traditional passive teaching practices, while long-standing, are not always the most effective nor the most favored by students. The majority of students who participated in this study indicated that the mastery learning approach had a positive impact on their learning. Students enjoyed the active learning strategies employed through the workshops, the reduced lecture time, and the opportunity to retake unit quizzes. They felt motivated by the goal of mastery and appreciated the invitation to make mistakes and learn from them. Students discussed things like test anxiety, demanding schedules, and pressure to perform when explaining their preference for the component of our model that allows students multiple attempts to achieve mastery. Students saw mastery as a chance to focus on learning in a low-stress, yet active, environment. Additionally, students seemed to appreciate the reduced lecture time and enjoyed the integration of more hands-on active learning activities. Instructors who are concerned with undergraduate education should prioritize alternative learning methods, such as the mastery model, over more traditional, lecture-heavy methods.

In addition to positive student perceptions of mastery, we found objective support for the mastery model. For instance, the vast majority (93.5 percent) of students performed better on the final exam than they did on the pretest. This indicates that knowledge growth did take place among mastery students. Interestingly, this knowledge growth was not limited to students of a particular academic level. The few students who performed worse or no different on the final exam as compared to the pretest represented a variety of academic ability levels (see Table 1). This finding directly challenges the suggestion from Lai and Biggs (1994) that mastery learning works best for “educationally disadvantaged students” at the detriment of higher-achieving students. These results provide support for mastery learning’s emphasis on prioritizing deep learning on a handful of concepts as opposed to surface-level learning on a vast array of concepts. The component of our approach to mastery that sets aside half of class time for an active learning exercise was also a likely factor in facilitating deep understanding of material (Baeten et al. 2010; Stes et al. 2012; Ellis 2016).

One of the most important findings of this study was the observation that a majority of students at each academic ability level earned grades greater than or equal to their GPA. Within our sample, mastery proved an effective approach for all students. This finding indicates that students are capable of surpassing expectations when provided the opportunity to do so. Additionally, exam performance improved significantly in mastery courses. Students in mastery courses on average scored 5 percent higher overall on exam questions than did those in the control group, demonstrating further support for our mastery model.

Although the ideas we present here are by no means “new” (the first explorations into mastery learning began over 50 years ago), we anticipate some will question the merit and value of these methods. Indeed, we have heard mastery criticized as merely testing students’ ability to take multiple-choice quizzes. We have also heard mastery referred to as too lenient because it allows students the opportunity to demonstrate knowledge multiple times. Some also worry that mastery rewards students for failing or incentivizes them to procrastinate on their studies. While we do not believe one must use mastery to be an effective instructor, we do find these particular criticisms misguided.

As we discussed earlier in this article, we constructed quizzes that varied from one attempt to another in terms of questions asked but retained an emphasis on the same concepts. This was done to prevent students from “mastering” (i.e., memorizing) the answers rather than the material. Moreover, quizzes went beyond multiple-choice and true/false questions. While students were often asked to respond to multiple-choice questions in mastery quizzes, they were also asked to solve problems, argue a perspective, or analyze a dilemma. Regarding the extent to which mastery incentivizes laziness or leniency, this can be true only if one believes our students’ ability to earn high marks the first time around is more important than their ability to learn the material. Mastery incentivizes persistence, continued study, and student accountability. In open-ended responses to the survey, for instance, students who achieved mastery after failing their first or second attempt indicated that the prospect of achieving mastery is what motivated them to return to material they did not grasp the first time around—to not give up.

Limitations and Future Research

Similar to most new endeavors, a learning curve existed. While the authors met regularly to discuss implementing mastery in the classroom, inevitably some inconsistencies occurred. The use of nonlecture class time was the primary source. For instance, individual assignments, group work activities, video clips, and interactive classroom participation were implemented differently across classrooms. The extent of differentiation was minor, however, as the main tenets of our new mastery approach were utilized in each classroom and deliberately kept in mind during course planning. Despite this, inconsistencies were unavoidable due to the six different social science courses taught. The use of varying classes within the study may be perceived as a strength, however, as mastery learning was tested across social science curricula. In addition to regular meetings, observing classes of instructors implementing mastery pedagogy may encourage a more consistent mastery approach across classrooms.

Regarding limitations for instructors, time management was difficult when mastery quiz retakes were not completed in class. When left to students’ individualized scheduling, a disproportionate number of students came into office hours at the end of the course attempting retakes for every unit throughout the semester. This issue was mitigated during the second semester as retakes were given either time in class or a deadline (in the form of an end date for attempts on a particular unit) for completion. Developing alternate versions of each quiz required further time investment by instructors. However, these alternate versions helped us to address a common critique of mastery quiz retakes, grade inflation. Grade inflation, defined as an increase in grades students receive without an increase in achievement (Bejar and Blew 1981), has long been a concern among educators. These discussions tend to be heightened regarding mastery learning. However, Boretz (2004) argues that not all upward shifts in grade distribution are grade inflation and attributes higher proportions of students earning As and Bs to (a) effective instructional strategies and (b) improved student support services. Based upon our own data, we do not perceive this as an issue. Instructors altered questions across quiz attempts, so while the concepts remained stable, the questions differed each time, thus ensuring that students did not memorize questions and, instead, learned concepts. Although creating alternative versions of each quiz requires an initial time investment for instructors, once the class is developed, it largely allows for consistent usage of materials in future semesters. Finally, difficulties arose in testing long-term retention. Follow-up surveys were emailed to students one month following the course; unfortunately, response rates were too low to be used for analysis.

Overall, traditional sage-on-stage methods have not been shown to improve student learning. Moreover, they are often not engaging, and lectures alone can be boring for student and teacher alike. Unfortunately, the lecture-heavy curriculum is often the most accessible for faculty and graduate instructors, who face the pressure to publish along with an expectation of service. In response to these issues, we developed and tested an approach to mastery learning that emphasizes active over passive learning. Our approach was found to significantly improve student performance and was positively received by the majority of our students. We believe that a shift in teaching from traditional sage-on-stage methods to more active learning techniques and strategies is needed within academia, and our approach represents one alternative that shows promise in improving student learning.

Footnotes

Editor’s Note

Reviewers for this manuscript were, in alphabetical order, Matthew T. Lee, Melinda Messineo, and Daphne Pedersen.

ORCID iD

Patrick R. Cundiff

Notes

Author Biographies

Patrick R. Cundiff is an assistant professor at Western Michigan University in the Department of Sociology. His research and teaching interests include criminology, delinquency, victimization, prisoner reentry, and criminal justice education.

Olivia McLaughlin is a doctoral candidate at Western Michigan University in the Department of Sociology. Her research and teaching interests include critical whiteness studies, intersectional perspectives on state crimes and state violence, and pedagogy.

Katherine Brown is a doctoral student at Western Michigan University in the Department of Sociology. With specialties in criminology, research methods, and intersectionality, her research consists of state/state-corporate victimization, pedagogical/andragogical techniques utilized in the undergraduate classroom, and domestic/sexual violence addressed with a trauma-informed lens.

Keiondra Grace is a doctoral candidate at Western Michigan University in the Department of Sociology. Her research and teaching interests include black feminism, criminology, and victimology.

References

Addy

Tracie M.

LePrevost

Catherine

Stevenson

Maura

. 2014. “Thinking Critically in Undergraduate Biology: Flipping the Classroom and Problem-Based Learning.” Double Helix 2:1–9.

Albers

Cheryl

. 2009. “Teaching: From Disappointment to Ecstasy.” Teaching Sociology 37(3):269–82.

Baeten

Marlies

Kyndt

Eva

Struyven

Katrien

Dochy

Filip

. 2010. “Using Student-Centered Learning Environments to Stimulate Deep Approaches to Learning: Factors Encouraging or Discouraging Their Effectiveness.” Educational Research Review 5:243–260.

Bejar

Isaac I.

Blew

Edwin O

. 1981. “Grade Inflation and the Validity of the Scholastic Aptitude Test.” American Educational Research Journal 18(2):143.

Block

James H.

Burns

Robert B

. 1977. “Mastery Learning.” Review of Research in Education 4:3–49.

Bloom

Benjamin S.

1968. “Learning for Mastery.” Evaluation Comment 1(2):1–11.

Bloom

Benjamin S.

1974. “Time and Learning.” American Psychologist 29(9):682–88.

Boretz

Elizabeth

. 2004. “Grade Inflation and the Myth of Student Consumerism.” College Teaching 52(2):42–46.

Carroll

John

. 1963. “A Model for School Learning.” Teachers College Record 64:723–33.

10.

Creswell

John W.

2009. Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. 3rd ed. Thousand Oaks, CA: Sage Publications.

11.

Ellis

Robert A.

2016. “Qualitatively Different University Student Experiences of Inquiry: Associations among Approaches to Inquiry, Technologies, and Perceptions of the Learning Environment.” Active Learning in Higher Education 17(1):13–23.

12.

Fisher

Elizabeth

Hanze

Martin

. 2019. “Back from ‘Guide on the Side’ to ‘Sage on the Stage’? Effects of Teacher-Guided and Student-Activating Teaching Methods on Student Learning in Higher Education.” International Journal of Educational Research 95:26–35.

13.

Garver

Ken

. 1998. “A Computerized Approach to Mastery Learning.” Journal of College Science Teaching 28(2):94.

14.

Gilmer

Penny J.

2010. Transforming University Biochemistry Teaching Using Collaborative Learning and Technology: Ready, Set, Action Research! Dordrecht, Netherlands: Springer.

15.

Guskey

Thomas R.

2007. “Closing Achievement Gaps: Revisiting Benjamin S. Bloom’s ‘Learning for Mastery.’” Journal of Advanced Academics 19(1):8–31.

16.

Guskey

Thomas R.

2010. “Lessons of Mastery Learning.” 2010. Educational, School, and Counseling Psychology Faculty Publications 68(2):52–57.

17.

Guskey

Thomas R.

Gates

Sally L

. 1986. “Synthesis of Research on the Effects of Mastery Learning in Elementary and Secondary Classrooms.” Educational Leadership 33(8):73–80.

18.

Guskey

Thomas R.

Pigot

Therese D

. 1988. “Research on Group Based Mastery Learning Programs: A Meta-analysis.” Journal of Educational Leadership 81(4):197–216.

19.

Haak

David C.

HilleRisLabers

Janneke

Pitre

Emile

Freeman

Scott

. 2011. “Increased Structure and Active Learning Reduce the Achievement Gap in Introductory Biology.” Science 332:1213–16.

20.

Handelsman

Ebert-May

Diane

Beichhner

Robert

Bruns

Perter

Chang

Amy

DeHaan

Robert

Gentile

Jim

. 2004. “Scientific Teaching.” Science 304:521–22.

21.

Hartley

James

. 1974. “Programmed Instruction 1954-1974.” Programmed Learning and Educational Technology 11(6): 278–291.

22.

Keller

Fred S.

1968. “Good-Bye Teacher . . . .” Journal of Applied Behavior Analysis 1(1):79–89.

23.

Kulik

Chen-Lin C.

Kulik

James A.

Bangert-Drowns

Robert L

. 1990. “Effectiveness of Mastery Learning Programs: A Meta-analysis.” Review of Educational Research 60:265–99.

24.

Lai

Patrick

Biggs

John

. 1994. “Who Benefits from Mastery Learning?” Contemporary Educational Psychology 19(1): 13–23.

25.

Lee

Matthew T.

Wrigley

Julia

Dreby

Joanna

. 2008. “The Research Article as a Foundation for Subject-Centered Learning and Teaching Public Sociology: Experiential Exercises for Thinking Structurally about Child Care Facilities.” Teaching Sociology 34:173–87.

26.

Linneman

Judith A.

2019. “Share, Show, and Tell: Group Discussion or Simulations versus Lecture Teaching Strategies in a Research Methods Course.” Teaching Sociology 47(1):22–31.

27.

Martinez

Monica R.

McGrath

Dennis.

2014. Deeper Learning: How Eight Innovative Public Schools Are Transforming Education in the Twenty-First Century. New York: New Press.

28.

Mostrom

Alison M.

Blumberg

Phyllis

. 2012. “Does Learning-Centered Teaching Promote Grade Improvement?” Innovative Higher Education 37(5):397–405.

29.

Nation

Jack R.

Roop

Stephen S

. 1975. “A Comparison of Two Mastery Approaches to Teaching Introductory Psychology.” Teaching of Psychology 2(3):108–11.

30.

Ozden

Mustafa

. 2008. “Improving Science and Technology Education Achievement Using Mastery Learning Model.” World Applied Sciences Journal 5(1):62–67.

31.

Pedersen

Daphne E.

2010. “Active and Collaborative Learning in an Undergraduate Sociological Theory Course.” Teaching Sociology 38(3):197–206.

32.

Roberts

Judith C.

Roberts

Keith A

. 2008. “Deep Reading, Cost/Benefit, and the Construction of Meaning: Enhancing Reading Comprehension and Deep Learning in Sociology Courses.” Teaching Sociology 36:125–40.

33.

Song

Yanjie

Kapur

Manu

. 2017. “How to Flip the Classroom: Productive Failure or Traditional Flipped Classroom Pedagogical Design?” Journal of Educational Technology and Society 20(1):292–305.

34.

Stes

de Maeyer

Gijbels

Van Petegem

. 2012. “Instructional Development for Teachers in Higher Education: Effects on Students’ Perceptions of the Teaching-Learning Environment.” British Journal of Educational Psychology 82(3):420–35.

35.

Tagg

John

. 2003. The Learning Paradigm College. Bolton, MA: Anker.