This article was downloaded by: [University of North Dakota]On: 09 October 2013, At: 13:54Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The Journal of Experimental EducationPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/vjxe20

The Interactive Effects of PersonalAchievement Goals and PerformanceFeedback in an Undergraduate ScienceClassKrista R. Muis a , John Ranellucci a , Gina M. Franco a & Kent J.Crippen ba McGill University , Montreal , Canadab University of FloridaPublished online: 11 Jul 2013.

To cite this article: Krista R. Muis , John Ranellucci , Gina M. Franco & Kent J. Crippen (2013)The Interactive Effects of Personal Achievement Goals and Performance Feedback in anUndergraduate Science Class, The Journal of Experimental Education, 81:4, 556-578, DOI:10.1080/00220973.2012.738257

To link to this article: http://dx.doi.org/10.1080/00220973.2012.738257

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

THE JOURNAL OF EXPERIMENTAL EDUCATION, 81(4), 556–578, 2013Copyright C© Taylor & Francis Group, LLCISSN: 0022-0973 print/1940-0683 onlineDOI: 10.1080/00220973.2012.738257

LEARNING, INSTRUCTION, AND COGNITION

The Interactive Effects of Personal Achievement Goals andPerformance Feedback in an Undergraduate Science Class

Krista R. Muis, John Ranellucci, and Gina M. FrancoMcGill University, Montreal, Canada

Kent J. CrippenUniversity of Florida

The authors explored whether manipulating feedback influenced cognition, motivation, and achieve-ment in an undergraduate chemistry course. They measured students’ (N = 250) achievement goals,test anxiety, self-efficacy, and metacognitive strategy use at the beginning and end of the semester.After completing the first set of questionnaires, students were randomly assigned to 1 of 4 con-ditions: (a) control, (b) mastery feedback, (c) performance-approach feedback, and (d) combinedmastery/performance-approach feedback. In each condition, students received a raw performancescore for each weekly quiz they completed online and, for the treatment conditions, additional feed-back reflective of that specific feedback condition. Results provide evidence for the multiple goalsperspective (specialized pattern) wherein performance-oriented feedback was beneficial for someoutcomes, whereas mastery feedback was beneficial for other outcomes.

Keywords feedback, metacognition, personal achievement goals, self-efficacy, test anxiety

OVER THE PAST THREE DECADES, A majority of the theoretical and empirical work con-ducted in the achievement motivation literature has concentrated on achievement goal theory(Hulleman, Schrager, Bodmann, & Harackiewicz, 2010). Goal theory has developed within asocial-cognitive approach to motivation that emphasizes cognitive factors, such as how individ-uals interpret situations, the events of situations, and how they process information about thesesituations (Dweck, 1986; Dweck & Leggett, 1988). Goals are one of the major determinants ofhow people feel about, react to, and cognitively process success or failure (Ames & Archer, 1988;

This research was supported by a grant to Krista R. Muis from the Canadian Council on Learning.Address correspondence to Krista R. Muis, Department of Educational and Counselling Psychology, McGill Univer-

sity, 3700 McTavish Street, Montreal, Quebec, H3A 1Y2, Canada. E-mail: krista.muis@mcgill.ca

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 557

Dweck, 1986). Given Hulleman and colleagues’ (2010) comprehensive meta-analysis on the con-ceptualizations of achievement goals, we adopt their definition. Accordingly, achievement goalsare defined as “a future-focused cognitive representation that guides behavior to a competence-related end state that the individual is committed to either approach or avoid” (Hulleman et al.,2010, p. 423).

Goal theorists conceptualize achievement goals within a trichotomous (e.g., Elliot & Church,1997; Middleton & Midgley, 1997; Pintrich, 2000) or 2 × 2 achievement goal framework (Elliot &McGregor, 2001; Elliot & Murayama, 2008) that distinguish goals from a mastery-performancedichotomy and an approach-avoidance dichotomy. Within the trichotomous framework, threedistinct achievement goal orientations have been proposed: mastery, performance-approach, andperformance-avoidance. A mastery goal orientation (or mastery-approach orientation, as it islabeled in the 2 × 2 framework) describes individuals who strive to develop competence andtask mastery. In contrast, a performance-approach goal orientation characterizes individualswho strive to demonstrate aptitude and seek favorable judgments; competence is self-evaluated incomparison to others. The third goal is a performance-avoidance orientation, whereby individualsstrive to avoid appearing incompetent and avoid negative judgments. Like the performance-approach orientation, comparisons of competence are made with other individuals. Last, the 2× 2 achievement goal framework adds a fourth goal orientation: mastery-avoidance orientation,whereby a learner’s goal is to avoid failure rooted in an intrapersonal perspective rather thanin comparison to others. For the mastery-avoidance goal construct, avoiding incompetence isthe focus (Elliot & McGregor, 2001). Because mastery-avoidance is a relatively new constructcompared with the other three (Elliot, 1999), few studies have explored this orientation andlittle empirical work has been conducted to examine its psychometric properties (Hullemanet al., 2010). Moreover, given the negative relations between performance-avoidance goals andvarious outcomes (see Hulleman et al., 2010), we focused solely on mastery (approach) versusperformance-approach goals at both the personal goals and induced goals levels (subsequentlydetailed).

As Hulleman and colleagues (2010) reported, the past 25 years of research on achievementgoal theory has resulted in a number of theoretical frameworks, yet theorists generally agree thatmastery goals and performance goals are two distinct ways to define competence. Traditionally,mastery goals were construed as having a positive association with adaptive motivational pro-cesses and outcomes, whereas performance goals were theoretically associated with less adaptivepatterns of learning and motivation (Ames, 1992b). As evidence accumulated, results with re-spect to performance goals were mixed wherein some positive patterns were found and otherswere neutral or negative. On the basis of conflicting evidence, goal theorists reconceptualizedperformance goals into performance-approach and performance-avoidance goals, and began toidentify ways in which performance-approach goals can combine with mastery goals to promoteoptimal motivation, cognition, and learning outcomes (Barron & Harackiewicz, 2001; Harack-iewicz, Barron, Pintrich, Elliot, & Thrash, 2002; Lau & Nie, 2008; Murayama & Elliot, 2009;Pintrich, 2000).

Today, researchers continue to explore conditions under which mastery goals, performance-approach goals, or a combination of the two result in better cognitive, motivational, and learningoutcomes. In particular, while the theoretical distinction between the normative goals (or masterygoals) perspective and the multiple goals perspective (Hulleman et al., 2010) is clear, empiricalwork that examines these two perspectives continues. For example, some scholars theoretically

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

558 MUIS ET AL.

favor the mastery goals perspective and argue that adopting a mastery goal results in bettermotivational and learning processes and outcomes (Kaplan & Middleton, 2002; Midgley, Kaplan,& Middleton, 2001), such as higher levels of cognitive engagement, self-reported self-regulatorystrategies (e.g., Pintrich, 2000), self-efficacy, interest, and value (e.g., Harackiewicz, Barron,Tauer, Carter, & Elliot, 2000; Wolters, Yu, & Pintrich, 1996), and higher positive affect andlower negative affect (e.g., Kaplan & Maehr, 1999; Middleton & Midgley, 1997), to name a few.According to this perspective, mastery goals are adaptive given that mastery-oriented individualsare said to seek out challenge and persist in the face of difficulty. In contrast, they suggestthat performance goals are typically maladaptive, which is supported by some evidence thatshows that performance goals are negatively related to self-efficacy (e.g., Skaalvik, 1997) andpositively related with negative affect and test anxiety (e.g., Kaplan & Maehr, 1999; Middleton& Midgley, 1997) and with avoidant help seeking (Ryan & Pintrich, 1998). In theory, thesepatterns are maladaptive given that individuals who adopt performance goals tend to avoidchallenge (especially if they perceive themselves as low in competence), as challenges threatenthe possibility of demonstrating ability and outperforming others (Dweck & Leggett, 1988).Moreover, rather than risk failure, they are likely to withdraw effort or give up in the face ofchallenge.

In contrast, the multiple goals perspective suggests that performance-approach goals do notalways result in negative effects, even for individuals with low perceived competence. Thatis, the positive effects of mastery goals do not automatically result in negative performance-approach goal effects (Harackiewicz, Barron & Elliot, 1998). As several studies have demon-strated, performance-approach goals are also associated with positive outcomes such as effectivecognitive engagement (e.g., Meece, Blumenfeld, & Hoyle, 1988), task value (Bong, 2001; Church,Elliot, & Gable, 2001), academic self-concept (Pajares, Britner, & Valiante, 2000), effort expen-diture (Elliot, McGregor, & Gable, 1999) and achievement (Bouffard, Vezeau, & Bordeleau,1998; Harackiewicz et al., 2000; Pintrich, 2000). Moreover, these theorists argue that masteryand performance-approach goals should not be construed as mutually exclusive (Harackiewiczet al., 1998); striving to outperform other students can be coupled with attempting to master a task.In this regard, adopting multiple goals of mastery and performance-approach may result in betterlearning outcomes (Barron & Harackiewicz, 2001; Pintrich, 2000). Central to the multiple goalsperspective, then, is the notion that endorsing performance-approach goals can be beneficial,especially when mastery goals are concurrently endorsed (Harackiewicz et al., 2002).

From an educational reform perspective, it is important for researchers to determine whichgoals teachers should induce in a classroom context. Should teachers develop classroom envi-ronments that foster mastery goals, or should teachers develop classroom contexts that includeelements of both mastery and performance-approach? Before attempting to answer this ques-tion, it is important to distinguish between induced goals in a learning situation and personalachievement goals (Harackiewicz & Barron, 2004). For example, students may come into aclassroom situation with personally adopted goals that may interact with situationally inducedor manipulated goals (Harackiewicz et al., 1998; Murayama & Elliot, 2009). Accordingly, itis important to consider both personal achievement goals as well as the contextual factors thatmay induce achievement goals (e.g., type of feedback received, instructions given, or statementsmade by teachers during lecture; Barron & Harackiewicz, 2001; Linnenbrink, 2005). We questionwhich goal or combination of goals results in better learning outcomes and under which types ofclassroom conditions.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 559

The purpose of this research was to further empirically evaluate the mastery versus multiplegoals perspective in the context of an authentic classroom-learning situation. Following Harack-iewicz and colleagues’ (1998) recommendations, because learners may adopt multiple goals, weevaluated the effects of personal and manipulated goals (induced through feedback) and theirpotential interactions on multiple outcomes. Specifically, we examined whether manipulatingfeedback influenced students’ test anxiety, and altered their personal mastery and performance-approach goals, academic self-efficacy, self-reported use of metacognitive self-regulatory strate-gies, and academic achievement in the context of a first-year undergraduate chemistry course.We examine how three feedback conditions used to induce specific types of achievement goals(mastery, performance-approach, combined mastery/performance-approach) relate to students’motivation and affect, learning strategies, and achievement. In the sections that follow, we ex-plore the multiple goals perspective and related empirical work to provide a basis from which todevelop specific testable hypotheses.

The Multiple Goals Perspective

As previously noted, although some studies have found that performance-approach goals canhave detrimental effects on learning, proponents of the multiple goals perspective assert thatperformance-approach goals can also have positive effects on motivation, learning, and achieve-ment. That is, performance-approach goals may be adaptive under certain situations, and maybe particularly adaptive when adopted in conjunction with a mastery goal. Moreover, severalresearchers have found that mastery and performance-approach goals are positively related (e.g.,Archer, 1994; Harackiewicz, Barron, Carter, Lehto, & Elliot, 1997), suggesting that students canand do pursue multiple goals and that these goals may interact and combine to promote motivationand achievement.

According to Harackiewicz and colleagues (Barron & Harackiewicz, 2001; Harackiewiczet al., 2002), to test the multiple goals perspective requires considering multiple outcomes alongwith tests of independent and interactive effects of mastery and performance-approach goals oneach of those outcomes. Specifically, they proposed four ways in which performance-approachand mastery goals may combine. First, the additive effect occurs when each goal is indepen-dently beneficial for a single outcome, which results in positive main effects for mastery andperformance-approach goals. Second, the interactive effect occurs when the adoption of bothgoals is more adaptive than adopting one goal alone for an outcome, which results in a positiveMastery × Performance-Approach interaction. For both of these effects, students who adopt bothgoals will have the most adaptive outcomes. That is, if the effects are additive, then positive effectsof performance-approach goals do not depend on a high level of mastery goals. In contrast, forthe interaction effect, positive performance-approach goals do depend on a high level of masterygoals (Harackiewicz et al., 2002).

The third pattern, specialized effects, occurs when there are unique effects of each goal acrossmultiple outcomes. That is, rather than promoting the same educational outcome, mastery andperformance-approach goals affect different outcomes. In this case, a positive main effect formastery goals on one outcome is found whereas a positive main effect for performance-approachon a different outcome would support the specialized effects pattern. Last, selective effects refer tosituations where learners focus on the goal orientation that is most relevant in a particular contextor learning situation. For example, a student may adopt mastery goals when given a writing

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

560 MUIS ET AL.

assignment, but then switch to adopt a performance-approach goal when given a multiple-choiceexam. Empirical evidence of this pattern requires multiple time points and multiple outcomes todemonstrate a mix of positive main effects and/or interactions between each type of goal acrossthe various outcomes (Harackiewicz et al., 2002).

In contrast with the multiple goals perspective, the normative goals perspective would pre-dict positive main effects for mastery goals and null or negative effects for performance-approach goals. In this regard, students who adopt high mastery goals and low performance-approach goals should result in the most adaptive pattern, whereas students who adopthigh mastery and performance-approach goals will experience a depression of the posi-tive effects of the mastery goals as a result of the adoption of high performance-approachgoals.

Research that has evaluated the mastery versus multiple goals debate has explored the effectsof personal (i.e., self-set) achievement goals on outcomes such as motivation, affect, strategy useand performance (e.g., Pintrich, 2000), cognitive, emotional, and achievement outcomes (e.g.,Daniels et al., 2008), and interest and academic achievement (e.g., Harackiewicz et al., 2002).Researchers have also examined the effects of manipulated goals on intrinsic motivation (e.g.,Harackiewicz & Elliot, 1993), or a combination of personal and manipulated achievement goalson various outcomes (e.g., Linnenbrink, 2005).

In particular, previous research with samples of undergraduate students has investigatedwhether achievement goals predict the four variables of interest in the present study: metacogni-tive strategies, test anxiety, self-efficacy, and academic achievement. Specifically, mastery goalstend to negatively predict test anxiety (Bandalos, Finney, & Geske, 2003) but positively pre-dict metacognitive strategies (Muis & Franco, 2009; Pintrich, 2000; Wolters et al., 1996) andself-efficacy (Bandalos et al., 2003; Coutinho & Neuman, 2008), whereas performance-approachgoals tend to negatively predict metacognitive strategies (Muis & Franco, 2009) but positivelypredict academic achievement (Curry, Elliot, Da Fonseca, & Moller, 2006: Senko & Harack-iewicz, 2005), self-efficacy (Bandalos et al., 2003; Coutinho & Neuman, 2008), and test anxiety(Bandalos et al., 2008; Pekrun, Elliot, & Maier, 2009).

As such, the conceptualization of achievement goals as antecedents to academic achievement,test anxiety, and metacognitive strategies is clear; however, the directionality between goals andself-efficacy is less clear. In the literature, self-efficacy has been reported as a precursor to personalgoals (Liem, Lau, & Nie, 2008; Wigfield, 1994) and as an outcome to personal goals (Bandaloset al., 2003; Coutinho & Neuman, 2008). Since self-efficacy is based on past performance andpredicts future performance, and performance itself is predicted by personal goals (Bandura,1993), it is reasonable to examine self-efficacy as either an antecedent or a consequence ofachievement goals. For the present study, we adopt the latter.

As Harackiewicz and colleagues (1998) suggested, when examining relations between achieve-ment goals and various outcomes, it is important to consider both personal achievement goals aswell as situationally induced goals that may be established in a classroom or learning context.Specifically, as Harackiewicz and colleagues highlight, it is pertinent to distinguish between per-sonal goals and induced goals, as there is not necessarily a one-to-one correspondence. Rather,these two types of goals may interact in ways that influence various outcomes. For example,Harackiewicz and Sansone (1991) proposed that achievement goals are most effective when anindividual’s personal goals match those of the learning context, like the type of environment ateacher establishes in his or her classroom.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 561

Of particular interest, several studies have examined how students’ personal achievementgoals are influenced by situational goals, and how these interact and relate to self-efficacy, self-regulatory skills, engagement, interest, effort withdrawal, avoidance coping, intrinsic motivation,academic self-concept, and achievement (Lau & Nie, 2008; Linnenbrink, 2005; Murayama &Elliot, 2009). To date, results have been mixed and theorists have proposed why students mightrespond differentially depending on the learning context. For example, Murayama and Elliot(2009) proposed three ways in which personal and situational goals may interact to influenceachievement-related outcomes. The first is what they call the direct effect, wherein classroom goalstructures directly influence relevant outcomes even after the effects of personal goals are con-sidered. The second is the indirect effect, wherein classroom goals indirectly influence outcomesthrough personal achievement goals. Last, the third model they propose is the interaction effect,wherein classroom goals moderate the influence of personal achievement goals on outcomes.

From a slightly different perspective, Linnenbrink and Pintrich (2001) proposed two competinghypotheses with regard to how classroom goals and personal achievement goals interact: thebuffering hypothesis, and the matching hypothesis. For the buffering hypothesis, they proposethat a personal mastery goal or mastery-oriented context will buffer the negative effects of personalperformance-approach goals. In contrast, the matching hypothesis suggests that goals establishedin the learning context that match students’ personal goals are most beneficial in terms of a varietyof learning outcomes. To test these two competing hypotheses, Linnenbrink (2005) examined theeffects of a quasi-experimentally manipulated classroom goal condition (mastery, performance-approach, and a combined mastery/performance approach) and personal goal orientations onmotivation, help-seeking, emotional well being, cognitive engagement, and achievement witha sample of upper-year elementary students. To objectively create the different classroom goalconditions, Linnenbrink altered the general classroom goal structure as well as the specific type offeedback that students received in small groups. The mastery goal condition focused on learningand improvement, whereas the performance-approach goal condition emphasized the importanceof demonstrating individual and group competence. For the combined mastery and performance-approach condition, the classroom and feedback given to students in their groups incorporatedboth elements of the mastery and performance-approach conditions.

At posttest, Linnenbrink (2005) found that students in the mastery condition reported higherpersonal mastery goals than students in the performance-approach condition. Similarly, at posttest,students in the performance-approach condition reported higher personal performance-approachgoals than students in the mastery goal condition. Moreover, students’ personal goals at thebeginning of the school year did not influence posttest outcomes, providing no support forthe buffering or matching hypothesis. It is interesting that the combined mastery/performanceapproach classroom goal condition had the greatest positive influence on achievement and helpseeking.

To interpret this outcome, Linnenbrink (2005) proposed that because students were placedinto small groups to create the various types of classroom goal conditions, the effects might havebeen a function of grouping. That is, consistent with Deutsch’s (1949a, 1949b) research, studentsmay have benefitted from being in groups; more engagement, more adaptive patterns of learning,and more social interactions are expected within a group when competing against other groups(between-group competition) as compared with when individuals are competing against eachother within a group (within-group competition). Accordingly, one important question to addressis whether similar beneficial patterns of outcomes are replicated when individuals are learning

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

562 MUIS ET AL.

independently in a more authentic learning situation, and whether this pattern is replicated at theundergraduate level of education. Our study responds to this call.

The Present Study

Our primary research question addresses the mastery goals versus multiple goals perspective:What learning context (mastery, performance-approach, or combined mastery/performance-approach) is more beneficial for reducing test anxiety and for promoting self-efficacy, cognitiveengagement (e.g., metacognitive self-regulation), and achievement? To address this question, wemanipulated feedback to induce specific achievement goals using an online quiz-taking envi-ronment in accordance with Ames’ (1992a, 1992b) TARGET framework. Ames (1992a, 1992b)identified six dimensions of a learning environment that may communicate messages about mo-tivation and achievement, and may influence the types of personal goals that students adopt:tasks, authority relations, recognition, grouping procedures, evaluation, and timing (TARGET).Similar to personal achievement goals, the feedback given to students can also be described interms of an orientation toward mastery or performance. For example, mastery feedback focuseson student mastery and self-improvement through tasks that involve variety and diversity, andrecognition and evaluative feedback that focuses on individual effort (to name a few of Ames’[1992a, 1992b] dimensions). In contrast, performance-approach evaluative feedback focuses onstudent competition and high achievement. Each of these types of feedback may influence thetypes of goals that students adopt in a learning context (Harackiewicz et al., 1998).

After completing a set of pretest questionnaires, undergraduate chemistry students were ran-domly assigned to one of four feedback conditions: a control condition, a mastery condition, aperformance-approach condition, and a combined mastery/performance-approach condition. Ineach condition, students received their actual raw performance score for each of the 14 weeklyquizzes they completed online and, for the treatment conditions, additional feedback reflective ofthat specific feedback condition. At the end of the semester, students completed a set of posttestquestionnaires and their final grades were collected as a measure of academic achievement.

Following Linnenbrink (2005), we present two plausible hypotheses, one for each of the posi-tions on goal adoption. First, from the mastery goals perspective, students in the mastery feedbackcondition will benefit most across all outcomes, whereas students in the performance-approachand combined conditions will not benefit given the detrimental effects of performance-approachgoals. In this regard, we expect positive main effects for the mastery condition, but neutral ornegative main effects for the performance-approach and mixed conditions. Alternatively, fromthe multiple goals perspective, three plausible outcomes may occur: the additive effect, the inter-active effect, or the specialized effect. The additive effect would result in positive main effectsfor mastery and performance-approach goals on the same outcome. The interactive effect resultswhen there is a positive mastery × performance-approach interaction on the same outcome. Last,the specialized effect occurs when there is a positive main effect for mastery goals on one outcomewhereas a positive main effect for performance-approach goals results on a different outcome.

Moreover, we measured students’ personal achievement goals at the beginning of the semesterand then again at the end of the semester to assess whether students’ personal achievement goalschanged as a function of their learning context and whether personal goals moderated the effects ofthe situationally induced goals to test the buffering versus the matching hypothesis. Accordingly,our second research question was: How do students’ personal achievement goal orientations

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 563

change and interact with situationally induced achievement goals? If the buffering hypothesis iscorrect, then students in the mastery condition should maintain their level of personal masterygoals, performance-approach goals should decrease, and these individuals should have the greatestbenefit compared with the other groups. In contrast, if the matching hypothesis is correct, thenpersonal achievement goals should moderate the effects in each of the manipulated conditions:students who adopt personal mastery goals at the beginning of the semester should have greaterbenefits in the mastery condition than students who adopt personal performance-approach goals,and vice versa for the performance-approach condition. Last, those in the combined conditionshould have equal benefits as both goals are induced in that condition.

METHOD

Participants

Participants were 250 (151 female, 99 male) undergraduate university students consented toparticipate in the study. Students were enrolled in a prerequisite qualifying first-year generalchemistry course for science majors. The average age was 20.10 (SD = 4.65) and 65% of thesample was majoring in biology, biochemistry, or chemistry.

Materials

Prior knowledge

To measure prior knowledge of course content, participants completed a prior knowledge test.This 10-item multiple-choice test measured students’ knowledge of various chemistry conceptsincluding properties of matter, electron configurations, chemical reactions, and stoichiometry(see Appendix A).

Personal achievement goals

To assess students’ personal mastery and performance-approach goal orientations, we usedthe student version of the Patterns of Adaptive Learning Scales (Midgley et al., 2000). It isa 14-item Likert-scale instrument designed to measure personal achievement goals within atrichotomous framework (i.e., mastery-avoidance goals are not measured); we chose it becauseit has demonstrated good psychometric qualities with similar samples (see Muis, Winne, &Edwards, 2009). We assessed mastery goals with items such as “One of my goals in chemistryclass is to learn as much as I can.” Performance-approach goals were assessed with items suchas “One of my goals is to show others that I’m good at my chemistry class work.” Items on thePatterns of Adaptive Learning Scales are anchored along a 5-point scale ranging from 1 (not atall true) to 5 (very true).

Self-efficacy, test anxiety, and metacognitive self-regulation

Items from the self-efficacy, test anxiety, and metacognitive self-regulation subscales of theMotivated Strategies for Learning Questionnaire (Pintrich, Smith, Garcia, & McKeachie, 1991)

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

564 MUIS ET AL.

were used. The questionnaire is a widely used 81-item self-report measure designed to assessundergraduate students’ use of varying learning strategies and motivational orientations for anundergraduate course. The self-efficacy for learning and performance subscale comprises eightitems and assesses expectancy for success and self-efficacy. Expectancy for success refers specif-ically to task performance, and self-efficacy includes judgments of one’s ability to successfullycomplete a task as well as one’s confidence in one’s skills to perform that task. Example itemsinclude “I believe I will receive an excellent grade in this class” (expectancy for success) and “I’mconfident I can learn the basic concepts taught in this course” (self-efficacy). The test anxietysubscale includes five items that measure a student’s level of anxiety for learning course contentand taking course exams. One example is “I have an uneasy, upset feeling when I take an exam.”Last, the metacognitive self-regulation subscale includes twelve items that measure processes ofplanning, monitoring and regulating cognitive activities. An example item is “When studyingfor this course I try to determine which concepts I don’t understand well.” Students rate eachstatement on a 7-point Likert scale ranging from 1 (not at all true of me) to 7 (very true of me).For all of these subscales, higher scores indicate greater agreement and thus greater strategyuse.

Achievement

We measured achievement using the overall course raw score, which was then converted intoa percentage (out of 100). This measure was computed as a nonweighted compilation of gradesfrom three exams, a final exam, a cumulative online quiz score, laboratory work, and a cumulativein-class quiz score.

The Learning Environment and Feedback Conditions

Students enrolled in this course attended face-to-face lectures and labs, and were required tocomplete 14 weekly quizzes online (quizzes began in the second week of a 15-week semester).Instruction was primarily lecture-based, but the delivery was interactive and included informal,formative assessment with feedback. The instructor’s goal was to develop a mastery-orientedclassroom and lab context for learning, and her philosophy was student-centered, supportive,and focused on individual achievement. For example, she used a small group problem solvingstrategy during nearly every session. During these problem-solving sessions, students workedcooperatively in heterogeneous groups of their own choosing to solve a well-structured problemprovided by the instructor. Problem solutions were submitted and graded as a group quiz whereineveryone in the group who was present received the same grade. The instructor used this strategyas a mechanism for improving student learning and retention of the material, as well as theopportunity to improve teamwork skills.

Course evaluation consisted of scores from three midterm exams, online quizzes, groupquizzes, laboratory work, and a final examination. Online content was provided through a com-mercial courseware system and was intended as a communication medium among students andthe instructor, as an access point for the quizzing system, and as a reference location for notes,homework assignments, extra practice problems, and solution sets. Each feedback condition (mas-tery, performance-approach, combined mastery/performance-approach, control condition) wasestablished using this system. Each week, students were required to complete a well-structured,

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 565

multiple-choice content quiz (online only). Quiz questions were developed using Bloom’s Taxon-omy to target application, synthesis, and evaluation levels of understanding. At any point duringthe week, students had access to their quiz answers and could modify their responses at any time.At the end of each week, quizzes were graded and students were given their actual achievementscore (a raw score out of 10, one point for each correct answer). If a score of at least 8 pointsout of 10 was not achieved, students were given the opportunity to complete a make-up quiz,which was available for four days following the original quiz. Moreover, based on the conditionunder which students were randomly assigned, each goal condition received additional feedbackabout their quiz performance (with the exception of the control condition). In all other aspects,the online content was identical.

Goal conditions were created via the type of feedback that students received on each weeklyquiz. Following Linnenbrink’s (2005) protocol, and in line with the evaluation dimension of Ames’(1992a, 1992b) TARGET framework, feedback in the mastery goal condition (n = 68) emphasizedthe importance of learning, understanding, and improvement. For example, each week, studentsin the mastery condition were provided textual information regarding the elements noted abovealong with graphical information that indicated how much they improved (or did not improve) onthat week’s assessment compared with previous weeks. In contrast, feedback in the performance-approach condition (n = 78) included textual messages that emphasized the importance ofdemonstrating individual competence, with a particular emphasis on competition for high scores.Students in this condition were also provided performance feedback in graphical form thatindicated how well they performed compared with other students. Moreover, their percentilerank was provided. Feedback in the combined mastery/performance-approach condition (n = 52)included elements of both the mastery and performance-approach conditions, with an emphasison doing better than others and trying to learn and understand. For this condition, students wereprovided textual and graphical information on their improvement from week to week as well ashow well they performed compared with others. Percentile rank was also provided. Last, with theexception of a single performance score for each quiz, the control condition (n = 52) received noadditional feedback. Scores were not compared with previous weeks’ scores or to other students’scores. Feedback for all other course assignments and exams, available through WebCampus,were reported as a raw score.

Procedure

Students were randomly assigned to one of the four conditions: mastery feedback, performance-approach feedback, combined mastery/performance-approach feedback, and a control group.Within the first week of the course, and prior to any assessments, students completed a de-mographics questionnaire, a prior knowledge test, and items from the Patterns of AdaptiveLearning Scales and the Motivated Strategies for Learning Questionnaire. Over the course of thesemester, students received weekly feedback about their performance on the weekly quizzes inthe manner described previously. After quizzes were graded, feedback was automatically dis-played upon login to ensure students received their performance evaluation. At the end of thesemester, before students completed the final exam, we asked them to complete the Patternsof Adaptive Learning Scales and the Motivated Strategies for Learning Questionnaire itemsagain.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

566 MUIS ET AL.

TABLE 1Descriptive Statistics and Reliability Coefficients for All Variables

Pretest Posttest

n M SD α M SD α

Prior knowledge score 250 3.34 1.79 — — — —High school GPA 237 3.67 0.49 — — — —Final grade (%) 250 — — — 76.07 13.52 —Self-regulation variablesa

Self-efficacyc 250 5.68 0.98 .91 5.33 1.15 .94Metacognitive self-regulationc 250 4.99 0.86 .80 4.79 0.97 .86Test anxiety 250 4.00 1.38 .78 4.14 1.32 .79

Personal goal orientationsb

Mastery goal 250 4.70 0.48 .84 4.57 0.63 .88Performance-approach goal 250 2.56 1.13 .86 2.47 1.14 .90

a1–7 scale.b1–5 scale.cSignificant difference (pretest to posttest) at the .01 or lower level.

RESULTS

Preliminary Analyses

All means, standard deviations, and reliability coefficients are presented in Tables 1 through 4for all variables at pretest and posttest, and correlations between all variables are presented inTable 5. Using a series of one-way analyses of variance, we then explored whether there wereany treatment group differences at pretest on prior knowledge, high school GPA, self-efficacy,test anxiety, metacognitive self-regulation, mastery goals, and performance-approach goals (seeTable 2). Results revealed no differences between groups on any of these variables (all ps > .05).Moreover, no gender differences were found on any of the pretest variables (all ps > .05), sogender differences were not explored across outcomes.

Effects of Personal Achievement Goals and Feedback Condition on Outcomes

To explore our two main research questions at the multivariate level, we examined the ef-fects of personal achievement goals and feedback on students’ academic performance (finalgrade), self-efficacy, test anxiety, and metacognitive self-regulation. We conducted an over-all repeated measures multivariate analysis of variance with (pretest) personal mastery goalsand personal performance-approach goals as moderators, feedback condition (control, mastery,performance-approach, and mastery/performance-approach) as the between-subjects independentvariable, time as the within-subjects variable (pretest-posttest), and self-efficacy, test anxiety, andmetacognitive self-regulation as the dependent variables (pretest-posttest). We conducted a sep-arate analysis of variance for achievement given that achievement (grade) was measured at onlyone time.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 567

TABLE 2Descriptive Statistics for Prior Knowledge, GPA, and Final Grade, by Condition

Prior knowledge n M SD

Control 52 3.44 2.02Mastery 68 3.37 1.67Performance 78 3.27 1.64Combined 52 3.29 1.96

High school GPAControl 50 3.69 0.48Mastery 62 3.58 0.53Performance 75 3.74 0.46Combined 50 3.69 0.48

Final gradeControl 52 73.02a 14.69Mastery 68 74.81 13.20Performance 78 79.43a 11.67Combined 52 75.74 14.62

Note: Significant difference at the .01 or lower level.

Using Pillai’s Trace, a significant omnibus F for was obtained for time, F (1, 244) = 7.09,p < .01, η2 = .07, scale (differences across each of the dependent variables), F (2, 244) = 5.22,p < .01, η2 = .04, and feedback, F (3, 244) = 3.21, p < .01, η2 = .05. Moreover, mastery goalsand performance-approach goals moderated the effects over time, F (1, 244) = 4.32, p < .05,η2 = .02, and F (1, 244) = 7.83, p < .01, η2 = .03, scale, F (2, 243) = 6.52, p < .01, η2 = .05,

TABLE 3Descriptive Statistics for Self-Regulated Learning Variables, by Condition

Pretest Posttest

n M SD M SD

Self-efficacy∗

Control 52 5.70 .90 5.22 1.24Mastery 68 5.78 .91 5.62 .99Performance 78 5.74 .97 5.05 1.30Combined 52 5.41 1.13 5.33 1.08

Metacognitive self-regulation∗

Control 52 5.04 .84 4.79 .83Mastery 68 4.87 .87 4.75 1.04Performance 78 5.06 .86 4.71 .97Combined 52 4.98 .88 4.96 1.01

Test anxietyControl 52 3.93 1.22 4.22 1.36Mastery 68 3.91 1.62 3.72 1.41Performance 78 3.96 1.18 4.32 1.11Combined 52 4.23 1.48 4.32 1.38

Note:∗All variables were reported on a 1–7 scale.

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

568 MUIS ET AL.

TABLE 4Bivariate Correlations Among All Pretest and Posttest Variables

Variable 1 2 3 4 5 6 7 8 9 10 11 12

1. Mastery pretest —2. Mastery posttest .36∗ —3. Performance-approach −.09 −.12 —

goals pretest4. Performance-approach −.06 −.16∗ .65∗ —

goals posttest5. Performance-avoidance −.11 −.15∗ .81∗ .58∗ —

goals pretest6. Performance-avoidance −.11 −.21∗ .54∗ .88∗ .58∗ —

goals posttest7. Self-efficacy pretest .19∗ .10 .08 .11 .03 .04 —8. Self-efficacy posttest .01 .32∗ −.01 −.01 −.06 −.06 .59∗ —9. Metacognitive self- .33∗ .30∗ .07 .07 .02 .01 .44∗ .37∗ —

regulation pretest10. Metacognitive self- .17∗ .36∗ −.08 −.04 −.12 −.10 .28∗ .58∗ .61∗ —

regulation posttest11. Test anxiety pretest −.03 .05 .27∗ −.20∗ .31∗ −.20∗ −.25∗ −.10 .06 .12 —12. Test anxiety posttest .02 .03 .22∗ −.35∗ .22∗ −.39∗ −.23∗ −.23∗ −.01 .03 .58∗ —13. Final grade −.10 .12 −.09 −.09 −.10 −.06 .29∗ .53∗ .11 .26∗ −.08 −.31∗

Note: ∗Significant at p < .01 level.

and F (2, 243) = 8.02, p < .001, η2 = .06, and feedback, F (1, 244) = 9.36, p < .01, η2 = .04,and F (1, 244) = 10.99, p < .001, η2 = .05, respectively.

These results suggest that the feedback condition had a significant effect on student out-comes, and that these effects were moderated as a function of the level of students’ mastery andperformance-approach goals at the beginning of the semester. Accordingly, follow-up repeatedmeasures analyses were conducted for each dependent variable to explore in more detail thepattern of results to test the mastery versus multiple goals perspective.

Self-Efficacy

For self-efficacy, the univariate analysis revealed a main effect for time, F (1, 244) = 26.93,p < .01, η2 = .03, but no main effect for feedback and no significant time × feedback interaction,which suggests that students across all conditions had a similar level of decline in self-efficacy.Moreover, students’ mastery goals moderated change in self-efficacy from pretest to posttest,F (1, 244) = 8.02, p < .005, η2 = .03, wherein higher levels of mastery goals were related togreater decreases in self-efficacy.

Metacognitive Self-Regulation

The univariate analysis for metacognitive self-regulation showed a main effect for time, F (1,244) = 7.82, p < .01, η2 = .03, and a significant time × feedback interaction, F (3, 244) = 2.62,

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 569

p < .05, η2 = .03. Moreover, both mastery and performance-approach goals moderated change instudents’ metacognitive self-regulation over time, F (1, 244) = 7.04, p < .01, η2 = .03, and F (1,244) = 7.61, η2 = .03 and mastery goals moderated the effect for feedback, F (1, 244) = 20.99,p < .001, η2 = .08. In particular, the main effect for time revealed that students had self-reportedhigher levels of metacognitive strategy use at pretest compared with posttest, and that the rate ofchange was greater for individuals with higher levels of mastery and performance-approach goals.Moreover, these changes over time varied as a function of feedback. Follow-up analyses revealedthat individuals in the control and performance conditions significantly dropped in metacognitiveself-regulation compared with the other two groups (p < .01), which did not change (p > .10).For feedback, moderation analyses revealed that individuals with higher levels of mastery goalshad greater rates of change than individuals with lower levels of mastery goals.

Test Anxiety

For test anxiety, unlike self-efficacy and metacognitive self-regulation, there was no main effectfor time or feedback, but there was a significant moderation effect for performance-approachgoals on feedback, F (1, 244) = 19.28, p < .001, η2 = .09. In particular, students with higherlevels of performance-approach goals at the beginning of the semester had higher levels of testanxiety (collapsed across time) compared with students who adopted lower levels of performance-approach goals. No other effects were found.

Academic Achievement

For achievement, the analysis of variance revealed a significant main effect for feedback, F(3,244) = 2.75, p < .05, η2 = .03, but no other significant effects were found. Post hoc analyses usingthe least significant difference procedure demonstrated that students in the performance-approachfeedback condition had the highest level of achievement, which differed significantly from thecontrol group (p < .01) and mastery group (p < .05). No other significant differences were found.

Effects of Feedback Condition on Students’ Entering Personal Achievement Goals

We conducted the last series of analyses to assess whether feedback influenced students’ enteringpersonal achievement goals. Accordingly, we conducted a repeated measures analysis for eachpersonal goal orientation (pretest-posttest) with feedback condition as the between-subjects inde-pendent variable. Means and standard deviations are presented in Table 5. For students’ personalmastery goals, using Pillai’s Trace, a significant main effect of time was found, F(1, 246) = 9.95,p < .01, η2 = .04, but there was no significant effect of feedback and no interaction. In particular,collapsed across feedback group, students’ mastery goals decreased from the beginning to theend of the semester.

In contrast, for students’ personal performance-approach goals, there was a main effect fortime, F(1, 246) = 3.90, p < .05, η2 = .02, and a main effect for feedback, F(1, 246) = 5.60,p < .05, η2 = .03, but no interaction. Over time, collapsed across feedback condition, students’performance-approach goals significantly decreased. For the main effect for feedback, post hocleast significant difference analyses revealed that students in the mastery feedback condition had

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

570 MUIS ET AL.

TABLE 5Descriptive Statistics for Goal Orientations, by Condition

Pretest Posttest

Variable n M SD M SD

Mastery goal∗

Control 52 4.71 .44 4.56 .56Mastery 68 4.75 .42 4.60 .60Performance 78 4.67 .51 4.54 .72Combined 52 4.69 .54 4.60 .60

Performance-approach goal∗

Control 52 2.81 1.23 2.57 1.31Mastery 68 2.25 1.09 2.00 .96Performance 78 2.61 1.11 2.79 1.12Combined 52 2.65 1.07 2.52 1.05

∗Note. All variables were reported on a 1–5 scale.

significantly lower levels of performance-approach goals compared with all other groups (allp < .05). No other differences were found.

GENERAL DISCUSSION

The purpose of this study was to explore the effects of manipulating feedback on students’personal achievement goals, motivation, self-reported strategy use and affect for learning sciencecontent, as well as their achievement in a first-year undergraduate chemistry course. Resultsfrom our study indicate that students’ self-reported levels of self-efficacy and metacognitiveself-regulation decreased significantly over time and that these decreases were moderated bystudents’ mastery goals (for self-efficacy) and both mastery and performance-approach goals (formetacognitive self-regulation). Moreover, decreases in metacognitive self-regulation occurredonly in the control and performance-approach feedback groups. In addition, students’ levelsof test anxiety were moderated by performance-approach goals wherein individuals with higherlevels of performance-approach goals also had higher levels of test anxiety. Last, for achievement,students in the performance-approach feedback group achieved the highest grades compared withthe other groups. We discuss the theoretical and educational implications of these findings inthe context of the various hypotheses that we tested for this research. We then conclude withlimitations and directions for future research.

The Normative versus Multiple Goals Perspective

To examine these two theoretical positions, we explored whether students in a mastery-orientedfeedback condition experienced more gains compared with students in a performance-approachcondition, a combined mastery/performance-approach condition, and a control condition, andwhether students’ personal entering goal orientations moderated the effects of the feedback

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 571

conditions. The normative goals perspective predicts that students in the mastery feedback con-dition would have the greatest benefit across all outcomes compared with the other conditions.In contrast, the multiple goals perspective would predict additive effects, interactive effects, orspecialized effects across the various outcomes.

Results from our study provide strong evidence for a specialized pattern of outcomes, whichsupports the multiple goals perspective. In particular, whereas students in the mastery feedbackand combined conditions did not self-report lower levels of metacognitive strategy use at posttest,students in the performance-approach and control conditions reported significantly lower levels.These results are consistent with previous research that has found that mastery goals positivelypredict metacognitive strategy use (Muis & Franco, 2009; Pintrich, 2000; Wolters et al., 1996),whereas performance-approach goals negatively predict their use (Muis & Franco, 2009). We takethis one step further and infer that the feedback students received in our study had a significantinfluence on students’ use of these strategies wherein mastery-oriented feedback served as aprotective mechanism; students who received that feedback maintained their level of strategyuse, whereas students who did not receive this type of feedback (e.g., they received normativeinformation or no information) significantly lowered their use of these strategies.

In contrast, students who received normative feedback (e.g., performance-approach and com-bined conditions) had the highest levels of achievement compared with students who did notreceive this type of feedback. This is also consistent with previous research that has demon-strated a positive relationship between performance-approach goals and academic achievement(Cury et al., 2006; Senko & Harackiewicz, 2005; Senko & Miles, 2008; Wolters et al., 1996). AsSenko and Miles (2008) demonstrated, although mastery goals are associated with deep learningstrategies and high course interest, mastery goals are rarely related to achievement. In contrast,performance-approach goals are associated with high levels of achievement, despite their rela-tions to surface learning strategies and low interest. It is important to note that results from theirstudy provide empirical evidence that mastery-oriented students, who were more interested inthe course content, were likely to focus their studying on more interesting content, while ignor-ing the boring material. In comparison, performance approach–oriented students, who were notinterested in the course content, were more likely to study more of the content, which resulted inhigher grades.

In our study, when placed into a normative feedback condition, students performed wellcompared with others, regardless of their entering personal achievement goals. As such, consistentwith Senko and Miles (2008), we infer that under noncompetitive learning conditions, masterygoals are not adaptive from an achievement perspective. However, when placed into a competitiveenvironment, mastery-oriented individuals can reap the benefit of that competitive environmentby focusing more on higher achievement outcomes. In this regard, it appears that performance-approach feedback elicits a positive influence on achievement.

On the basis of these results, as did Linnenbrink (2005), we interpret that performance-approach learning environments may be beneficial for some learning outcomes, like academicachievement, but not so beneficial for other outcomes like metacognitive strategy use. In thecontext of the types of feedback that students received in our study, we posit that a visual displayof one’s performance over time can have important influences on students’ personal judgments oftheir competence. For example, for students in the mastery group, seeing one’s performance overtime compared with others can provide valuable information with regard to learning outcomes.That is, feedback provides information about competence or incompetence, and goal orientation

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

572 MUIS ET AL.

influences how that information is interpreted (Bobko & Colella, 1994). In the face of negativefeedback (e.g., receiving lower scores than others in the class), individuals with a mastery goalorientation may increase effort or revise tactics and strategies (Dweck & Leggett, 1988; Elliott& Dweck, 1988). In contrast, when presented with information about how well they are doingin comparison to others, individuals with high performance-approach goals are more likelyto perceive negative information as evaluative or judgmental (Ames, 1992b). This evaluativeinformation is likely threatening, resulting in negative affect like higher levels of anxiety (Elliot& Harackiewicz, 1996), which is what we also found in our study. In particular, students withhigher performance-approach goals had higher levels of test anxiety.

Last, similar to Linnenbrink (2005), given the mixed results, our results do not provide clearsupport for the buffering or matching hypothesis. Although mastery-oriented individuals benefitedfrom being in a mastery feedback condition for metacognitive self-regulation, this was not thecase for their academic achievement or self-efficacy. Moreover, normative feedback benefitted allstudents in that particular condition, and students’ personal goals did not moderate the effect ofthat feedback, as one might expect with the matching hypothesis. This complex pattern of resultssuggests that perhaps other mechanisms or interactions are occurring that are not captured byeither of these two hypotheses. We find this particularly noteworthy; as Linnenbrink suggestedfor her study, perhaps group dynamics interacted with the induced goal context. For our study,perhaps the interaction or complex patterns were a result of the content and quizzes that wereavailable online. It is important to note that it is possible that the combination of more traditionalin-class lecture, labs, and exams coupled with the online content and quizzes may have been animportant factor that influenced the results of our study. To date, we do not know of any studiesthat have explored relations between these variables in a mixed environment such as the onedescribed here. In this regard, there may be other important environmental factors that need to beexplored in future work.

The Effects of Feedback on Personal Achievement Goals

Previous research that has examined the effects that learning environments have on students’personal achievement goals, self-regulatory strategies, self-efficacy, intrinsic motivation, aca-demic self-concept, and achievement has typically found that a combined mastery/performance-approach classroom goal orientation is related to students’ adoption of mastery and performance-approach goals (e.g., Church et al., 2001; Murayama & Elliot, 2009). Based on these previousstudies, we hypothesized that students in the mastery feedback condition would experience in-creases in personal mastery goals and decreases in performance-approach goals, whereas studentsin the performance-approach feedback condition would experience decreases in mastery goalsand increases in performance-approach goals. Last, we expected that students in the combinedcondition would experience increases in both mastery and performance-approach goals.

Results from our study only partially supported these hypotheses. Students in the masteryfeedback condition had the lowest level of performance-approach goals compared with the othergroups. However, regardless of their feedback condition, all students experienced significantdecreases in their mastery and performance-approach goals over time. From a goal stability per-spective, we find these results particularly noteworthy. That is, recent studies that have examinedthe stability of undergraduate students’ achievement goal orientations have found significant de-creases in students’ mastery goals (e.g., Fryer & Elliot, 2007; Muis & Edwards, 2009; Senko &

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 573

Harackiewicz, 2005) and performance-approach goals (Senko & Harackiwicz, 2005). From thesestudies, one may infer students’ mastery goals decrease over the course of the semester, evenin classrooms that are primarily mastery oriented (Muis & Edwards, 2009). However, we alsonote that, in our study, students’ initial mastery goals level was quite high (the mean was 4.70),with little opportunity to move upward. Alternatively, perhaps students initially report an idealwith regard to what goals they would like to adopt for their course, an ideal that changes as thatcourse becomes a reality. Of course, these are merely speculative ideas; ones that require furtherempirical scrutiny. Either way, these results have important educational implications, which wediscuss next.

Educational Implications

Given increasing demands for scientists (National Science Board, 2006) and the growing con-cerns of the dropout rates for undergraduate students enrolled in the sciences (Daempfle, 2004),it is pertinent that educators implement classroom interventions that foster student learning, mo-tivation, and achievement in science. We responded to this pressing concern by implementing asystem that manipulated the types of feedback students received on their online course quizzes.Our goal was to create different feedback environments that mirrored the types of classroom goalstructures that researchers suggest might foster improved learning, motivation, and achievement.By creating environments that focus on varying feedback, we aimed to increase student cognition,motivation, and achievement in a first-year undergraduate chemistry course.

Because of the positive outcomes with regard to metacognitive self-regulation with the masterygoal condition, and the positive effects with regard to achievement for the performance-approachcondition, we posit that our intervention is a positive first step toward developing effectiveinterventions designed to address student achievement at the undergraduate level of education.Specifically, provincially or state-funded colleges and universities across North America typicallyhave large first-year undergraduate enrollments in the sciences and mathematics. Given the largenumber of students in these courses, it is particularly challenging for professors to provideimmediate feedback and feedback that provides more than raw score information. In a time ofbudget cuts and reductions in teaching assistantships, professors are left to grapple with largeamounts of grading and little time to complete it. By implementing a system that providesstudents with immediate feedback on their performance coupled with information that focuseson understanding and improvement, comparative performance, or both, there is more opportunityfor students to learn from that immediate feedback, which may subsequently improve studentoutcomes (Muis, 2007; Winne & Hadwin, 1998).

Limitations and Future Directions

Although some interesting patterns of results emerged from our research, results must be inter-preted with caution, as there are a number of limitations that need to be addressed. First, it ispertinent to assess the effectiveness of the intervention in future studies. In particular, we planto assess treatment fidelity by directly asking students about the nature of the feedback they re-ceived, their perceptions about that feedback, and whether and how the feedback influenced theirthinking about their own learning. By better understanding whether and how students attended to

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

574 MUIS ET AL.

the feedback, researchers may be able to develop instructional interventions that more effectivelyaddress students’ needs in terms of bolstering their learning, motivation, and achievement.

In addition, future studies could examine different dimensions of classroom goal structuresto assess whether some dimensions are more effective than others with regard to influencingstudents’ cognition, motivation, and academic achievement. In the present study, by manipulatingfeedback structures, we focused on one dimensions of Ames’ (1992a, 1992b) TARGET system:the evaluation dimension. Results from our study suggest that attending to this dimension ofthe classroom goal structure can lead to changes in students’ personal goals, self-efficacy, testanxiety, metacognitive self-regulation, and achievement. However, given the modest effect sizesassociated with these changes, we wonder if there are other dimensions that may have a greatereffect on student outcomes. That is, how would results have differed if our intervention focusedon manipulating different dimensions of the TARGET framework, such as authority relations?If educators could only feasibly enhance only one or two dimensions of their classroom goalstructures, which dimensions should they focus on? We anticipate exciting new directions alongthis line of work not only with students enrolled in science courses, but also for students enrolledin other courses from other disciplines including mathematics and the social sciences.

AUTHOR NOTES

Dr. Krista R. Muis is an Associate Professor at McGill University in the Department of Ed-ucational and Counselling Psychology. Her research interests are in the areas of self-regulatedlearning, epistemic beliefs, achievement motivation, emotions, and mathematics and sciencelearning. She is interested in how students’ epistemic beliefs and emotions influence variousfacets of learning, motivation, and academic performance. She also examines how teachers’ be-liefs about knowledge and knowing influence their practices and how those practices affect studentoutcomes. Moreover, all lines of research are brought together under contemporary models ofself-regulated learning. John Ranellucci is a Ph.D. candidate in the Learning Sciences Programat McGill University. He is interested in achievement goals and emotions, and is exploring theeffects of an intervention designed to address interest-based learning in mastery-oriented indi-viduals. Dr. Gina M. Franco completed her Ph.D. under the supervision of Dr. Muis at McGillUniversity. She is interested in the role of epistemic beliefs in students’ learning and conceptualchange. Dr. Kent J. Krippen is an Associate Professor in STEM Education at the Universityof Florida. His research interests include the development and use of technology to improveinstruction and maximize learning.

REFERENCES

Ames, C. (1992a). Achievement goals and the classroom climate. In D. H. Schunk & J. L. Meece (Eds.), Studentperceptions in the classroom (pp. 327–348). Hillsdale, NJ: Erlbaum.

Ames, C. (1992b). Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology, 84,261–271. doi: 10.1037//0022-0663.84.3.261

Ames, C., & Archer, J. (1988). Achievement goals in the classroom: Students’ learning strategies and motivation processes.Journal of Educational Psychology, 80, 260–270. doi: 10.1037/0022-0663.80.3.260

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 575

Archer, J. (1994). Achievement goals as a measure of motivation in university students. Contemporary EducationalPsychology, 19, 430–446. doi: 10.1006/ceps.1994.1031

Bandalos, D. L., Finney, S. J., Geske, J. A. (2003). A model of statistics performance based on achievement goal theory.Journal of Educational Psychology, 95, 604–616. doi: 10.1037/0022-0663.95.3.604

Bandura, A. (1993). Perceived self-efficacy in cognitive development and functioning. Educational Psychologist, 28,117–148. doi: 10.1207/s15326985ep2802 3

Barron, K. E., & Harackiewicz, J. M. (2001). Achievement goals and optimal motivation: Testing multiple goal models.Journal of Personality and Social Psychology, 80, 706–722. doi: 10.1037//0022-3514.80.5.706

Bobko, P., & Colella, A. (1994). Employee reactions to performance standards: A review and research propositions.Personnel Psychology, 47, 1–28. doi: 10.1111/j.1744-6570.1994.tb02407.x

Bong, M. (2001). Between- and within-domain relations of academic motivation among middle and high school students:Self-efficacy, task value, and achievement goals. Journal of Educational Psychology, 93(1), 23–34. doi: 10.1037/0022-0663.93.1.23

Bouffard, T., Vezeau, C., & Bordeleau, L. (1998). A developmental study of the relation between combined learningand performance goals and students’ self-regulated learning. Journal of Educational Psychology, 68, 309–319. doi:10.1111/j.2044-8279.1998.tb01293.x

Church, M. A., Elliot, A. J., & Gable, S. L. (2001). Perceptions of classroom environment, achievement goals, andachievement outcomes. Journal of Educational Psychology, 93, 43–54. doi: 10.1037/0022-0663.93.1.43

Coutinho, S. A., & Neuman, G. (2008). A model of metacognition, achievement goal orientation, learning style andself-efficacy. Learning Environments Research, 11, 131–151. doi: 10.1007/s10984-008-9042-7

Cury, F., Elliot, A. J., Da Fonseca, D., & Moller, A. (2006). The social-cognitive model of achievement motivationand the 2 × 2 achievement goal framework. Journal of Personality and Social Psychology, 90, 666–679. doi:10.1037/oo22.3514.90.4.666

Daempfle, P. A. (2004). An analysis of the high attrition rates among first-year college science, math, and engineeringmajors. Journal of College Student Retention, 5, 37–52. doi: 10.2190/DWQT-TYA4-T20W-RCWH

Daniels, L., Haynes, T., Stupnisky, R., Perry, R., Newall, N., & Pekrun, R. (2008). Individual differences in achieve-ment goals: A longitudinal study of cognitive, emotional, and achievement outcomes. Contemporary EducationalPsychology, 33, 584–608. doi: 10.1016/j.cedpsych.2007.08.002

Deutsch, M. (1949a). An experimental study of the effects of co-operation and competition upon group process. HumanRelations, 2, 197–292. doi: 10.1177/001872674900200301

Deutsch, M. (1949b). A theory of cooperation and competition. Human Relations, 2, 129–152. doi: 10.1177/001872674900200204

Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist, 41, 1040–1048. doi: 10.1037//0003–066X.41.10.1040

Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation and personality. Psychological Review,95, 256–273. doi: 10.1037/0033–295X.95.2.256

Elliot, A. J. (1999). Approach and avoidance motivation and achievement goals. Educational Psychologist, 34, 149–169.doi: 10.1207/s15326985ep3403 3

Elliot, A. J., & Church, M. A. (1997). A hierarchical model of approach and avoidance achievement motivation. Journalof Personality and Social Psychology, 72, 218–232. doi: 10.1037/0022-3514.72.1.218

Elliot, A. J., & Harackiewicz, J. M. (1996). Approach and avoidance achievement goals and intrinsic motivation: Ameditational analysis. Journal of Personality and Social Psychology, 70, 461–475. doi: 10.1037/0022-3514.70.3.461

Elliot, A. J., & McGregor, H. (2001). A 2 × 2 achievement goal framework. Journal of Personality and Social Psychology,80, 501–509. doi: 10.1037//0022-3514.80.3.501

Elliot, A. J., McGregor, H., & Gable, S. (1999). Achievement goals, study strategies, and exam performance: A medita-tional analysis. Journal of Educational Psychology, 76, 628–644. doi: 10.1037//0022-0663.91.3.549

Elliot, A. J., & Murayama, K. (2008). On the measurement of achievement goals: Critique, illustration, and application.Journal of Educational Psychology, 100, 613–628. doi: 10.1016/j.ijer.2004.06.003

Elliott, E. S., & Dweck, C. S. (1988). Goals: An approach to motivation and achievement. Journal of Personality andSocial Psychology, 54, 5–12. doi: 10.1037/0022-3514.54.1.5

Fryer, J. W., & Elliot, A. J. (2007). Stability and change in achievement goals. Journal of Educational Psychology, 99,700–714. doi: 10.1037/0022-0663.99.4.700

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

576 MUIS ET AL.

Harackiewicz, J. M., & Barron, K. E. (2004).Conducting social psychological research in educational settings: “Lessonswe learned in school.” In A. T. Panter, C. Sansone & C. C. Morf (Eds.), The Sage handbook of methods in socialpsychology (pp. 471–484). Thousand Oaks, CA: Sage.

Harackiewicz, J. M., Barron, K. E., Carter, S. M., Lehto, A. T., & Elliot, A. J. (1997). Predictors and consequences ofachievement goals in the college classroom: Maintaining interest and making the grade. Journal of Personality andSocial Psychology, 73, 1284–1295. doi: 10.1037//0022-3514.73.6.1284

Harackiewicz, J. M., Barron, K. E., & Elliot, A. J. (1998). Rethinking achievement goals: When are they adaptive forcollege students and why? Educational Psychologist, 33, 1–21. doi: 10.1207/s15326985ep3301 1

Harackiewicz, J. M., Barron, K. E., Pintrich, P. R., Elliot, A. J., & Thrash, T. M. (2002). Revision of achievementgoal theory: Necessary and illuminating. Journal of Educational Psychology, 94, 638–645. doi: 10.1037//0022-0663.94.3.638

Harackiewicz, J. M., Barron, K. E., Tauer, J. M., Carter, S. M., & Elliot, A. J. (2000). Short-term and long-term conse-quences of achievement goals: Predicting interest and performance over time. Journal of Educational Psychology,92, 316–330. doi: 10.1037/0022-0663.92.2.316

Harackiewicz, J. M., & Elliot, A. J. (1993). Achievement goals and intrinsic motivation. Journal of Personality and SocialPsychology, 65, 904–915. doi: 10.1037//0022-3514.65.5.904

Harackiewicz, J. M., & Sansone, C. (1991). Goals and intrinsic motivation: You can get there from here. Advances inMotivation and Achievement, 7, 21–49. doi: 10.1016/S0022-1031(03)00013–1

Hulleman, C. S., Schrager, S. M., Bodmann, S. M., & Harackiewicz, J. M. (2010). A meta-analytic review of achievementgoal measures: Different labels for the same constructs or different constructs with similar labels? PsychologicalBulletin, 136, 422–449. doi: 10.1037/a0018947

Kaplan, A., & Maehr, M. L. (1999). Achievement goals and student well-being. Contemporary Educational Psychology,24, 330–358. doi: 10.1006/ceps.1999.0993

Kaplan, A., & Middleton, M. (2002). Should childhood be a journey or a race? Response to Harackiewicz et al. (2002).Journal of Educational Psychology, 94, 646–648. doi: 10.1037//0022-0663.94.3.646

Lau, S., & Nie, Y. (2008). Interplay between personal goals and classroom goal structures in predicting student outcomes: Amultilevel analysis of person-context interactions. Journal of Educational Psychology, 100, 15–29. doi: 10.1037/0022-0663.100.1.15

Liem, A. D., Lau, S., & Nie, Y. (2008). The role of self-efficacy, task value, and achievement goals in predicting learningstrategies, task disengagement, peer relationship, and achievement outcome. Contemporary Educational Psychology,33, 486–512. doi: 10.1016/j.cedpsych.2007.08.001

Linnenbrink, E. A. (2005). The dilemma of performance-approach goals: The use of multiple goal contexts to promotestudents’ motivation and learning. Journal of Educational Psychology, 97, 197–213. doi: 10.1037/0022-0663.97.2.197

Linnenbrink, E. A., & Pintrich, P. R. (2001). Multiple goals, multiple contexts: The dynamic interplay between personalgoals and contextual goal stresses. In S. Volet & S. Jarvela (Eds.), Motivation in learning contexts: Theoretical andmethodological implications (pp. 251–269). Amsterdam, The Netherlands: Pergamon Press.

Meece, J. L., Blumenfeld, P., & Hoyle, R. (1988). Students’ goal orientations and cognitive engagement in classroomactivities. Journal of Educational Psychology, 80, 514–523. doi: 10.1037/0022-0663.80.4.514

Middleton, M. J., & Midgley, C. (1997). Avoiding the demonstration of lack of ability: An unexpected aspect of goaltheory. Journal of Educational Psychology, 89, 710–718. doi: 10.1037/0022-0663.89.4.710

Midgley, C., Kaplan, A., & Middleton, M. (2001). Performance-approach goals: Good for what, for whom, under whatcircumstances, and at what cost? Journal of Educational Psychology, 93, 77–86. doi: 10.1037//0022-0663.93.1.77

Midgley, C., Maehr, M. L., Hruda, L., Anderman, E. M., Anderman, L., Freeman, K. E., . . . Urdan, T. (2000). Manualfor the Patterns of Adaptive Learning Scales (PALS). Ann Arbor, MI: University of Michigan.

Muis, K. R. (2007). The role of epistemic beliefs in self-regulated learning. Educational Psychologist, 42, 173–190. doi:10.1080/00461520701416306

Muis, K. R., & Edwards, O. (2009). Examining the stability of achievement goal orientation. Contemporary EducationalPsychology, 34, 265–277. doi: 10.1016/j.cedpsych.2009.06.003

Muis, K. R., & Franco, G. M. (2009). Epistemic beliefs: Setting the standards for self-regulated learning. ContemporaryEducational Psychology, 34, 306–318. doi: 10.1016/j.cedpsych.2009.06.005

Muis, K. R., Winne, P. H., Edwards, O. V. (2009). Modern psychometrics for assessing achievement goal orientation: ARasch analysis. British Journal of Educational Psychology, 79, 547–576. doi: 10.1348/000709908×383472

Murayama, K., & Elliot, A. J. (2009). The joint influence of personal achievement goals and classroom goal structureson achievement-relevant outcomes. Journal of Educational Psychology, 101, 432–447. doi: 10.1037/a0014221

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

ACHIEVEMENT GOALS AND PERFORMANCE FEEDBACK 577

National Science Board. (2006). National Science Board commission on 21st century education in science, technology,engineering, and mathematics. Reston, VA: Author.

Pajares, F., Britner, S., & Valiante, G. (2000). Writing and science achievement goals of middle school students.Contemporary Educational Psychology, 25, 406–422. doi: 10.1006/ceps.1999.1027

Pekrun, R., Elliot, A. J., Maier, M. A. (2009). Achievement goals and achievement emotions: Testing a model of their jointrelations with academic performance. Journal of Educational Psychology, 101, 115–135. doi: 10.1037/a0013383

Pintrich, P. R. (2000). Multiple goals, multiple pathways: The role of goal orientation in learning and achievement.Journal of Educational Psychology, 92, 544–555. doi: 10.1037//0022-0663.92.3.544

Pintrich, P. R., Smith, D. A., Garcia, T., & McKeachie, W. J. (1991). A manual for the use of the Motivated Strategies forLearning Questionnaire (MSLQ). Ann Arbor, MI: University of Michigan.

Ryan, A. M., & Pintrich, P. R. (1998). Achievement and social motivational influences on help seeking in the classroom.In S. A. Karabenick (Ed.), Strategic help seeking: Implications for learning and teaching (pp. 117–139). Mahwah,NJ: Erlbaum.

Senko, C., & Harackiewicz, J. M. (2005). Achievement goals, task performance, and interest: Why perceivedgoal difficulty matters. Personality and Social Psychology Bulletin, 31, 1739–1753. doi: 10.1177/0146167205-281128

Senko, C., & Miles, K. M. (2008). Pursuing their own learning agenda: How mastery-oriented students jeopardize theirclass performance. Contemporary Educational Psychology, 33, 561–583. doi: 10.1016/j.cedpsych.2007.12.001

Skaalvik, E. M. (1997). Self-enhancing and self-defeating ego orientation: Relations with task and avoidance orientation,achievement, self-perceptions, and anxiety. Journal of Educational Psychology, 89, 71–81. doi: 10.1037/0022-0663.89.1.71

Wigfield, A. (1994). Expectancy-value theory of achievement motivation: A developmental perspective. EducationalPsychology Review, 6, 49–78. doi: 10.1007/BF02209024

Winne, P. H., & Hadwin, A. F. (1998). Studying as self-regulated learning. In D. J. Hacker, J. Dunlosky, & A. C. Graesser(Eds.), Metacognition in educational theory and practice (pp. 277–304). Hillsdale, NJ: Erlbaum.

Wolters, C., Yu, S., & Pintrich, P. R. (1996). The relation between goal orientation and students’ motivational beliefs andself-regulated learning. Learning and Individual Differences, 8, 211–238. doi: 10.1016/S1041-6080(96)90015-1

APPENDIX

Prior Knowledge Questionnaire

1. An increase in the temperature of a solution usually . . . {Solutions}decreases the solubility of a solid solute in the solution.Increases the boiling point.Decreases the solubility of a liquid solute in the solution.Increases the solubility of a gas in the solution.

∗ increases the solubility of a solid solute in the solution.2. The correct answer for the addition of 7.5 g + 2.26 g + 1.311 g + 2 g is . {Significant Figures}

13.071 g∗ 13 g

13.0 g10 g13.1 g

3. A pure substance is matter with a composition that . . . {Properties of Matter}always contains oxygen

• is fixed in a definite proportion at all timesalways contains two or more substancesvaries according to the amount of water presentdepends on the temperature

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3

578 MUIS ET AL.

4. Identify the metalloid in the following list. {Periodic Table}sulfur (S)

• germanium (Ge)silver (Ag)copper (Cu)fluorine (F)

5. The abbreviated electron configuration for a boron atom (Z = 5) is . {Electron Configuration}• [He] 2s2 2p1

1s2 2s2 2p1[Ne] 2s2 2p1[H] 2s2 2p11s2 2s2 2p6

6. Gold (III) bromide has the following correct formula. {Formula Writing}Au3BrAuBr3Au3Br2AuBr

∗ Au3Br37. What is the coefficient of hydrogen, H2, when the following equation is balanced? {Chemical Equations}

Al + H2SO4 → Al2(SO4)3 + H2• 3

5214

8. The following reaction is an example of a reaction. {Chemical Reaction Types}2C2H6 + 7O2 -→ 4CO2 + 6H2Osingle replacementdecompositiondouble replacementdisplacement

∗ combustion9. When 2.50 mol of Mg3N2 are allowed to react according to the following equation, how many moles ofH2O also react? {Stoichiometry}

Mg3N2 + 6H2O - → 3Mg(OH)2 + 2NH31.25 mol2.50 mol9.00 mol

∗ 15.0 mol6.00 mol

10. The volume of a gas with a pressure of 1.2 atm increases from 1.0 L to 4.0 L. What is the final pressureof the gas, assuming constant temperature? {Gas Laws}

1.0 atm1.2 atm4.8 atm3.3 atm

∗ 0.30 atm

Dow

nloa

ded

by [

Uni

vers

ity o

f N

orth

Dak

ota]

at 1

3:54

09

Oct

ober

201

3