Environmental influences on tertiary students' motivation to learn: Analyses of quantitative and qualitative data. Dr Jennifer Archer University of Newcastle Ms Jill Scevak University of Newcastle Paper presented at the annual meeting of the Australian Association for Research in Education, University of Newcastle, November 27 - December 1, 1994. Abstract Approximately 400 students were enrolled in a child and adolescent development subject. The subject was taught by a group of lecturers who t ook turns to present a massed lecture to all students and then conducted weekly two hour tutorial sessions with smaller groups of students. Mid way through the year long subject a series of changes was instituted in a n attempt to increase students' motivation to learn. These included res ubmission of the major assignment after feedback from lecturers, the right to work cooperatively, choice of assignment topics, and the provision of a booklet for writing good assignments. Students completed a questi onnaire almost half way through the year before the changes were introdu ced, and a second questionnaire at the end of the year. In addition, 54 of these students were interviewed about their reactions to the changes introduced to the subject. Students' perceptions of the motivational " climate" of the subject were delineated in terms of mastery and performa nce achievement goals. Comparing the questionnaires, students perceived a stronger mastery climate at the end of the year and a weaker performa nce climate, they rated their ability higher, and they indicated greater enjoyment of the tutorials and the massed lectures. Analysis of the in terview data provided further insight into students' reactions to the ch anges introduced in the subject. Introduction An emerging theory of motivation focuses on the achievement goal or goals that aÏperso n holds (Ames, 1992; Blumenfeld, 1992; Dweck, 1986; Nicholls, Patashnick , &ÏNolen, 1985). As Ames (1992, p. 261) describes it, a goal "defines a n integratedÏpattern of beliefs, attributions, and affect that produces t he intentions of behavior ...Ïrepresented by different ways of approachi ng, engaging in, and responding toÏachievement„type activities." The ad option of a goal, then, sets in motion aÏparticular way of interpreting a nd responding to the world. Two types of goal inÏparticular have been p roposed. Those who hold a performance (or ego„involved)Ïgoal are concer ned primarily with demonstrating their ability (or concealing a lack ofÏa bility) to others, and this is shown to best advantage by out„performing others,Ïparticularly if success is achieved with little effort. Those w ho hold a mastery (orÏlearning, or task„involved) goal want to develop t heir competence on a task orÏincrease their understanding of a topic, and are prepared to work hard to achieve their goal. Adoption of a mastery goal has been associated with the desire to gain in understanding of a topic (Ames & Archer, 1987), the choice of more diff icult ratherÏthan easy tasks (Ames & Archer, 1988; Elliott & Dweck, 1988 ), focusing onÏattributions to effort rather than attributions to abilit y (Ames & Archer, 1988;ÏNicholls et al., 1985), and reported use of more effective learning strategies (Ames &ÏArcher, 1988; Meece, Blumenfeld & Hoyle, 1988). In all, the adoption of a masteryÏgoal "encourages child ren to explore, initiate and pursue tasks that promoteÏintellectual grow th" (Dweck, 1986, p.1043). On the other hand, adoption of a performance goal has been associated with aÏtendency to avoid challenging tasks (El liott & Dweck, 1988), negative feelings suchÏas shame and embarrassment f ollowing poor performance (Elliott & Dweck, 1988),Ïand use of "surface" s trategies such as rote learning (Meece et al., 1988). ÏSusceptibility to these maladaptive attitudes and behaviours is most pronounced inÏstuden ts who feel they lack ability (Dweck, 1986). Because the focus of aÏper formance orientation is demonstrating competence to others, the fear of a ppearingÏincompetent (particularly pronounced in westernised countries) i mpels students to useÏbehaviours such as cheating or rote learning that m ight hide incompetence in theÏshort term but does little to help learnin g in the long term.ÌÌOrientation towards a goal has been demonstrated to be affected by individualÏdifferences or to be induced by situational c ues (Ames, 1992; Dweck, 1986). InÏlaboratory situations, signals or cue s to subjects to subjects encouraging the adoptionÏof one goal or the ot her can be presented clearly and unambiguously. In classrooms,Ïhowever, students may receive contradictory or confusing cues. For example, aÏt eacher may exhort all students to work hard but at the same time single o ut theÏmore able students by treating them in a particularly friendly an d informal manner. ÏIn addition, students within the same class may not a ttend to the same cues or mayÏreact differently to the same cues (Weinst ein, 1989).ÌÌResearch has shown that teachers are perceived by their stu dents to encourage theÏadoption of a mastery goal, a performance goal, o r both goals (though, as noted, notÏall students perceive similarly). S tudents' perceptions that the teacher is encouragingÏa mastery goal has b een linked with enjoyment (Ames & Archer, 1988; Archer,Ï1992, 1993), wil lingness to tackle difficult tasks (Ames & Archer, 1988), andÏreported u se of effective metacognitive strategies (Ames & Archer, 1988; Archer,Ï1 992, 1993; Nolen & Haladyna, 1990; Meece et al., 1988). The present stu dy wasÏan attempt to increase students' perception of a mastery climate o perating within aÏuniversity subject by making a series of changes to it s operation. Research methods Both quantitative and qualitative data were gathered for the present study. TheÏqualitative data, in addition to quantitative questionnaire data, added richer and moreÏdetailed information about the social and p sychological processes engendered by theÏchanges (Neuman, 1994). In rec ent years, the so„called paradigm wars have abated,Ïwith recognition on b oth sides that the complementarity of the two approachesÏ"serves better, fuller, and more satisfying understanding" (Salomon, 1991, p. 16). InÏh er review of achievement goal theory, Blumenfeld (1992) calls for resear ch thatÏcontains "thick description" in which qualitative methods (the s ystemic approach inÏSalomon's terms) supplement quantitative methods (th e analytic approach inÏSalomon's terms).ÌÌEducational researchers engage d in questionnaire studies tend to isolate variables andÏsearch for caus al relationships between a limited number of variables, and in soÏdoing a ttempt to reduce the influence of context. However, a classroom full of Ïstudents presents a cluster of interrelated variables that change over t ime. The wholeÏis more than the sum of its component variables (Salomon , 1991), with change in one exerting a pervasive effect on the cluster as a whole. The prese nt studyÏrepresents an attempt to benefit both from the careful isolatio n of variables affordedÏby the analytic approach and the authenticity an d emphasis on context afforded byÏthe systemic approach. Method and Instruments The subjects of the study were students enrolled in a Bachelor of Ed ucation course atÏthe University of Newcastle in a variety of specialisa tions (including early childhood,Ïprimary, and secondary). In their fir st year of the course, all students (approximatelyÏ400) take a year long subject in child and adolescent development, commonlyÏreferred to as Ed ucation 1. The subject is run by a group of lecturers (eight duringÏthe time of the present study) who take turns to present a massed lecture a ttended byÏall students. In addition to the massed lecture, there is a t wo hour tutorial held everyÏweek for smaller groups of students (approxi mately 25 in each group) conducted byÏone of the lecturers. Towards the end of the first semester 1993, students were asked to complete aÏquest ionnaire about their Education 1 subject. There were 354 students in th e firstÏsample, including 110 males and 244 females. The contents of th e questionnaire areÏdetailed below. After students completed the first q uestionnaire, a series of changesÏ(listed below) were introduced into th e subject, changes designed to increase theÏnumber of mastery goal cues to students. (1) The major assignment, unlike the minor assignment comp leted during the firstÏsemester, was to be submitted twice: the first s ubmission would result in a mark outÏof 10 and written feedback about ho w the assignment could be improved; the secondÏsubmission would result i n a mark out of 20, and therefore a total mark out of 30. ÏStudents also were required to submit their plan for their assignment. (2) Each stud ent was given a 33 page booklet (written by the present authors)Ïentitled "How to write an essay in TE115E." (3) Subjects were given the cho ice of working alone on the major assignment orÏworking with a partner, u nlike the minor assignment which had to be completedÏindividually. (4) S tudents were given a choice of topics for the major assignment, unlike t heÏminor assignment where no choice was given. Towards the end of the second semester, students were asked to respond to the sameÏquestio nnaire they completed at the end of the second semester. There were som eÏadditional questions concerning the changes to the subject that were i ntroduced. ÏThere were 319 students in the second sample, including 98 m ales and 221 females. ÏThere were 270 students who completed both questi onnaires. The great bulk ofÏstudents at both times of testing were aged between 18 and 20, with a tail of older students. Achievement goals This set of items was designed to assess students' perceptions of the mastery and performance dimensions of the tutorial sessions of Ed ucation 1. AÏmodified version of the scales developed by Ames and Arche r (1988) was used. ÏItems were prefaced with the heading "In this Educat ion 1 course" and students ratedÏeach item on a five point Likert scale r anging from "do not agree at all" (1), "agree"Ï(3), to "strongly agree." For the mastery scale, ten items were selected. TheÏcoefficient alpha for the scale was .79 for the first data set and .82 for the second. ÏF or the performance scale, eight items were selected. The coefficient al pha for theÏscale was .72 for the first data set and .66 for the second. The low alpha for theÏsecond data set point to a less than adequate sc ale and any statistics derived from itÏmust be accepted with caution. M ore details on the construction of the two scalesÏare available from the first author. The mastery and performance scales were notÏsignificantl y correlated in the first sample (r=.07), but correlated in the second sample (r=.14, p<.05). The items comprising the mastery and performa nce scales areÏshown in Appendix 1. Learning strategies Students' reported use of effective strategies in studying for Education 1 was assessed using 18 items. Strategies included plannin g activities,Ïelaboration strategies which focus on integrating new info rmation with what has beenÏlearned previously, and monitoring strategies which focus on checking understanding. ÏThe items were selected as stra tegies that are generic to the process of learning,Ïrather than specific to a particular discipline. The coefficient alpha for the scale wasÏ.8 4 for the first data set and .84 for the second. The items in the scale were prefacedÏwith "In this course" and students rated each item on fiv e point Likert scales fromÏ"not at all typical of me" (1) to "very typic al of me" (5). The following variables all were measured by single items using five point Likert scales, with 5 the positive end of the scale. Relevance Students were asked how relevant they found Education 1. Interest Students were asked how interesting they found Education 1. Enjoyment In two separate items, students were asked how enjoyable they found both the tutorial sessions and the massed lectures of Education 1. Perceived ability Students were asked to rate their ability in the subject compared with other students in their tutorial group. Re-submission In two separate items, students were asked how willing they would be to re- submit future assignments, and if they found re-submission h elpful in understanding the content of the assignment. Booklet In two separate items, students were asked how carefully they read the booklet, and how useful they found the booklet. Cooperation Students were asked if they thought the choice of working with a partner or working alone was a good idea. Choice of topics Students were asked if having a choice of topics for the assignment was a good idea. Interviews A sub-sample of 54 students were asked to participate in an audio-taped interview lasting approximately 40 minutes in which questions were as ked aboutÏtheir motivational orientation in subjects at high school and a t university, the sorts ofÏstudy strategies they employed to complete wo rk in these subjects, their attributionsÏfor success and failure, and th eir reactions to the changes instituted in the EducationÏ1 subject. The students were drawn from all specializations within the BEd degreeÏand r epresented a wide range of achievement levels. For the present study, o nly dataÏrelating to students' reactions to the four changes were analysed. Results This study was concerned with individual students' perceptions of the attitudes andÏbehaviour of their tutorial lecturer, and the relation b etween these perceptions andÏstudents' reported attitudes and behaviour. Because of this, all analyses wereÏconducted on the sample as a whole, not by tutorial group. Descriptive statistics Table 1 shows the means and standard deviations for variables that were measured at both times of testing. Table 2 shows the means andÏstandard deviations for subjects' reactions to the change s introduced between the firstÏand second times of testing. All the cha nges were well received, having a choice ofÏtopics and re„submitting the assignment receiving the highest ratings. Correlational analyses For both times of testing, students' scores on the mastery and performance goal scales were correlated with their reported use o f learningÏstrategies, interest level of the subject, relevance of the s ubject, enjoyment ofÏtutorials and massed lectures, and their perceived a bility. Results are shown in Table 3. As the correlations show, when the students perceived the lecture r to beÏencouraging a mastery goal, they reported greater use of effecti ve learning strategies,Ïmore interest in the subject, more relevance of t he subject, and more enjoyment ofÏthe tutorials and the massed lectures. The one significant correlation of perceptionÏof a performance goal wa s with reported use of strategies at the first time of testing. ÏAlso, t here were significant correlations between perceived ability and percept ion ofÏa mastery climate at both times of testing. These correlations w ith perceived abilityÏwill be discussed in the next section. Regression analyses It might be expected that students who perceive themselves among the best students in the tutorial group would be the ones to us e effectiveÏlearning strategies and to demonstrate a more enthusiastic a pproach to their workÏthan students who see themselves as less capable. T o demonstrate that students'Ïperceptions of a mastery climate predicted u se of learning strategies, interest,Ïrelevance, and enjoyment over and a bove that of perceived ability, a series ofÏhierarchical regressions was conducted, for both times of testing. Perceived abilityÏwas entered fi rst into the regression equation, followed by a performance goal, andÏla stly by a mastery goal. The results, produced in Table 4, show that, fo r the fourÏdependent variables, a mastery goal makes a significant indep endent contribution toÏthe total R squared, over and above the the contr ibution of perceived ability and aÏperformance goal. The independent co ntribution of perceived ability is mostÏpronounced for reported use of l earning strategies. Changes from Time 1 to Time 2 It had been anticipated that making changes to the structure of the Education 1 subject would increase students' per ception of aÏmastery climate and a more positive approach to learning: allowing re-submission ofÏthe major assignment following feedback should e mphasise the importance of effortÏin gaining success and the attitude th at mistakes can be corrected; providing a "how-to" booklet should emphasise that there are appropriate strategies that can be learnedÏfo r writing a good assignment; allowing students to work with a partner sh ouldÏencourage cooperation rather than competition among students; and p roviding aÏchoice of topics (of relatively equal level of difficulty) sh ould allow students a someÏsense of control of their learning and the ch ance to select a topic of personal interestÏto them. For these analyses , the sample was restricted to students who hadÏcompleted questionnaires at both times of testing (n=270). Dependent measures t-tests were conducted on students' perception of mastery and performance goals, theirÏreported use of learning strategies, interest in the subject, relevance of the subject,Ïenjoyment of the tutorials and the massed lectures, and perceived ability. SignificantÏresults emerged for a mastery goal ( t=4.40, p<.001), a performance goal (t=-2.06,Ïp<.05), enjoyment of tutorials (t=2.49, p<.05), enjoyment of massed lectures (t=4.82,Ïp<.001), and perceived ability (t=2.68, p<.01). The results for interest almostÏrea ched significance (p=.05). Interview data Tables 5, 6, 7, and 8 show students' reactions to the four changes introduced midway through Education 1. For re-submission of the majo r assignmentÏ(Table 5), the reactions were overwhelming positive, with c omments about theÏusefulness of receiving feedback, additional motivatio n, and help in understandingÏthe topic of the assignment. For the strat egies booklet (Table 6), the great majorityÏof students found it useful, particularly for referencing procedures and appropriateÏstructuring of a n academic essay. For working with a partner (Table 7), students'Ïreact ions were mixed, with more than half not taking up the option of working withÏa partner. A common reason for working alone was preferring to ta ke soleÏresponsibility for success or failure of the assignment, not wan ting to let down, or beÏlet down by, other students. Students who chose to work with a partner citedÏreasons such as reducing workload or gaini ng additional ideas to their own. ForÏchoice of topics (Table 8), almos t all students thought it was a good idea because itÏgave them the chanc e to choose something of interest to them, something they wouldÏbe keen t o work on. Discussion Analyses of quantitative data The results demonstrate that students' perceptions of a climate of a subject, as encouraged by its lecturing staff, can be d elineated in termsÏof a mastery goal orientation using an internally rel iable scale. However, it wasÏmore difficult to delineate a performance g oal climate. This will be discussed inÏmore detail later. Students' per ception of a mastery climate was linked to reportedÏuse of effective lea rning strategies, interest, relevance, and enjoyment of tutorials andÏma ssed lectures. As such, the criterion validity of the mastery goal scal e has beenÏstrengthened by significant relationships with variables that point to an adaptiveÏapproach to learning. The link between a mastery climate and reported use of effective learning strategiesÏis of particula r interest because students may choose not to use strategies with whichÏt hey are familiar. This may reflect an attempt to protect a sense of sel f„worth. ForÏexample, writing an assignment at the last minute so that t here is no time to reviewÏthe work or make revisions means that a good m ark can be attributed to high ability,Ïwhile a poor one can be attribute d to lack of effort. However, if the student's goal isÏto understand a t opic (a mastery goal), then it makes sense to use strategies such asÏrev iew and revision. There is no sense of shame attached to working hard. The regression analyses demonstrated that a mastery goal orientation less ens theÏimpact on students of their perceived level of ability. Seeing o neself as among theÏtop students in a tutorial group was not the most im portant reason for finding theÏsubject enjoyable, interesting, and relev ant. For use of learning strategies, thoughÏperceived ability was an im portant predictor, a mastery goal also exerted aÏsignificant and indepen dent influence. This study and others (see Nicholls et al.,Ï1989) sugge st that a mastery goal orientation lessens the impact of perceived abili tyÏon attitudes and behaviour. The inability to form a robust performance goal scale may reflect the lack of aÏstrong competitive focus within E ducation 1 (a similar problem occurred with theÏprevious cohort describe d in Archer, 1993). Even though grades are awarded at theÏend of the ye ar (high distinction, distinction, credit, pass, fail), lecturers tend t o makeÏlittle reference to grading practices during the year. Also, unl ike some universityÏsubjects where more students are admitted into first year than can be accommodatedÏin subsequent years, it is poor performan ce alone that leads to failure. LittleÏinformation about performance re lative to others is available to students. It may beÏthen that a robust performance scale did not emerge because lecturers involved in theÏsubj ect provided few cues to students to adopt a performance goal. Indeed, n one ofÏthe items comprising the performance scale refers directly to the lecturer thoughÏthere were five items in the questionnaire that did (eg , "the lecturer gives moreÏattention to students who do better on tests a nd assignments than other students"). ÏThe performance goal scale perhap s should be seen as indicating a personalÏorientation rather than an ass essment of a lecturer's orientation. Students' perception of a mastery climate did increase significantly from the firstÏtime of testing to the s econd, while there was a less marked decrease in students'Ïperception of a performance climate. It is argued that the increase in perception of aÏmastery climate was the result of the changes introduced between the f irst andÏsecond time of testing. There also was an increase in the enjo yment of the subjectÏand this could be attributed to an increased master y orientation. The increase inÏstudents' perceived ability is more diff icult to explain. It may have been the result ofÏgreater confidence eng endered by requiring re„submission of the assignment and aÏsubsequent in crease in the mark for the assignment. Apart from enjoyment and ability , there were no significant changes in variables Ïsuch as use of learnin g strategies that were expected to increase. Perhaps the changes made in the subject were too isolated. Ames (1992) argues th at forÏmotivation of students to increase, change must be effected simul taneously to threeÏaspects of the learning environment: the nature of t asks; the way tasks are evaluatedÏand students recognised for good work; and the amount of autonomy accordedÏstudents. With change in some of t hese aspects but not in others the motivationalÏoutcomes may be confused . Analyses of qualitative data Analyses of interviewed students' reactions to the four changes produced some interesting and at times unexpected results that provided aÏricher picture of the effect of the changes. Students' reac tions did not alwaysÏprovide support for those predicted by achievement g oal theory, and at times theyÏprovided additional insights. Also, with f our changes introduced into Education 1,Ïanalysis of interview data sugg ested that some changes engendered more motivationÏto learn in students t han others. The re-submission of the major assignment highlighted two different notions ofÏassessment: assessment as ranking versus asse ssment as informative feedback. ÏMany of the students mentioned the usef ulness of receiving feedback about theirÏwork - its strengths and its weaknesses - and therefore where future effort should beÏexpended. Not on ly was fairly detailed feedback provided, they had to pay attentionÏto i t to prepare their assignment for re-submission. For quite a few of the students,Ïthis presented a contrast to their typical approach to assess ment which was to findÏthe mark they received, glance at the comments (i f there were any), and then "throwÏaway" the assignment. That is, assessment was used only to get a sense of theirÏranking within the class - one of the top students, about in the middle, or down atÏthe bottom. One student remarked: "The essays shouldn't be about the final mark, it sh ould beÏabout what you really know." Another said: "I hate it when you do an exam, andÏyou hand it in, you get a mark back, and they don't eve n give you your exam. TheyÏsay it was fair enough, but it's confidentia l. But you don't know which questionsÏyou got right and which questions you got wrong, so you can't improve upon them. ÏYou just keep making th e same mistakes over and over again. This way, when youÏget the feedbac k, at least you know what you're doing wrong. So you can pick upÏyour a ct a little bit." Another interesting aspect which emerged about the re-submission of the majorÏassignment was the sense of guidance, of knowin g what was expected, that detailedÏcomments provided. For students who w ere nervous and unsure of their ability inÏthe subject matter, this reli eved some of their tension and suggested ways in whichÏweaknesses could b e overcome. This brings to mind the work of cognitiveÏpsychologists suc h as Resnick (1989) who argue that learners benefit from theÏguidance of experts who provide support, or scaffolding, which gradually can beÏrem oved as the learners grow in competence and confidence. In keeping with theÏbuilding analogy is the term cognitive apprenticeship, where the ap prentice practisesÏunder the eye of an expert. One student said: "If y ou get to re„submit it, you sort ofÏlearn how to rectify that problem, s o the next time, you can look and say, well sheÏtold me that I did this w rong, and that's how I rectified it, so I'll do that again." A simila r emphasis on guidance and the comfort of having clear expectationsÏemer ged in students' comments about the strategies booklet: they knew what w asÏexpected of them. Without guidance in a relatively new field, many s tudents worryÏabout their ability to cope. One student described use of the booklet this way: "YouÏcould look at it and think, well yeah that' s what I have to do. And then you can putÏyour work in and say, well it doesn't quite look the same. Why not? And you canÏrefer back and have something to compare it to. It's all very well to say to a child,Ïlook , you put a full stop after the last word in a sentence, and you put a c apital atÏthe beginning. But if they can't visualise it, then it's just like us. For me particularly, if I've got something there I can have a look at, I fee l more confident." ÏHowever, not all students felt this way. Some said t hey paid little attention to theÏbooklet because they felt confident tha t they knew already how to write academicÏassignments. Students' overw helming response to having a choice of topics for the majorÏassignment w as that they could select a topic of interest to them. In most instance s,Ïtoo, students drew connections between interest in a topic, willingne ss to expendÏeffort, and to search for meaning. For example: "If you c an choose somethingÏyou're interested in, you're more likely to read thr ough it, find out more information. ÏI mean, you're expanding your knowl edge. Whereas for something you really don'tÏwant to do, then you'll ju st do what you have to and not go any further. And youÏwon't learn anyt hing." Also of interest was the finding that few students selectedÏwhat they saw as the easiest topic to do (there was no deliberate attempt on the partÏof lecturers to provide topics varying in difficulty). In fac t, a number deliberatelyÏchoose what they perceived to be the hardest or the topic about which they had theÏleast amount of prior knowledge. It had been anticipated that providing choice would give students a greate r sense ofÏpersonal control over their work. However, there were few co mments to supportÏthis. In retrospect, this anticipation may have been u nrealistic. The majorÏassignment was the only aspect of Education 1 whe re students could make a choice.ÏAll other aspects of assessment were fi xed, and students were not asked to assessÏtheir own work or the work of others. We had considered allowing students toÏassess their own assign ments, but decided that with their lack of expertise inÏdevelopmental ps ychology (this was a first year subject) this would be too difficultÏfor them to do. Students' comments about producing the major assignment wi th the help of a partnerÏpointed to differing perspectives on collaborat ive work. In fact, more than half theÏstudents interviewed worked alone . For some of these, this was a practical solutionÏbecause they lived t oo far away from possible partners. But many did not want toÏwork with a partner. There was a moral overtone to their reasons (for a discussion Ïof the moral aspect to cooperative groupings see Ames, 1984): they did n't want toÏlet a partner down, they didn't want to be let down by a par tner, they preferred toÏtake sole responsibility for the success or failure of the assignment. Of the students who did work with a partner, a n umber said they did this to cutÏdown on the work involved. On the face o f it, this may be seen as a maladaptiveÏstrategy to reduce the complexit y of a task. In a another sense though it is anÏadaptive strategy. Mas tery„oriented students, with other tasks to be done, might useÏthis to m aximise limited time. Other students said they worked collaboratively s oÏthey could benefit from the ideas of other students. For example: "s o you canÏincorporate both sorts of perspectives and end up with a bette r product", and "IÏthought it would be interesting to see how she goes a bout it...we bounced off eachÏother with ideas." It also was interestin g to note that some parts of the collaborativeÏexercise worked better th an others. Joining forces to gather information in theÏlibrary and to d iscuss the broad outline of the assignment tended to be positiveÏtimes. W riting collaboratively often did not work well, with some students admit tingÏto coming close to blows. Another somewhat negative response to co llaboration emerged in the interviews. ÏSome students admitted openly (o r in a more indirect fashion) that they didn't workÏwith a partner becau se they were rejected by possible partners or that they felt theyÏwere p oor students who would let down the team. For example: "I didn't want t oÏdrag the other person down", "there's five of us that sit together in o ur EducationÏclass. And the others picked the other person for convenie nce more than anything. ÏAnd then I had a major panic attack about the a ssignment. I suppose I choose not to work with someone else because I wasn't competent in my own work, and I didn'tÏwant to disable anyone else by being like that", "there's are a lot of people that areÏgood friends with someone, but I'm not like that , so there wasn't anyone I felt IÏshould ask." Ames (1981, 1984) has in vestigated experimentally students' reactionsÏto success and failure in c ooperative structures, demonstrating the strong negativeÏemotions that c ome with group failure. The more competent members of the groupÏare loo king for someone to blame, while the less competent members feel remorse Ïfor letting down the team. It seems obvious that these students had be en involved inÏgroup work in the past and were aware of its ability to p roduce these emotions. ÏThere also is the public humiliation of having a n offer of partnership rejected. Conclusion The present study has demonstrated that the way lecturers approach their teaching - the attitudes and behaviour they display - is related to students'Ïmotivation to learn. Further, making changes (in line wit h achievement goal theory)Ïto the structure of a university subject can i ncrease students' perception of a masteryÏclimate with its emphasis on e njoyment and understanding operating within theÏsubject. In general, the interviews with a sub-sample of the students involved in the studyÏpro vide support for the questionnaire data, that the changes introduced did enhanceÏstudents' motivation to learn. This was particularly clear for three of the fourÏchanges: re„submission of the assignment, provision o f a how„to strategies booklet,Ïand a choice of topics for the assignment . The fourth change, the choice of workingÏalone or with a partner on t he assignment, produced more equivocal results, not onlyÏin terms of the negative emotions that come with group failure or the humiliation of being rejected as a partner, but also with the finding that some tasks may be more amenable to a group approach than others. With cooperative groupin g of students aÏwidespread educational practice these days, it might be w ell to bear in mind thatÏthere is a "down side." The motivational effec ts of enforced collaboration or evenÏencouragement to collaborate requir e more thorough investigation. References Ames, C. (1981). Competitive versus cooperative reward structures: The influenceÏof individual and group performan ce factors on achievement attributions and affect. American Education al Research Journal, 18, 273-278. Ames, C. (1984). Competitive, cooperative, and individualistic goal structures: AÏcognitive-motivational analysis. I n R. Ames and C. Ames (Eds.), Research on motivation in education, Volume 1 (pp.177-208). Orlando, FL: Academic Press. Ames, C. (1992) Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology, 261-271. Ames, C., & Archer, J. (1987). Mothers' beliefs about the role of ability and effortÏin school learning. Journal of Educational Psychology, 79, 409- 414. Ames, C., & Archer, J. (1988). Achievement goals in the classroom: StudentÏlearning strategies and motivation processes. Journal of Education Psychology, 80, 260-267. Archer, J. (1992). Students' perceptions of classroom climate: Implications for motivation. Paper presented at the annual meeting of AARE, Deakin University, Victoria, November. Archer, J. (1993). Students' perceptions of a mastery climate operating within a university course: Motivational Implications. Paper presented at the annual meeting of AARE, Fremantle, Western Australia. Archer, J. (1994). Achievement goals as a measure of motivation in university students. Contemporary Educational Psychology, 19, 430-446. Blumenfeld, P.C. (1992). Classroom learning and motivation: Clarifying and expanding goal theory. Journal of Educational Psychology, 84, 271-281. Dweck, C.S. (1986). Motivational processes affecting learning. American Psychologist, 41, 1040-1048. Elliott, E.S., & Dweck, C.S. (1988). Goals: An approach to motivation and achievement. Journal of Personality and Social Psychology, 54, 5-12. Meece, J.L., Blumenfeld, P.C., & Hoyle, R.K. (1988). Students' goal orientationsÏand cognitive engagement in classroom activities. Journal of Educational Psychology, 88, 514-523. Neuman, L. (1994). Social research methods (second edition). Boston: Allyn & Bacon. Nicholls, J.G., Patashnick, M., & Nolen, S.B. (1985). Adolescents' theories ofÏeducation. Journal of Educational Psychology, 77, 683-692. Nicholls, J.G., Cheung, P.C., Lauer, J., & Patashnick, M. (1989). Individual differences in academic motivation: Perceived ability , goals, beliefs, and values. Learning and Individual Differences, 1 63-84. Nolen, S.B., & Haladyna, T.M. (1990). Personal and environmental influences on students' beliefs about effective study strategies. Contemporary Educational Psychology, 15, 116-130. Resnick, L.B. (1989). Introduction. In L.B. Resnick (Ed.), Knowing, learning, and instruction (pp.1-24). Hillsdale, NJ: Erlbaum. Salomon, G. (1991). Transcending the qualitative-quantitative debate: The analytic and systemic approaches to educational research. Educational Researcher, 20, 10-18. Weinstein, R.S. (1989). Perceived classroom processes and student motivation. In C. Ames & R. Ames (Eds.), Research on motivation in education, Volume 3 (pp. 187-221). New York: Academic Press. Appendix 1 Items comprising the Mastery goal scale: The lecturer wants us to en joy learningÏabout educational psychology/ The lecturer gives us interes ting work to do/ TheÏlecturer makes sure we understand our work/ The lec turer wants us to learn how toÏsolve problems on our own/ The lecturer p ays attention to whether I'm improving inÏmy work/ The lecturer encourag es trying even though we make mistakes/ StudentsÏare given a chance to c orrect their mistakes/ The lecturer stresses that hard work isÏthe key t o success in this course/ We work hard because the lecturer wants us toÏk now more about child psychology/ The lecturer wants us to try new and di fficultÏtasks. Items comprising the Performance goal scale: Students compete against each other toÏget high marks/ Students feel bad when they d o not do as well as others/ DoingÏbetter than others is important to me/ Only a few students can get top marks/ÏStudents don't care about the ma rks others students get (reversed)/ Students competeÏto see who can do t he best work/ Students feel embarrassed to make mistakes inÏclass/ Stude nts want to know how others score on assignments and tasks.