Christy M.Y. Siu
The Chinese University of Hong Kong
Shatin, N.T., Hong Kong
A consistent body of research conducted in North America indicates that trait intrinsic motivation facilitates creativity and academic performance, whereas extrinsic motivation hinders creativity but has no effect on academic performance. We examined the effects of trait intrinsic and extrinsic motivations in Hong Kong college students. In study 1, on a sample of 127 year-1 students, trait intrinsic motivation correlated negatively with year-1 self-reported GPA, whereas trait extrinsic motivation correlated positively. In study 2, on a sample of 38 students engaged in experimental conditions resembling those of an in-class creative writing task, trait intrinsic motivation correlated positively with creativity of the story, whereas trait extrinsic motivation did not. Findings suggest that the Hong Kong college environment (a) constrains and/or penalizes the expression of intrinsic motivation and, thus, creativity, and (b) facilitates and/or rewards the expression of extrinsic motivation and, thus, means-end opportunism. The negative consequences of this situation on students' development and quality of learning are discussed, and possible remedies are outlined.
motivation, extrinsic motivation, state, trait, reward, Hong Kong Advanced
Level Examination, GPA, algorithmic task, heuristic task, creativity
How do people who are high in intrinsic motivation differ from those who are low? They do not differ in selfishness, readiness to cooperate with others, or obedience. Yet, they behave in very different ways when they face a special type of situation. When given an interesting task or assignment without being promised reward or punishment upon completion of the task, persons high in intrinsic motivation continue working on the task, whereas persons low in intrinsic motivation stop working. Thus, intrinsic motivation is analogous to an emergency battery that provides energy to action in case of blackout. Furthermore, in absence of incentives, persons high in intrinsic motivation proactively explore the environment seeking interesting stimuli and opportunities for action, whereas persons low in intrinsic motivation come to a halt and remain passive until the environment provides them with incentives.
How do people who are high in extrinsic motivation differ from those who are low? People who are high in extrinsic motivation typically do not enjoy what they do while they are doing it and, thus, enjoyment does not energize their work. Yet, their minds look ahead and anticipate the rewards or punishments (the “carrot” or the “stick”) that will be received upon completion of the task or failure to complete the task, respectively. The anticipation of the consequences of one’s actions is the “fuel” used by the extrinsically motivated person. Persons high in extrinsic motivation tend to engage the shortest and easiest path to the end, and enjoy work only after its completion, when they can savor the reward or celebrate the avoidance of punishment. By contrast, persons low in extrinsic motivation tend to ignore incentives and are less likely to undertake actions in order to obtain rewards or avoid punishments.
As state variables, intrinsic and extrinsic motivations have been traditionally considered opposite to each other. In particular, the introduction of extrinsic incentives, such as money and praise, in interesting tasks has been systematically found to reduce intrinsic motivation (see review by Deci & Ryan, 1985). However, as trait variables, intrinsic and extrinsic motivations are reciprocally independent (Amabile, Hill, Hennessey, & Tighe, 1994); so that, some individuals are high in both, others low in both, and some high in one and low in the other one.
There is incomplete empirical evidence on how trait intrinsic and extrinsic motivations interact with each other in driving a person's behavior. Conceptually both forms of motivation should result in higher overall performance, be it in school or work. Thus, from the point of view of performance, the single most negative personal disposition is the absence of both intrinsic and extrinsic motivation, that is, the lack of motivation.
Our goal was to examine whether intrinsic and extrinsic motivations are as predictive of academic performance and creativity in Hong Kong as they are in North America. In study 1, we examined the relationship between motivations and academic performance by conducting an observational study of year-1 college students. In study 2, we examined the relationship between motivations and creativity among college students in experimental conditions resembling those of an in-class creative writing task.
In 1999/2000, 127 year-1 undergraduate students, 40 males and 87 females, at the Chinese University of Hong Kong were recruited through the General Psychology classes from a wide range of departments. Participants were first asked to report their attained grades in the Hong Kong Advanced Level Examination (HKALE). Then, they completed a kit of personality inventories. In the following months, participants were contacted by phone and asked to report the grades they attained in the first and second semester of their first year in college.
The kit of personality inventories contained, among others, a Chinese translation (So, 1999) of the Work Preference Inventory (WPI) (Amabile et al., 1994), measuring trait intrinsic and extrinsic motivations, and a Chinese translation (So, 1999; Moneta & Wong, in press) of selected scales from the Personality Research Form, Form E (Jackson, 1989), measuring other trait motivations including the need for achievement, the aspiration to accomplish difficult tasks and willingness to put forth effort to attain excellence (McClelland, 1985).
Based on an analysis of the construct validity of the Chinese translation of the WPI (Moneta, 2001; Moneta & Wong, 2001), we selected a subset of 18 items out of the 30 contained in the WPI, of which 9 measured trait intrinsic motivation (e.g., "I enjoy tackling problems that are completely new to me") and 9 trait extrinsic motivation (e.g., "I am strongly motivated by the recognition I can earn from other people") rated on a scale ranging from 1 ("Never or almost never") to 4 ("Always or almost always").
In the statistical analysis, we first estimated the correlation coefficients involving trait intrinsic and extrinsic motivations, on one hand, and year-1 GPA, on the other. Then, we performed a regression analysis wherein year-1 GPA is the dependent variable and trait intrinsic and extrinsic motivations, achievement motivation, and GPA in HKALE were the independent variables. This second analysis allowed us to estimate the linear relationships between the two key motivations and GPA controlling for the possible confounding of overall academic ability and achievement motivation.
Surprisingly, intrinsic motivation correlated negatively with year-1 self-reported GPA (r= -.24, p<.009), while extrinsic motivation correlated positively (r=.33, p<.001). These results indicate that the more intrinsically motivated a student is, the lower his or her academic performance, and the more extrinsically motivated a student is, the higher his or her performance. Thus, the students who attain the highest levels of academic performance are those who are simultaneously low in intrinsic motivation and high in extrinsic motivation, and the students who attain the lowest levels of academic performance are those who are simultaneously high in intrinsic motivation and low in extrinsic motivation.
Can these paradoxical results be an artifact of confounding by academic ability or achievement motivation? For example, could it be that the students who score higher in intrinsic motivation are also less academically skillful? If so, that confounding could explain why intrinsic motivation is associated with lower performance.
We investigated this possibility first by estimating simple correlations. GPA in HKALE was positively and significantly correlated with extrinsic motivation (r=.38, p<.001) and negatively but nonsignificantly with intrinsic motivation (r=-.12, p<.17). In turn, GPA in HKALE was positively and significantly correlated with year-1 GPA (r=.38, p<.001). This pattern of correlations reveals that there is some degree of confounding.
However, when we controlled statistically for GPA in HKALE and achievement motivation by regression analysis the results did not change: intrinsic motivation still was a significant, negative predictor of GPA, and extrinsic motivation still was a significant, positive predictor of GPA. In addition, GPA HKALE was also positively and significantly associated with GPA, whereas achievement motivation was positively but nonsignificantly associated with GPA.
This study provided findings that are diametrical to those obtained in North American colleges (Amabile et al., 1994). Whereas in North America trait intrinsic motivation is conducive to higher pre-enrollment academic ability scores and course grades, in our college environment intrinsic motivation is unrelated to pre-enrollment academic ability scores (perhaps negatively related) and is conducive to lower course grades. Furthermore, whereas in North America extrinsic motivation is unrelated to both pre-enrollment academic ability scores and course grades, in our college environment extrinsic motivation is conducive to higher pre-enrollment scores and course grades. These findings suggest that Hong Kong college environment penalizes self-motivation and rewards "carrot-stick" motivation.
Is this problem specific to college or was it inherited from secondary school? It is important to note that trait extrinsic motivation was also significantly correlated with the HKALE composite score of pre-enrollment academic ability. This finding indicates that trait extrinsic motivation is rewarded by the education system prior to entering college. The positive effect of trait extrinsic motivation on college GPA was estimated controlling statistically for the HKALE score. Thus, the positive effect of trait extrinsic motivation in college is independent, additional to the positive effect that trait extrinsic motivation has prior to entering college. In sum, college seems to reward extrinsic motivation to a greater extent than secondary school.
It is also important to note that trait intrinsic motivation was not significantly related to the HKALE score, while it predicted negatively college GPA. Thus, secondary school seems neither reward nor penalize trait intrinsic motivation, whereas college seems to penalize trait intrinsic motivation.
On the whole, the findings are surprising and counterintuitive. If they are indeed valid and replicable, they unequivocally indicate the presence of problems in our college environment. Yet, before engaging in speculative interpretations, we wanted to be more confident about the validity of our study. The crucial question was: Did we actually measure intrinsic and extrinsic motivations?
Although prior validation work indicated that the connotations of trait intrinsic and extrinsic motivations in Chinese are very similar to the connotations that these traits have in North America (Moneta, 2001; Moneta & Wong, 2001), still nothing was known about their functional significance. What is the main function of intrinsic motivation in North American students? That of facilitating creativity. Thus, we set out to investigate whether or not trait intrinsic motivation facilitates creativity also in Chinese students.
The participants of study 1 were a sub-sample of a larger pool of 292 undergraduate students from all levels of seniority that participated in the College Experience Study (CES) (Moneta, 2000). For all 292 students, we had measures of trait intrinsic and extrinsic motivations. In the fall 2000/2001, we conducted an experiment with a sub-sample of the CES of 38 students. In order to optimize statistical efficiency, we selected participants that were maximally different in the two motivations forming four groups: high/high, high/low/ low/high, and high/high.
Participants were invited individually to a lab, shown a 7' movie clip, and asked to write a story "as dramatic and creative as you can" based on it in 35'. Participants were asked to tell what has led up to the event shown in the movie clip, to describe what was happening at the moment, what the characters were feeling and thinking, and to give an outcome.
The movie clip was extracted from Tom & Viv (Samuelson, Samuelson, Kass, & Gilbert, 1996), and portrayed a marital discussion followed by a sequence of surprising and amusing events open to multiple interpretations. We selected this movie clip because of its non-academic, relational, and real life content. Thus, we presumed that traditional facets of academic ability, such as competence and intelligence, would not have markedly influenced the creativity of the story. Furthermore, we selected this movie clip because it leaves ample space for conjectures as to what caused the marital problem and how the life story of this couple may evolve and end. Thus, we presumed that this stimulus was interesting and ambiguous enough to allow the expression of intrinsic motivation and, thus, creativity.
Participants typed their stories on a word processor, and the entire process was recorded to allow a wide range of analyses fully reported in Siu (2001). In this paper, we focus only on the end product of the process, the creativity of the stories.
In assessing the creativity of the stories we adopted Amabile's (1996) consensual definition of creativity according to which a product is creative if independent experts concur in recognizing it as creative, and the task is heuristic rather than algorithmic. A task is algorithmic if it has a clearly identifiable goal and a straightforward path to a solution. A task is heuristic if it has neither a clearly identifiable goal nor a straightforward path to a solution. In a heuristic task, people have to figure out on their own the exact goal and the path to a solution. The nature of the task depends on the person: if the person is not informed of the existing goal and algorithmic solution, it is still considered a heuristic task.
In North American studies, experts' ratings typically exceed correlation coefficients of .8. In this study, however, we were concerned with breadth of creative expression more than with consistency between judges. We therefore used two as independent as possible judges: the first, non-Chinese author of this paper and a Chinese undergraduate student who took one class in creativity and studied independently two books on the topic (Amabile, 1996; Csikszentmihalyi, 1997).
Stories were evaluated blindly and independently by the teacher and the peer on the item "How individual originality compare with other students in the group" ranging from 1 ("Low") to 5 ("High"). As inter-rater agreement was high (r= .70, p<.001), we based the analysis on the average of the two ratings.
Trait intrinsic motivation correlated positively with creativity (r=.36, p<.026), while extrinsic motivation did not correlate. Thus, students who scored higher in intrinsic motivation also tended to produce more original stories.
This study reinforced our confidence in the validity of the measurement. As it is with North American students, also with Chinese students trait intrinsic motivation facilitates creativity. Thus, also in Chinese students trait intrinsic motivation is a positive feature, a potential that, if encouraged and let free to express, produces what is perhaps the utmost human ability.
The observational study provided findings that are diametrical to those obtained in North America, while the experimental study, wherein students tackled a creative task and were validly assessed, confirmed the expected relationship between intrinsic motivation and creativity. On the whole, this study suggests that the Hong Kong college environment discourages intrinsic motivation and creativity, and it encourages extrinsic motivation.
Before analyzing the practical implications of these findings, we emphasize two main limitations of this study. First, the observational study was conducted in 1999/2000. As it is for vintage, it could be that that year was a bad one for our college. We have extended the observational study to additional 88 year-1 students who entered college in 2000/2001. We are now in the process of collecting the GPA of their first year of study. We will soon be able to test whether or not the results reported in this paper are confirmed one year later.
Second, the observational study focused only on year-1 GPA. A key question is: Do intrinsic and extrinsic motivation keep having the same effects on academic performance in the following years? We keep monitoring all the participants and, in two years time, we will be able to answer the question.
While we wait for the evidence, however, we are compelled to take the findings seriously and raise several questions. The first question is: Why are these findings disturbing? We propose three answers. First, they are disturbing because they reveal a fundamental injustice in the system. Why should intrinsic motivation, an extremely positive human potential that promotes exploration and innovation, be penalized? Why should extrinsic motivation, an opportunistic tendency to exploit the environment by seizing all possible shortcuts to a personal end, be rewarded? It just does not look right.
The second answer concerns the effects that penalizing intrinsic motivation and rewarding extrinsic motivation have on the motivational development of our students. We lack scientific evidence as to (a) the extent to which trait intrinsic and extrinsic motivation change over long periods of time, and (b) whether grades received affect this change. It could be that students do not change their intrinsic and extrinsic orientations, and that the effect of receiving good or bad grades "washes out" by itself. Yet, there is a large body of evidence (see review by Deci & Ryan, 1985) indicating that state intrinsic motivation can be easily turned off by simple but very effective methods. For example, intrinsic motivation is reduced by rewards administered for completing a task no matter how, rather than for completing the task well. Thus, our findings are disturbing because they open the possibility that the current reward system may actually turn off intrinsic motivation and promote extrinsic motivation.
The third answer concerns the effects that penalizing intrinsic motivation and rewarding extrinsic motivation have on our students' quality of learning. Kahoe and McFarland (1975) found that trait intrinsic and extrinsic motivations interact with perceived difficulty of course in determining GPA: whereas intrinsic motivation predicts higher GPA in high challenge courses, extrinsic motivation predicts higher GPA in low challenge courses. Thus, it is possible that intrinsic motivation facilitates performance when facing complex learning tasks, whereas extrinsic motivation facilitates performance when facing simple learning tasks. Furthermore, temporarily induced states of extrinsic motivation result in poorer concept attainment (McCullers & Martin, 1971) and impaired problem solving (Glucksberg, 1962). Pooling these findings together suggests that learning is more efficient, deeper, and enjoyable when we are intrinsically motivated than when we are extrinsically motivated. Thus, our findings are disturbing because they open the possibility that the current reward system may actually encourage surface learning and discourage higher-order learning.
The second question is: Why does our college environment penalize intrinsic motivation and reward extrinsic motivation? What is the cause of the problem? Our findings do not give us specific indications. However, based on what is known about the link between motivations and performance in study and work contexts, we can identify four areas of potential problems: course interestingness-complexity, heuristic value of assignments, completeness-validity of assessments, and shortcuts-to-grades. First, if the curriculum and teaching methods fail arousing students' interest, then intrinsic motivation is a useless asset. Complexity is an important component of "interestingness". If the course contents are too easy, given students' ability, then learning becomes fundamentally uninteresting. If that is the case, extrinsic motivation is the only source of energy that students can draw and benefit from. Thus, it could be that, at least in the first year of college, courses tend not to be sufficiently challenging.
Second, if the assignments do not allow sufficient autonomy and exploration, and require mostly execution of step-by-step procedures, then the more intrinsically motivated students will not have opportunities to profit from their orientation. They may actually experience a certain frustration, redirect their intrinsic motivation toward extra-curricular activities and, thus, attain a comparatively lower academic performance. Thus, it could be that, at least in the first year of college, assignments tend to be too algorithmic and not sufficiently heuristic in Amabile's (1996) terminology.
Third, if the assessments do not validly recognize and reward students' creative output, then the practical value of intrinsic motivation is wasted. More dramatically, an excessive focus on "model answers" penalizes students' creative output because, by definition, any novel idea involves some degree of deviation from the most common answer or the ideal answer expected by the teacher. If this penalization actually happens, it contributes to the found negative relationship between intrinsic motivation and academic performance. Thus, it could be that, at least in the first year of college, course assessments tend to be biased toward memorization of factual content at the expense of creativity.
Last, if the course assignments and assessments allow shortcuts, the more extrinsically motivated students will not "waste" time trying to develop a deeper understanding, and will be more likely to engage the shortcuts and attain the end as efficiently and smoothly as possible. Thus, it could be that, at least in the first year of college, course assignments and assessments provide students with too many shortcuts to good grades through surface learning.
The third and last question is: What can we do to transform our college environment so that it no longer penalizes intrinsic motivation and rewards extrinsic motivation? What can we do to turn things around? We propose that the wisest way to approach this issue is first to verify empirically whether and to what extent the four areas of potential problems that we have highlighted (course interestingness-complexity, heuristic value of assignments, completeness-validity of assessments, and shortcuts-to-grades) are actually problematic. It can be that the scope of the problem is narrow, limited to one problem area only (e.g., there are too many shortcuts to good grades). If, however, the scope of the problem is broad, generalized to all four problem areas, there would indeed be need for a drastic change of attitudes in both educational institutions and students.
The Hong Kong education system of the 80s and early 90s was depicted as highly competitive, examination oriented, characterized by large classes, expository teaching, and excessive amounts of homework (Biggs, 1992; Salili, 1994). These and other features of the Hong Kong culture seem to have fostered students that are less creative than other Asian and Western students (Spinks, Yu-Ku, Shek, & Bacon-Shone, 1996). Due to the limitations of our data and the enormous difficulties in conducting valid cross-cultural comparisons of abilities, we can say nothing as to whether Hong Kong college students are less self-motivated and creative than North American students. What we can say from study 2 is that some students wrote amusing and imaginative stories, whereas others wrote boring and unimaginative stories. Thus, there was a variation in creativity as we would expect to find anywhere else in the world.
However, in study 1 we did find something that makes the Hong Kong college environment quite unique: whereas the North American environment rewards intrinsic motivation, the Hong Kong environment penalizes it. Thus, we can say that the Hong Kong college environment appears to have a problem. If indeed Hong Kong college students are less creative than their North American counterparts, then this cross-cultural difference could be caused at least in part by the problem affecting the Hong Kong college environment.
We point out that if maximization of self-motivation and creativity in the college population is a target, the shortcut of selection is not a valid tool to achieve the goal. It would be tempting to select students who are high in intrinsic motivation and exclude those who are low. This strategy, however, would be doomed to fail. This is because trait intrinsic and extrinsic motivations are independent of each other: we all possess both to some extent. The reward system does not operate on separate groups of individuals; rather, it operates on separate parts of the same individual, by promoting one tendency at the expense of the other. Thus, if the college environment does not change, and keeps penalizing intrinsic motivation and rewarding extrinsic motivation, there is a concrete risk that those students who were high in intrinsic motivation at enrollment time will be low in intrinsic motivation and high in extrinsic motivation by the end of their college years.
Whereas a business organization is relatively free of adopting no matter what reward structure for the sake of optimizing profit, educational institutions have the moral obligation of promoting a balanced, harmonious development of each student. The Hong Kong Education Reform states that "the priorities should be accorded to enabling our students to enjoy learning, enhancing their effectiveness in communication, and developing their creativity and sense of commitment" (Education Commission, 1999, p. 1). Based on what is known about the determinants of joyful learning, commitment, and creativity, these broad goals clearly suggest that Hong Kong colleges should protect and promote students' intrinsic motivation and, perhaps, discourage their extrinsic motivation to some extent.
Amabile, T. M. 1996. Creativity in context. Boulder, CO: Westview Press.
Amabile, T. M., Hill, K. G., Hennessey, B. A., Tighe, E. M. 1994. The work preference
inventory: Assessing intrinsic and extrinsic motivational orientations. Journal of Personality and Social Psychology, 66, 950-967.
Biggs, J. B. 1992. Learning and schooling in ethnic Chinese: An Asian solution to a Western
problem. Unpublished manuscript, University of Hong Kong.
Csikszentmihalyi, M. 1997. Creativity: Flow and the psychology of discovery and invention. New York: Harper Collins.
Deci, E. L., Ryan, R. M. 1985. Intrinsic Motivation and Self-determination in Human Behavior. New York: Plenum Press.
Education Commission. 1999. Education Commission's Revised Draft Proposal on the Aims of Education (As of 22.9.99) [On-line]. Available: http://www.e-c.edu.hk/eng/what.html.
Glucksberg, S. 1962. The influence of strength of drive on functional fixedness and perceptual recognition. Journal of Experimental Psychology, 63, 36-41.
Jackson, D. N. 1989. Personality Research Form Manual. 4th Edition. Port Huron, MI: Sigma Assessment Systems, Inc.
McClelland, D. C. 1985. Human Motivation. Glenview, IL: Scott, Foresman.
McCullers, J. C., & Martin, J. A. G. 1971. A reexamination of the role of incentive in children's discrimination learning. Child Development, 42, 827-837.
Moneta, G.B. 2000. The College Experience Study (CES): Study design and variables. Technical report, Department of Psychology, Chinese University of Hong Kong, Hong Kong.
Moneta, G. B. 2001. C-WPI version 1: Psychometric properties. Technical report, Department of Psychology, Chinese University of Hong Kong, Hong Kong.
Moneta, G. B.,Wong, F.H.Y. In press. The construct validity of the Chinese adaptation
of four thematic scales of the Personality Research Form. Social Behavior and Personality, 29.
Moneta, G.B., Wong, F.H.Y. 2001. Intrinsic/extrinsic motivations, self-construal, and the daily experience of engagement in Hong Kong college students. Abstract accepted for presentation at VIIth European Congress on Psychology, 1-6 July 2001, London, Symposium The Flow Experience Across Cultures.
Salili, F. 1994. Age, sex and cultural differences in the meaning and dimensions of
achievement. Personality and Social Psychology Bulletin, 2, 635-648.
Samuelson, P., Samuelson, M. , Kass, H. (Producers), Gilbert, B. (Director). 1996. Tom
& Viv [Film]. (Available from Buena Vista Home Video, United Kingdom).
Spinks, J. A., Yu-Ku, H. H. Y., Shek, D. T. L., Bacon-Shone, J.H. 1996. 'Recent developments in the area of creative thinking in Hong Kong.' Ed S. Dingli. Creative thinking: New perspectives. Malta: Malta University Press.
 Corresponding author. E-mail: firstname.lastname@example.org. URL: http://www.geocities.com/gbmoneta