The purpose of this research was to examine students’ achievement goals and how they related to their engagement, concerns about cheating and perceived success under two learning conditions resulting from the measures implemented during the COVID-19 pandemic. In this discussion, we focus on how our findings can expand the current understanding of achievement goals under different learning conditions, in particular, when shifting to remote online delivery and a CR/NC grading system. Specifically, we attend to three findings with relevant theoretical and practical implications. First, we discuss how students’ achievement goals, engagement and perceptions changed across the two learning conditions. Second, in line with much of the achievement goal theory literature we describe the importance of mastery-approach goals. Third, we highlight the unexpected relationship between performance-avoidance goals and success under remote learning conditions. In closing, we also discuss the limitations of our research and recommendations for future research.
All self-report variables changed
Based on the variables we collected, our results suggest that “everything” changed following the shift to remote learning conditions. We saw a uniform decrease in all four achievement goals and all three forms of engagement. Moreover, the C1 scores on achievement goals and engagement were correlated with C2 scores on the same measure, yet, they were not always the stronger correlations. This is highly uncommon from a measurement perspective in which the best predictor of a later variable tends to be an earlier score on that variable. Instead, our results highlight that the change in learning conditions had a meaningful impact on students’ achievement goals and their self-reported engagement. One limitation of the current study is that we did not collect information about students’ personal wellbeing or hardships related to COVID-19. There may have been many personal and health-related factors that exacerbated the effects of the remote learning condition. As instructors continue to design courses for remote delivery and particularly during COVID-19, they may want to keep in mind that these are not only novel instructional conditions but many life conditions may make it harder for students to pursue their achievement goals and engage as fully as they could have in more traditional learning conditions.
Mastery-approach goals are important
Despite the finding that MAP goals had the largest mean decrease from C1 to C2 in the paired-samples t-tests, they continued to significantly and positively predict four of our five outcome measures, suggesting their importance for all three types of student engagement and perceptions of success. This is consistent with previous research that highlights the positive outcomes associated with MAP goals (Huang 2012; Senko and Dawson 2017). This is encouraging for instructors to know that the benefits of MAP can be realized even at lower mean levels. It also suggests that instructors may want to be particularly mindful of cultivating and maintaining MAP goals in either learning condition as a means of supporting students’ engagement and perceptions of success.
To aid instructors in support MAP goals in the classroom, we draw on the universal design for learning (UDL) guidelines. In particular, multiple means of engagement provide suggestions for recruitment of interest and sustaining effort and persistence (CAST, 2018), both of which are important for MAP goals. Indeed, recruiting interest can include providing students with choice and autonomy to support their engagement with the learner outcomes for the course. When students have a choice, they can pick topics or activities of interest to them, supporting not only their MAP goals but also their engagement, behaviorally, emotionally and cognitively. Indeed, research by Jang and colleagues (2016) found perceived autonomy support was associated with student engagement. When instructors move their courses online, this can open up a variety of online learning opportunities that students can choose from, which supports their goals of learning and understanding the course content.
Moreover, the UDL guidelines note the importance of optimizing relevance, value and authenticity (CAST 2018). This guideline highlights the importance of utility-value when it comes to motivation and engagement, that is, when students choose to complete a task because they find it useful or relevant to their short- or long-term goals (Eccles and Wigfield 2020). When designing their instruction and assessment approach, instructors should be mindful of how these course components are viewed by students to increase their perceived utility-value and by extension, student engagement. This could involve providing an explanatory rationale statement at the beginning of assignments, or highlighting how the information will be useful in their future career when giving examples in online lectures.
Interestingly, emotional engagement decreased the most in C2 compared to C1 and it was the only regression analysis where engagement levels at C1 predicted C2. One reason for this finding could be that students’ emotional engagement is related to course factors outside of instruction and assessment formats, including course content. Indeed, student interest is an important factor in determining their level of engagement (Kahu et al. 2017). Interest may be more prevalent when students enjoy the course content rather than when they feel forced to take mandatory courses they find little interest in, which could also impact their endorsement of MAP goals. One way to facilitate emotional engagement is through the use of humor. With that in mind, we turn to the work of McCabe et al. (2017), who discuss how humor can support the development of relationships with students and increase their learning. Nevertheless, future research is needed in this area to further explore additional course components to better understand the association between achievement goals and emotional engagement.
Concerns with cheating and perceived success
Students’ concerns about cheating in their courses in C2, were not related to their achievement goals, but rather predicted by the number of students in the course and their perceptions of cheating in C1. In short, the larger the course, the more concerned students were with cheating – and students who were concerned about cheating in a traditional learning condition remained concerned in remote learning conditions. One explanation for this finding is that larger classes are often more impersonal and anonymous (Cash et al. 2017), perhaps allowing students to believe that cheating will go unnoticed. To remedy this, instructors can draw on the relatedness aspects of basic psychological needs (Ryan and Deci 2017) to develop a sense of community within their courses regardless of class size. This can be accomplished through breaking the class to work on tasks online together or creating study group options via online platform. Moreover, Cueso (2007) suggests that larger courses have more frequent violations of academic integrity, including cheating. This could be related to feelings of anonymity, mentioned above, but it may also be simple probability: having larger courses increases the odds that there are more likely to be students who cheat present.
Students’ concerns with cheating in C1 were also a significant predictor of their concerns with cheating at C2. This finding may speak to the assessment practices of the course. According to the International Centre for Academic Integrity (2020), 39% of undergraduate students admit to cheating on tests and 62% admit to cheating on written assignments. If instructors were simply changing assessment to a CR/NCR system, the different assessment types may still warrant similar discrepancies in cheating behaviour and therefore, concerns with cheating would remain consistent. Contrary to this, our results show that students’ concerns with cheating between C1 and C2 significantly increased. It seems that concerns with cheating exist regardless of whether or not specific normative grades are awarded. This provides an excellent opportunity for researchers and instructors to rethink summative assessment practices from a motivation perspective (Daniels et al. 2020).
Students’ feelings about their success in C2 were positively predicted by MAP and PAV goals in C2. Said differently, when students held goals to learn as much as possible and to avoid doing worse than other students, they perceived themselves as successful after the learning condition changed. This pairing of goals is particularly interesting because it suggests that success under the remote learning conditions came from both MAP goals and their intraindividual perspective on competence and PAV goals and their interindividual perspective on not losing to others. Without grades, it seemed that students had to look to other means for defining competence and the strongest indicators came from both approach and avoidance perspectives.
Similar to concerns with cheating, instructors could draw on the relatedness component of basic psychological needs (Ryan and Deci 2017). Building a sense of community in the course might alleviate some of the concern with losing to others, consistent with PAV goals. Moreover, two additional UDL guidelines can be implemented to support students in their perceptions of academic success (CAST 2018). Consistent with the notion of relatedness, the UDL guidelines highlight the importance of fostering collaboration and community within courses. Indeed, research by Sadera and colleagues (2009) found strong associations between learner interaction and engagement, sense of community and success in online learning. Moreover, the UDL guideline of increased mastery-oriented feedback, highlights that this type of feedback guides students towards mastery rather than performance goals (CAST 2018). Therefore, instructors when providing feedback to their students should be mindful that the feedback, is relevant, constructive, accessible, important and timely (CAST 2018).
Limitations and future directions
Although our findings provide important insights into the achievement goals of students under different instructional and assessment conditions, it is important to consider the following four limitations. First, participants represent a convenience sample of university students who utilized social media platforms and/or responded to email correspondence for their university. As such, these findings cannot generalize to other groups of students and post-secondary institutions. Rather, the results provide a snapshot of some students' experiences during the changes to assessment and instruction as a result of the COVID-19 pandemic. Nonetheless, the proximity of the research to their actual learning experiences may somewhat compensate for the narrow sample. Future research could strategically target students from specific faculties or expand data collection to additional postsecondary institutions.
Second, to capture students’ perceptions of their learning condition the items focused students on both the changes to instruction (e.g., online) and assessment (e.g., CR/NCR). An important consideration when designing questionnaire items is to limit each question to one idea (Mertens 2014), however, instruction and assessment together authentically represent the learning condition that students experience. Balancing these perspectives, we felt students would be unable to separate their responses to specifically changes in instruction or assessment and instead combined the two in the survey instructions. Future research interested in examining changes to instruction and assessment will also have to balance the authenticity of the learning experience with the precision of questionnaire items and may decide to separate the domains to gain additional information on students’ perspectives. Moreover, future research in this area should make the distinction between online learning and “emergency remote teaching” (Hodges et al. 2020). Online learning is designed to be remote from the beginning, with online assessment methods planned in advance. This could result in different outcomes for students’ motivation, engagement and perceptions when compared to an abrupt change to instruction and assessment. This type of research will continue to emerge as researchers navigate the education impact of the COVID-19 pandemic.
Third, the data we collected were correlational and retrospective meaning that students provided answers in a single sitting for both their motivation, engagement and perceptions before changes to instruction and assessment were made and then again after. It is possible that responses to the “before” questions may be slightly more positive had they been collected in live time; however, the mean scores are actually quite similar to what is typically found in university samples. The opposite is also true, that students’ responses to the “after” questions may be more negative because of their proximity to the “before” responses. Even if this were to be the case, the decreases were so marked and consistent that we believe the changes to instruction and assessment did have a clear uniformly negative association with motivation, engagement and perceptions of success. Despite these limitations, the information garnered is nonetheless important and increases our understanding of how the change to remote learning impacted university students’ motivation, engagement and perceptions.
Finally, we utilized a single item to measure students’ feelings of success and their concerns regarding cheating in their courses. While multi-item scales are often seen as more advantageous, there have been arguments made for the adequacy of single items in similar domains such as job satisfaction (e.g., Dolbier et al. 2005) or achievement emotions (Gogol et al. 2014). Nevertheless, future research could implement multi-item scales to measure these constructs so long as it did not increase participant fatigue and reduce participation.