The Effect of Specifications Grading on Students’ Learning and Attitudes in an Undergraduate-Level Cell Biology Course

ABSTRACT

Specifications (specs) grading is a grading system in which mastery of specific educational outcomes is the basis for the final grade a student earns in the course. Implementation of the types of assessments used for specs grading has shown to be beneficial for student learning and motivation compared to traditional grading systems. We designed a specs grading strategy in an undergraduate Cell Biology course, creating 20 individual learning outcomes (LOs). The grade earned in lecture depended on the number of LOs the student mastered. If students were unable to master the content on their initial attempt, they could earn retakes for each LO assessment by completing an assignment associated with the information covered in that LO. A student’s final class grade was dependent on the number of LOs mastered combined with the grade earned on their final exam. Here, we present how specifications grading was implemented in Cell Biology, differences in overall grade distribution between grading systems, improved performance on content-related assessment questions in sections using specifications grading, and more-positive attitudes for sections using specifications grading than for traditionally graded sections.

INTRODUCTION

Biology educational research (BER) has advanced in the past few decades to determine methods for supporting student engagement in the classroom. A standard measurement of success in these studies included increases in students’ academic performance, engagement, and attitudes toward the new teaching methods (1). Dirks completed a meta-analysis of approximately 200 BER studies from 1990 to 2010 and identified three major categories of research studies: student learning and performance, student attitudes about learning biology, and testing and validating instruments (1). Insight into the impact of different pedagogical and grading strategies that promote a positive learning environment and student success are integral for student success and retention in the undergraduate classroom.

Analysis of multiple studies concluded that focusing on student mental health and academic success would include pedagogical approaches that foster time management skills, provide timely formative feedback, and promote a growth mindset in the classroom (2). Instructional approaches that foster student learning and retention of course content include, but are not limited to, implementation of active learning strategies, providing timely feedback in formative assessment, and reducing high-stakes assessments. Hsu et al. noted that these strategies promote reduced stress and anxiety in students (2). Implementing low-stakes assessments by having multiple quizzes covering less course content at any given time, rather than large exams, has contributed to student academic performance in the classroom. For example, there is correlation between student performance on course content with frequent low-stakes quizzing and well-timed formative assessment in conjunction with increased active learning strategies in the classroom (3). This finding was consistent with studies that suggest that repeated quizzing and testing with timely feedback promote an increase in working memory capacity, enhancing retention of material (4, 5). Thus, continued training on foundational science concepts improves retention and mastery of the material. Bloom reasoned that if students who are typically distributed based on aptitude are all afforded the same educational opportunities and instruction, very few of these students will achieve mastery because only a few would have the capacity to do so (6). Alternatively, if students are afforded different opportunities and instruction to enhance their aptitude for the material over time, far more students could achieve mastery. The basic framework for mastery learning is as follows: students must clearly understand the learning objectives of the course, objectives should be broken into small, specific learning units, teachers must provide specific feedback on knowledge deficiencies, students need to be allowed time to revisit their deficiencies, additional or alternative learning opportunities should be given to allow for greater understanding of concepts, and students should be encouraged to work together in small groups to review and identify their own strengths and weaknesses (6). This method encourages individuals to recognize that they can achieve their goals or outcomes with hard work and effort, reinforcing that there are multiple paths that they can take to be successful as long as they give themselves time and are diligent in their studies. This allows the student to view challenges with the hopefulness of possible success instead of hopelessness and inevitable failure (7). Additionally, interventions that motivate students and improve their perception of academic setbacks, such as failing an exam, increase persistence toward graduation in STEM fields (8).

Assessment of student learning is theoretically a way to measure a student’s competency or mastery of course content. The traditional grading method focuses on a single numerical score on an exam that encompasses multiple topics or chapters of content. These assessments tend to foster the student’s dependence on the teacher instead of encouraging the student to be in charge of their own learning (9). Additionally, few students use traditional exams to enhance their learning (7, 10), perhaps because they perceive the grade on a particular summative assessment as permanent and unimprovable, thus disincentivizing the learning and metacognitive processes. Students may view extrinsic factors, such as a bad day or perhaps unrealistic expectations set by the professor, as the reason for a lower-than-expected score, rather than view more preparation or mastery of the content as the mechanism to increase their scores. For disadvantaged or minority students, students may experience feelings of not belonging or not being capable (7). This demonstrates a student’s view that their aptitude for a subject is unchangeable and any achievements are obtained from innate talent with no effort associated (7). Teaching the students that their intelligence and academic success are flexible increases academic progress and greater engagement in minorities and nonminorities (11). Specifications grading is a form of mastery learning or standards-based grading that encompasses many of these strategies. Using this method, students rely on intrinsic factors for mastery rather than extrinsic ones, thus improving student motivation and interest (12), and it can result in more equitable grading (13). Variations of standards-based grading have an emphasis on assessment in the teaching and learning process, and although implementation of these progressive grading strategies has been challenging for educators (14), it makes instruction clearer and more purposeful, resulting in a positive learning environment for students (15), and therefore is a successful alternative to more traditional grading practices (16, 17).

Specifications grading implements two design principles that motivate students: (i) detailed feedback of what was completed well and what needed further practice or review and (ii) additional chances to practice skills that are challenging but attainable (18). In this system, students can retake individual assessments to achieve mastery, increasing student effort and retention and giving them more control over their grade (12). Since the introduction of specifications (specs) grading, it has been implemented in various STEM courses, including Mathematics, Physics, Anatomy and Physiology, and Chemistry (19–23). In each of these publications, it was noted that even though the development of the many assessments was time consuming, the grading of these assessments saved time because no partial credit was awarded. The grading became simpler for both the instructors and students to identify areas of needed improvement (19–23). Faculty reported fewer students failing or withdrawing from these courses than in previous years, suggesting that students were performing better in these courses overall (19, 21–23). Additionally, faculty reported that students were using the learning outcomes in the courses to focus their studying. They developed more self-reflective learning behaviors and took control of their learning, a habit that both Bloom and Nilson (and many others) suggest should occur in mastery learning courses (6, 12, 19–22).

In this study, faculty evaluated the course content, types of detailed feedback, and assessment techniques before implementing specifications grading. Cell biology was chosen because it is an essential prerequisite and foundational course for upper-level biology coursework. Faculty reviewed the course content and developed narrower learning outcomes that allowed the material to be assessed in smaller quantities (Table S1). The 20 learning outcomes (LOs) that were developed provided students the ability to learn and apply their knowledge of smaller portions of information before moving to the next concept. This aligns with the finding that structuring courses with low-stakes assessments over smaller amounts of information can reduce failure rates in the classroom (24). The strategy of having more-defined learning outcomes has previously been followed in classes using specifications grading (12), and the LOs designed for Cell Biology align with the American Society of Cell Biology’s (ASCB) Cell Biology Learning Framework (https://www.coursesource.org/courses/cell-biology). In addition to smaller summative assessments, faculty developed various formative assessments for students to earn retakes, ensuring that each LO was assessed using both lower-level and higher-level Blooms taxonomy questions (25).

This study aimed to implement specifications grading in an undergraduate cell biology course to increase student learning and retention of essential material and sought to help faculty more transparently represent student learning outcomes to promote student success. It was hypothesized that more students would successfully complete the course because they were afforded multiple opportunities to show mastery, had low-stakes assessments in the form of quizzes rather than large exams, were given timely feedback, and were provided additional learning tools to supplement and enhance their ability to master course content. Finally, once students understood how specifications grading worked, it was expected that they would have more-positive attitudes toward both how their grades were determined and their level of understanding of the material presented in the course.

MATERIALS AND METHODS

Implementation of specifications grading in Cell Biology

To implement specifications grading in Cell Biology, faculty determined 20 specific learning outcomes (LOs) that were narrower and could be taught and assessed at any given time (Table S1).

In sections that implemented specifications grading, the grade for the lecture portion of Cell Biology consisted of points earned by passing quizzes associated with each LO. Students could earn up to 20 LO points by mastering the content for each LO, where mastery was defined as achieving an 80% or higher on the assessment (LO quiz) associated with the specific material covered in that portion of the course. Because there were 20 learning outcomes, each LO point represented 5% of the students’ final grade (Table S2). The class size for Cell Biology at Georgia Gwinett College (GGC) is capped at 24 students, allowing instructors to provide specific individualized feedback to students on each LO quiz attempt.

Mastery of course content was also expected on a cumulative midterm and final exam for students to maintain the grade earned during the semester. Students who showed mastery of overall course content, achieving an A or B on the exam, could increase their grade in the course. Conversely, if students failed the final exam with a D or an F, they could lose LO points toward their overall final grade (Table S3).

In addition to assessing smaller portions of course content, students were given the opportunity to have additional attempts to earn LO points for each LO quiz if they did not master the material on the first try. The course was designed specifically to allow students up to two retakes for each individual LO quiz (three attempts total). To earn a retake opportunity, students were required to complete an assignment associated with the content covered on that quiz. Examples of assignments that students could do to earn retakes included worksheets, online quizzes, and providing summaries of the specific material covered in the associated LO.

Statistical analysis of course grades

Grade distribution in traditional versus specifications grading classes was analyzed using data from spring 2018 (prior to the implementation of specifications grading), fall 2018, spring 2019, and fall 2019 (some sections using traditional grading and some using specifications grading). A chi-square test was performed to compare passing (A, B, or C) versus nonpassing (D, F, or withdrawal) rates and to compare the grade distribution for individual course letter grades. There were, in total, 236 students in traditional sections and 252 students in specifications grading sections.

Statistical analysis of survey data

Surveys were distributed among all sections of Cell Biology during the spring 2019 and fall 2019 semesters after specifications grading had been fully implemented in the course. The survey included 34 questions related to content knowledge and 10 questions related to student attitudes about the course. The specific wording for the questions about student attitudes can be found listed in Table 1. Surveys were excluded for students who opted out of being included in the analysis, were under the age of 18, and/or did not respond correctly to a question inserted to check for random responses. There were only two sections that had not implemented specifications grading by this time; this resulted in only 29 surveys from students in sections using the traditional grading strategy and 112 surveys from students in sections using the specifications grading strategy. The Georgia Gwinnett College Institutional Review Board approved this assessment methodology (IRB proposal no. 17119).

TABLE 1.

Student perceptions of grading method^a

Question no.	Survey question	Traditional grading percentage	Specifications grading percentage
1	The method in which the lecture portion of the course was graded helped my understanding of the course content.	59%	77%
2	The method in which the lecture portion of the course was graded helped my retention of the course content.	45%	61%
3	The method in which the lecture portion of the course was graded increased my ability to relate the structure of cellular components to their functions.	59%	73%
4	The method in which the lecture portion of the course was graded increased my ability to describe the regulation of the cell cycle.	61%	69%
5	The method in which the lecture portion of the course was graded increased my ability to explain and classify various methods of membrane and intracellular transport.	62%	69%
6	The method in which the lecture portion of the course was graded increased my ability to describe the molecular events involved in the regulation of gene expression.	61%	65%
7	The method in which the lecture portion of the course was graded increased my ability to compare concepts of intercellular and intracellular signaling in cell function.	59%	68%
8	The method in which the lecture portion of the course was graded increased my ability to analyze the processes of cell specialization.	59%	66%
9	The method in which the lecture portion of the course was graded increased my ability to examine cell biological concepts and techniques in scientific research and real-world problems.	62%	71%
10	As a result of my effort in this class my enthusiasm for cell biology increased.	48%	71%

^aPercentages shown are percent positive (strongly agree/agree) for each question. n = 29 for traditional grading and n = 112 for specifications grading. These increases seen were not found to be statistically significant.

Because a Shapiro-Wilk test suggested that the student performance data for content-related questions were not normally distributed, a Kruskal-Wallis test was used to compare student performance on content questions between traditional and specifications grading sections. A Fisher exact test was performed on the student attitudinal responses.

Safety issues

There was not a laboratory component involved in this research study.

RESULTS

To implement specifications grading in Cell Biology, 20 specific LOs were designed collaboratively by the faculty teaching Cell Biology and generally align with the ASCB’s Cell Biology Learning Framework (https://www.coursesource.org/courses/cell-biology). Specifications grading was implemented in all five sections of Cell Biology in spring 2019 and three of the six sections in fall 2019. The traditionally graded sections that did not adopt specifications grading had a range of instructors and did not use the 20 defined LOs. Common to all traditionally graded sections were broadly defined course learning objectives and exams covering multiple chapters, and sections used the same textbook in any given semester. To determine whether there were differences in the overall grade students received in the course, we assessed the grade distribution between traditionally graded sections and those using specifications grading. GGC does not have a +/− grading system; thus, grades earned were A, B, C, D, F, or W. Comparing the proportions of students who passed (earning an A, B, or C) and did not pass (earning a D, F, or W [withdraw]), no significant difference was observed between specs and traditional grading [χ²(1, N = 488) = 0.017; P = . 9] (Fig. 1). However, when analyzing course letter grades achieved, rather than grouping as pass versus not pass, a significant difference was observed in the overall grade distribution between specifications and traditional grading [χ²(5, N = 488) = 42.5, P < 0.001] (Fig. 2). The effect size for this analysis (Cramer’s V = 0.30) is medium (26). Visual inspection of the overall grade distribution reveals that among passing students (A, B, C), the specs grading students earned a larger proportion of A grades and a smaller proportion of B and C grades than students in classes with traditional grading. Similarly, among nonpassing students (D, F, W), a larger fraction of students in specifications grading courses withdrew (W) or earned a D, and a smaller percentage earned an F than in the traditionally graded sections.

FIG 1.

Pass rates in courses using traditional and specifications grading. Grade distribution in classes with the traditional grading scheme (black) versus specifications grading (gray) was assessed. A chi-square test was performed showing that there is no significant difference between passing scores (A, B, or C) versus nonpassing scores (D, F, or withdrawal) rates. n = 236 for traditional grading and n = 252 for specifications grading.

FIG 2.

Couse letter grade distribution in courses using traditional and specifications grading. Grade distribution in classes with the traditional grading scheme (black) versus specifications grading (gray) was assessed. A chi-square test was performed showing that there was a significant difference, P < 0.001, in the overall grade distribution between these groups. n = 236 for traditional grading and n = 252 for specifications grading.

In addition to the analysis of grade distribution, surveys administered at the end of each semester included questions addressing course content and student attitudes toward the grading methods used in the course. It was hypothesized that students in the specs courses would perform better on course content than those in traditionally graded sections, since these students were encouraged to revisit the content and permitted to retake LO quizzes. Because the student score data did not follow a normal distribution (Shapiro-Wilk test, W = 0.98, P = 0.01), a Kruskal-Wallis test was used to compare student performance on content questions. Students in the specs grading sections scored significantly better than those in the traditionally graded sections, with a small effect size (26) [H(1) = 4.84; P = 0.03; η² = 0.03]. In sections using specifications grading, the median and maximum scores achieved were higher, demonstrating higher levels of knowledge of the content assessed (Fig. 3).

FIG 3.

Student performance on assessment questions associated with course content. Student performance on assessment questions associated with LO content was analyzed and assessed for differences in performance between students in sections using traditional grading methods versus those using specifications grading. The division in the box plots indicates the median percentage correct. The top (black) indicates the top 75th percentile, and the bottom (gray) shows the bottom 25th percentile. Student in the specs grading sections scored significantly better than those in the traditionally graded sections, with a small effect size (20) H(1) = 4.84, P = 0.03, η² = 0.03.

Data were collected to determine if students in the specs-graded course have more-positive attitudes toward the way the course was graded than those in the traditionally graded course. A five-point Likert scale was used with the options of strongly agree, agree, neutral, disagree, and strongly disagree. There were 10 questions associated with student attitudes, and these were worded so that if a student selected agree or strongly agree, the response denoted positive feedback toward grading in the course. The questions on student attitudes toward grading included those addressing whether the method increased or helped their (i) understanding of course content, (ii) retention of course content, (iii) ability to relate cellular components to their function, (iv) ability to describe the regulation of the cell cycle, (v) ability to explain methods of membrane and intracellular transport, (vi) ability to describe the events involved in the regulation of gene expression, (vii) ability to compare intercellular and intracellular signaling in cell function, (viii) ability to analyze the process of cell specialization, (ix) ability to examine cell biology concepts in research and real-world problems, and (x) enthusiasm for cell biology (Table 1). In sections of Cell Biology that implemented specifications grading, there was an increase in the percentage of students who responded positively to each survey question related to student attitudes compared to students in traditionally graded classes, but these increases were not found to be statistically significant. Specifically, in response to whether the grading strategy used helped with student understanding of course content, 77% of students in classes using specifications grading selected agree or strongly agree compared to 59% of those in traditionally graded classes. In response to whether the grading strategy used helped with student retention of course content, 61% of students in classes using specifications grading selected agree or strongly agree compared to 45% in traditionally graded classes. Additionally, in response to the question of whether the grading strategy used increased student enthusiasm for cell biology, 71% of students in classes using specifications grading selected agree or strongly agree compared to 48% of those in traditionally graded classes (Table 1). Detailed data for all responses on the Likert scale can be found in Figure S1.

DISCUSSION

Specifications grading was implemented in Cell Biology to promote student learning and successful completion of the course. Assessment of overall grade distribution showed a significant difference between classes using traditional grading methods and those using specifications grading (Fig. 2). One such difference was a higher percentage of A’s in specifications-graded courses. One possible reason for this could be that those students who would typically have received a B or C could continue working on the material that was more challenging for them. They submitted associated assignments and earned the ability to retake individual LO quizzes, collectively resulting in an opportunity to earn a higher course grade. Additionally, more students received a D or withdrew from the specifications grading course, and fewer students received an F. Some students who normally would have received an F may have increased their grades by retaking LO quizzes, resulting in a D, but were unable to achieve a passing grade. The increase in withdrawals could be attributed to the following: (i) students may have withdrawn from the course simply because it used a grading strategy different from what they were familiar with, perhaps intimidating them, or (ii) in courses using specifications grading, students should have an acute understanding of their standing in the class based on individual LO quiz scores, allowing them to decide to withdraw before the withdrawal date rather than receive a failing grade. These possibilities are only assumptions at this point. We are currently working on designing studies to send follow-up surveys to students who received W’s to assess their decisions to withdraw from either the traditionally graded sections or those using specifications grading.

One noteworthy result obtained was the difference in student performance on content-related questions from the end-of-semester survey. Students enrolled in specifications-graded courses performed significantly better on the content assessment than those enrolled in traditional courses, demonstrating higher levels of knowledge of the content being assessed (Fig. 3), which suggests that the use of specifications grading in the course promotes student understanding and retention of course material. Studies comparing content gains in traditional and specifications grading in STEM courses reported similar trends (27, 28), with specifications grading students making more considerable gains than their traditionally graded counterparts. The reasons for these increased gains may be attributed to students spending more time studying and practicing challenging concepts, making it easier for them to recall and apply information during assessments (29).

Attitudinal survey responses from students in specifications grading sections of Cell Biology were overwhelmingly positive. Collectively, these students reported feeling optimistic about the impact of this grading system on their understanding and retention of course information and their enthusiasm for the topic (Table 1 and Fig. S1). From a faculty perspective, these responses reinforce that the effort they put into implementing specifications grading in the course was valued and positively affected their students. Considering the growing body of literature linking academic success and mindset (30, 31), it may be possible that positive student attitudes in the specifications grading courses may have contributed to their higher end-of-semester grades and larger content gains. These gains in student attitudes and perception of specifications grading in the classroom may be related to the lower-stakes assessments and may foster more of a growth mindset within the classroom. Therefore, as a future endeavor, we are interested in investigating student attitudes (e.g., self-efficacy and academic mindset) on performance in a specifications grading course. Moreover, since GGC’s student body is diverse, we are also interested in assessing how alternative grading systems affect the retention and persistence of students from populations that are historically excluded in STEM.

Overall, there is an increase in performance on content-related questions (Fig. 3) and in positive student attitudes toward specifications grading (Table 1 and Fig. S1). Based on these findings, instructors are confident that implementing this grading strategy has resulted in positive gains for students taking Cell Biology. This grading strategy has been successfully implemented in various forms at GGC (17) and other institutions (27, 28, 32, 33). It could be used as a template for other biology courses at GGC and institutions across the country to promote student understanding and enthusiasm for a variety of classes.

Variations of mastery learning improve student academic performance. Studies have demonstrated that frequent, low-stakes assessments and the delivery of timely feedback correlate with gains in student performance in the classroom (7, 24). Faculty teaching Cell Biology implemented specifications grading to increase learning, retention of content, and student success in Cell Biology and future coursework. The specific aspects of specifications grading that were adopted included assessing students on smaller portions of course content (LOs) throughout the semester and allowing students the opportunity to retake individual assessments should they not master the content on the first attempt. Timely feedback was used to allow students time to retake quizzes within a specified time frame and to foster student success (7). This feedback allowed students to learn from their mistakes while the content was still fresh in their minds and helped prevent students from falling behind on new material while working to master previously covered material. In addition, providing students the ability to retake the lower-stakes LO quizzes provided students time to revisit material they did not master on the first attempts, which may correlate with reduced stress and anxiety, potentially resulting in greater enthusiasm for learning (7, 18).

Positive student gains were seen after implementation of specifications grading in an undergraduate Cell Biology course, including increased academic performance on content-related questions and a higher proportion of A’s compared to those in sections that used traditional grading methods. Additionally, student attitudes toward specifications grading were positive, suggesting that this methodology can foster student success in the classroom. The framework developed and implemented at GGC can be used as a model for cell biology classes at other institutions and can also be adapted for other courses in the biological sciences to promote student success and retention.