Motivating students
What is the effect of the BSA system on students?
By Sam Verbraak 10012540
1. Introduction
Motivating students to work harder has been something researched many times before (Angrist 2009, Leuven 2010, 2011). Why? Because students tend to take longer than the time prescribed to complete their study. In this, the economics track at the University of Amsterdam (UvA) is no exception. A little under seven and a half thousand students joined the economics and business course of the UvA in the period between 1997 and 2010. At the end of 2011 only 35% got their bachelor degree and only 53% had passed all classes of their first year. This goes hand in hand with very high dropout rates; almost 50% of the students quitted the study without a bachelor degree, and of these dropouts over 60% drops out in the first year. So what did the UvA, or more specifically the FEB (Faculty of Economics and Business), do in order to increase the completion rate? They introduced a selection mechanism, called the BSA (Bindend Studie Advies or Obligatory Study Advice) in the first year of the economics and business course. In this paper I want to explore the effect this BSA has had on the student behaviour at the FEB1.
The reasoning behind the implementation of the BSA was that first year students who collect more ECTS (European Credit Transfer System) in their first year, continue to do so in the following years (Booij, 2008). So in order to reduce the number of students who take too long to complete the study, the FEB introduced the BSA as a threshold. The BSA is an amount of ECTS a student must collect in order to be allowed to enrol for the second year. If they do not reach this threshold they are expelled from the study and are not allowed to start it again for three years. While the nominal amount of credits to be obtained in the first year is 60 ECTS, the BSA was set a lower amount. It was first introduced with the threshold set at 30 points in 2002, later this was raised to 35 in 2006 and to 45 in 2009. The hypothesis was that with the introduction and each increase of the BSA, the students who succeeded to reach the threshold would have a higher bachelor completion rate within four years.
But does the BSA influence the study behaviour of the students? Since actual study behaviour is hard to measure, I use ECTS collected to assess the effectiveness of the BSA system. The effect of the BSA is measured on four different outcomes. First I look at the effect of the BSA on students in their first year; do they collect more ECTS? Second comes the effect on the completion rate of the bachelor within four years; does this increase? Third is the effect on dropout rates; do students drop out less after they have passed the BSA in their first year? Lastly I look at the negative effect the BSA might have on student behaviour, now that students have to pass a threshold they might be less able to focus on the courses specifically. Thereby neglecting their grades because in order to reach the BSA threshold they only focus on passing their courses, and not on how well they know the material. Passing a course with the minimum grade still awards the same amount of credits as passing it with a high grade; so do the grades drop after the implementation of the BSA?
1 This research will be most applicable for the FEB, because the motivation to choose a certain study is influenced by many factors: Actual interest for the field of the study, entry requirements, perceived easiness and job perspectives of the study to name a few. Since it is the motivation one has for the study that is the main focus of this paper, these results may not be applicable to other studies with a similar problem (students acquire too little ECTS).
The effect of the BSA seems to be ambiguous. While there is both a negative and a positive effect of the BSA on the first year ECTS collection, the effect of the BSA seems strictly
positive on bachelor completion rate and reducing the dropout rates, but strictly negative on the grades obtained. These results suggest that the BSA in itself is just a guideline and a selection mechanism and it is not a tool that can be used to motivate students to work harder and to collect more ECTS.
This paper is further organised as follows: The second part starts with a description of what motivation is, then continuous with relevant literature about incentivising students and ends with the general effectiveness of threshold performance, the third part presents a more detailed description of the results and is followed by the conclusion.
2. Literature review
In this paper I explore the effect of the BSA system introduced by the FEB on their students. The way the BSA is supposed to change the students their behaviour is one of motivation. In this paragraph a description of motivation and it’s link to the BSA are first addressed, after which I continue with research done in the direction of rewarding students for good results and why this is not interesting for the FEB and lastly I go more in depth about why punishing them could do the trick.
The problems the FEB wishes to address is that students take too long in order to complete their bachelor and very high dropout rates. The issue could be that there is a lack of
motivation for these students to study in the regular term. While some students are able to find the motivation to complete all, or at least most of their courses, there is a large group that does not complete their courses. This suggests that while it is possible to complete all of the courses, some students do not find the motivation to do so, leaving the ability to do so aside. Motivation itself can be roughly distinguished into two categories, intrinsic and extrinsic motivation (Ryan, 2000). Intrinsic motivation refers to motivation that comes from one’s own interests/desires; a student studies economics because he or she is inherently interested to know how things work within economics. With extrinsic motivation someone does a task not for the task itself but for the possibilities completion provides, for instance that a bachelor degree is needed to be able to apply for a specific desired job.
The BSA system of the FEB is not only a criterion to select the truly motivated students but also a punishment for not passing courses. This can be translated into an extrinsic motivation system in which, in order to be able to avoid punishment, a student will have to work harder and thereby pass more courses. So not only does the BSA select the intrinsically motivated students, it could also motivate some of the less motivated students to start working harder. If this is the case this system will incentivise some students to perform better than they would have done otherwise.
Strategies to influence behaviour extrinsically are either rewarding good behaviour or punishing bad behaviour. While the BSA system is implemented as a punishment, there is also research done in the direction of a system with rewards. This mostly focusses on paying students if they get good grades, thereby incentivising them to work hard. An example of this is given by Angrist, Lang and Oreopoulos (2009) in Canada. They executed an experiment
where students could get a monetary reward if they performed above a threshold GPA (grade point average). Second to this test group, the researchers also had a group which could apply for support services (extra aide by upper-class students or help with coping to study in general) and a third group which could apply for both support and a monetary reward. Compared to the control group, GPA’s did not rise for male students in any test group, but it did for females, with the greatest effect in the group with both aid and a monetary reward (Angrist, 2009). The effects however where small, thereby suggesting that students hardly react to rewards. Secondly, since relatively more males apply to the study economics at the UvA this research suggests that rewarding them would not trigger the preferred behaviour.
Two researches conducted at the UvA itself find similar results to the one of Angrist et all.. In the academic year 2001-2002, Leuven, Oosterbeek and van der Klaauw ran an experiment in which students would receive a monetary reward if they completed all their first year
courses. Their main result is that the effect of the monetary reward is small and insignificant (Leuven, 2010). While Angrist et all. then differentiate on gender to see if there is a
difference, Leuven et all. look at the difference between low and high ability students. With this distinction they find that high ability student do react positively, while low ability students react negatively (Leuven, 2010). So on the high ability side these results could prove interesting, but in order to also motivate the low ability student this is not and effective system. Then, in the period of 2004-2006, Leuven et all try a different reward system, here students had to select themselves in a tournament in which the student with the highest grade would win a price (Leuven 2011). This time not for all their courses but a single course, which was mandatory in the first year. The stakes in this experiment were high; students could win up to 5000 euro, to create maximal motivational effect. The effects were not observed though. Even with these high rewards students in the test group did not perform significantly better than the control group (Leuven 2011).
So let’s have a look at punishing bad behaviour. There hasn’t been done much research in this direction, because of ethical reasons, punishing students is tricky. Physical and monetary punishments may prove too drastic in altering the students their behaviour, which could lead to students no longer registering at the FEB at all. The new system by the FEB however does punish students for their bad behaviour, but within the boundaries of ethical human behaviour. This because the punishment introduced by the FEB is based on threshold performance. Students have to reach this threshold in order to avoid the punishment of being expelled. The general effectiveness of threshold performance on motivating people is mostly discussed within a business setting. The effects of threshold performance here are mixed, some claim that goal setting is effective in improving performance within a firm and some contradict this (both goal setting and threshold performance refer to the same optimal state required to succeed). Locke and Latham claim in their article that:
“So long as a person is committed to the goal, has the requisite ability to attain it, and does not have conflicting goals, there is a positive, linear relationship between goal difficulty and task performance.” (Locke 2006).
Thereby suggesting that goal setting within a firm does indeed improve work performance. Opponents of goal setting theory claim that it could cause more problems than it solves. In the paper by Ordóñez et al. (2009) goal setting theory is questioned in its effectiveness and they present several flaws that it could have. Most of the flaws centre on the point that people lose sight of the bigger picture when they need to reach a goal. If a goal is too specific or too narrow employees could focus only on the goal and not on what is good for the
company. Secondly if a goal is too challenging it may cause people not to try and reach it at all or to make them take excessive risk in order to reach it.
Both the benefits and the flaws of goal setting are less present in the student situation. The described benefits are less useful because the BSA is an all or nothing situation. If a worker of a firm does not reach the goal set by the firm but he does produce more then he would have done if there was no goal, the firm still benefits. If a student however gets more ECTS but not enough to pass the BSA he is still expelled from the study. The same holds for the flaws stated by Ordóñez et al.. These problems could cause serious harm to a firm but are of less a concern in the student situation. First of all the goal stated by the FEB might be too challenging, but this is not a problem. The FEB wishes to reduce the number of students who take too long to complete the study. If students cannot reach the threshold set by the FEB in their first year, this signals that it is most likely that the rest of the study will also prove difficult (Booij, 2008). Therefore the FEB wishes to expel these students in their first year instead of later on. Then about the taking of excessive risk, the only way in which students could take extra risk is by cheating on their exams. But to control for this there is data needed on the amount of cheating done at exams. Since this is not yet available, new research should be conducted in order to acquire these and to see if it significantly changes after the introduction of the BSA. Next to the cheating of students, it could also be the case that teachers become more excusing or permitting. This by giving students a higher grade then they would have done without a BSA, but do now, in order to allow a student to pass to the second year. Another reason for the teachers to be more lenient could be that dropout rates are used in the ranking of departments and teachers want to favour their standings. But no data is available on this either. Lastly, it could be the case that the BSA causes students to focus only on reaching the necessary amount of credits no matter the cost, where the cost in this situation are grades. This translates to the hypothesis that students do try to reach the threshold, but at the expense of lower grades. This would be a serious negative side-effect of the BSA, therefore this hypothesis is tested for in this paper.
Table 1: Differences between the BSA systems
Note; These are the differences between the BSA systems in which the old values are subtracted from the new.
-- significance level
Three stars *** significant at the 1% level No stars not significant
3. Results
As stated earlier I will look at the amounts of ECTS collected to see the effect of the BSA system. A consequence that could arise with the introduction of the BSA system is that different people/students apply for the study. People who would have chosen to study economics without a BSA but now choose not to do so, or vice versa. This would bias the sample and no longer allow for comparisons between different years. To test for this I look at three characteristics; age and gender for general comparison and the average grade on high school level (vwogpa) to control for skill level. It turns out that the high school GPA and the age of the students are not normally distributed, so instead of a t-test I conducted a Wilcoxon rank sum test. In the case of the high school GPA the student population does not significantly changes, the age however does change. Between the years of ’02-’05 and ’06-’08 the age of the students drop significantly. This states that comparisons between these two groups are biased, since age is an explanatory factor for all the sub questions. Although this is unfortunately the remaining groups still provide enough room to draw conclusions from. The t-test for the gender of the student population shows that the proportion of males and females doesn’t significantly change.
Next to the necessity of the student population to remain unchanged, it is also of
importance that the surroundings remains the same. For the sake of the research I assume that this is the case. However, around the year 2008 two major events may have an impact on the students their behaviour/motivation. First there is the economic crisis starting in 2008. This global meltdown of the economic system and the rise in costs and unemployment could have been reasons for the students starting after 2008 to start working harder.
Secondly the Dutch government was at that time looking over the possibilities to implement a fine for students who were studying for more than four years. These “lang studeerders” (students who needed more than four years to complete their bachelor) could be fined for 3000 euro, and this will also have created a strong incentive for the students to work harder. These are factors that are not controlled for within this research. Another research would be needed to see what the precise effect of government policy or of the business cycle is on student behaviour.
The data that I therefore use are all the first year students of both the economics and business track and the econometrics track from 1997 until 2010 at the University of
30 vs 0 35 vs 30 45 vs 35 Age (years) -0.036 -0.381*** -0.044
Vwogpa 0.005 0.002 -0.027
Table 2: Student population for the different BSA systems
Amsterdam. There is also data available on the academic year of 2011, but while from 1997 until 2010 the program students follow remains roughly the same, the program changes dramatically in 2011, so this data is not used for comparison. During this time 7434 students registered for the two studies, this student population is summarised in table 2 according to the different BSA systems. Here we can see that the amount of students who applied at the FEB increases over time. This is not reflected by the amount of students in each BSA cohort because the duration of each BSA level becomes shorter over time. The average amount of credits obtained shows a drop after the introduction of the BSA but rise with the last increase of the threshold, the same holds for the average grade obtained in the first year.
In the introduction four questions were presented:
1. Do the students collect more ECTS in their first year? 2. Does the bachelor completion rate at four years increase?
3. Do students drop out less now they have passed the BSA in their first year? 4. Do the grades drop after implementing the BSA?
The results are presented over the next four subsections.
3.1 Do students collect more ECTS in their first year?
I start with the analysis of the first year effects. From table 2 we can conclude that the average amount of credits obtained actually drops after the first two BSA systems and to rise again with the third. So overall performance does not increase with the implementation of the BSA system. But does the number of students who pass the respective thresholds increase?
The data shows that this hardly happens (see table 3). Compared here are the students before and after the implementation or rise of the BSA. The first column shows the percentage of students who reached more or an equal amount of credits with a BSA of thirty. This amount rises, so there are more students that collect thirty or more credits after the BSA is introduced. Which is what should be happening if the students are motivated by the BSA. The effect is however not significant and when the BSA is raised to thirty-five the amount of students who acquire thirty-five or more ECTS even drops, though once again, insignificantly. The last rise of the BSA does seem to have effect. Here the amount of people who pass the threshold increases significantly.
BSA 0 30 35 45 Student total 1842 1856 1996 1740 Age (years) 19.8 19.8 19.4 19.4 Male (%) 0.73 0.71 0.71 0.71 VWO GPA (1-10) 6.76 6.76 6.76 6.74 Credits (0-60) 32.7 32.0 30.5 34.6 GPA (1-10) 5.74 5.36 5.40 5.59 Years '97-'01 '02-'05 '06-'08 '09-'10
Table 3: Analysis of the differences between BSA systems
Note; The table compares the amount of students who were able to acquire the amount of ECTS set by the latest BSA. The comparison is made three time, once when the BSA was introduced, then when it was raised to 35 and the last time when raised to 45.
-- significance level
Three stars *** significant at the 1% level
No starts not significant.
Through a regression analysis, the results are similar, with one exception. When controlling on age, gender and vwogpa the effects of the BSA at thirty is now significantly negative. The constructed regression in table 4 shows the chance a student passes the respective BSA thresholds. The first column shows that the need to reach a BSA of thirty significantly lowers the chance of reaching thirty or more ECTS. At a BSA of thirty-five the effect is insignificantly negative and at forty-five significantly positive. Age and being a man always lower the chance of passing the threshold, while a higher GPA in high school contributes to a higher chance. These results therefore do not point in one general direction, so the BSA does not seem to be that effective in motivating students in their first year performance.
Table 4: Regression analysis
Note; -- significance level
Three stars *** significant at the 1% level Two stars ** significant at the 5% level One star * significant at the 10% level No stars not significant
Starting years compared '97-'01 '02-'05 '02-'05 '06-'08 '06-'08 '09-'10
Number of students 1842 1856 1856 1996 1996 1740
BSA 0 30 30 35 35 45
Acquired the ECTS required
by the last BSA or more (%) 0.596 0.601 0.525 0.513 0.319 0.507
Difference 0.005 -0.012 0.19*** variable ≥ 30 ≥ 35 ≥ 45 constant 0.2901* -0.719*** -1.449*** (-0.1495) (0.1567) (0.1623) age -0.0566*** -0.0467*** -0.0331*** (0.0051) (0.0178) (0.0058) male -0.0612*** -0.0737*** -0.0628*** (0.0184) (0.0179) (0.0173) vwogpa 0.2194*** 0.3247*** 0.36*** (0.0141) (0.0149) (0.0147) BSA 30 -0.0378** (0.0161) BSA 35 -0.0238 (0.0158) BSA 45 0.1893*** (0.0152)
Now there are two things which require a closer look. First there is the sudden change of effectiveness of the BSA when set at forty-five ECTS and secondly the difference between the increase of students who reach thirty ETCS or more with a BSA of thirty, while the effect of this BSA is significantly negative. So the amount of students who reach the threshold of forty-five is almost 60% higher if the BSA is forty-five, compared to a BSA of thirty-five. This while the effect of the BSA on credits obtained at the levels of thirty and thirty-five was small and insignificant. There is no clear reason for this large increase in effectiveness. While a lower BSA is easier to acquire, the amount of students who pass this lower threshold even drops. This suggests that students do not pay that much attention to the BSA. Perhaps the rise with the BSA of forty-five is there because of different reasons, like the ones stated in the beginning of this chapter, but within this research I’m not able to conclude this. Then to the difference between a higher percentage of students who reach thirty or more ECTS and the significantly negative effect of the BSA. This difference is explained by a group of
students who do register for the study, but never make even a single exam. For the time period of ’97-’01 (no BSA) there are 182 students who do this, and for the period of ’02-’05 (a BSA of 30) there are 146. If these students are left out of the equation the percentage of students who reach thirty or more ECTS becomes 66.1% without a BSA and 65.2% with a BSA of thirty. This is in line with a significantly negative result for the BSA in the regression
without the register only students (-0.0495*** (0.0163)). It therefore could prove useful to leave these students entirely out of the research, but their amount does not remain
constant over time, therefore they are still included.
3.2 Does the bachelor completion rate at 4 years increase?
But does the BSA make students study faster? Besides the effect in the first year does the BSA reduce the number of years students need to complete their bachelor? The FEB
implemented the BSA to make students complete their bachelor faster. They do this because they claim that a student who collects more credits in his first year keeps on doing this in the years that follow (Booij, 2008). All but the last group of BSA-students (at a BSA of forty-five) have had the possibility, with this dataset, to complete their bachelor in four years.
If the comparison is made between the different BSA systems it seems to prove the reason why the FEB has implemented the BSA. The completion rate of students within four years increases with each increase of the BSA, even while there was no effect at the first year level. The amount of students who complete their bachelor in four years increases drastically, especially from no BSA to a BSA of thirty. In table 5 it is shown that the percentage of students complete their bachelor in four years increases from 9.5% up to 23.2%, an increase of 144%. This trend continues with the rise of the BSA to thirty-five, were the amount keeps growing to a total of 30.7%. Although for the last rise, the age of the students has changed, so I cannot base a conclusion on that result. But the large increase at a BSA of thirty does suggest that the BSA succeeds in rising the bachelor completion rate within 4 years.
Note; The division is shown twice, once with the students who only registered and once without them. The very first column shows year of entry, and the first column of each division shows the percentage of students who dropped out in their first year, the second shows the students who dropped out after their first year and the last column
Table 6: Division of dropout rates per year Note; -- significance level
Three stars *** significant at the 1% level
Table 5: Percentage of students who completed bachelor (BA) in four years or less
3.3 Do students drop out less now they have passed the BSA in their first year?
So more students complete their bachelor within four years. But the BSA could have a side effect, it could also reduce dropout rates at later years. Because the students who do not reach the BSA limit and would have tried the study for another year, if it was still a
possibility, are now forced to leave the study after their first year. This effect does take place and is best represented by the dropout rates per year. In table 6 the dropout rates of the different starting years are divided into a group that quitted in their first year and a group who quitted after their first year. The implementation of the BSA system greatly shifts the year of quitting forward. Students hardly dropout in their second or third year at a BSA of forty-five while most of the students drop out after their first year if there is no BSA. This division is shown twice, because the students who only register naturally drop out in their first year. Even if these student are taking out of the comparison there are less students dropping out in the later years with the introduction of the BSA. So not only does the BSA make more students complete their bachelor within four years, it also reduces the dropout rates in later years.
BSA no BSA 30 30 35
Number of students 1842 1856 1856 1996 BA complete (%) 0.095 0.232 0.232 0.307
Difference 0.137*** 0.075***
With register only Without register only cohort first year later years total first year later years Total
1997 21% 79% 241 13% 87% 220 1998 25% 75% 180 14% 86% 157 1999 35% 65% 173 21% 79% 143 2000 43% 57% 151 30% 70% 122 2001 62% 38% 235 43% 57% 156 2002 76% 24% 191 71% 29% 155 2003 68% 32% 241 62% 38% 204 2004 78% 22% 244 74% 26% 201 2005 75% 25% 252 72% 28% 225 2006 83% 17% 315 81% 19% 274 2007 85% 15% 351 84% 16% 319 2008 88% 12% 363 86% 14% 308 2009 95% 5% 439 94% 6% 388 2010 100% 0% 408 100% 0% 348
3.4 Do the grades drop after implementing the BSA?
The last problem concerned the possible negative effect of the BSA system. Which was represented by the effect of the BSA on the grades the students obtained. In the
introduction the concern was mentioned that the increased pressure to obtain more credits could cause students to only focus on passing a course no matter the grade. But two of the BSA systems (30 & 35) did not cause a significant change in the amount of ECTS collected. Therefore I first look at the last group of students who did get more ECTS. On the right side of table 7 there is the comparison between the students who had more than forty-five ECTS but who had a different BSA. The first group had a BSA of thirty-five and the second one a BSA of forty-five. The grades obtained by the last group are significantly lower than that of the first group. So while these students do get more ECTS, their grades go down. This suggests that the BSA does indeed create extra pressure to pass courses no matter the grade.
The effect is even stronger for the entire group, while the effect of the BSA seems to be generally absent on the amount of ECTS collected in the first year, there does seem to be a clear effect on the grades. In graph one the average grades obtained by students in their first year are divided over the amount of ECTS they collected. They are then pooled by blocks of 5 ECTS. If we compare the students without a BSA to the group that does have a BSA it can be seen that the grades have declined. This effect is significant, as seen in table 7 on the left side. So the BSA did not only lead to the decline of the grades for the last group, but did this for all the groups. This suggests that while the BSA does not lead to more ECTS collected by the students who had a BSA of thirty or thirty-five, they do feel extra pressure and therefore have lower grades.
Table 7: Analysis of the grade differences
Starting years compaired 97-'01 02-'10 06-'08 09-'10 Number of students 1616 5128 636 882
Average grade 5.74 5.44 6.67 6.53
Difference -0.3*** -0.14***
Note; compared are the grades of students with a different BSA. The first comparison is between students who did not have any BSA and those who did. The second is between the students who acquired more than 45 ECTS but who had a different BSA,
respectively 35 and 45. -- significance level
4. Conclusion
The effect of the BSA on student performance is mixed. On first year performance it is both positive and negative, on bachelor duration and dropout rates positive and on grades negative. It seems that the BSA is therefore not a straightforward tool in order to achieve better studying behaviour. The BSA could have had a motivational and a selective effect on students. However the results showed that the BSA does not seem to be an extrinsic motivation for the students.
The selection mechanism of the BSA however does prove to be effective. Dropouts hardly occur after the cut in the first year and students who are allowed to continue complete their bachelor faster. This proves the claim made by Booij (2008) that students who collect more credits in their first year, continue to do so in their later years. While these results may also be influenced by government policy or the business cycle, as the increase of the
effectiveness of the BSA at a level of forty-five, they already appear with the first
introduction of the BSA and before the suggested government and economic interferences. The effect of the BSA on the grades seems, opposite to that on the dropout rates and completion time, to be negative. The grades are significantly lower for the students who do need to pass a BSA threshold. This suggest that the concerns stated by Ordóñez et all. (2009) are justified and that threshold performance has a negative effect on the quality of the work by the students.
Even with the drop in the age of the students, by which I cannot compare between the BSA of thirty and thirty-five, the conclusions remain the same. The effect of the BSA appears to be one mostly of selection. Through this mechanism of only allowing the motivated students to continue, the FEB is able to improve it standings. But it comes at a price, the students do experience extra pressure to perform and therefore get lower grades. This feeds on the discussion of the university as a knowledge institute or as a production firm. Does the quantity or the quality count the most in providing education to the next generation of students? 0 1 2 3 4 5 6 7 8 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60
no BSA BSA 1 BSA 2 BSA 3
References
Angrist, J., Lang, D., & Oreopoulos, P. (2009). Incentives and services for college
achievement: Evidence from a randomized trial. American Economic Journal: Applied
Economics, 1(1), 136-163.
Booij, A. (2008) Increasing the BSA: improving future performance. Unpublished working
paper.
Leuven, E., Oosterbeek, H., & van der Klaauw, B. (2010). The effect of financial rewards on students’ achievement: Evidence from a randomized experiment. Journal of the
European Economic Association, 8(6), 1243-1265.
Leuven, E., Oosterbeek, H., Sonnemans, J., & van der Klaauw, B. (2011). Incentives versus sorting in tournaments: Evidence from a field experiment. Journal of Labor Economics,
29(3), 637-658. doi:10.1086/659345
Locke, E. A., & Latham, G. P. (2006). New directions in goal-setting theory. Current Directions
in Psychological Science, 15(5), 265-268.
Ordóñez, L. D., Schweitzer, M. E., Galinsky A. D. & Bazerman M. H. (2009) Goals gone wild: The systematic side effects of over-prescribing goal setting. Unpublished working paper. Harvard Business School
Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25(1), 54-67.
