Most studies of extrinsic incentives and intrinsic motivation, including those men- tioned earlier, used as controls subjects who received no rewards or feedback, apparently on the assumption that un- der these conditions original levels of intrinsic motivation would be maintained.

The role of the availability of such information was studied in comparison with conditions of nonreceipt of any information and of receipt of normative evaluation.

We expected that receipt of indiviaualized, specific, non-normative information about task performance, includ- ing both positive and negative comments, would maintain or even enhance subsequent motivation.

The present research was thus designed to study the ef- fects of different feedback conditions on intrinsic motiva- tion.

One aspect that merits further study is the search for information about one's com- petence and success in a task (Festinger, 1954; Suls & Miller, 1977). Such information seems vital to a sense of mastery and self-determination because without it one can- not assess one's mastery in any given task. Thus one would expect the availability (and/or expectation) of feedback to be an important factor in task motivation in general and in determining interest, or intrinsic motivation, in particular. Specifically, one would expect intrinsic motivation to be greater for tasks perceived as supplying information about competence and to be undermined when no such informa- tion is expecte

InastudybySwann&Arthurs(1998),alargenumberoftheirstudentsseemedtotakeaninstrumentalviewoflearning,conceivingassessmenttasksasobstaclestoovercomeinthepursuitofgrades.Formativefeedbackwasviewedasameanstonegotiatetheseobstacles.InanearlierstudybyBeckeretal.(1968)ofAmericancollegelife,assessmentdemandswereubiquitous,andstudentbehaviourreectedtheinstrumentalandpragmaticstrategiestheyadoptedtocopewiththeparticularteachingandassessmentpracticesimposedonthem.Butisthistruefortoday’sstudentinthecontextoftheUK?Amajorityofthestudentsinourstudyperceivehighereducationasa‘service’,andfeltthatfeedbackconstitutespartofthatservice.Asonestudentnoted:TheywayIseeitiswe’repaying£1,000.It’smoreofaservicenow.Ifhighereducationisviewedasaservice,thenstudentsarearguablytheconsumersofthatservice.Butwhatdotheyexpecttheservicetoconsistof?Moststudentsinourstudylinkfeedbacktoattainingbettergrades.Thesestudentsperceivefeedbackcommentsasidentify-ingwhattheyaredoingrightandwrongand,therefore,helpingthemtoimprovetheirperformanceinsubsequentassessedassignmentsandexaminationsinordertoraisetheirmarks:Partofwritingtheessayquestionintheexamishavingtherighttechnique,andwhilstitwouldbeusefultosaythat‘yeah,you’rebringingingoodpartsoutsidethesubjectandit’sgoodthatyou’vebroughtinthis’,itwouldalsobegoodtoknow‘well,don’teverusethislanguageintheexam’causeit’sgoingtocountagainstyou’

80%disagreedwiththestatement‘Feedbackcommentsarenotthatuseful’

ormativefeedbackcommentscanonlybeeffectiveifstudentsreadandmakeuseofthem.MostofthestudentsinvolvedinstudiesbyHyland(2000)andDing(1998)seemedtoreadtutors’comments.Ourquestionnairedatareectthis(seeTableI).Thetimespentreadingcommentsvaries,withthemajorityofstudentsclaimingtospendlessthan15minutesdoingso(although,ofcourse,ourdatadonottelluswhenthistakesplaceorwhetherstudentsreturntolookattheirfeedbackonmorethanoneoccasion).But,overall,97%ofstudentsindicatedthattheyusually‘read’thewrittenfeedbacktheyreceive.Furthermore,wecanseefromTableIIthat82%ofthestudentsclaimedto‘paycloseattention’tofeedback.Theinterviewdataalsosupportthis:Ialwayslookforwardtoseeingwhattheyhadtosay.NormallyIgetthegradeandthenlookthroughtheself-assessmentandthetutor’sassessment,readthecommentsand...seewhatcommentshe’smadeontheessay.ThisndingisreinforcedbyHyland’s(2000)study.Henotedthatthemajorityofthestudentsinvolved(fromarangeofinstitutions)seemedtotry(evenifonlyoccasionally)tousecommentsforfutureassignments

Afurtherbarriertotheuseofformativefeedbackmaybethatsomestudentsincreasinglyfailtounderstandthetaken-for-grantedacademicdiscourseswhichunderpinassessmentcriteriaandthelanguageoffeedback(Hounsell,1987).AccordingtoEntwistle(1984,p.1),‘effectivecommunicationdependsonsharedassumptions,denitions,andunderstanding’.ButastudyatLancasterUniversityfoundthat50%ofthethird-yearstudentsinoneacademicdepartmentwereunclearwhattheassessmentcriteriawere(Baldwin,1993,citedinBrown&Knight,1994).Asoneofourstudentsnoted:‘Ihaven’tgotacluewhatI’massessedon’

hesecommentssuggestthatstudentsinourstudyperceivefeedbacknegativelyifitdoesnotprovideenoughinformationtobehelpful,ifitistooimpersonal,andifitistoogeneralandvaguetobeofanyformativeuse

Black&Wiliam’s(2000)developingtheoreticalframeworkofformativeassessmentempha-sisestheinteractionsbetweenteachers,pupilsandsubjectswithin‘communitiesofpractice’

The Conscientious Consumer: Reconsidering therole of assessment feedback in student learning

value. The final model was significant, F(l, 349) = 52.80, p < .001 (R2 = .51). There was a significant gender main effect, ? = 08, ?(349) = 1.98, p < .05, indicating that females' intrinsic value was higher at Time 2 than males' intrinsic value, controlling for intrinsic value at Time 1. T

The Journal of Experimental Education, 2005, 73(4), 333-349 Changes in Self-Efficacy, Challenge Avoidance, and Intrinsic Value in Response to Grades: The Role of Achievement Goa

n addition, we examined whether there were differences in reactions to grades for papers versus exams. The two-way interactions between goals and type of feedback and the three-way interactions among goals, types of feedback, and grades were tested. Neither the main effects nor the interaction terms were significant, and including these terms did not alter the results for goals and goal interaction term

. Even in the context of high grades, a per formance-avoidance goal is related to decreased motivation, highlighting just how aversive evaluative situations are with this goal orientation. Pe

ulnerability to helpless behavior in the context of failure on experimental tasks. In contrast, and also consistent with Dweck's prior work (Diener & Dweck, 1978, 1980; Dweck & Leggett, 1988; El liott & Dweck), mastery goals were associated with enhanced motivation during the critical phase of processing evaluative information on an important task, and this effect was not moderated by the level of the gra

Achievement goals were important to changes in motivational constructs around the receipt of grades in the classroom. As expected, the effects of a per formance-approach goal on changes in motivational constructs were moderated by grades. When students received high grades, a performance-approach goal was unrelated to changes in self-efficacy, desire to avoid challenge, or intrinsic value. However, when students received low grades, a performance-approach goal was related to decreased intrinsic value and increased desire to avoid chal lenge. Thus, although a performance-approach goal does not seem to have draw backs in the context of success, there are drawbacks when students experience setbacks

in recent years, some researchers have concluded that it is only perfor mance-avoidance goals that have drawbacks and that performance-approach goals promote high achievement and do not affect motivation and engagement negatively

Discussion

Intrinsic value. We adapted items developed by Eccles (1983) to assess students' intrinsic value regarding their academic work in the class. This section of the sur vey asked students, "What is your opinion of this class along the following di mensions?" Students rated the "enjoyment of work" on a scale ranging from 1 (not at all enjoyable) to 7 (very enjoyable), "interest in the work" on a scale rang ing from 1 (very boring) to 7 (very interesting), and "liking what is learned" on a scale ranging from 1 (a little) to 7 (a lot). Alpha coefficients for the three items were .92 at Time 1 and .93 at Time 2

Preference to avoid challenging work. Preference to avoid challenging work (4 items) assesses students' desires for easy, familiar tasks (Urdan, Ryan, Ander man, & Gheen, 2002). Sample items are "I prefer doing work that does not make me think too hard" and "I prefer assignments that I know I can do rather than those that are a challenge." The measure was found to be reliable in our sample (a at Time 1 = .85; Time 2 = .85)

rades. Students self-reported their grades. Grades for the English and history classes were the grades students received on the first major paper of the semester. Grades for the other courses were the grades students received on the first major exam of the semester. Grades were converted into a 13-point scale with F = 1 and A+ =13. We also assessed students' perceptions of their grades to address the pos sibility that it is the students' perceptions of their grades as successful, or not, that are more important than their actual grades. After reporting their exam or paper grades on the survey, students were asked, "In your opinion, how well did you do on the exam or paper?" The format for this item was 1 (not at all well) through 7 (very well). Students' perceptions of how well they did were highly correlated with their reported exam or paper grades (r = .77, p < .001), indicating that stu dents who received higher grades tended to feel they did w

In summary, our main goal was to examine how students' achievement goals are related to changes in self-efficacy, preference to avoid challenge, and intrin sic value in the face of evaluation. Early in the semester, we assessed students' achievement goals, self-efficacy, desire to avoid challenge, and intrinsic value. We assessed students' self-efficacy, desire to avoid challenge, and intrinsic value again immediately after they received their grades on their first major exam or paper. This design allowed us to examine the role of goals in the change in mo tivational constructs associated with performance feedback. Our main hypothe ses were (a) a mastery goal will be associated with enhanced motivation around receipt of grades (i.e., increased efficacy and value and lower preference for chal lenge avoidance); (b) a performance-avoidance goal will be associated with di minished motivation around receipt of grades (i.e., decreased efficacy and value and increased preference for challenge avoidance); and (c) the effects of a per formance-approach goal on changes in motivation will be moderated by grades. When students encounter low grades, a performance-approach goal will be relat ed to diminished motivation. When students receive high grades, a performance approach goal will be unrelated to changes in motivation.

Shim & Ryan 337 Furthermore, we expected a performance-avoidance goal to be associated with declines in motivational constructs, even in the context of high grades. A perfor mance-avoidance goal brings about negative achievement-related processes re garding evaluation. A performance-avoidance goal is associated with construing exams as a threat; incurring negative emotions, such as worry, fear, and anxiety; and the desire to escape exam situations (McGregor & Elliot, 2002). A perfor mance-avoidance goal, undergirded by a fear of failure, inherently involves a focus on a negative outcome (Elliot, 1999). With a performance-avoidant frame work, positive feedback is interpreted as "not failing" or "not being the worst." Al though such an assessment satisfies a performance-avoidance goal, it is unlikely to boost motivation, as the absence of something negative is not evidence of some thing positive. Thus, we expected a performance-avoidance goal to be associated with diminished motivation, regardless of whether grades are high or

Performance goals are associated with the belief that intelligence is fixed (a

uccess and failure are attrib uted to effort (Ames, 1992). Even in the face of failure or obstacles, a mastery goal is associated with persistence (Ames, 1984; Diener & Dweck, 1978; Hong, Chiu, & Dweck, 1995; Stipek & Kowalski, 1989). When oriented toward a mas tery goal, success (in the case of high grades) bolsters motivation, whereas a lack of progress (in the case of low grades) signals more effort is needed and thus does not diminish motivation. Therefore, we expected a mastery goal to be associated with increases in self-efficacy, intrinsic value, and a decrease in preference to avoid challenge regardless of high or low grades.

We expected a mastery goal to be associated with increases in motivation in response to grades. A mastery goal is associated with construing exams as a chal lenge and incurring positive emotions, such as eagerness, hopefulness, and ex citement (McGregor & Elliot, 2002). It is also associated with the belief that in telligence is a malleable attribute that can be developed through effort (an incremental theory of intelligence; Dweck, 1999).

n recent years, some research has indicated that performance-ap proach goals are beneficial for achievement and do not affect motivation nega tively (see Harackiewicz, Barron, Pintrich, Elliot, & Thrash, 2002). In particular, when the approach versus avoidance nature of performance goals is considered, performance-avoidance goals are maladaptive, whereas performance-approach goals are often positively associated with achievement and show a positive or neutral relation to motivation

iener and Dweck (1978) demonstrated that performance-oriented students exhibited a learned helpless pattern after ex periencing failure.

In the present research, we examined how students' goals relate to changes in self-efficacy, desire to avoid challenge, and intrinsic value in response to grades in the classroom

performance-approach goal has been positively associated with self-efficacy (Bong, 2001; Pajares et al; Skaalvik; Wolters, Yu, & Pintrich, 1996), desire to avoid challenging work (Meyer, Turner, & Spencer, 1997; Middleton & Midgley, 2002), and task value (Bong; Church, Elliot, & Gable, 2001; Wolters et al). However, some researchers have found no relation between a performance-ap proach goal and self-efficacy (Middleton & Midgley, 1997; Pajares et al.) or task value (Lopez, 1999; Tanaka & Yamauchi, 2001), so it is not clear whether this is always the case

udy, a mastery goal is positively associated and a performance-avoidance goal is nega tively associated with self-efficacy, challenge-seeking, and intrinsic value (Mid dleton & Midgley, 1997; Pajares, Britner, & Vahante, 2000; Skaalvik, 1997).

contrast, a performance goal concerns a focus on demonstrating competence. Performance goals can be distinguished as either approach or avoidant (Elliot & Church, 1997; Middleton & Midgley, 1997; Skaalvik, 1997). A performance-approach goal concerns a focus on gaining favorable judgments of one's ability, and a performance-avoid ance goal concerns a focus on avoiding negative judgments of one's ability. Achievement goals represent disparate purposes for involvement regarding aca demic tasks and, as such, have been linked to different achievement-related processes and outcomes

performance goal concerns a focus on demonstrating competence. P

mastery goal concerns a focus on developing competence and gaining understanding or mastery. I

chievement goals capture meaningful distinctions in how individuals orient themselves to achieving competence in the academic setting (Ames, 19

Our purpose in this study was to examine the role of college students' achieve ment goals in changes in their self-efficacy, preference to avoid challenge, and intrinsic value in response to grades on an important task in the classroom (first major exam or paper of the academic yea

rades are widely recognized as important in our society and carry importance for students. Grades are the basis for many ca reer decisions, including how students perceive themselves and their schoolwork. The potential undermining effect of grades for student motivation has been dis cussed often (Deci & Ryan, 2002; Kohn, 1993). Despite the importance of grades for student motivation, there has been little research examining how classroom grades relate to changes in students' motivation. Prior research has more often fo cused on different facets of performance feedback in the lab setting (Deci, Koest ner, & Ryan, 2001; Eisenberger, Pierce, & Cameron, 1999; Kamins & Dweck, 1999; Lepper, Henderlong, & Gingras, 1999; Mueller & Dweck, 1998). T

Manual for thePatterns of Patterns of Adaptive Adaptive Learning ScalesLearning Scales

. First,although grading has been shown to provoke challenge avoidance(Harter, 1978

Importantly, both Elliot (1999) and Pintrich (2000) have arguedthat strong situational cues or contexts can override the effects ofchronic goal orientation. In the case of graded assessments, whenis the situational cue most salient? According to Harackiewicz,Manderlink, and Sansone (1992), when a performance-contingentreward such as a grade is attached to performance, performanceevaluation is anticipated and can arouse motivational and emo-tional consequences in the individual prior to engaging in a task.Thus, pretask performance goals would seem to be a fruitful stategoal area to explore, particularly as pretask performance goaladoption has important consequences on task performance andrelated outcomes (Elliot & McGregor, 2001; McGregor & Elliot,2002).

Kluger and DeNisi (1996), in their presentationof feedback intervention theory, used grading as an example of anexternal intervention likely to render metatask processes concern-ing self-goals, such as performance goals, salient at the expense oftask-related goals.

achievement goal theory (Ames, 1992; Butler,1987; Nicholls, 1984) has consistently argued that grades engendera performance focus, as performance achievement goals are ge-nerically considered as normative in content (Elliot & Murayama

Grades, a value-laden symbol indicating the rela-tive quality of a performance that is a regular feature of school life(Pope, 2001), are positioned firmly on the evaluative side of thistypology. The value of a grade reflects certain norms, and thegrade attributed reflects the degree to which normatively deter-mined standards have or have not been attained

a typology of assessment feedback, Tunstall and Gipps(1996)

Dependence on those who distribute valued resources isthe equivalent of powerlessness, and powerlessness has been as-sociated with a basic inhibition or avoidance motivational orien-tation (Keltner, Gruenfeld, & Anderson, 2003

Evaluation is an inescapable feature of academic life with regular grading and performance appraisals atschool and at university. Although previous research has indicated that evaluation and grading inparticular are likely to have a substantial impact on motivational processes, little attention has been paidto the relationship between grading and approach versus avoidance achievement goals, 2 fundamentalconcerns whenever evaluation is at stake. Three experiments, carried out in professional schools, revealedthat expectation of a grade for a task, compared with no grade, consistently induced greater adoption ofperformance-avoidance, but not performance-approach, goals. Experiments 2 and 3 revealed that expec-tation of a grade, compared with no grade, consistently induced greater adoption of performance-avoidance goals even when grading was accompanied by a formative comment. Furthermore, Experiment3 showed that reduced autonomous motivation measured after having completed a task for a grade versusno grade mediated the relationship between grading and adoption of performance-avoidance goals in asubsequent task. Results are discussed in the light of achievement goal and self-determination th

Why Grades Engender Performance-Avoidance Goals:The Mediating Role of Autonomous Motivation

Brookhart (2008, p.8) concludes, “the grade ‘trumps’ the comment” and “com-ments have the best chance of beingreadas descriptive if theyare not accompanied by a grade.”

Butler R, Nisan M (1986). Effects of no feedback, task-related com-ments, and grades on intrinsic motivation and performance. J EducPsychol78,

Teaching More by Grading Less (or Differently)

Grades appear to play on students’ fearsof punishment or shame, or their desires to outcompete peers,as opposed to stimulating interest and enjoyment in learningtasks (Pulfreyet al., 2011)

Swinton (2010) additionally foundthat a grading system that explicitly rewarded effort in ad-dition to rewarding knowledge stimulated student interestin improvement. This implies that balancing accuracy-basedgrading with providing meaningful feedback and awardingstudent effort could help avoid some of the negative conse-quences of grading.

icol and Macfarlane-Dick argue that studentshave been deprived of opportunities to become self-regulatedlearners who can detect their own errors in thinking. T

ssessing ef-fort and participation can happen in a variety of ways (Beanand Peterson, 1998; Rocca, 2010). In college biology courses,clicker questions graded on participation and not correct-ness of responses is one strategy. Additionally, instructors canhave students turn in minute papers in response to a ques-tion posed in class and reward this effort based on submissionand not scientific accuracy. Perhaps most importantly, biol-ogy instructors can assign out-of-class work—case studies,concept maps, and other written assignments—that can pro-mote student practice and focus students’ attention on keyideas, while not creating more grading work for the instruc-tor. Those out-of-class assignments can be graded quickly(and not for accuracy) based on a simple rubric that checkswhether students turned the work in on time, wrote the re-quired minimum number of words, posed the required num-ber of questions, and/or included a prescribed number ofreferences. In summary, one strategy for changing grading isto balance accuracy-based grading with the awarding of someproportion of the grade based on student effort and partic-ipation. Changing grading in this way has the potential topromote student practice, incentivize in-class participation,and avoid some of the documented negative consequencesof grading.

ne strat-egy for focusing students on the importance of effort andpractice in learning is to provide students opportunities toearn credit in a course for simply doing the work, completingassigned tasks, and engaging with the material. A

Multiple research studies described above suggest that theevaluative aspect of grading may distract students from afocus on learning. While evaluation will no doubt alwaysbe key in determining course grades, the entirety of stu-dents’ grades need not be based primarily on work thatrewards only correct answers, such as exams and quizzes.Importantly, constructing a grading system that rewards stu-dents for participation and effort has been shown to stimulate

In part, grading practices in higher education have beendriven by educational goals such as providing feedback tostudents, motivating students, comparing students, and mea-suring learning. However, much of the research literature ongrading reviewed above suggests that these goals are oftennot being achieved with our current grading practices. Ad-ditionally, the expectations, time, and stress associated withgrading may be distracting instructors from integrating otherpedagogical practices that could create a more positive and ef-fective classroom environment for learning.

Rather than motivating students to learn, grading appearsto, in many ways, have quite the opposite effect. Perhaps atbest, grading motivates high-achieving students to continuegetting high grades—regardless of whether that goal alsohappens to overlap with learning. At worst, grading lowersinterest in learning and enhances anxiety and extrinsic moti-vation, especially among those students who are struggling

High-achieving students on initial graded assignments ap-pear somewhat sheltered from some of the negative impactsof grades, as they tend to maintain their interest in completingfuture assignments (presumably in anticipation of receivingadditional good grades; Butler, 1988). Oettinger (2002) andGrant and Green (2013) looked specifically for positive im-pacts of grades as incentives for students on the thresholdbetween grade categories in a class. They hypothesized that,for example, a student on the borderline between a “C” anda “D” in a class would be more motivated to study for a finalexam than a student solidly in the middle of the “C” range.However, these studies found only minimal (Oettinger, 2002)or no (Grant and Green, 2013) evidence that grades moti-vated students to perform better on final exams under theseconditions

Rather than seeing low grades as an oppor-tunity to improve themselves, students receiving low scoresgenerally withdraw from class work (Butler, 1988; Guskey,1994). While students often express a desire to be graded,surveys indicate they would prefer descriptive comments togrades as a form of feedback (Butler and Nisan, 1986).

Even providingencouraging, written notes on graded work does not appearto reduce the negative impacts grading exerts on motivation(Butler, 1988

Grades can dampen existing in-trinsic motivation, give rise to extrinsic motivation, enhancefear of failure, reduce interest, decrease enjoyment in classwork, increase anxiety, hamper performance on follow-uptasks, stimulate avoidance of challenging tasks, and heightencompetitiveness (Harter, 1978; Butler and Nisan, 1986; But-ler, 1988; Crooks, 1988; Pulfreyet al., 2011)

rather than stimulating an interest in learning, grades pri-marily enhance students’ motivation to avoid receiving badgrades (Butler and Nisan, 1986; Butler, 1988; Crooks, 1988;Pulfreyet al., 2011)

ur current“A”–“F” grading system was not designed with the primaryintent of motivating students. Rather, it stemmed from effortsto streamline communication between institutions and di-minish the impacts of unreliable evaluation of students fromteacher to teacher (

Grading does not appear to provide effective feedback thatconstructively informs students’ future efforts. This is partic-ularly true for tasks involving problem solving or creativity.Even when grading comes in the form of written comments, itis unclear whether students even read such comments, muchless understand and act on them.

arbleet al.,1978; Butler 1988; Pulfreyet al., 2011

Anecdotal accounts, as well as some studies, indicate thatmany students do not read written feedback, much less useit to improve future work (MacDonald, 1991; Crisp, 2007). I

Butler and Nisan (1986) compared theimpacts of evaluative feedback, descriptive feedback, and nofeedback on student achievement in problem-solving tasksand in “quantitative” tasks (e.g., those requiring quick, timedwork to produce a large number of answers). They foundthat students receiving descriptive feedback (butnotgrades)on an initial assignment performed significantly better onfollow-up quantitative tasks and problem-solving tasks thandid students receiving grades or students receiving no feed-back. Students receiving grades performed better on follow-up quantitative tasks than students receiving no feedback,but did not outperform those students on problem-solvingassignments. In other words, providing evaluative feedback(in this case, grades) after a task does not appear to enhancestudents’ future performance in problem solving.

Feedback is generally divided into two categories: eval-uative feedback and descriptive feedback. Evaluative feed-back, such as a letter grade or written praise or criticism,judges student work, while descriptive feedback provides in-formation about how a student can become more competent(Brookhart, 2008, p. 26

Higginset al., 2002

Assessment and Classroom Learning

This manuscript provides a critical examination of the formative assessment literature in particular issues related to the formative assessment lexicon, Black and Wiliam’s (1998) seminal work, and more recent research. Finally, this manuscript provides the foundation for a series of manuscripts on “best practices” for evaluating student achievement through the use of formative assessment.

A Critical Review of Research on Formative Assessment: The Limited Scientific Evidence of the Impact of Formative Assessment in Education

professor’s perspective in a particular cla

ConclusionsThis paper finds that the effort grade affects theknowledge grade positively and significantly across allspecifications. This is strong evidence that more studenteffort does lead to increase learning

The following examples illustratesome choices a professor can make. Suppose a professordecides to maximize student’s grades by giving all studentsthe best possible grade. If the professor does this, studentshave no incentive to put forth effort. Even worse, the sig-nal, which is just the grade, does not tell future employersanything about the students’ ability.If a professor decides instead to maximize the effortgiven, the professor gives all the credit to effort as opposedto knowledge. This maximizes the effort expended by thestudents, however, the signal only tells employers how thestudent is at giving effort and not what the student knows.If a professor instead decides to give credit only forknowledge as opposed to effort, this may cause studentsbelow a certain ability level to give no effort. This policy sig-nals to future employers which students have low or highability. However, the policy will not induce all students togive effort.9Therefore, the problem is how do you induceall students to give effort without lowering the amount ofknowledge gained by the students or weakening or mini-mizing the signal that is sent to future employe

d. A stu-dent’s grade is a signal to future employers about thestudents’ ability, therefore, students want to maximize thegrade that they receive in a class. The goal of the pro-fessor/administrator, hereafter called the professor, is toassign grades that maximize the informational learning ofthe students. Students generally want to put in the leastamount of effort to earn a given grade—the measure ofthe knowledge that the student has mastered in the class.8The professor wants to maximize the signal that the gradegives to future employers, i.e. to have the student earn thehighest grade, but does not want to lower the amount ofknowledge that the students are required to lear

s. Thismeasure includes class attendance, whichRomer (1993),Dobkins, Gil, and Marion (2010)and others, find is relatedto performance

endogeneit

study time did have an impact ongrades, whereas class attendance did not. Running a regres-sion on separate categories of students, they found thatstudents who study throughout the week derived morebenefits than crammers. In contrast to Schuman et al., theyfound that an increase in academic effort was rewardedwith higher grades.Rau and Durand (2000)also tested theeffects of effort on college grades. They found that studentswho studied daily and have better study habits performedbetter on tests, even if they had lower standardized testscores. They concluded that effort made a difference atIllinois State University.Schuman (2001)replied toRauand Durand (2000)that the difference in results were duemainly to the fact that Rau and Durand used somewhatdifferent measures of effor

at study time during theweek was not related to any measure of college grades,while study time on the weekend was significantly and pos-itively correlated with all measures of college grad

The authors found no significant increase inthe relationship between time studied and GPA. An organicchemistry lab was also studied. For this specific class, therewas even less evidence that the amount of time spentstudying affected achievement as measured by grades. Thepaper concluded that study time and grades may have asubstantial relationship, but the measures of effort in anindividual study day may not capture the variation thatoccurs from day to day

Conventional wisdom suggests that the more effort astudent puts forth, the better grade the student will earn.However, researchers find mixed results when examiningthis relationship. This finding may be due to the less thanideal manner in which effort has been measured in theliterature. Moreover, existing estimates regarding the linkbetween effort and grades are biased because the estimatesignore the stochastic nature of effor

The unique aspect of this policy is that, for freshmanand sophomore level courses, effort is a separate compo-nent part of a student’s grade.3The SE2policy requires aprofessor to report two grades to the registrar for studentstaking freshman and sophomore level courses: effort andcontent learning (knowledge). The administration weightsthe two grades differently for freshman and sophomorecourses.Tables 1 and 2present the final grade outcomes forfreshman and sophomore courses, respectively. Roughlyspeaking, knowledge and effort grades are weighted 40%and 60%, respectively, for freshman courses and 60% and40% in sophomore course

The effect of effort grading on learnin

THE GRADING OF STUDENT

WhereAIsOrdinary:TheEvolutionofAmericanCollegeandUniversityGrading,1940-2009

Ifyouhavefoundthecouragetogoawholesemesterorcoursewithoutanygrades,youwillbefacedwithapotentiallyparalyzingproblem:Howdoyoucomeupwithareportcardgradewhenyouhavenogradebook?WhenIfirstfacedthisproblem,IfoundmyselfwritingtoAlfieKohn,whereIoutlinedthatIwashappytoreportthatIhadreplacedeverydaygradingwithrealcommentsandconstructivefeedback,butthatIwasstrugglingwithhowIcouldconvertorsymbolizeallthefeedbackasagrade,notbecauseIwantedtobutbecauseIhadto.Kohn’sreplywastremendouslyhelpful:Myprimaryanswertoyourquestionis“Bringthekidsinonit.”Thisshouldbeadecisionyoumakewiththem,notforthem.Thatgoesforthegeneralclasspolicy(andtherationaleforit)aswellthespecificgradegiventoeachstudent.Someteachersmeetwitheachstudentindividuallyanddecidetogetherwhatthefinalgradewillbe.Others,whoaremorewillingtogiveupcontrolandempowerstudents,simplyletthestudentdecide.Theyinvariablyreportthatstudentsenduppickingthesamegradethattheteacherwouldhavegiven,andsometimestheyevensuggestalowerone.Buttheadvantagesoflettingthekidsdecideareincalculable,andtheprocessalsohasthesalutaryeffectofneutralizingthedestructiveeffectsofhavingtogivegradesinthefirstplace.(citedinBower,2010)Yearslater,Istillabidebythisprofoundadvice.Bringthekidsinonitremainsattheheartofmyanswertoanyonewhoasks,“Howdoyougradewithoutgrading?”First,evenifagradeismandatedforthereportcard,itmakesverylittlesensethattheonlywaytocomeupwithafinalgradewouldbetotakealistofotheraveragesandaveragethemtogethertogetafinalaverage(Wormeli,2006).

Infact,whenGallimoreandRolandactuallyrecordedandcategorized2,326ofCoachWooden’sactsofteaching,theyfoundthatonly6.6%wereactsofdisapprovalwhile6.9%wereactsofpraise(Coyle,2009).Thatmeansthemajorityofhisinteractionswithhisathleteswerejudgmentfreestatementsofinformation

IhavecometolivebywhatAlfieKohnhasidentifiedasBruner’sLaw,whichistosaythatweshouldtryandcreateanenvironmentwherestudentscan“experiencesuccessandfailurenotasrewardandpunishment,butasinformation”(ascitedinKohn,1999e,p.191)

OnceIwasabletomovepastsimplyasking“HowdoIgradebetter?”IrealizedthatIneededtofocusmoreon“WhydoIgradeatall?”Thereisalotofcommonsensearoundwhywegradestudents,butifyoulookclosely,theconventionalwisdomdoesnotmakeallthatmuchsenseandisunfortunatelyalltoocommon

haddecidednottogradetheiressays.Iwasbeamingwithexcitement;theywerenot.Suddenly,theairbeneathmywingshaddisappeared.Myexcitementwaslostonthem,andIwasdisheartened.Butwhathappenednextbothappalledandenlightenedme.Istoodthereatthefrontoftheclassandheardwhatsoundedlikeall30ofthemyellinunison,“Youmeanwedidthisallfornothing?”InitiallyIfeltlikeIhadbeenkickedinthegut,butthenIfeltliketheGrinchwhenhisheartgrewthreesizesthatday.

IremembersearchingtheWebforalternativestogradingandfindinganarticletitled“TheCostsofOveremphasizingAchievement”(Kohn,1999b).ItwasthefirstAlfieKohnarticleIhadeverread,butitwouldbefarfromthelast,anditprovedtobethepedagogicalpillthatIhadbeenlookingfortocuremyailmentsforgrading.Irememberdevouringthearticleandreturningtomygrade8classroomthenextdaywithasenseofrevitalizedurgencythatIcouldnotwaittosharewithmystudents.Thatyearmyteachingassignmentincludedtwoclassesofabout30grade8studentswhomItaughtlanguageartsandscience.Inanefforttointegratethetwosubjects,Ihadassignedmystudentstowriteanessayontheparticlemodelofmatter.Asfarastheyknew,Ishouldhavebeengradingtheirpapers,butIwasabouttoblowtheirminds.Iwalkedintocl

Thepeoplewhohaveahardtimecomprehendinghowchildrencouldlearnwithoutextrinsicmanipulatorsconcernmethemost.Theyaresoinvestedintraditionalschoolingthattheyhaveneverquestioneditsfoundation.Unfortunately,somehaveadistrustfulviewofthenatureofchildren;theybelievethatwithoutgradingtherewouldbenothingtostopchildrenfromrunningamok

Fortoolong,Iwaslettingschoolinggetinthewayofmyteachingandtoomanyofmyteachingpracticeswerebasedonpedagogythatwasatbestunhelpfulandatworstharmfultomylongtermgoals.Throughcriticalquestioningandextensiveresearch,Icametotheconclusionthatmypedagogyhadtorevolvearoundonepriority:learning.Iftherewerethingsthatworkedtosabotagelearning,thenitwasmyprofessionalresponsibilitytoremovethem

IamnotthesameteacherIusedtobe.WhenIstarted,Iwasfocusedonpowerandcontrol.Iassignedloadsofhomework,dishedouthugepenaltiesforlateassignments,assignedpunishmentsforrulebreakingbehavior,andaveragedmarkstodeterminethestudents’finalgrade.IdidsomeofthesethingsbecauseIwastrainedtodosoinuniversity.However,mostoftheseteachingstrategieswerebeingdonemindlesslyand,forthemostpart,IwassimplyteachingthewayIwastaught.

IttookonlysixyearsbeforeIwantedtoquitteaching.Ihadbecomeincreasinglyunhappywithmyteachingandmystudents’learning.Iwastiredoflaboringthroughhoursandhoursofmarking,andIhatednaggingkidstocompletetheirhomework.Insteadofstudentsasking“Whatisthisquestionworth?”Iwantedthemtoactuallygetexcitedaboutthecontent.Iwantedchange,andIcameclosetothinkingthatchangerequiredmetoleavetheprofession

Bruner’s Law -we want kids to regard success and failure as information not as reward and punishment.

It was of interest that all attribute categories of un- creative characteristic~ and almost all attribute catego- ries of creative traits (39 of 42) were suggested by both male and female teachers

The respondents also de- scribed a creative person as one who has a collectivistic orientation, such as one who "inspires people," "has contribution to the progress of society" and "is appreciated by others." These descriptions, found in this sample of Chinese people, did not occur in U.S. investigations (Rudowicz et al., 1995

onventional, " "timid, " "lack of conjidence. "and "conforming. "

imaginative, " "always ques- tioning, " "quick in responding. " "active, " and "high intellectual ability, "

Implicit Theories of Creativity: Teachers' Perception of Student Characteristics in Hong Kong

"self-directed," "curious," "original," "artistic," "intel- ligent," "interested in many things," "exploratory," "unique," "innovative," "flexible," "imaginative," "al- ways questioning," "nonconforming," "challenging," "uninhibited," "independent," "sensitive," "expres- sive," "inventive," and "good at designing."

nalogies between teaching and various aspects of show business or guidance counseling are more often than not excuses for having abdicated the task

White (1984, cited by Vaughan, 1991) reported on a study conducted at California State University in which two essays were tucked into a huge sample of essays and read a year apart by the same readers using a 6-point scale. The reading a year later produced scores that were identical to the first in only 20 per cent of the cases. The scores differed by one point or less in 58 per cent of cases and 2 points or less in 83 per cent of the cases. As White points out, a 1-point difference is generally considered unproblematic, but on a 6-point scale the difference between a 3 and a 4 is the difference between a pass and a fail. Obviously, then, changes in examiner severity/leniency over-time have implications for maintaining standards, and must be monitored. Research has been conducted into variations in examiner severity/leniency during the marking of a particular allocation of scripts, a marking period, and over more extended periods of time.

According to Stemler, consistency estimates of interrater reliability assume that it is not necessary for judges to share a common meaning of the rating scale, so long as each judge is consistent in their classifications.

(2004) notes that most research papers describe interrater reliability as though it is a single, universal concept. He argues this practice is imprecise and potentially misleading. The specific type of interrater reliability being discussed should be indicated. He categorises the most common statistical methods for reporting interrater reliability into one of three classes: consensus estimates; consistency estimates; and measurement estimates.

in the latter both the wide differential in manuscript rejection rates and the high correlation between refereerecommendations and editorial decisions suggests that reviewers and editors agree more on acceptance than on rejection.

In the former there is also much more agreement on rejectionthan acceptance

. Referees ofgrant proposals agree much more about what is unworthy of support than about what does have scientific value. In

A system that assumes a "quite good" institution is unable to get better, and thus denies them the funds that would enable them to get better, is probably not an optimal system for promoting merit. A system that rewards in proportion to merit would at least be able to recognise and reflect the dynamism of university research; research groups wax and wane as people come, go, get disheartened, get re-invigorated.

it could be argued that we don’t just need an elite: we need a reasonable number of institutions in which there is a strong research environment, where more senior researchers feel valued and their graduate students and postdocs are encouraged to aim high. Our best strategy for retaining international competitiveness might be by fostering those who are doing well but have potential to do even better

Manso (2011)

Lerner and Wulf’s (2007) study of corporate R&D lab heads.They show that higher levels of deferred compensation are associated with the production ofmore heavily cited patents, whereas short-term incentives bear no relationship to firm innovativeperformance

Tian and Wang (2010) show that start-up firms backed by morefailure-tolerant venture capitalists are more innovativeex post

Lazear (2000)

If incentives play an important role in theproduction of novel ideas, this heroic story might be atypical. In this article, we provide empiricalevidence that nuanced features of incentive schemes embodied in the design of research contractsexert a profound influence on the subsequent development of breakthrough ideas.

Mario Capecch

excellencetorecognizeexcellence

. I consider that my job, as a philosopher, is to activate the possible, and not to describe the probable, that is, to think situations with and through their unknowns when I can feel them

5

They have an obligation to develop and maintain competence and effectiveness within their area of expertise, to conscientiously prepare and organize their subjec

VotingFour (4) persons selected through procedures established by the Faculty Council and approved by the General Faculties Council. The procedures shall provide for a system of alternates. Alternates shall replace regular members whose schedules would cause unreasonable delay in a committee’s proceedings or who would have a conflict of interest

Begg, C. B., & Berlin, J. A. (1988). Publication bias: A problem in interpreting medical data.Journal of theRoyal Statistical Society A, 151, 419–463.

Gerber, A. S., & Malhotra, N. (2008). Publication bias in empirical sociological research: Do arbitrarysignificance levels distort published results?Sociological Methods & Research, 37, 3–30

Gilbody, S. M., Song, F., Eastwood, A. J., & Sutton, A. (2000). The causes, consequences and detection ofpublication bias in psychiatry.Acta Psychiatrica Scandinavica, 102, 241–249

Greenberg, S. A. (2009). How citation distortions create unfounded authority: Analysis of a citation net-work.British Medical Journal, 339, b2680

Hochberg, M. E., Chase, J. M., Gotelli, N. J., Hastings, A., & Naeem, S. (2009). The tragedy of the reviewercommons.Ecology Letters, 12, 2–4

Kennedy, D. (2004). The old file-drawer problem.Science, 305, 45

Koletsi, D., Karagianni, A., Pandis, N., Makou, M., Polychronopolou, A., & Eliades, T. (2009). Are studiesreporting significant results more likely to be published?American Journal of Orthodontics andDentofacial Orthopedics, 136, 632e1

Krzyzanowska, M. K., Pintilie, M., & Tannock, I. F. (2003). Factors associated with failure to publish largerandomized trials presented at an oncology meeting.Journal of the American Medical Association,290, 495–501

Levine, T., Asada, K. J., & Carpenter, C. (2009). Sample sizes and effect sizes are negatively correlated inmeta-analyses: Evidence and implications of a publication bias against non-significant findings.Communication Monographs, 76, 286–302

Marsh, H. W., Bornmann, L., Mutz, R., Daniel, H. D., & O’Mara, A. (2009). Gender effects in the peerreviews of grant proposals: A comprehensive meta-analysis comparing traditional and multilevelapproaches.Review of Educational Research, 79, 1290–1326

Paris, G., De Leo, G., Menozzi, P., & Gatto, M. (1998). Region-based citation bias in science.Nature, 396,6708

Rosenthal, R. (1979). The file drawer problem and tolerance for null results.Psychological Bulletin, 86,638–641

Song, F. J., Parekh-Bhurke, S., Hooper, L., Loke, Y. K., Ryder, J. J., Sutton, A. J., et al. (2009). Extent ofpublication bias in different categories of research cohorts: A meta-analysis of empirical studies.BMCMedical Research Methodology, 9, 79

Sterling, T. D. (1959). Publication decisions and their possible effects on inferences drawn from tests ofsignificance—Or vice versa.Journal of the American Statistical Association, 54, 30–34

Osuna, C., Crux-Castro, L., & Sanz-Menedez, L. (2011). Overturning some assumptions about the effects ofevaluation systems on publication performance.Scientometrics, 86, 575–592

Pautasso, M. (2010). Worsening file-drawer problem in the abstracts of natural, medical and social sciencedatabases.Scientometrics, 85(1), 193–202

Schmidt, S. (2009). Shall we really do it again? The powerful concept of replication is neglected in thesocial sciences.Review of General Psychology, 13(2), 90–100.

Shelton, R. D., Foland, P., & Gorelskyy, R. (2009). Do new SCI journals have a different national bias?Scientometrics, 79(2), 351–363. doi:

Silvertown, J., & McConway, K. J. (1997). Does ‘‘publication bias’’ lead to biased science?Oikos, 79(1),167–168.

Yousefi-Nooraie, R., Shakiba, B., & Mortaz-Hejri, S. (2006). Country development and manuscript selec-tion bias: A review of published studies.BMC Medical Research Methodology, 6, 37

Evanschitzky, H., Baumgarth, C., Hubbard, R., & Armstrong, J. S. (2007). Replication research’s disturbingtrend.Journal of Business Research, 60(4), 411–415. doi

Jeng, M. (2006). A selected history of expectation bias in physics.American Journal of Physics, 74(7),578–583

Ioannidis, J. P. A. (2008a). Perfect study, poor evidence: Interpretation of biases preceding study design.Seminars in Hematology, 45(3), 160–166

Feigenbaum, S., & Levy, D. M. (1996). Research bias: Some preliminary findings.Knowledge and Policy:The International Journal of Knowledge Transfer and Utilization, 9(2 & 3), 135–142.

Song, F., Parekh, S., Hooper, L., Loke, Y. K., Ryder, J., Sutton, A. J., et al. (2010). Dissemination andpublication of research findings: An updated review of related biases.Health Technology Assessment,14(8), 1–193. doi

De Rond, M., & Miller, A. N. (2005). Publish or perish—Bane or boon of academic life?Journal ofManagement Inquiry, 14(4), 321–329. doi:

Several possible problems have been hypothesised, including: undue proliferation ofpublications and atomization of results (Gad-el-Hak2004; Statzner and Resh2010);impoverishment of research creativity, favouring ‘‘normal’’ science and predictable out-comes at the expense of pioneering, high-risk studies (De Rond and Miller2005); growingjournal rejection rates and bias against negative and non-significant results (because theyattract fewer readers and citations) (Statzner and Resh2010; Lortie1999); sensationalism,inflation and over-interpretation of results (Lortie1999; Atkin2002; Ioannidis2008b);increased prevalence of research bias and misconduct (Qiu2010). Indirect empiricalevidence supports at least some of these concerns. The per-capita paper output of scientistshas increased, whilst their career duration has decreased over the last 35 years in thephysical sciences (Fronczak et al.2007). Rejection rates of papers have increased in thehigh-tier journals (Larsen and von Ins2010; Lawrence2003). Negative sentences such as‘‘non-significant difference’’ have decreased in frequency in papers’ abstracts, while catchyexpressions such as ‘‘paradigm shift’’ have increased in the titles (Pautasso2010; Atkin2002). No study, however, has yet verified directly whether the scientific literature isenduring actual changes in conten

Formann, A. K. (2008). Estimating the proportion of studies missing for meta-analysis due to publicationbias.Contemporary Clinical Trials, 29(5), 732–739. doi

Fronczak, P., Fronczak, A., & Holyst, J. A. (2007). Analysis of scientific productivity using maximumentropy principle and fluctuation-dissipation theorem.Physical Review E, 75(2), 026103. doi:10.1103/PhysRevE.75.026103.

Atkin, P. A. (2002). A paradigm shift in the medical literature.British Medical Journal, 325(7378),1450–1451

Bonitz, M., & Scharnhorst, A. (2001). Competition in science and the Matthew core journals.Sciento-metrics, 51(1), 37–54

The results presented here suggest that competition among researchers haspronounced effects on the way science is done. It affects the progress of sciencethrough secrecy and sabotage and interferes with peer review and other universal-istic merit-review systems. It twists relationships within a field and can increase thelikelihood of a scientist engaging in misconduct. None of the focus-groupparticipants made reference to positive effects of competition on their work,despite the fact that the focus-group questions dealt in a general way with scientists’work and the norms of conduct that govern that work. If the protocol questions hadasked explicitly about competition, doubtless there would have been somediscussion about the positive aspects of science. In the context of the generalquestions, though, the scientists referred to competition as a constant and negativeforce that interferes with the way science is done. It is disconcerting to ponder theconsequences of competition, such as mistrust and defensive posturing, for acommunity that has long been committed—in principle—to shared ideas andcollegiality.

Competition is also manifested in scientists’ pressured haste, leading tocarelessness, which can verge on questionable behavior. One discussant talkedabout scientists ‘‘cutting a little corner’’ in order to get a paper out before others orto get a larger grant, and another said that she once published a result that she gotthree times in one week but could not replicate the following week

But there’s, I think there is a question of how you interpret the data, even ... ifthe experiments are very well designed. And, in terms of advice—not that I’mgoing to say that it’s shocking—but one of my mentors, whom I very muchrespect as a scientist—I think he’s extraordinarily good—advised me to alwaysput the most positive spin you can on your data. And if you try to present, like,present your data objectively, like in a job seminar, you’re guaranteed tonotgetthe job

You are. And you know what the problems are in doing the experiments. And ifyou, in your mind, think that there should be one more control—because youknow this stuff better than anybody else because you’re doing it, you know—you decided not to do that, not to bring up what the potential difficulties are, youhave a better chance of getting that paper published. But it’s—I don’t think it’sthe right thing to do.

dishonesty occursmore with the postdoc. Because they want to get the data—whereas if they don’tpublish, they don’t move on. And they, I think, are more likely to sort of fudge alittle bit here and there if they need to get the data done. Unless, like you say, youwatch them.

One mid-career scientist told a story of how he and others in his lab counteractedan abuse of power by his mentor, a senior scientist, while he was in training. Hismentor received a manuscript to review that was authored by a ‘‘quasi-competitor.’’It presented results of experiments similar to those that were going on in thementor’s lab. The scientist continued, ‘‘That paper ... basically would have beat usto the punch. They would have published these results before us, and they wouldhave gotten credit, and not us. And my mentor, God bless him, sat on the paper.’’The mentor not only delayed writing the review but asked someone working in thelab to write it (a move of questionable ethicality in itself). That lab person and ourrespondent decided, in response, to stall their own work, so that their lab would nothave an unfair advantage over the group who submitted the paper for review. In theend, the original group got credit for the findings, while the respondent’s lab wasalso able to publish their slightly different findings. He ended his story with,‘‘Sometimes you’re in an awkward position, and you try to do the best thing you canunder the circumstances, within your own internal ethical clock or whatever. Andsometimes it’s ugly and it’s imperfect, but it’s the only thing you can do. If we hadgone to the mentor and voiced this objection, our careers would have been over. Ifwe had approached the journal—God forbid, forget it.’’ The speaker qualified thisstory by saying that it made him sound much more ethical than he actually is.

The focus-group discussions showed, however, that scientists see peer review asaffording a unique, even protected opportunity for competitors to take advantage ofthem. In this sense, competition infects the peer review process, not only throughscientists’ competition with other applicants, but also through scientists’ distrust ofthe reviewers themselves, as competitors. The following exchange among mid-career discussants shows their sense of vulnerability

By contrast, others use it in thepublication process solely to maintain their competitive bid for priority in a line ofresearch inquiry, as a way to sabotage others’ progress. A scientist in an early-career group acknowledged the need to make results reproducible by telling people‘‘the whole recipe of the whole method’’ if asked directly, but then she talkedabout the ‘‘little trick’’ of not including all the details in a publication orpresentation. Like the scientists in a different group quoted above, she mentionedothers’ practice of taking photographs of poster presentations in order then topublish the results first. She said that people, in defense, ‘‘omit tiny little details’’:‘‘But sometimes in the publication, people, just to protect themselves, will not giveall the details. It’s always right, but maybe it’s not totally complete—to protectthemselves. Because your ideas get stolen constantly, and it’s so competitive ifyou’re a small lab.

A more deliberate form of not sharing is the omission of critical details inpresentations, papers and grant proposals so that others will have difficultyreplicating and extending one’s own research.

I know a large number of people in that category, in my own experience, who ...opted out because they didn’t want to play. They didn’t want to play the kind ofgames that have to be played to be successful, and in bringing in money and gettingthe papers out. There’s so much more than just doing good science that comes intoit. There’s so much communication and there’s salesmanship that has to go on

I really hate to admit this, but you do the same thing with your competitors asyou do with grant agencies. You sucker-punch them. You might have—when Isubmit a paper, I already have the next two or three papers’ worth of data. Imean, I know what I’m going to be publishing a year from now, mostly. But thepaper that comes out of my lab is Part A. Parts B and C are mostly on my desk.And I’ve put things in part A to basically entice my competitors into making anass out of themselves, or to second guess, or say, ‘‘Oh that must be wrongbecause of that, or something.’

You submit the first grant, youpropose the novel thing. You know damn well any study section that’s evenmildly conservative is going give you, ‘‘Well, it sounds promising.’’ Theymight give you a good score, you hope for a good score, but it’s not going toget funded, because it’s too novel, it’s too risky, it’s too blah blah. But youalready have the damn data. You know on the second resubmit, you’re goingto say, ‘‘Good point! We took that to heart. Oh, what a wonderful suggestion!We will worry about this too. Guess what? Here’s the data!’’ Shove it downtheir throat. And then it’s funded. Because, wow, you flagged them, yousucker-punched them. They said, ‘‘This is really novel, blah, blah. Boy if youcould only do that, that would be a great grant.’’ Well, you alreadydiddo it,and that’s the point. And you basically sucker-punch the study section intogiving you the money by default. They have to at that point. They don’t havea choice.

Competition for funding, publications, scientific priority and overall career successleads scientists to describe their work in terms of strategies one might use in a game.Focus-group participants revealed that, like it or not, working within the scientificcommunity requires artful maneuvering and strategizing.

To publish. And sometimes publish in the right journals.... In my discipline ...there’s just a few journals, and if you’re not in that journal, then yourpublication doesn’t really count

In addition to that, the other thing that they focus on is science as celebrity.... Sothe standards are, ‘‘How much did it cost, and is it in the news?’’ And if it didn’tcost much and if it is not in the news, but it got a lot of behind-the-scenes talkwithin your discipline, they don’t know that, nor do they care

You’ve got to have a billionpublications in my field. That is the bottom line. That’s the only thing that counts.You can fail to do everything else as long as you have lots and lots of papers

In short, there are many people (the oversupply factor) competing for prestigious,desirable and scarce rewards and resources (the funding factor), in a struggle thatbestows those rewards disproportionately on those of marginally greater achieve-ment (the tournament factor). This situation is supported to the detriment of that‘‘legion of the discontented’’ and to the benefit of senior investigators, because it‘‘generates good research by employing idealistic young graduate students andpostdoctoral fellows at low cost’’ [26]. In other words, the benefits accrue to fundingand employing institutions. This paper explores some of the costs that accompanythese benefits

Richard B. Freeman and colleagues [28] havecharacterized the problem as follows: ‘‘Research in the biosciences fits a tournamenteconomic structure. A tournament offers participants the chance of winning a bigprize—an independent research career, tenure, a named chair, scientific renown,awards—through competition.... It fosters intense competition by amplifying smalldifferences in productivity into large differences in recognition and reward. Well-structured tournaments stimulate competition. Because the differences in rewardsexceed the differences in output, there is a disproportionate incentive to ‘win’’’(p.2293). Research environments in which only small numbers of scientists have theopportunity to gain significant attention increase the competitive stakes: playing thegame may be a gamble, but the payoff for winning is significant [28,36]

Another perspective sees competition as a function not just of funding, but of thebalance between supply and demand of resources, particularly human resources. Inthe current competitive system, young scientists are pitted against one another forattractive career opportunities that are becoming increasingly scarce [28].Researchers, feeling the pressure to be first to present findings in their fields,employ armies of graduate students and postdoctoral fellows and strive to maketheir laboratory groups the smartest and the fastest. The result is a ‘‘postdocbottleneck’’ [29] where the supply for highly educated and trained researchers farexceeds the demand [30–33]. In concrete terms, Donald Kennedy and colleagues[34] have described the structural problem as a source of excess supply of humancapital: ‘‘We’ve arranged to produce more knowledge workers than we can employ,creating a labor-excess economy that keeps labor costs down and productivity high’’(p. 1105). The system produces, they claim, a ‘‘legion of the discontented’’ [34].They argue that institutional and policy decisions about training scientists should becoupled to placement histories of recent graduates, numbers of intellectual offspringof faculty, and job markets for scientists. Roger L. Geiger [35] has suggested thatthe imbalance between supply and demand is due in part to deficiencies in graduate

Analysts differ as to the reasons why competition has intensified. Some see thesituation in terms of money. Tempering the effects of competition is not a primeimpetus behind calls by the National Science Board [26] and by a recent coalition of140 college presidents and other leaders [27] for more federal funding for scientificresearch; however, some scientists see such advocacy movements in terms of easingcertain aspects of competition that are worsened by tight dollars. More money, morepositions, and overall expansion of the research enterprise would improve thesituation

here are indications, however, that the natureof competition has changed in recent years. Goodstein [25] argues that this shift islinked to negative outcomes:Throughout most of its history, science was constrained only by the limits ofits participants’ imagination and creativity. In the past few decades, however,that state of affairs has changed dramatically. Science is now held back mainlyby the number of research posts and the amount of research funds available.What had been a purely intellectual competition has become an intensestruggle for scarce resources. In the long run, this change, which is permanentand irreversible, will probably have an undesirable effect on ethical behavioramong scientists. Instances of scientific fraud will almost surely become morecommon, as will other forms of scientific misconduct (p. 31)

It is negatively correlated with subscription tonormative systems (either traditional or alternative) and sense of community

Melissa S.Anderson [20] furthermore found that a competitive departmental environment inscience is positively correlated with exposure to misconduct, fears of retaliation forwhistle-blowing, and conflic

Empirical findings show a strong, positive relationship between the level ofperceived competition in an academic department and the likelihood that depart-mental members will observe misconduct among their colleagues [19]