GRADING: A CORRELATIONAL STUDY by Camille Phinney A project submitted in partial fulfillment of the requirements for the degree of MASTER OF EDUCATION IN CURRICULUM AND INSTRUCTION WEBER STATE UNIVERSITY Ogden, UT November 27, 2018 Approved _____________________________ Louise R. Moulding, Ph.D. ______________________________ Penée W. Stewart, Ph.D. ______________________________ Darlene Sangiorgio, M.Ed. MINDSET, ANXIETY, AND GRADING 2 Acknowledgements I would like to express appreciation for all the people who made this project possible. For Dr. Moulding and Dr. Stewart who guided my thoughts and my feelings about my students toward a topic I was really interested in studying. Also, for Dr. Moulding for never giving up on my project and having faith in me even when I was not so certain of myself. For Darlene Sangiorgio for supporting my idea and giving me access to resources I needed to complete my project. Finally, for my wonderful husband and children who surprisingly survived while I was at class and away from home writing this project. Without all their support, none of this would be possible. MINDSET, ANXIETY, AND GRADING 3 Table of Contents NATURE OF THE PROBLEM.......................................................................................... 7 Literature Review.................................................................................................... 9 Defining Mindset ...................................................................................... 10 Consequences of Holding a Fixed Mindset .............................................. 12 How Traditional Grading Systems Contribute to a Fixed Mindset .......... 18 Student Anxiety ........................................................................................ 21 Consequences of Test Anxiety.................................................................. 23 How Traditional Grading Systems Contribute to Test Anxiety ................ 30 Standards-Based Grading as an Alternative ............................................. 31 Summary ............................................................................................................... 34 PURPOSE ......................................................................................................................... 37 METHODS ....................................................................................................................... 38 Participants ............................................................................................................ 38 Instruments ............................................................................................................ 39 Procedure .............................................................................................................. 42 RESULTS ......................................................................................................................... 44 DISCUSSION ................................................................................................................... 50 Limitations ............................................................................................................ 57 Recommendations ................................................................................................. 58 REFERENCES ................................................................................................................. 60 APPENDICES .................................................................................................................. 65 Appendix A: Student Questionaires...................................................................... 66 MINDSET, ANXIETY, AND GRADING 4 Appendix B: Student Attitude and Motivation Survey Excerpt (Mannix, 2014) ............................................................................... 71 Appendix C: Institutional Review Board Approval from Weber State University .......................................................... 72 Appendix D: Approval Letter from School District ............................................. 73 Appendix E: Parent Letter and Consent Form ...................................................... 74 Appendix F: Information Sheet for Students on Improving Growth Mindset and Reducing Test Anxiety .................................. 77 Appendix G: Full Table of Correlations ............................................................... 79 MINDSET, ANXIETY, AND GRADING 5 List of Tables Table 1. Descriptive Statistics of Mindset and Anxiety Questionnaire ............................. 44 Table 2. Correlations of Mindset, Test Anxiety, and Items on a Grading Questionnaire 46 Table 3. Descriptive Statistics for Mindset and Anxiety by Gender ................................. 48 Table 4. Descriptive Statistics for Mindset and Anxiety by Grade Level ......................... 49 MINDSET, ANXIETY, AND GRADING 6 Abstract Traditional grading systems may reduce student cognitive well-being by promoting fixed mindset and reduce student emotional well-being by promoting test anxiety. This correlational study intended to determine the relationship between mindset, test anxiety, and attitudes toward grading and to find relationships between these measures and demographic factors including gender and grade level. One-hundred ninety-six high school students participated by filling out three questionnaires which were analyzed for correlations. The results showed a significant negative relationship between growth mindset and test anxiety; however, when separated by gender, this result was only true for females, not for males. The study showed a small significant positive correlation between mindset and grading items related to fairness, ability to learn, and the importance of homework and classwork while the results showed a significant negative correlation between test anxiety and the same grading items. The study showed no significant correlations between mindset and test anxiety across grade levels. These results were consistent with prior research that suggested that a focus on evaluation could contribute to lower mindset scores and higher test anxiety. Similarly, lower mindset scores and higher test anxiety are both associated with lower levels of achievement which may explain the more negative attitudes toward grading demonstrated in this study by students with fixed mindset and higher test anxiety. Prior research demonstrated higher test anxiety among females, consistent with the findings in this study. This study indicates that there is a relationship between mindset, test anxiety, and attitudes toward grading which should be studied further. MINDSET, ANXIETY, AND GRADING 7 NATURE OF THE PROBLEM Most schools use traditional grading systems in which teachers assign students’ scores on assignments and tests which are then averaged over the length of a semester or term. Sometimes grades are used as a stick or a carrot approach to motivate students where good grades indicate success whereas bad grades indicate failure. Consequently, good grades result in rewards, such as privileges or praise whereas bad grades result in punishments, such as embarrassment or taking away privileges (Guskey, 2011). This emphasis on grades convinces students that getting good grades is more important than learning; the grade becomes the goal instead of mastering the skills and content knowledge from the class (Long, 2015). Focusing on a goal, in this case the letter grade, instead of mastery is called performance orientation, and this type of motivation promotes what Dweck (2008) called the fixed mindset, where students believe that their intelligence cannot be changed over time through effort (Dweck, 1986, 2008). When students with a fixed mindset receive a grade, they stop trying to gain new information or try new strategies because they believe there is nothing they can do to affect that outcome (Dweck, Mangels, & Good, 2004). Therefore, fixed mindset can lead students to develop maladaptive motivational patterns where they give up when they are challenged which contributes to anxiety and learned helplessness (Dweck, 1986). Students shy away from new challenges for fear of negatively affecting their grade in a course and therefore limit their learning because the grade seems more important than developing new skills (Dweck, 2008). As a result, traditional grading practices encourage students to limit their learning opportunities once they have achieved the highest grade they think possible. MINDSET, ANXIETY, AND GRADING 8 The majority of high school students feel anxious about school with an even larger number reporting anxiety related to tests (Deb, Strodl, & Sun, 2015). This anxiety is partially a result of increased pressure to achieve high marks in traditional grading systems as good grades are associated with success leading to rewards such as getting into college, achieving scholarships, or looking good compared to others (Flannery, 2018). Because tests are one of the primary determinants of grades, more students are developing test anxiety (Deb et al., 2015; Von Der Embse & Hasson, 2012) which is associated with lower life satisfaction (Steinmayr, Crede, McElvany, & Wirthwein, 2016). Therefore, focusing on the importance of evaluations, like tests, leads to increased test anxiety and decreased emotional well-being. Traditional grading systems contribute to this test anxiety by averaging grades, particularly on evaluations like tests, over a term thereby punishing students for early mistakes when they were practicing new and challenging material. The more students are tested, the less time they have to practice between evaluations, and the more anxious they may become about evaluations because they may not feel sufficiently prepared (Hembree, 1988). This cycle may result in lower scores on evaluations which, when averaged over a term, may result in a lower overall grade thereby increasing test anxiety at each evaluation opportunity. Traditional grading practices can decrease student emotional well-being by increasing test anxiety and reduce effective cognition by encouraging students to develop performance goals and consequently a fixed mindset about their intelligence (Dweck, 1986; Steinmayr et al., 2016). Both performance goals and test anxiety inhibit learning and accurate evaluation of that learning, respectively. For example, students who develop performance goals will give up when faced with challenging material or stop trying to MINDSET, ANXIETY, AND GRADING 9 learn after they have received feedback (Dweck, 1986; Dweck et al., 2004), which demonstrates that they no longer value the process of learning. Likewise, test anxiety is negatively associated with performance on evaluations, and test anxiety accounts for some of the variance in scores on tests (Steinmayr et al., 2016; Von Der Embse & Hasson; 2012), which demonstrates that test anxiety can interfere with accurate evaluation. Grading practices that focus on evaluation and performing well contribute to these negative cognitive and emotional outcomes for students. Therefore, traditional grading practices need to be reconsidered in light of evidence that they promote poor cognitive and motivational skills and increase student anxiety related to school (Miller, 2013). Literature Review Historically, in the United States, students have been evaluated by averaging points and percentages over the length of a term on a variety of tasks determined by the teacher. These tasks may include, but are not limited to: homework, classwork, exams, essays, participation, and attendance. In traditional grading systems, these tasks may or may not directly link to standards of mastery expected for the course which therefore dilutes the meaning of the evaluation (Iamarino, 2014; Miller, 2013). Student evaluation methods are important because classroom environment and evaluation practices can affect student goals, motivation, and anxiety, which in turn affects student performance (Boaler, 2013; Church, Elliot, & Gable, 2001; Deb et al., 2015; Dweck, 2008; Hembree, 1988). This literature review will first explain how traditional grading practices may be linked to negative cognitive outcomes for students by encouraging students to develop a fixed mindset which has numerous negative cognitive consequences. Next, the literature MINDSET, ANXIETY, AND GRADING 10 review will explain how traditional grading practices may be linked to increased student anxiety, particularly test anxiety, by focusing on the importance of evaluation at any given moment instead of learning over time. Last, the literature review will propose an alternative to traditional grading practices, standards-based grading, and explain possible concerns of teachers and students with the alternative system. Defining Mindset Grading systems used in most schools today may reinforce an entity view of intelligence wherein students believe that their intellectual ability is fixed and there is little or nothing they can do to improve their inherent ability (Dweck, 1986, 2008). This entity view of intelligence is what Dweck (2008) termed a fixed mindset as opposed to a growth mindset wherein students believe their intelligence is malleable and can be changed over time with effort (Boaler, 2013; Dweck, 1986, 2008). Mindset exists along a continuum between a fixed mindset and a growth mindset with most people falling somewhere along the spectrum as opposed to holding either a purely fixed or purely growth mindset (Dweck, 2008). Researchers using mindset questionnaires to determine which set of beliefs a person holds, a more fixed mindset or a more growth mindset, have found that approximately 40% of the population hold a fixed mindset, about 20% hold mixed views, and approximately 40% hold a growth mindset (Boaler, 2013; Dweck, 2008). The type of mindset a person develops is influenced by a variety of factors including the messages they receive from others about the importance of inherent ability or the importance of effort (Dweck, 1986, 2008). These messages, in turn, influence the types of goals people assume and their motivation for achievement (Boaler, 2013; MINDSET, ANXIETY, AND GRADING 11 Dweck, 1986, 2008). Students with fixed mindsets usually choose performance goals that focus on outcomes of ability, such as success or failure, instead of processes like learning or developing ability (Dweck, 1986). In contrast, students with growth mindsets usually choose process goals that focus on increasing competence or mastering something new (Dweck, 1986). There is a reciprocal influence between the types of goals students adopt and their mindset; mindset influences goal preferences which in turn reinforces mindset. For example, a student with a fixed mindset will usually measure their success by achieving an outcome, such as a good grade. That performance outcome will then reinforce the student’s entity view of intelligence. In other words, a student with a fixed mindset might say, “I am inherently smart, so I should get good grades.” When the student achieves a good grade, he may attribute his achievement to being smart which reinforces his entity view of intelligence and therefore contributes to his fixed mindset (Dweck, 1986, 2008). Performance goals can lead to a fixed mindset because students are mostly interested in the result, such as a grade in a course, which confirms or challenges their sense of intellectual ability. Students who have adopted these performance goals do not see the process of learning as equally valuable to the outcome because they have come to believe that their intelligence is fixed and cannot be changed, and their intelligence is evident to others by their performance (Dweck et al., 2004). Students with performance goals may believe that they are successful learners only if they achieve a desired outcome, such as an expected grade. They often do not define learning as improving their process, using better strategies, or challenging themselves (Dweck 1986, 2008). Students who develop a fixed mindset are more concerned with appearing intelligent than in MINDSET, ANXIETY, AND GRADING 12 putting forth the effort to do well because they believe that effort will not help them (Dweck, 2008). Consequences of Holding a Fixed Mindset Students can be successful at school whether they hold a fixed mindset or a growth mindset. Many top-performing students hold a fixed mindset that they have developed over years of being praised for their intelligence (Dweck, 2008). However, when students are given challenging tasks, asked to put forth effort, or when students make mistakes the disparity between mindsets may become apparent (Dweck 1986, 2008). Avoiding challenges. In these challenge situations where students must put forth effort or may make mistakes, educational research has demonstrated many negative consequences of developing a fixed mindset. In a review of research about the role of goals in task pursuit, Dweck (1986) found that performance goals led children in elementary school to shy away from challenges because they did not want to fail and appear unintelligent. Since the goal is to maintain a favorable judgment of ability, students would rather sacrifice learning opportunities than seek challenges that might result in failure and a low judgment of ability. In another study almost 20 years later, Grant and Dweck (2003) found a similar pattern among college students in pre-med organic chemistry. College students in the class that had a fixed mindset (performance goal orientation) usually failed to recover from a poor grade and used more surface processing strategies as opposed to deep processing strategies (Grant & Dweck, 2003). Surface processing strategies were defined by memorizing facts and re-reading notes which were not as effective at helping students remember material later on. By contrast, MINDSET, ANXIETY, AND GRADING 13 deep processing strategies were defined by their metacognitive nature where students would think about the theories presented in class and develop their own ideas about the theories. Surface processing strategies are linked to poorer performance on exams (Elliot, McGregor, & Gable, 1999; Grant & Dweck, 2003). In both cases, the research shows that students with a fixed mindset will either avoid challenge completely or struggle when faced with challenges. This also means that students with a fixed mindset may limit their own learning opportunities if they believe a subject or class is too challenging because they do not want to perform poorly (Dweck, 1986; Elliot & Dweck, 1988). Limiting learning could lead to lower outcomes over time. A correlational study done with elementary school students in grades three through six over two years demonstrated that having a fixed mindset was correlated to a greater rate of decline in standardized achievement scores over the course of the study (McCutchen, Jones, Carbonneau, & Mueller, 2016). The participants in the study were 419 students, all of whom were eligible for free lunch, from 28 classrooms from seven different schools in a southern city in the United States. The researchers had the students fill out a mindset questionnaire (see Dweck, 1999) in the fall and spring semesters for the two years. McCutchen and colleagues then correlated mindset scores to achievement scores in reading and math from the Iowa Test of Basic Skills – Form C (ITBS-C). Over the course of the study, students’ achievement decreased on the ITBS-C, but the rate of decline was dependent on their initial mindset. Students with a fixed mindset at the beginning of the study had a higher rate of decline than those that initially demonstrated a growth mindset (McCutchen et al., 2016). This finding could be in part due to students with fixed mindset shying away from challenges. If a student with a fixed mindset MINDSET, ANXIETY, AND GRADING 14 already believed he was poor at test taking, he might accept a lower score as confirmation of his perceived low ability of his test taking ability rather than try harder on the test (McCutchen et al., 2016). Therefore, students with fixed mindsets tend to demonstrate lower outcomes, especially when faced with challenges. Less effective cognitive strategies. Part of the reason for the discrepancies in outcomes found between students with fixed mindset and growth mindset might be due to the different strategies that they use when attempting to solve problems. Students with a growth mindset are more likely to attempt new strategies and put forth more effort when faced with challenges whereas students with a fixed mindset are more likely to give up (Dweck, 2008). Two studies found that a fixed mindset leads to more surface processing of material for college students and less effective problem-solving strategies in elementary students (Elliot & Dweck, 1988; Grant & Dweck, 2003). Grant and Dweck (2003) found that college students of chemistry at an elite university with a growth mindset were more likely to attribute failure (a poor performance on an exam) to lack of effort and take steps to fix the problem in the future. In contrast, students with a fixed mindset were more likely to attribute a series of setbacks to a lack of ability, ruminate on that failure, and disengage from the course by using more surface processing strategies for learning material, thereby leading to a lower overall grade in the course (Grant & Dweck, 2003). Nevertheless, the result of disengagement and a lower course grade was only true for students with a fixed mindset who experienced multiple failures. Students with a fixed mindset who still believed success was possible or who did not experience failure did not demonstrate lower course grades (Grant & Dweck, 2003). This indicates MINDSET, ANXIETY, AND GRADING 15 that differences in outcomes for students with a fixed mindset compared to a growth mindset only become apparent in challenge or failure situations (Dweck, 1986, 2008). The problem of ineffective strategies was not only apparent in college students. Elliot and Dweck (1988) gave 101 fifth-grade students a difficult pattern recognition test. Some of the students were led to believe that they had high ability while others were led to believe they had low ability. Also, some of the students were led to focus on performance goals while others were led to focus on learning goals. The researchers then monitored the students on the task observing the sophistication of their problem-solving strategies (Elliot & Dweck, 1988). Both ability groups performed equally well on the initial easier tasks, but when the tasks became more challenging, those who believed they had low ability and were being measured by performance goals demonstrated a decline in the sophistication of their problem-solving strategies when told they were wrong (Elliot & Dweck, 1988). Although the high-ability performance-goal group persisted with better strategies, they also passed up an opportunity to increase their skills on a task that involved public mistakes (Elliot & Dweck, 1988). Thus, both these studies show that a fixed mindset can lead to poor learning strategies and fear of failure that debilitates the student and makes the student believe he should no longer put forth effort. Another way fixed mindset appears to inhibit cognitive strategies is by reducing student ability to transfer what they have learned to new situations. In two reviews of research, Dweck (1986) and Dweck et al. (2004) cited the same study of transfer done by Farrell and Dweck (1985). In this study, junior high students were taught a challenging new science unit and then given a transfer task (Farrell & Dweck, 1985). Students learned the material equally well, regardless of whether they had a fixed or growth mindset; MINDSET, ANXIETY, AND GRADING 16 however, students with a fixed mindset tried fewer transfer strategies, produced less work, and were less successful at the task (Dweck, 1986; Dweck et al., 2004). This means that students with a fixed mindset limited their thinking compared to their growth mindset peers even when they had equal ability to begin with (Dweck, 1986; Dweck et al., 2004). Therefore, it appears that students at the elementary, junior high, and college level have all been shown to demonstrate poor cognitive strategies in response to a fixed mindset or performance goal orientation in three different studies from 1985 to 2003. Achievement gaps. Not only does growth mindset affect cognitive strategies, but the difference in these cognitive strategies can lead to achievement gaps between students with growth mindset and students with fixed mindset over time (Grant & Dweck, 2003). For example, college chemistry students at Columbia University with a fixed mindset demonstrated lower final grades in the semester-long course than their growth mindset counterparts (Grant & Dweck, 2003). Two other longitudinal studies of elementary students and secondary students demonstrate similar findings that students with fixed mindset fall behind in achievement compared to students with growth mindset over longer periods of time than just one semester (Blackwell, Trzesniewski, & Dweck, 2007; McCutchen et al., 2016). A similar longitudinal study was performed with 373 junior high students entering the 7th grade in a junior high in New York City (Blackwell et al., 2007). Blackwell and colleagues (2007) used four successive seventh-grade classes entering the school and tracked student mindsets using Dweck’s mindset questionnaire (1999) and achievement using the Citywide Achievement Test (CAT) for math. The school was varied in ethnicity, achievement, and socioeconomic status (55% African American, 27% South MINDSET, ANXIETY, AND GRADING 17 Asia, 15% Hispanic, and 3% East Asian and European American; Blackwell et al., 2007). The researchers found that a fixed mindset predicted lower math grades over the course of junior high (Blackwell et al., 2007). These studies were done with very different levels of students (college, junior high, and elementary), very different backgrounds (elite university, low socioeconomic elementary schools, and a varied junior high), and different time lengths (one semester, two years, and five years). Despite the differences, all the aforementioned studies showed that a fixed mindset is associated with lower achievement outcomes over time. Not all research agrees that students who hold fixed mindsets will achieve lower outcomes over time. Li and Bates (2017) studied 433 impoverished students in China to determine if there was a correlation between mindset and GPA over the course of two semesters. Using Dweck’s standard mindset questionnaire (1999) to assess mindset at the beginning of the study and comparing the results to initial GPA at the beginning of the year and GPA at the end of the year, the researchers found no statistically significant correlation between mindset and change in GPA. Even when controlling for initial ability, mindset did not predict GPA in direct contrast with other mindset research (Li & Bates, 2017). The researchers believed cultural differences did not account for the difference in outcomes between this study and most other mindset research because some mindset research already includes Chinese subjects, and they suggested further replicating other mindset studies to see if the results were similar (Li & Bates, 2017). Li and Bates’ research also directly contrasts with a much larger correlational study done with all the 10th grade public students in Chile which included over 160,000 students (Claro, Paunesku, & Dweck, 2016). Claro and colleagues used a dataset of all MINDSET, ANXIETY, AND GRADING 18 10th grade public school students in Chile which included the results of standardized tests that measured mathematics and language skills as well surveys of students, students’ families, and each school. In 2012, the survey included a measure of student mindset that was a short version of the standard instrument used by Dweck (1999). Claro and colleagues found that mindset accounted for 11.8% of the variance in student-level scores, the largest student-level explanation for variance (Claro et al., 2016). Fixed mindset was associated with lower standardized test scores over all socioeconomic levels. (Claro et al., 2016). Therefore, although not all of the research on the effects of mindset on outcomes is consistent, the majority of studies support the finding that a fixed mindset can lead to lower outcomes on measures of achievement (Blackwell et al., 2007; Claro et al., 2016; Grant & Dweck, 2003; McCutchen et al., 2016). How Traditional Grading Systems Contribute to Fixed Mindset A fixed mindset is not beneficial to students in challenging learning situations because it makes them focus more on their ability to perform and be successful rather than on the process of learning, which requires that students make mistakes (Dweck, 1986). Traditional grading systems can contribute to developing fixed mindsets because mistakes are punished in the form of lower grades, which makes students want to avoid mistakes, avoid challenge, and set performance goals. Students internalize grades as a measure of their intellectual ability in a subject, the outcome of their knowledge, as opposed to where they are in a process of learning (Dweck, 2008). This is often reinforced by traditional grading systems where grades on assignments and tests are averaged over the length of a term. This practice punishes students for making mistakes early in the learning process. For students who already have a fixed mindset, “school may MINDSET, ANXIETY, AND GRADING 19 often be perceived by students as a place where they are tested and judged – and, understandably, they are often resistant to this” (Dweck, 2008, p. 12). Furthermore, school systems often use grades or achievement tests to group students, which reinforces to students that their intellectual capability is fixed. Once students are placed into groups or tracks, it is very difficult for them to change to a different track affecting their beliefs about their ability and their intelligence. Even very young students are aware of what groups of different abilities means for them- that some students are smart and some are not (Boaler, 2013). Grouping by ability based on grades or achievement is not only bad for the students in the low-ability groups by reinforcing that they are not as smart as the high-ability group, but ability grouping also has negative results for members of the high-ability group by reinforcing the idea that they need to maintain an image of smartness by not appearing to work too hard or put forth too much effort (Boaler, 2013; Dweck, 1986, 2008). Evaluation Focus Finally, traditional grading practices can lead to a focus on evaluation which influences students to develop performance goals and fixed mindsets. In two correlational studies of college chemistry students, Church and colleagues (2001) demonstrated that classroom environment was correlated to the types of goals students adopted. Those goals, in turn, affected their outcomes on the final exam in the course. Specifically, the researchers found that there was negative correlation between mastery goals (growth mindset) and evaluation focus and harsh evaluation (Church et al., 2001). This means that the more the students perceived the professor to focus on evaluation and the harsher the students perceived that evaluation to be, the less likely they were to develop mastery MINDSET, ANXIETY, AND GRADING 20 goals. Similarly, evaluation focus and harsh evaluation were positively correlated with performance-avoidance goals, meaning that students who perceived the professor as focusing on evaluation and who perceived that evaluation as being harsh were more likely to focus on the goal of avoiding failure (Church et al., 2001). The students were more concerned about the evaluation than they were about learning. Therefore, focusing on evaluation led students to develop a fixed mindset. The types of evaluations and problems asked of students can contribute to fixed mindset. In this era of high-stakes testing, students can come to believe that their intelligence is determined by whatever the standardized test indicates and that it cannot be improved. Students believe that the test is a measure of their performance, their ability that is unchangeable (Dweck, 2008). Even in everyday assessment, if the teacher selects short, closed questions that have right and wrong answers, they are communicating to students that education is about being right or wrong instead of being a process through which one works over time (Boaler, 2013). Traditional grading systems, then, only provide feedback to students in the form of how right or wrong they were on the assignments and tests at the end which gives students no opportunity to improve. The grades students receive on evaluations and assignments are only a measure of how many mistakes they did, or did not make, over the course of the unit, term, or semester. This type of “right or wrong” feedback contributes to fixed mindset by convincing students that there is no room for improvement so effort is futile because intelligence is fixed. Dweck et al. (2004) explained a reason why type of evaluation might affect fixed mindset in a review of research on the effects of motivational patterns on attention and cognitive processing. The review focused on studies of people who were asked general MINDSET, ANXIETY, AND GRADING 21 information questions and then given ability-relevant feedback first (whether the answer was right or wrong) and then learning-relevant feedback next (what the correct answer was; Dweck et al., 2004). The studies measured attention using electroencephalography (EEG); the researchers were specifically looking for a type of brainwave called a stimulus preceding negativity (SPN) wave which is associated with attention. The researchers found that people with a fixed mindset focus their attention on ability-relevant feedback but not the learning-relevant feedback, whereas people with growth mindsets show the SPN attentional spike for both ability- and learning-relevant feedback (Dweck et al., 2004). This finding indicates that people with fixed mindsets focus their attention only on ability and not on what they can learn to be more success next time (Dweck et al., 2004). These studies of attention provide evidence for why a teacher focus on evaluation (Church et al., 2001), high stakes testing (Dweck, 2008), and closed feedback (Boaler, 2013) contributes to students’ development of a fixed mindset. Traditional grading systems like those most often used in the United States contribute to fixed mindset by focusing on evaluation instead of the process of learning over time; traditional grading systems punish mistakes which influences students to avoid failure by avoiding challenges further reducing their ability to learn over time. Student Anxiety High school student anxiety has been increasing in recent years. In 2015, approximately 6.3 million teenagers reported having an anxiety disorder at some point in their lives (Schrobsdorff, 2016). Although there are many and diverse causes of anxiety in general, pressure to do well in school is at least one factor that accounts for student worry (Flannery, 2018; Schrobsdorff, 2016). Pressure to do well can lead to anxiety over MINDSET, ANXIETY, AND GRADING 22 examinations, especially if they are the primary determinant of grades. In the study of 190 Indian students, 81.6% reported anxiety related to examinations (Deb et al., 2015). Traditional grading systems that average grades on examinations over the course of semester and therefore punish students for mistakes while they are still learning the material can lead to test anxiety which is associated with negative emotional outcomes for students (Deb et al., 2015; Steinmayr et al., 2016). Although anxiety over tests appears to be a problem that is increasing in the United States (Hembree, 1988), the majority of test anxiety research occurs outside of the United States (Von Der Embse, Barterian, & Segool, 2013). For example, the aforementioned survey of academic stress in India demonstrated that 63.5% of students felt stressed because of academic pressure and 66% felt parental pressure to do better in school (Deb et al., 2015). In this survey, 190 11th and 12th grade students were asked to complete two questionnaires including the General Health Questionnaire (which is designed to diagnose minor psychiatric disorders in the general population) and a structured questionnaire that asked about demographic and socio-economic information, stress, exam anxiety, future aspirations, and extra-curricular activity. Although the sample size was relatively small compared to all the high school students in India and the Indian school systems focus more on examinations that determine student’s future career paths than do the school systems in the United States, the high levels of concern over academic success appears to be an international problem. Additionally, 32.6% of students in the survey scored above the threshold on the General Health Questionnaire to be diagnosed with psychiatric caseness. This means that almost a third of the students surveyed qualified as having an anxiety disorder (Deb et al., 2015). MINDSET, ANXIETY, AND GRADING 23 This rate of anxiety disorders found in the Indian survey is very similar to the rate found in teenagers in the United States (Schrobsdorff, 2016). Another international study that demonstrates how pressure to be successful can increase student anxiety came from Chan (2012) in his study of gifted Chinese children. In this survey, Chan (2012) asked students to fill out four questionnaires including: the Almost Perfect Scale- Revised, Satisfaction With Life Scale, Oxford Happiness Questionnaire, and Mindset Rating Form. Chan (2012) found that students who develop unhealthy perfectionism (as measured by a focus on discrepancy between their personal standards and their actual achievements in the Almost Perfect Scale) had a lower happiness rating as measured by the Oxford Happiness Questionnaire compared to their healthy perfectionist and non-perfectionist peers. This means that when students focused on not achieving as well as they expected, they were more likely to be unhappy (Chan, 2012). Interestingly, students who were described as unhealthy perfectionists also correlated with having a fixed mindset which might further increase their anxiety over evaluation (Church et al., 2001). Because this study was done with only gifted students identified by their teachers and it included both primary and secondary school students, it may not generalize to the high school population in the United States. Nevertheless, it provides further evidence that student anxiety is a problem which is at least partially influenced by concern over academic success, especially when student expectations are unmet. Consequences of Test Anxiety Test anxiety is associated with a multitude of negative outcomes for students including: lower moods (Steinmayr et al., 2016), lower self-esteem (Hembree, 1988), and MINDSET, ANXIETY, AND GRADING 24 lower academic achievement (Chapell et al., 2005; Hembree, 1988; Osborne, 2001; Von Der Embse & Hasson, 2012). A correlational study of German students suggests that test anxiety is associated with lower levels of well-being. (Steinmayr et al., 2016). In this study, 290 German high schoolers were asked to fill out the Habitual Subjective Well-Being questionnaire that has a mood component and a life satisfaction scale. They were also asked to fill out a short version of the German Test Anxiety Questionnaire which measures both a cognitive component of test anxiety (worry) and an emotional component (physiological response). They were tested once in 11th grade and then given the same questionnaires in 12th grade to measure any changes over that time. The researchers found that worry from the test anxiety questionnaire was related to negative changes in both parts of subjective well-being (Steinmayr et al., 2016). This means that students who were worried about doing well on tests had lower mood scores and had lower life satisfaction than those who were less worried. The researchers, however, did not find a correlation between the emotional component of test anxiety and either part of the subjective well-being (Steinmayr et al., 2016). This means that physiological responses such as an increased heart rate or the feeling of “butterflies” in the stomach did not affect student mood or life satisfaction in this study. This seeming contradiction could be explained by Hembree’s (1988) meta-analysis of 562 studies on test anxiety. In the meta-analysis, he found evidence to suggest that test anxiety can be facilitative or debilitative depending on the student’s interpretation of their emotional symptoms (Hembree, 1988). For example, a student who experiences an increased heart rate before a test and interprets this as nervousness might be debilitated by this fear and do worse on the test. Opposingly, a student who MINDSET, ANXIETY, AND GRADING 25 experienced an increased heart rate before a test and interprets this as excitement might be facilitated by this feeling and do better on the test. The concept of debilitative and facilitative anxiety in the same situation based on interpretation is consistent with psychological theories of motivation (King, 2103; Myers, 2005). Consequently, the emotional component may not have a significant effect on subjective well-being because some students do not negatively interpret these emotions (Hembree, 1988; Steinmayr et al., 2016). Hembree’s (1988) research further explains why the cognitive component of test anxiety is associated with lower levels of subjective well-being. Worry is directly related to negative expectations about outcomes; students who worry have already interpreted their anxiety as debilitative which means that worry will result in negative feelings and beliefs about testing (Hembree, 1988; King, 2013; Myers, 2005). In addition to lower life satisfaction, test anxiety is also associated with lower self-esteem. In his meta-analysis, Hembree (1988) found that students with high test anxiety were more negative toward themselves and toward evaluation than students with low test anxiety. Also, students with high test anxiety demonstrated strong fears of negative evaluation (Hembree, 1988). This further corroborates Steinmayr and colleagues (2016) findings that test anxiety can lead to negative emotional outcomes for students. Despite the similarities, these two studies disagree about how to best reduce test anxiety among students. Steinmayr and colleagues (2016) recommended that schools should develop test-anxiety interventions that focus on worry because if students are worried, some of their cognitive capacity is used attending to information other than test content. Hembree (1988), however, found that intervention strategies that focused on treating the emotional component of test anxiety were effective at treating both worry and MINDSET, ANXIETY, AND GRADING 26 emotionality whereas interventions that focused on treating worry were not effective at reducing either component. This contradiction could be at least partially explained by the nearly 30-year gap between the two studies. Perhaps students are more worried about tests today than they were in 1988 (Von Der Embse & Hasson, 2012) or perhaps test anxiety interventions that focus on the cognitive component have improved over time. In either case, both studies agree that test anxiety leads to negative outcomes and should be reduced if possible. Another consequence of test anxiety is lower academic achievement. Hembree’s (1988) meta-analysis of 562 studies of test anxiety showed that all relationships between test anxiety and performance were inverse and negative. The relationships were significant from grades 3-12 (Hembree, 1988). Also, the relationship between test anxiety and test scores and the relationship between test anxiety and course grades were negative (Hembree, 1988). This indicates that test anxiety is associated with lower academic performance. Other researchers found a similar connection between test anxiety and lower performance in college students (Chapell et al., 2005). In a study of 4000 undergraduate and 1414 graduate students from three universities in the eastern United States, researchers asked students to complete questionnaires including Spielberg’s (1980) Test Anxiety Inventory, a self-reported GPA, and a socioeconomic section assessing the mean educational level of their parents. The researchers found that test anxiety was significantly correlated to lower GPAs at both the graduate and undergraduate levels (Chapell et al., 2005). Although the student GPA differences between those with high- and low-test anxiety was significant, it was also small. The average GPA for students MINDSET, ANXIETY, AND GRADING 27 with lower levels of test anxiety was approximately a B+ whereas the average GPA for students with higher levels of test anxiety was approximately a B (Chapell et al., 2005). The researchers questioned whether such a small overall difference in GPA indicated a problem that should receive clinical attention (Chapell et al., 2005). Nevertheless, the college students showed the same trends of poorer academic performance among students with high anxiety as the high school students in the meta-analysis. More studies corroborate this trend. For example, Von Der Embse & Hasson (2012) studied 75 tenth graders, 35 from a suburban school and 40 from an urban school, hypothesizing that test anxiety had a significant relation with high-stake tests. The researchers administered the Friedben Test Anxiety Scale a week before the students took the Ohio Graduation Test. The Ohio Graduation Test was considered a high-stakes exam because a passing score is required to graduate. The researchers found a negative correlation between the anxiety questionnaire and score on the Ohio Graduation Test (Von Der Embse & Hasson 2012). This means that students with higher levels of anxiety did poorer on this high stakes exam. The sample size was very small compared to the other studies of anxiety mentioned, but it provides further evidence that test anxiety is associated with lower academic achievement on tests in high school students. Further research has demonstrated that test anxiety can explain some of the variance observed on tests. For example, in the Von Der Embse & Hasson (2012) study, 15% of the variance observed on the math portion of the Ohio Graduation Exam could be attributed to test anxiety. Additionally, 9% of the variance on the social studies portion and 7% of the variance on the science portion on the Ohio Graduation Exam could be MINDSET, ANXIETY, AND GRADING 28 attributed to test anxiety (Von Der Embse & Hasson, 2012). This is important because it indicates that student scores might be more accurate if students had less test anxiety. A similar study of stereotype threat shows that test anxiety contributes to variance (Osborne, 2001). Stereotype threat is the idea that people feel at risk of conforming to a stereotype about a group to which they belong (Osborne, 2001). Stereotype threat is used in education to partially explain achievement gaps between majority culture and minority culture students. The topics most often studied under stereotype threat are achievement gaps between Whites and African Americans, Whites and Latinos, Whites and Native Americans, and math achievement gaps between males and females (Osborne, 2001). Osborne (2001) proposed that stereotype threat might increase student test anxiety which thereby might explain lower achievement test scores. The study included 15561 seniors from high schools around the United States as identified by the High School and Beyond study. The participants took achievement tests in various subjects (vocabulary, reading, and mathematics were the only subjects used in the final correlation), then, were asked to fill out an anxiety questionnaire about how they felt while taking the tests, a questionnaire about academic preparedness, and a questionnaire on demographic variables which included gender, race, and ethnicity (Osborne, 2001). Osborne (2001) used the demographic variables and separated students into the most homogenous groups by race and ethnicity. This excluded about half of the participants because they did not fit neatly into one of the pre-determined demographic categories (White, African American, Latino, or Native American). Osborne (2001) then analyzed the data looking for how much anxiety could explain achievement differences among these groups. He found that up to 23% of the variance in achievement test scores were due to race and that up to 41% MINDSET, ANXIETY, AND GRADING 29 of these effects could be explained by differing levels of anxiety (Osborne, 2001). His study provides evidence for stereotype threat as a result of the variance due to race. Osborne’s (2001) evidence also shows that test anxiety contributes to achievement differences among students. Osborne (2001) also compared math achievement scores among males and females, but he found no significant differences to support stereotype threat in that situation. This opposes other evidence that females have higher general test anxiety than males. Both Hembree’s (1988) meta-analysis and Chapell and colleagues’ (2005) study found that females tend to have higher test anxiety scores than males. This does not necessarily mean, however, that females always score lower than males on tests. Chapell and colleagues (2005) discovered the opposite; although female students had higher test anxiety, they also had a higher average GPA. The researchers suggested further studies into this particular outcome as it was not previously found in other literature on test anxiety (Chapell et al., 2005). The research seems to indicate that test anxiety is associated with lower academic achievement for both males and females, particularly on tests (Chapell et al., 2005; Hembree, 1988; Von Der Embse & Hasson, 2012). There is still debate whether test anxiety is the cause or the result of the poorer performance on tests. Hembree (1988) found in his meta-analysis that ability levels affect test anxiety. Students with high ability usually had lower test anxiety and vice versa (Hembree, 1988). Steinmayr and colleagues (2016) proposed the opposite. These researchers found no correlation between GPA and test anxiety which means, according to the researchers, that poor performances do not increase the risk of a student developing test anxiety (Steinmayr et al., 2016). The meta-MINDSET, ANXIETY, AND GRADING 30 analysis is compelling because of the number of studies it used, but the age might call its validity into question today. As a result, the research indicates that test anxiety is a significant problem among high school students. It can lead to negative emotional outcomes such as more worry and fear of negative evaluation (Deb et al., 2015; Hembree, 1988; Steinmayr et al., 2016). Test anxiety is also associated with negative outcomes on tests which leads to lower academic achievement (Chapell et al., 2005; Hembree, 1988; Von Der Embse & Hasson, 2012). How Traditional Grading Systems Contribute to Test Anxiety Traditional grading systems that focus on points and percentages on assignments and that average grades over the length of a semester punish students for imperfections while they were still learning the material (Iamarino, 2014). Mistakes are part of the learning process. If teachers punish students with lower scores on assignments whenever they make mistakes, students will quickly learn that the only thing that matters is being right, not the process of learning (Boaler, 2013). Consequently, students will start to fear evaluation because they do not know if they will be able to achieve their own or others’ expectations (Chan, 2012; Hembree, 1988). Punishing every mistake by taking away points can lead to test anxiety among students. Next, because of accountability-based education, students are experiencing more and more high-stakes tests (Von Der Embse & Hasson, 2012). The more tests students are given, the more opportunities students have to become anxious. Students start to see school as a place where they are constantly evaluated, not a place where they come to learn (Dweck, 2008). When students believe that performance in the form of good grades MINDSET, ANXIETY, AND GRADING 31 is the only thing that matters, they naturally become anxious about achieving that goal (Scriffiny, 2008). Finally, traditional grading systems can increase student anxiety by putting pressure on teachers to help their students achieve higher and higher test scores. Hembree (1988) found that teacher anxiety correlated with student test anxiety. This means that the more teachers are worried about evaluation, the more anxious the students become. Also, harsh evaluation by teachers was associated with performance-avoidance goals in Church and colleagues’ (2001) study of undergraduate students which means that students who felt the evaluation method was harsh were more likely to fear failing and looking incompetent compared to others. Thus, teacher evaluation method does influence student motivation and anxiety. Standards-Based Grading as an Alternative If traditional grading practices can negatively impact students, then an alternative system for evaluating students’ needs to be considered. Standards-based grading focuses on mastery of very specific content objectives as opposed to assigning points to assignments which are weighted and averaged over the course of a term or semester (Miller, 2013; Scriffiny, 2008). In standards-based grading, students have to know what the objectives of the course are and how they can demonstrate mastery of each objective (Miller, 2013; Munoz & Guskey, 2015). When they are evaluated on each objective, students are given feedback on how their work and understanding approaches the expectation (for example, “advanced,” “proficient,” “basic”) and given detailed descriptions on what was done well or what could be improved (Miller, 2013). MINDSET, ANXIETY, AND GRADING 32 Standards-based grading helps student cognitive well-being because it makes them more responsible for their own learning. The purpose of this grading system is to give students more feedback on their learning process and, in some cases, more opportunities to revise their understanding (Iamarino, 2014; Miller, 2013; Scriffiny, 2008). Process feedback is more likely to develop a growth mindset as opposed to a fixed mindset in students because they can feel successful by progressing toward the ultimate goal even if they do not have a complete or proficient understanding of a specific objective yet (Dweck, 2008; Masters, 2013; Munoz & Guskey, 2015). Since learning is a process that requires mistakes, standards-based grading allows students to make mistakes without fear that it will ruin their grade at the end of the term (Scriffiny, 2008). If students are not afraid of making mistakes, they are more likely to challenge themselves and achieve better than the teacher expected (Iamarino, 2014). Standards-based grading can also improve student emotional well-being by reducing anxiety over grades (Masters, 2013; Miller, 2013). Many students feel that grades are arbitrary measures; students think that if they comply with teacher requests, then they should automatically get good grades in a class regardless of how well they understand the content (Iamarino, 2014; Miller, 2013; Scriffiny, 2008). Students in a traditional grading system that assigns points for assignments like homework or even non-academic factors such as attendance can be confused when they score well on everything that involves participation but then become anxious when they fail a test because they do not know the content (Scriffiny, 2008). When the objectives for the course are clear and the evaluation method is known, students can have less anxiety because they know exactly how they will be graded and what the expectation is for MINDSET, ANXIETY, AND GRADING 33 success (Miller, 2013; Munoz & Guskey, 2015). With less anxiety over grades, students are better able to use their attentional resources to understanding the content to the best of their ability. Despite the benefits of standards-based grading, both teachers and students have concerns if they are used to a traditional grading system and they switch to a standard-based grading system. Teacher concerns tend to focus around student motivation; some teachers believe that students will not learn unless motivated by grades, both good and bad (Miller, 2013; Scriffiny, 2008). Miller (2013) addresses this concern when she writes, “we can trust students to want to learn when they are sure it is their learning rather than their compliance that is our first concern” (p.112). This indicates that teachers can motivate students if they focus on the process of learning rather than the product. Student concerns focus around the new responsibility that they are required to take for their own learning. In traditional grading systems, students are more concerned about their grade than how to understand the content better (Iamarino, 2014). In standards-based grading systems, students are sometimes expected to revise, edit, and try again multiple times in order to show mastery of the same objective (Iamarino, 2014; Miller, 2013). Although this process of revision is more typical of what happens in life outside of school, students are not used to it, and so may resist standards-based grading at first (Iamarino, 2014). Nevertheless, when students understand the purpose of standards-based grading, they are usually more willing to learn and to take risks because they know that is a requirement for learning (Iamarino, 2014; Miller, 2013). Standards-based grading should be considered as an alternative to traditional-grading systems because it is more likely to improve student cognitive and emotional well-being by promoting growth mindset and MINDSET, ANXIETY, AND GRADING 34 reducing test anxiety. There is a lack of research on how mindset, test anxiety, and grading may be related which indicates a need for further research on these topics. Summary Overall, traditional grading systems can decrease student emotional and cognitive well-being. The focus on points and percentages instead of mastery of content entices students to limit their own learning opportunities and avoid challenges (Scriffiny, 2008). Averaging grades over the length of a term or semester punishes students for making mistakes which increases their anxiety, reduces their achievement, and encourages them to fear evaluation (Boaler, 2013; Chapell et al., 2005; Hembree; 1988; Steinmayr et al., 2016; Von Der Embse & Hasson, 2012). Traditional grading systems can decrease student emotional well-being by promoting test-anxiety. Students who worry about tests have lower mood levels and life satisfaction than their peers without this component of test anxiety (Steinmayr et al., 2016). Also, students with high levels of test anxiety tend to have negative feelings toward themselves, toward tests, and to fear being negatively evaluated (Hembree, 1988). All this negativity can spill over into other aspects of their lives as test anxiety is strongly correlated with trait anxiety (Hembree, 1988). When students fear they cannot live up to expectations set for them, they can develop unhealthy perfectionism which is associated with lower life satisfaction (Chan, 2012). At the same time, traditional grading systems can decrease student cognitive well-being by promoting fixed mindset and test anxiety, both of which are associated with decline in academic achievement. Students tend to believe that tests are about fixed ability and not about where they are at this moment (Dweck, 2008). This belief can MINDSET, ANXIETY, AND GRADING 35 influence students to develop a fixed mindset which has been linked to an overall decline in achievement over time (Blackwell et al., 2007; Dweck et al., 2004). Fixed mindset is also associated with lower levels of intrinsic motivation (Church et al., 2001) which can lead students to avoid challenges and limit their learning opportunities (Boaler, 2013; Dweck, 2008). By focusing only on whether answers are correct or incorrect, traditional grading systems can influence students to only attend to the emotional response of being right or wrong which therefore uses precious attentional resources on thoughts not related to learning (Dweck et al., 2004). Test anxiety has a similar effect because when students worry about a test, they are using their attentional resources on their worry as opposed to recalling course content for an exam (Hembree, 1988). This is the premise behind the test anxiety caused by stereotype threat (Osborne, 2001). Traditional grading systems can also promote test anxiety which is negatively associated with academic performance (Chappell et al., 2005; Hembree, 1988; Osborne, 2001; Von Der Embse & Hasson, 2012). Consequently, traditional grading systems that promote fixed mindset and test anxiety can reduce student cognition. Classroom environment and evaluation practices can affect student goals, motivation, and anxiety which in turn affects student performance. Teacher feedback, whether process or performance oriented, can also affect what types of goals students focus on (Dweck, 2008). Similarly, school systems and teachers that require rote memorization likely increases student test anxiety (Deb et al., 2015). Teachers that focus on evaluation can influence students to develop performance goals which are associated with a fixed mindset (Church et al., 2001). Likewise, teacher anxiety over evaluation tends to influence student test anxiety (Hembree, 1988). Teachers and classroom MINDSET, ANXIETY, AND GRADING 36 environment matter because interventions to teach growth mindset can be effective (Blackwell et al., 2007) as are interventions to treat test anxiety (Hembree, 1988). Teachers, then, have a great responsibility to encourage practices that will help their students succeed not only academically but also cognitively and emotionally. Standards-based grading may be a better alternative to traditional grading practices (Iamarino, 2014; Masters, 2013; Miller, 2013; Munoz & Guskey, 2015; Scriffiny, 2008). Standards-based grading may help students cognitively by encouraging growth mindset and emotionally by reducing test anxiety. MINDSET, ANXIETY, AND GRADING 37 PURPOSE Traditional grading practices can decrease effective cognition by encouraging students to develop performance goals which are associated with fixed mindset (Dweck, 1986, 2008). Students with a fixed mindset tend to shy away from challenges and sometimes limit their own learning opportunities because the outcome, such as a grade, seems more important than developing new skills (Dweck, 2008; Long, 2015). Traditional grading practices may also decrease student emotional well-being by increasing test anxiety which is associated with lower life satisfaction (Deb et al., 2015, Steinmayr et al., 2016; Von Der Embse & Hasson; 2012). Test anxiety may also interfere with accurate evaluation because test anxiety is negatively associated with performance on evaluations, and test anxiety accounts for some of the variance in test scores (Steinmayr et al., 2016; Von Der Embse & Hasson; 2012). Therefore, traditional grading practices need to be reconsidered as a result of evidence that they promote poor cognitive and motivational skills and increase student anxiety related to school (Miller, 2013). The goal of this correlational study was to better understand the relationship between fixed mindset, test anxiety, and attitudes toward grading among various demographic groups. The research questions to be answered were as follows: 1) What was the relationship between student scores on a mindset questionnaire, a test anxiety questionnaire, and an attitudes questionnaire about grading? 2) What were the patterns of correlation between the questionnaires and demographic factors including gender and grade level? MINDSET, ANXIETY, AND GRADING 38 METHODS This project was a correlational study of growth mindset, test anxiety, attitudes toward grading, and student demographic information. The purpose was to determine if there were any statistically significant relationship between growth mindset, test anxiety, and attitudes grading, and if the relationship among these factors were influenced by demographics. To complete the study, students at a suburban high school filled out Dweck’s (1999) growth mindset questionnaire, Nist and Diehl’s (1990) test anxiety questionnaire, and attitudes toward grading questionnaire designed by a master’s student at the University of Montana (Mannix, 2014). Student responses on each questionnaire were recorded and the students also provided the following demographic information: gender and grade level. After the responses and demographic information was collected, the data was uploaded into a statistical software (SPSS) and analyzed for correlations. Participants The participants were students from a large suburban high school in northern Utah. Students at this school were in grades 10-12, ages 15-18, and enrollment was approximately 2000 students. According to the Utah State Board of Education (Utah State Board of Education, 2017) the student body was approximately 88% Caucasian, 8% Hispanic, and 1% or less American Indian, Asian, Pacific Islander, or African American at the time of the study. About 17% of the student body qualified for free or reduced lunch and approximately 11% of the student body qualified for special education. The study was explained to the students in their FLEX class (study-hall) by their FLEX teacher, and a consent form was sent home for parents and students. All volunteers that returned the form were allowed to participate. Of the 91 teachers at the school, 54 agreed MINDSET, ANXIETY, AND GRADING 39 to announce the opportunity to participate to their FLEX students. The average size of a FLEX class at the school is 30 students. Of those 54 teachers that agreed to announce the opportunity, 46 teachers had one or more students who chose to participate in the study. One-hundred ninety-six students, of which there were 126 females, 67 males, and 3 who selected other, chose to participate by obtaining permission from their parents to fill out the questionnaires and demographic information. Of those 196 students, there were 75 from 10th grade, 68 from 11th grade, and 53 from 12th grade. Instruments Growth mindset was assessed using Dweck’s Mindset Instrument (DMI) which was created by Carol Dweck (1999) to assess how students view their own intelligence and talent (see Appendix A). This instrument was selected because it was the most commonly used instrument to test mindset in the research literature. It is most often used because of its high validity and reliability with other measures of mindset (Dweck, Chiu, & Hong, 1995) which means that if the instrument ranks people as having a fixed mindset, they actually demonstrate those characteristic thinking patterns in their lives. The DMI consisted of 16 items that students rank on a Likert-scale of 1 to 6 with 1 corresponding to “strongly agree” and 6 corresponding to “strongly disagree.” Some of the statements represented an entity view of intelligence (fixed mindset) while the other statements represented an incremental view of intelligence (growth mindset). In order to average the scores for the 16 items, the numbers a student selected on the Likert-scale for the incremental statements were reversed. This means that a student who selected 6 or “strongly disagree” for an incremental item was scored as if he or she had selected 1 or “strongly agree” for an entity item. Therefore, lower scores on the DMI correspond to a MINDSET, ANXIETY, AND GRADING 40 student having more of a fixed mindset. Items 1-8 on the DMI assessed views toward intelligence while items 9-16 assessed views toward talent. The responses to the first eight items were averaged with the incremental item scores being reversed. Then, the last eight items were averaged with the incremental item scores being reversed. Finally, the averages of views toward intelligence and view toward talent were averaged to get an overall growth mindset score. Students who score 1-3 were described as having a fixed mindset because they tend to agree more with the entity view statements. Students who score 4-6 were described as having a growth mindset because they tend to agree more with the incremental view statements. A score between 3 and 4 indicated the student has mixed views about growth or fixed mindset. Test anxiety was assessed using Nist and Diehl’s (1990) Test Anxiety Questionnaire (see Appendix A). This questionnaire was selected because it is relatively short, consisting of only ten items, and because it represents the most commonly accepted psychological framework that some anxiety can be beneficial while too much can inhibit performance (King, 2103; Myers, 2005). For this questionnaire, students read each statement and ranked each statement on a Likert-scale from 1 to 5 with 1 corresponding to “Never” and 5 corresponding to “Always.” Each ranking was added together to get a final score that can range from 10-50. Scores between 10-19 indicated low or no test anxiety. Scores between 20-35 indicated moderate or healthy test anxiety. Scores above 35 indicated high test anxiety that might interfere with performance. Student attitudes toward grading were assessed using an excerpt from a student attitudes and motivation survey developed by a master’s student at the University of Montana while studying the effects of standards-based grading on his classroom MINDSET, ANXIETY, AND GRADING 41 (Mannix, 2014). The survey was modified for the purposes of this study by removing seven items (items 13-19 in the original questionnaire) that did not relate to student attitudes toward grading. The seven items that were removed focused on student motivation specifically in science because that was the subject taught by the original developer. The remaining 12 items were reworded to represent an ideal by adding the word “should” because the students in the original research had been exposed to standards-based grading practices and were being asked to rate them as they had already occurred. Most of the students in this study have not had experience with standards-based grading, so this questionnaire was reworded to represent student attitudes toward grading possibilities in general making it more authentic for students who have not had experience with standards-based grading. For example, one item in the original study stated, “My grade is fair.” For the purposes of this study, the item was reworded to state, “My grade should be fair.” Finally, item 8 in the Mannix (2014) questionnaire was split and reworded to show two different ideas about the use of classwork. The original stated, “I do classwork because I think it is important.” In the new questionnaire this item was reworded to, “Doing classwork is important because it is worth points and improves my grade.” The researcher also added one item to the end of the revised questionnaire that state, “Doing classwork is important because it helps me learn the material,” to determine if there was a discrepancy between student opinions on these two statements. The 13 items on the revised questionnaire were presented to students to rank each statement on a Likert-scale from 1 to 5 with 1 corresponding to “Strongly Disagree” and 5 corresponding to “Strongly Agree.” The original questionnaire can be found in Appendix MINDSET, ANXIETY, AND GRADING 42 B. The updated questionnaire that was presented to students in this study can be found in Appendix A. Demographic information including FLEX teacher name, gender, and grade level were also collected from students in addition to responses to the three questionnaires. The demographic questions were used to disaggregate the data to look for patterns. The FLEX teacher names were collected to verify consent. Procedure 1. The researcher obtained permission from the Institutional Review Board at Weber State University (see Appendix C) and from the school district in which the high school being studied belonged (see Appendix D). 2. The researcher explained the study to FLEX teachers at the school by individually visiting teachers in their classroom. FLEX was a study-hall type class where students could work on homework or get extra help from teachers. Most students at the school were enrolled in a FLEX class. Fifty-four teachers agreed to announce the study to their FLEX students. 3. The FLEX teachers were given a parent letter attached to a consent form (see Appendix E), and questionnaires (see Appendix A) to give to students who returned the consent form. The FLEX teachers who agreed to announce the study explained the opportunity to participate to the students and sent home a parent letter and consent form with each student. 4. Students who obtained permission and returned the consent form were asked to fill out the student questionnaire at that time during FLEX class. After the students completed the questionnaire, they returned it to their FLEX teacher immediately, and MINDSET, ANXIETY, AND GRADING 43 their FLEX teacher gave them a package of gum and an information sheet on how to improve growth mindset and reduce test anxiety (see Appendix F). 5. FLEX teachers were given one week to collect student responses after which the researcher collected the responses. Student responses were uploaded into a spreadsheet. 6. After the researcher received the information, the researcher uploaded the data into a statistics software (SPSS) and then analyzed the data looking for correlations among growth mindset, test anxiety, and attitudes toward grading as well as how student demographics influence these relationships. MINDSET, ANXIETY, AND GRADING 44 RESULTS Traditional grading practices may reduce student cognitive and emotional well-being by encouraging students to develop a fixed mindset and by increasing test anxiety both of which are related to lower levels of academic achievement (Blackwell et al., 2007; Chapell et al., 2005; Claro et al., 2016; Grant & Dweck, 2003; Hembree, 1988; McCutchen et al., 2016; Osborne, 2001; Von Der Embse & Hasson, 2012). The purpose of this correlational study was to determine possible relationships between measures of student mindset, test anxiety, and attitudes toward grading. The study also sought patterns of correlation between the measures and demographic factors including gender and grade level. A total of 196 students participated in this study. More than 80% of the responding students scored within the range of growth mindset on the mindset questionnaire. On the test anxiety questionnaire, the average score was within the range of moderate test anxiety while 83.3% of student responses ranged from moderate to high test anxiety (see Table 1). Table 1 Descriptive Statistics of Mindset and Anxiety Questionnaire Questionnaire Minimum Maximum Mean Standard Deviation Mindset 2.5 6.0 4.68 .72 Anxiety 11 49 30.33 8.30 MINDSET, ANXIETY, AND GRADING 45 Research Question One To answer the first research question “What was the relationship between student scores on a mindset questionnaire, a test anxiety questionnaire, and an attitudes questionnaire about grading?” correlation coefficients were calculated for the mindset and text anxiety questionnaires. The researcher found a small yet statistically significant negative correlation (r=-0.164, p<.05) between scores on a mindset questionnaire and test anxiety questionnaire. This meant that higher mindset scores, which indicated a stronger growth mindset, were associated with lower levels of test anxiety. There was a small positive correlation between scores on a mindset questionnaire and items 1, 3, 5, 7, 9, 12, and 13 of a grading questionnaire (see Table 2 for correlations and Appendix A for the full grading questionnaire) meaning that students with higher mindset scores (indicating a growth mindset) were more likely to agree with these items. Conversely, there were small negative correlations between these seven items and test anxiety scores, meaning that the higher a student’s test anxiety score the less likely the student was to agree with the seven items (see Table 2). Items 1 and 5 were about equity and fairness in grading with item 1 stating, “My grade should be fair,” and item 5 stating, “My grade should accurately reflect what I know.” Students with higher scores on the mindset questionnaire were more likely to agree with these statements while students higher test anxiety were less likely to agree (see Table 2). Items 3 and 12 from the grading questionnaire were about being able to learn class material and improve a class grade with item 3 stating, “I should be confident that I can improve my grade,” and item 12 stating, “I should be able to learn even the hard concepts if I don’t give up and have enough time.” Students with higher mindset MINDSET, ANXIETY, AND GRADING 46 scores were more likely to agree with these statements while students with higher test anxiety scores were more likely to disagree with these statements (see Table 2). Items 7, 9, and 13 related to student attitudes toward homework and classwork with item 7 stating, “Doing homework helps me learn the class material,” item 9 stating, “Doing homework is worth it because it is worth points and improves my grade,” and item 13 stating, “Doing classwork is important because it helps me learn the material.” Students with a higher mindset score were more likely to agree with these statements while students with a higher test anxiety score were more likely to disagree with these statements (see Table 2). Table 2 Correlations of Mindset, Test Anxiety, and Items on a Grading Questionnaire Grading Item Number Mindset Correlation Test Anxiety Correlation 1 .262** -.272** 2 .053 -.110 3 .150* -.185* 4 .137 -.072 5 .182* -.190** 6 .070 -.132 7 .173* -.287** 8 .114 -.172* 9 .209** -.176 10 .128 -.221** 11 -.096 .108 12 .169* -.178* 13 .147* -.245** Note. *p < .05 **p < .01 MINDSET, ANXIETY, AND GRADING 47 Although there was no statistically significant relationship between mindset and items 8 and 10 on the grading questionnaire, there was a significant negative correlation between test anxiety and items 8 and 10 (see Table 2). Therefore students with higher test anxiety were less likely to agree with items 8 and 10. Item 8 stated, “Doing classwork is important because it is worth points and improves my grade,” and item 10 stated, “Completing homework thoroughly and on time should be an important part of my grade.” Finally, there was no significant correlation between either the mindset score or test anxiety score and items 2, 4, 6, or 11 of the grading questionnaire (see Table 2). Research Question 2 To answer the second research question, “What were the patterns of correlation between the questionnaires and demographic factors including gender and grade level?” mean and standard deviation were used to compare demographics. Correlation coefficients were also found to determine the relationship between the questionnaires and demographic factors. The average mindset score for males and females was very similar in this study with males scoring very slightly higher, on average, than females (see Table 3). Because only three students selected “other” as their gender, these results were not used for correlations as the sample size was too small to be reliable. For both genders analyzed the average fell within the growth mindset range. The average test anxiety score for females in this study was much higher than that of males. The females average test anxiety was slightly less than one standard deviation above the average for males although both averages were within the range of moderate test anxiety. Responses by males were MINDSET, ANXIETY, AND GRADING 48 toward the lower end of the range and female average scoring toward the top of the moderate range (see Table 3). Table 3 Descriptive Statistics for Mindset and Anxiety by Gender Gender Males Females Mindset Mean 4.73 4.68 Mindset Standard Deviation .69 .71 Anxiety Mean 25.61 32.83 Anxiety Standard Deviation 7.40 7.64 When mindset and anxiety were analyzed for correlations by gender, there was no significant correlation between mindset and anxiety for males (r=.006, p>.05). However, there was a significant negative correlation between mindset and anxiety for females (r=-.293, p<.01). When the data were analyzed based on grade level, the average mindset scores and the average test anxiety scores were very similar for 10th, 11th, and 12th grade students (see Table 4). The average mindset scores for each grade level fell within the growth mindset range. The average test anxiety scores for each grade level fell within the moderate test anxiety range. When the data were analyzed for correlations among mindset and test anxiety by grade level, the researcher found no significant correlations between the measures and MINDSET, ANXIETY, AND GRADING 49 grade level (see Appendix G for full table of correlations). Thus, grade level was not related to mindset nor test anxiety. Table 4 Descriptive Statistics for Mindset and Anxiety by Grade Level Grade 10 11 12 Mindset Mean 4.70 4.61 4.75 Mindset Standard Deviation .72 .74 .69 Anxiety Mean 29.99 30.88 30.11 Anxiety Standard Deviation 7.92 8.25 8.99 Note. Mindset scores ranged from 2.5-6. Anxiety scores ranged from 11-49. Because gender was the only demographic factor that linked mindset and anxiety in this study, gender was also correlated with each of the items on the grading questionnaire. Females had statistically significant lower agreement to only items 1 and 13 on the grading questionnaire (see Appendix A for grading questionnaire and Appendix G for full table of correlations). MINDSET, ANXIETY, AND GRADING 50 DISCUSSION The purpose of this study was to determine if there were relationships between mindset, test anxiety, and attitudes toward grading. In this study, 80% of participants scored within the growth mindset range with only three of the 196 participants scoring within the range of fixed mindset. This contrasts sharply with other research on mindset which demonstrates that in the general population 40% of people hold a fixed mindset, 40% of people hold a growth mindset, and 20% of people hold mixed views (Boaler, 2013; Dweck, 2008). The difference between the mindset scores in this study and prior research may signal a selection threat to the validity of the conclusions. The participants may not be representative of the population as a whole because those who chose to participate had to put forth effort to get a consent form signed, bring it back, and use their own time to participate in the study. Because effort is associated with growth mindset (Boaler, 2013; Dweck, 1986, 2008; Dweck et al., 2004), the procedures used to comply with ethical standards may have unintentionally led to a sample with more growth-mindset oriented sample than the population. Another possible explanation is that the population as a whole is becoming more growth-mindset oriented. The previous studies were conducted before mindset was a well-known topic. Students today may be more familiar with the benefits of fostering a growth mindset because they have exposure to the concept and therefore are more growth mindset-oriented as a population. Further research could be done to determine if the same percentages for mindset groups still hold true today. Alternatively, more exposure to the benefits of growth-mindset may have preset students to believe that growth-mindset responses are more desirable thereby influencing them to choose growth-mindset responses even if they do not truly believe in the incremental view of intelligence. MINDSET, ANXIETY, AND GRADING 51 Despite the more limited range of mindset scores, the test anxiety scores covered almost the entire spectrum of test anxiety from virtually no test anxiety to very severe test anxiety. Nevertheless, a large majority of students (83.3%) indicated at least moderate test anxiety or above which corroborates with measures in other countries such as the study of Indian high school students where 81.6% of students reported anxiety related to examinations (Deb et al., 2015). These results also indicated that test anxiety affected most of the students in the study, which supports the existing literature that posits that test anxiety continues to be a growing problem in the United States (Hembree, 1988; Von Der Embse & Hasson, 2012; Von Der Embse et al., 2013). The inverse relationship between mindset and test anxiety was somewhat expected because prior research has shown that students with fixed mindset have many of the same characteristics and reactions to evaluation as students who have high test anxiety, though there has not been a previous study linking the two concepts. For example, students who focused on the outcome of an evaluation as opposed to the process of learning tended to have lower mindset scores which were associated with a fixed mindset (Church et al., 2001; Dweck, 1986, 2008). Similarly, students who focused on evaluation or perceived the teacher to focus on evaluation also tended to have higher test anxiety (Hembree, 1988). In addition, attentional focus might be another factor that contributed to the inverse relationship between mindset and test anxiety. For example, students with lower mindset scores were likely to focus their attention on closed feedback (whether they were right or wrong) but less likely to focus attention on information that would help them get a correct answer the next time (Boaler, 2013; Dweck et al., 2004). This attentional focus on being right or wrong might be associated with test anxiety MINDSET, ANXIETY, AND GRADING 52 because the more students feel they are being evaluated, the more anxiety they feel about evaluation (Hembree, 1988; Von Der Embse & Hasson, 2012). The relationship between females and test anxiety was also an expected outcome. Females consistently demonstrated higher test anxiety than males in both a meta-analysis of test anxiety (Hembree, 1988) and a study of test anxiety among college students (Chapell et al., 2005). Thus, higher test anxiety among females in this study was consistent with other research on the topic. There could be many factors that influence this difference among genders. First, females tend to have higher levels of trait anxiety and are more likely to develop anxiety disorders than males (Myers, 2005), which is correlated with test anxiety (Hembree, 1988). Second, females may feel more pressure to do well in school than males leading to higher levels of anxiety because students who feel pressure to do well tend to have higher anxiety (Deb et al., 2015; Flannery, 2018; Schrobsdorff, 2016). Finally, stereotype threat might also contribute to higher test anxiety among females if they believe that, in general, females perform more poorly than their male counterparts on tests. Osborne (2001) demonstrated in his research on stereotype threat that females showed higher average test anxiety than males, but that higher test anxiety did not necessarily result in statistically significant lower achievement scores. Therefore, this study was consistent with other research in demonstrating higher test anxiety among females. Interestingly, when the results were separated by gender, female scores for mindset and test anxiety were negatively correlated, but male scores show no significant correlation for mindset and test anxiety. This could be because there were almost twice as many females as males who responded giving a larger, and perhaps more representative, MINDSET, ANXIETY, AND GRADING 53 sample size. The correlation was strong enough for females that when the scores were combined, the overall population showed a statistically significant correlation between the mindset and test anxiety. Further research might want to replicate the study with a more even number of males and females to see if the correlations are the same. In this study, grade level was not significantly related to mindset nor test anxiety. This is logical because prior research has shown no correlation between grade level and mindset. Students can have any range of mindset at the elementary level (McCutchen et al., 2016), the secondary level (Blackwell et al., 2007), and beyond (Boaler, 2013; Church et al., 2001). Similarly, there was no statistically significant difference between test level and anxiety, which is also logical because test anxiety has been found to increase from grades 2-5 and then remain constant until a small decline in college (Chapell et al., 2005; Hembree, 1988). Therefore, this study was consistent with prior research about grade level, mindset, and anxiety. The relationships between mindset, test anxiety, and items on the grading scale were also interesting to consider. Items 1, 3, 5, 7, 9, 12, and 13 were all positively correlated with mindset and negatively correlated to test anxiety. Grouping these items by theme allowed the researcher to gain additional insight into student attitudes toward grading specifically attitudes toward fairness of grades, attitudes about ability to improve, and attitudes about the importance of homework and classwork. For example, students with stronger growth mindset were more likely to see grades as being fair whereas students with high test anxiety showed the opposite attitude. Students with stronger growth mindset were more likely to believe that they could improve and learn hard material whereas students with higher test anxiety were less likely to show those beliefs. MINDSET, ANXIETY, AND GRADING 54 Additionally, students with higher growth mindset were more likely to view homework and classwork as beneficial to learning whereas students with higher test anxiety were less likely to hold this view. Items 1 and 5 were about equity and fairness in grading; the items asked if students believed their grades should be fair and accurately reflect what they know. Students with higher mindset scores might agree with these statements because they feel that learning is a process and grades are a way to evaluate their progress throughout the process (Dweck, 2008; Munoz & Guskey, 2015). Conversely, students with lower mindset scores might feel that grades are less fair or inaccurate because these students might see grades being as a tool of judgment which they resist (Boaler, 2013; Dweck, 2008). Students with higher test anxiety might be more likely to disagree with these items because test anxiety is associated with lower achievement scores (Chapell et al., 2005; Hembree, 1988; Osborne, 2001; Von Der Embse & Hasson, 2012), so students with high test anxiety might feel that their grades are inaccurate because test anxiety interferes with their evaluation. Items 3 and 12 were about being able to learn class material and improve a class grade. Improving and learning as a process is the essential characteristic of growth mindset (Dweck 1986, 2008), so these questions have an understandable connection to mindset score. The connection between these statements and test anxiety may be similar to what was mentioned in the previous paragraph. Test anxiety is associated with lower achievement (Chapell et al., 2005; Hembree, 1988; Osborne, 2001; Von Der Embse & Hasson, 2012) which could mean that students with higher test anxiety feel like they cannot improve and learn because their grades do not represent that idea. Students with MINDSET, ANXIETY, AND GRADING 55 high test anxiety may give up on the learning process because they have become conditioned to fear tests for the negative evaluations that they provide. Test anxiety may be a conditioned response, even if students feel confident on a particular exam. However, the physiological response may make well-prepared students feel a sense of lower self-efficacy in evaluation situations. Items 7, 9, and 13 were about student attitudes toward homework and classwork; items 7 and 13 were about homework and classwork helping students learn material, and item 9 was about homework being worth points and improving a student’s grade. Students with higher mindset scores likely agree with these statements because they see learning as a process (Dweck, 1986, 2008; Dweck et al., 2004) where homework and classwork are an integral part of the process. They might believe that improving their grade (by doing homework) is a reflection of the learning process (Masters, 2013; Munoz & Guskey, 2015). Students with higher test anxiety scores might be more likely to disagree with these statements because their achievement may not be reflected well with homework and classwork. This is an unexpected finding because it would seem that students who did not perform well on tests would put extra emphasis on homework and classwork to make up for missed points on exams. One explanation is that students who feel that tests make up the most important part of their grade may not see the importance of doing homework and classwork because their achievement, when it really matters on exams, is influenced by their test anxiety (Chapell et al., 2005; Hembree, 1988; Osborne, 2001; Von Der Embse & Hasson, 2012). The large sample of females may also have contributed to the correlations among these items. In the experience of the researcher, females tend to turn in homework and classwork more often than their male peers which MINDSET, ANXIETY, AND GRADING 56 could be a confounding variable that explains this correlation. Nevertheless, there was no statistically significant correlation between gender and responses to these three items about grading found in this research. Further study with a larger and more equal sample size among genders might be useful to clarify these results. It is interesting to note that these seven items have the same correlational pattern as the correlation between mindset and test anxiety. The overall data showed a negative correlation between mindset and anxiety which then was reflected in these seven items about grading. This could mean that these grading items were predictive of the overall connection between mindset and test anxiety. Two of the items on the grading questionnaire were related to test anxiety but not to mindset. They were item 8 which stated, “Doing classwork is important because it is worth points and improves my grade,” and item 10 which stated, “Completing homework thoroughly and on time should be an important part of my grade.” These items are likely related to test anxiety for the reasons already mentioned. Students with higher levels of test anxiety likely do not think that classwork and doing homework on time matter because homework and classwork do not impact their grades enough to compensate for lower achievement on tests resulting from higher test anxiety. Test anxiety research has demonstrated that students with test anxiety tend to have more negative attitudes toward evaluation (Hembree, 1988) which negative attitudes may have overshadowed the importance of homework and classwork for students who score high on test anxiety. Mindset may not have been related to item 8 because perhaps students do not see classwork as being as difficult as homework because the teacher is there to help scaffold the material, and mindset differences become most apparent in challenge situations MINDSET, ANXIETY, AND GRADING 57 (Dweck, 1986, 2008; Grant & Dweck, 2003; McCutchen et al., 2016). Mindset may not be related to item 10 because perhaps valuing punctuality is unrelated to views of intelligence and more dependent on personality traits such as conscientiousness. Finally, only two of the grading items were related to gender. Females were less likely to agree with item 1 about grades being fair, females were less likely to agree with item 13 about classwork helping learn the material. Attitudes toward item 1 may be the result of stereotype threat (Osborne, 2001) where females do not think their grades are fair because of a perceived stereotype about female performance in school. Responses to item 13 which was about classwork helping students learn the material were likely influenced by the high levels of test anxiety reported by females in the study. As previously mentioned, students with high test anxiety may not feel that classwork is important because it does not reflect on their grade as strongly as tests while at the same time their test achievement may be inhibited by test anxiety. Limitations Many of the limitations to the study were a result of sample size and necessary procedure. The sample size was relatively small as it represented slightly less than 10% of the population of the school being studied. A larger sample size would have been beneficial because it might have better reflected the population as a whole particularly related to the mindset questionnaire. The procedure which required the students to get parental consent for ethical reasons may have unintentionally skewed the variety of students who participated by forcing them to put forth personal effort to participate. This might have led to the unexpectedly high average on the mindset questionnaire thereby influencing the results. MINDSET, ANXIETY, AND GRADING 58 Another limitation was disparity between the number of males and females that chose to participate. There are almost twice as many females as males who chose to fill out the questionnaires which may have influenced the results as prior research has shown that females tend to have higher test anxiety than males (Chapell et al., 2005; Hembree, 1988; Osborne, 2001). A final limitation of the study was that only one high school was used to gain student responses. Because the high school was a large suburban school with a relatively small minority population, student responses may not be typical of all high school students in the United States or even in the state of Utah. Recommendations The study was intended to show possible correlations, so there is no way to determine whether fixed mindset causes test anxiety or vice versa or if there are other confounding causes involved. This study only demonstrates a relationship between the two measures. Further research might be useful to determine the cause of the correlation and to determine if the correlation holds true in a more typical population that demonstrated less of a bias toward growth mindset. Further research might also want to replicate the study with a larger sample size, a sample that had a larger proportion of students that fell within the fixed mindset range, or a more even distribution of males and females. This would help determine if the correlations found by the researcher hold true to other samples that might be more diverse. A further area of study that could add to this research would be the use of interventions to either reduce test anxiety or build growth mindset. If growth mindset MINDSET, ANXIETY, AND GRADING 59 interventions were also effective at reducing test anxiety, then it might provide further evidence that mindset and test anxiety were linked. The same would be true if test anxiety interventions were associated with higher mindset scores. Additionally, further research could investigate the effect of grading interventions in which students are exposed to certain grading practices (traditional practices vs. standards-based practices) and measure student mindset and test anxiety to corroborate or modify the results found in this study. Generally, any further study of how mindset, test anxiety, and attitudes toward grading could be associated with one another would be beneficial to the educational community, to individual teachers, and to students by showing the skills and attitudes that help students best thrive cognitively and emotionally. MINDSET, ANXIETY, AND GRADING 60 REFERENCES Blackwell, L. S., Trzesniewski, K. H., & Dweck, C. S. (2007). Implicit theories of intelligence predict achievement across an adolescent transition: A longitudinal study and an intervention. Child Development, 78(1), 246–263. Boaler, J. (2013). Ability and mathemathics: The mindset revolution that is reshaping education. Forum, 55(1), 143–152. Chan, D. W. (2012). Life satisfaction, happiness, and the growth mindset of healthy and unhealthy perfectionists among Hong Kong Chinese gifted students. Roeper Review, 34, 224–233. doi:10.1080/02783193.2012.715333 Chapell, M. S., Blanding, Z. B., Silverstein, M. E., Takahashi, M., Newman, B., Gubi, A., & McCann, N. (2005). Test anxiety and academic performance in undergraduate and graduate students. Journal of Educational Psychology, 97(2), 268–274. doi: 10.1037/0022-0663.97.2.268 Church, M. A., Elliot, A. I., & Gable, S. L. (2001). Perceptions of classroom environment, achievement goals, and achievement outcomes. Journal of Educational Psychology, 93(1), 43–54. doi:10.1037//0022-0663.93.1.43 Claro, S., Paunesku, D., & Dweck, C. S. (2016). Growth mindset tempers the effects of poverty on academic achievement. PNAS, 113(31), 8664-8668. Deb, S., Strodl, E., & Sun, J. (2015). Academic stress, parental pressure, anxiety and mental health among Indian high school students. International Journal of Psychology and Behavioral Sciences, 5(1), 26–34. Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist, 41(10), 1040–1048. MINDSET, ANXIETY, AND GRADING 61 Dweck, C. S. (2008). Mindsets and math/science achievement. The Opportunity Equation, 1–17. Dweck, C. S., Chiu, C., & Hong, Y. (1995) Implicit theories: Elaboration and extension of the model. Psychological Inquiry, 6(4), 322–333. Dweck, C. S., Mangels, J. A., & Good, C. (2004). Motivational effects on attention, cognition, and performance. In Dai, D. Y. & Sternberg, R. J. (Eds.), Motivation, cognition, and emotion (pp. 41–55). Retrieved from https://zodml.org/sites/default/files/%5BDavid_Yun_Dai,_Robert_J._Sternberg%5D_Motivation,_E_0.pdf#page=56 Elliot, A. J., McGregor, H. A., & Gable, S. (1999). Achievement goals, study strategies, and exam performance: A mediational analysis. Journal of Educational Psychology, 91(3), 549–563. Elliot, E. S., & Dweck, C. S. (1988). Goals: An approach to motivation and achievement. Journal of Personality and Social Psychology, 54(1), 5–12. ProQuest. Web. 30 Mar. 2018. Flannery, M. E. (2018). The epic of anxiety among today’s students. neaToday. Retrieved from http://neatoday.org/2018/03/28/the-epidemic-of-student-anxiety/ Grant, H., & Dweck, C. S. (2003). Clarifying achievement goals and their impact. Journal of Personality and Social Psychology, 85(3), 541–553. doi:10.1037/0022-3514.85.3.541 Guskey, T. R. (2011). Five obstacles to grading reform. Educational Leadership, 69(3), 16–21. MINDSET, ANXIETY, AND GRADING 62 Hembree, R. (1988). Correlates, causes, effects, and the treatment of test anxiety. Review of Educational Research, 58(1), 47–77. Iamarino, D. L. (2014). The benefits of standards-based grading: A critical evaluation of modern grading practices. Current Issues in Education, 17(2), 1–12. King, L. A. (2013). Motivation and emotion. In Mallory, S. & Colwell, S. (Eds.), Experience psychology (pp. 324–360). New York: McGraw Hill. Li, Y., & Bates, T. C. (2017). Does growth mindset improve children’s IQ, education attainment or response to setbacks? Active-control interventions and data on children’s own mindsets. Unpublished manuscript, Department of Psychology, University of Edinburg, Edinburgh, UK. Long, C. (2015). Are letter grades failing our students? neaToday. Retrieved from http://neatoday.org/2015/08/19/are-letter-grades-failing-our-students/ Mannix, L. D. (2014). What are the effects of standards-based grading on student learning and behavior in the secondary science classroom? (Unpublished master’s thesis). Montana State University, Bozeman, Montana. Masters, G. N. (2013). Towards a growth mindset in assessment. ACER occasional essays. Melbourne: Australian Council for Educational Research (ACER). Retrieved from https://research.acer.edu.au/cgi/viewcontent.cgi?referer =&httpsredir=1&article=1017&context=ar_mis McCutchen, K. L., Jones, M. H., Carbonneau, K. J., & Mueller, C. E. (2016). Mindset and standardized testing over time. Learning and Individual Differences, 45, 208–213. doi:10.1016/j.lindif.2015.11.0271041-6080 MINDSET, ANXIETY, AND GRADING 63 Miller, J. J. (2013). A better grading system: Standards-based, student-centered assessment. English Journal, 103(1), 11 –118. Munoz, M. A. & Guskey, T. R. (2015). Standards-based grading and reporting will improve education. Phi Delta Kappan, 96(7), 64–68. Myers, D. G. (2005). Motivation. In Fleming, N. & Brune, C. (Eds.), Exploring psychology (pp. 340–377). New York: Worth Publishers. Osborne, J. W. (2001). Testing stereotype threat: Does anxiety explain race and sex differences in achievement? Contemporary Educational Psychology, 26, 291–310. Schrobsdorff, S. (2016). Anxiety, depression, and the modern adolescent. TIME, 188(19), Retrieved from http://time.com/magazine/us/4547305/november-7th-2016-vol-188-no-19-u-s/ Scriffiny, P. L. (2008). Seven reasons for standards-based grading. Educational Leadership, 66(2), 70–74. Steinmayr, R. Crede, J., McElvany, N., & Wirthwein, L. (2016). Subjective well-being, test anxiety, academic achievement: Testing for reciprocal effects. Frontiers in Psychology, 6, 1–13. doi: 10.3389/fpsyg.2015.01994 Von Der Embse, N., Barterian, J., & Segool, N. (2013). Test anxiety interventions for children and adolescents: A systematic review of treatment studies from 2000-2010. Psychology in the Schools, 00(0), 1–15. doi:10.1002/pits.21660 Von Der Embse, N., & Hasson, R. (2012). Test anxiety and high-stakes test performance between school settings: Implications for educators. Preventing School Failure, 56(3), 180–187. doi:10.1080/1045988X.2011.633285 MINDSET, ANXIETY, AND GRADING 64 Utah State Board of Education. (2017). Weber High School [School Details]. Retrieved from https://datagateway.schools.utah.gov/Schools/35704 MINDSET, ANXIETY, AND GRADING 65 APPENDICES Appendix A: Student Questionnaires Appendix B: Student Attitude and Motivation Survey Excerpt (Mannix, 2014) Appendix C: Institutional Review Board Approval from Weber State University Appendix D: Approval Letter from School District Appendix E: Parent Letter and Consent Form Appendix F: Information Sheet for Students on Improving Growth Mindset and Reducing Test Anxiety Appendix G: Full Table of Correlations MINDSET, ANXIETY, AND GRADING 66 Appendix A- Student Questionnaires Consent Form Returned Relationships Between Growth Mindset, Test Anxiety, And Grading Student Questionnaires By choosing to fill out the following questionnaire, you are choosing to be a participant in a study on relationships between growth mindset, test anxiety, and grading. Your answers will only be used if your consent form is attached with both your signature and your parent’s signature. Part 1 Instructions: Fill out the following information. FLEX teacher: ________________________________ Gender (circle one): Male Female Other Grade level (circle one): 10 11 12 MINDSET, ANXIETY, AND GRADING 67 Part 2- Dweck Mindset Instrument Directions: Read each sentence below and then mark the corresponding box that shows how much you agree with each sentence. There are no right or wrong answers. 1=strongly agree, 2=agree, 3=mostly agree, 4=mostly disagree, 5=disagree, 6=strongly disagree 1 Strongly Agree 2 Agree 3 Mostly Agree 4 Mostly Disagree 5 Disagree 6 Strongly Disagree 1) You have a certain amount of intelligence, and you really can’t do much to change it. 2) Your intelligence is something about you that you can’t change very much. 3) No matter who you are, you can significantly change your intelligence level. 4) To be honest, you can’t really change how intelligent you are. 5) You can always substantially change how intelligent you are. 6) You can learn new things, but you can’t really change your basic intelligence. 7) No matter how much intelligence you have, you can always change it quite a bit. 8) You can change even your basic intelligence level considerably. 9) You have a certain amount of talent, and you can’t really do much to change it. 10) Your talent in an area is something about you that you can’t change very much. 11) No matter who you are, you can significantly change your level of talent. 12) To be honest, you can’t really change how much talent you have. 13) You can always substantially change how much talent you have. MINDSET, ANXIETY, AND GRADING 68 14) You can learn new things, but you can’t really change your basic level of talent. 15) No matter how much talent you have, you can always change it quite a bit. 16) You can change even your basic level of talent considerably. MINDSET, ANXIETY, AND GRADING 69 Part 3- Nist and Diehl Test Anxiety Questionnaire Directions: To complete this evaluation, read through each statement and reflect upon past testing experiences. You may want to consider all testing experiences or focus on a particular subject (history, science, math, etc.) one at a time. Indicate how often each statement describes you by a number from one to five as outlined below. 1 Never 2 Rarely 3 Sometimes 4 Often 5 Always 1) I have visible signs of nervousness such as sweaty palms, shaky hands, etc. right before a test. 2) I have “butterflies” in my stomach. 3) I feel nauseated before a test. 4) I read through the test and feel that I do not know any of the answers. 5) I panic before and during a test. 6) My mind goes blank during a test. 7) I remember the information I blanked on once I get out of the testing situation. 8) I have trouble sleeping the night before a test. 9) I make mistakes on easy questions or put answers in the wrong places. 10) I have trouble choosing answers. MINDSET, ANXIETY, AND GRADING 70 Part 4- Revised Attitudes Toward Grading Questionnaire Directions: For each of the following statements, select how strongly you agree or disagree. 5 Strongly Agree 4 Somewhat Agree 3 Neutral 2 Somewhat Disagree 1 Strongly Disagree 1. My grade should be fair. 2. I should understand how my grade is determined. 3. I should be confident that I can improve my grade. 4. My grade should motivate me to work harder in class. 5. My grade should accurately reflect what I know. 6. I should know how to improve my grade. 7. Doing homework helps me learn the class material. 8. Doing classwork is important because it is worth points and improves my grade. 9. Doing homework is worth it because it is worth points and improves my grade. 10. Completing homework thoroughly and on time should be an important part of my grade. 11. If I know the material, my homework completion and punctuality (turning it in on time) shouldn’t matter. 12. I should be able to learn even the hard concepts if I don’t give up and have enough time. 13. Doing classwork is important because it helps me learn the material. MINDSET, ANXIETY, AND GRADING 71 Appendix B-Student Attitude and Motivation Survey Excerpt (Mannix, 2014) Excerpt from Student Motivation and Attitude Survey (Mannix, 2014, pp.58–59) For each of the following statements, select how strongly you agree or disagree. Strongly Agree Somewhat Agree Neutral Somewhat Disagree Strongly Disagree 1. My grade is fair. 2. I understand how my grade is determined. 3. I am confident that I can improve my grade. 4. My grade motivates me to work harder in class. 5. My grade accurately reflects what I know. 6. I know how to improve my grade. 7. I do homework because it helps me learn the material. 8. I do classwork because I think it is important. 9. I do my homework because it is worth points and improves my grade. 10. Completing homework thoroughly and on time should be an important part of my grade. 11. If I know the material, my homework completion and punctuality (turning it in on time) shouldn’t matter. 12. I can learn even the hard concepts in science if I don’t give up and have enough time. 13. I don’t try very hard in science. 14. My science skills and knowledge have improved a lot in this class. 15. Some science concepts are just too hard for me to learn. 16. I enjoy learning 17. I am motivated to work hard in science class. 18. I care about my science grade, it is important to me. 19. I enjoy science class. MINDSET, ANXIETY, AND GRADING 72 Appendix C-Institutional Review Board Approval from Weber State University MINDSET, ANXIETY, AND GRADING 73 Appendix D- Approval Letter from School District October 15, 2018 Camille Phinney Weber State University Ogden, UT Dear Ms. Phinney: Weber School District is committed to quality education and supports the efforts made by all individuals and groups who wish to improve it through research and study. As a state agency charged with the education and safety of the students who attend our school, we take a close look at the many proposals that are sent to us. Please recognize our commitment to children and teachers, and know that all decisions we make are made with them in mind. We wish you the best and hope for your continued progress in education. Your project has been approved and you may begin your research immediately. When your research has concluded, you will need to provide a copy of the summary data to my office. Your project is approved __X__ Date: October 15, 2018__ Your project is rejected ______ Date: ________________ Continued Success, Sheri W. Heiter Curriculum Director Assessment and Accountability Director Weber School District MINDSET, ANXIETY, AND GRADING 74 Appendix E-Parent Letter and Consent Form MINDSET, ANXIETY, AND GRADING 75 MINDSET, ANXIETY, AND GRADING 76 MINDSET, ANXIETY, AND GRADING 77 Appendix F- Information Sheet for Students on Improving Growth Mindset and Reducing Test Anxiety Tips for a Growth Mindset How do you interpret success? Do you think people are successful because they are born that way or because they worked hard and put in the effort to succeed? People who say you are born with it likely have a fixed mindset which means they think that you cannot change your ability over time. People who say you can be successful if you work hard and put in the effort likely have a growth mindset which means they think you can change your ability over time. Who is right? Psychological science indicates the latter to be true; our brains are capable of doing and learning almost anything that we practice consistently over time. Sure, some people start out with more ability at some tasks, but each person’s brain is constantly changing and each person can decide what they want that change to be. So…what do you want to be good at? What are you willing to work hard to be successful at? Why does it matter? People with growth mindset have lots of benefits. They are more likely to:  Try challenging tasks because they are not afraid to try even if they fail.  Use better thinking strategies because they care about learning, not about being wrong.  Be resilient and try again if something did not work.  Have higher achievement than those who think that their ability is fixed. What can you do? If you want to change or improve your mindset, try these four steps: 1. Learn to hear your fixed mindset “voice.” When you start hearing that voice in your head tell you that you don’t have enough ability or talent, recognize it as a fixed mindset. 2. Recognize that you have a choice. Realize that how you interpret challenge, setbacks, and criticism is a choice. You can choose how you are going to let this event affect you. 3. Talk back to it with a growth mindset voice. So, when you hear the fixed mindset voice, talk back to it with a growth mindset. For example, if you hear, “Are you sure you can do that? You’re not smart enough.” Say back to it, “I’m not sure I can do it now, but I think I can learn with time and effort.” If you hear, “This would have been easy if you had the talent.” Say back to it, “That is so wrong. Basketball wasn’t easy for Michael Jordan and science wasn’t easy for Thomas Edison. They had a passion and put in tons of effort.” 4. Take the growth mindset action. When you start thinking the right words, the actions will come naturally. Over time you will choose to: take on challenges, overcome setbacks and try again, and hear and respond to criticism without being crushed. For more information see https://mindsetonline.com/index.html MINDSET, ANXIETY, AND GRADING 78 Tips for Reducing Test Anxiety Have you ever felt worried about taking a test? This worry could be manifest in many ways. You could have physical symptoms like sweaty palms, “butterflies” in your stomach, or trouble sleeping the night before. You could also have mental symptoms like your mind going blank, making simple mistakes, or having trouble choosing answers. While a little bit of nervousness could help you focus your efforts for the test, too much can lead to many negative results like:  Lower scores on tests  Lower moods  Lower life satisfaction  Lower self-esteem What can you do about it? 1. Try relaxation exercises before or during a test. If you can feel the stress, it’s time to take a deep breath. Try closing your eyes and breathing in for 8 counts and holding for 4 counts. Then, breathe out for 8 counts and hold for 4 counts. Count the numbers in your head and try not to think about anything else. Repeat five times or until you feel your body calming down. 2. Fuel your brain. Your brain needs a few things to function at its best: food, water, and exercise. Try getting all three before you test. The best kind of food is something with natural sugar like a fruit. Drinking water is always good. Exercise could be as simple as walking quickly to class or running up the stairs before your test. 3. Get good rest before. Try to get sufficient sleep the night before the test. Your brain forms new memories while you are sleeping. Staying up late to study will probably not help your score. 4. Study beforehand but do not cram. Stay organized and plan out a certain amount of time to study each day before your test. Do not try to learn it all the night before because you will not be able to remember it as well. 5. Start with a question you know. It can ease your anxiety if you start with something you already know how to do. It will reinforce the idea that you do know the material. Find a question that is easy for you and start with that one. 6. Focus on yourself and no one else. Try to ignore others around you while testing. It does not matter if they are writing a lot. Focus on yourself and what you know. Similarly, do not worry about other people turning in their test. Take the time you need. There are no bonus points for finishing first. Work at your speed and do your best. For more information see http://www.studygs.net/tstprp8.htm. MINDSET, ANXIETY, AND GRADING 79 Appendix G- Full Table of Correlations Correlations Dweck_total Anxiety_total gender_num Dweck_total Pearson Correlation 1 N 196 Anxiety_total Pearson Correlation -.164 Sig. (2-tailed) .021 N 196 gender_num Pearson Correlation -.035 .415 1 Sig. (2-tailed) .629 .000 N 193 193 193 G1 Pearson Correlation .262 -.272 -.153 Sig. (2-tailed) .000 .000 .035 N 193 193 191 G2 Pearson Correlation .053 -.110 -.126 Sig. (2-tailed) .466 .126 .080 N 195 195 193 G3 Pearson Correlation .150 -.185 -.088 Sig. (2-tailed) .036 .010 .224 N 195 195 193 G4 Pearson Correlation .137 -.072 -.057 Sig. (2-tailed) .056 .322 .431 N 194 194 192 G5 Pearson Correlation .182 -.190 -.057 Sig. (2-tailed) .011 .008 .429 N 194 194 192 G6 Pearson Correlation .070 -.132 -.093 Sig. (2-tailed) .336 .068 .201 N 192 192 190 G7 Pearson Correlation .173 -.287 -.036 Sig. (2-tailed) .016 .000 .615 N 194 194 192 G8 Pearson Correlation .114 -.172 -.043 Sig. (2-tailed) .114 .017 .550 N 194 194 192 MINDSET, ANXIETY, AND GRADING 80 G9 Pearson Correlation .209 -.176 .057 Sig. (2-tailed) .003 .014 .431 N 195 195 193 G10 Pearson Correlation .128 -.221 .017 Sig. (2-tailed) .076 .002 .816 N 194 194 192 G11 Pearson Correlation -.096 .108 -.123 Sig. (2-tailed) .183 .132 .090 N 195 195 193 G12 Pearson Correlation .169 -.178 -.092 Sig. (2-tailed) .018 .013 .205 N 195 195 193 G13 Pearson Correlation .147 -.245 -.146 Sig. (2-tailed) .040 .001 .043 N 195 195 193 Note. For gender, males=0 and females=1 for the sake of analysis