Jeremiah Dow .com



Jeremiah Dow

06/13/10

Competitive Destruction in Higher Education: A New Paradigm

Today’s academic environment is perhaps more competitive than ever before. Higher enrollment coupled with rapidly rising tuition rates ensures students demand maximum results for their dollar. Whereas stock prices and price earning ratios signal to the investor his or her returns per dollar of capital, grades provide similar information to the student, as they act as a sort of currency when signaling the student’s “intellectual capital” to other universities and future employees.[1] However, such competitive pressures have bread new problems within the halls of higher education, altering student behavior. Over the previous four decades, students have been out-performing their predecessors, yet, some question the causes of steadily increasing grades.[2] While there is widespread disagreement on whether grade inflation actually exists, the data do well to call attention to the possibilities of grade inflation’s increasing presence. By attempting to restrict the number of A’s awarded to students, grade inflation policies such as that implement by Princeton University[3] increase student incentive to engage in cheating, showing the potential for a paradigmatic shift in student behavior that renders academic misconduct a necessary condition.

Grade inflation is a new challenge that falls on both the University and faculty, but it has important implications for students as well. I will attempt to demonstrate that grade inflation policies, by restricting the number of A’s awarded—in effect imposing a grade ceiling—increases a student’s incentive to cheat. After an overview of grade inflation itself and the theoretical implication on incentives and competitive behavior, I will construct a model based in competitive game theory that will represent both zero-sum and non-zero-sum game scenarios. I will then use these models to demonstrate that implementing grade inflation policies in a way that imposes a grade ceiling changes the structure of the game from a non-zero-sum to a zero-sum game, thereby altering the players’ behavior. By doing so, I wish to establish a theoretical link between a rise in competitive behavior and the subsequent increase in students’ incentives to cheat. Due to the recent development and implementation of such policies, literature attempting to compare pre and post-policy rates of cheating is lacking. Therefore, my analysis will compare the theoretical validity of the model’s response to the policy change with other studies that have examined the behavioral patterns of students’ incentives to engage in academic misconduct.

There is consensus among some scholars of the grade inflation phenomenon, namely that rising grades over the last forty years is absent a like increase in student performance.[4] However, although numerous studies empirically demonstrate a rise in student GPA’s over time, other studies show the opposite.[5] It is no surprise then, that given such a contentious issue, data is clouded and little certainty as to causal factors exists. But it is not causes I wish to explore yet, but rather grade inflation’s perceived hold on higher education. While Hu’s discussion of rising grades in higher education cites numerous studies that both support and refute the prevalence of grade inflation, he provides empirical data that show not only a substantive increase in student GPA’s over time, but that demonstrate such an increase is more prevalent in private institutions.[6] For example, from 1973 to 1997, Princeton students saw a rise in their median GPA from 3.08 to 3.42. And by 2003, A’s made up 47 percent of undergraduate grades compared to only 31 percent in 1970.[7] Princeton’s report duplicated theses results, showing that between 2001-2004, the three year period prior to their new grading policy, A’s accounted for 47 percent of all grades awarded.[8] In addition, Harvard saw an increase of students graduating with honors from 50 percent in 1961 to 91 percent in 2001.[9] Such statistics make it difficult to argue against the possibility of artificially high grades, yet, not all scholars agree that a rise in student performance is indeed absent.

Despite the consensus reached by many scholars that student performance is not keeping pace with grades in the classroom, some argue just the opposite. Opponents of the grade inflation myth argue, among other things, that a rise in grades is reflective of better performance by students and improvement in pedagogy.[10] Moreover, many claim that grade inflation is an empirically unfounded phenomenon based on faith. Two studies cited by Hu[11] posit that perceptions of grade inflation are falsely shaped by “anecdotes from a few elite institutions such as Harvard and Princeton.”[12] However, data also exists from more typical universities—those that provide a cross-section of the more average American student. Fairfield University in Southern New England, with nearly 200 full-time faculty and 3,000 full-time undergraduate students exhibited an increase in average grades from 83 percent in 1986 to 86 percent in 2002. More interesting is that SAT scores of Fairfield students were lower in 2003 than in the previous fourteen years.[13] And that SAT/ACT scores and high school grades are “consistently the strongest predictors of student GPA in college,”[14] suggests a disparity between performance and grades. Moreover, Barriga’s study of a small Catholic, liberal arts university with 61 full-time faculty and 1,347 undergraduate students duplicated the results found by Abbott. Over the course of seven academic years, Barriga found an increase in the average course grades from 3.33 in the 1994/1995 academic year to 3.47 in 2000/2001. However, a decrease in SAT scores from 976 to 937 over the same period was also prevalent.[15] In addition to lower grading standards, Abbott notes that selectivity shows a lag in collegiate standards. While only 37 percent of Fairfield applicants were accepted in 1987, despite lower SAT scores 49.4 percent were admitted in 2003.[16] Both pieces of data indicate not only a decrease in student performance, but also a decrease in grading standards and selectivity among universities. This, coupled with rising GPAs, lends credence to the grade inflation myth.

The above data are important for other reasons besides exploring the validity of the grade inflation phenomenon. In order that we, as students, faculty, administrators, and scholars are capable of mitigating its debilitating effects, we must track its source—the fundamental variable that leads to the subsequent factors of grade inflation. The variable in question, I claim, is incentive. But it is only through analysis of grade inflation’s effects that one can shed light on its substantive causes. That said, at a modest attempt at reverse engineering the development of grade inflation, attention must first fall on the concept of incentives and how they play their part.

The implications for students, faculty, and institutions of higher education are many, but they all stem from incentives. One may argue that human behavior is based within the realm of economics because behavior itself is highly incentivized. As Steven D. Levitt stated in Freakonomics, “Economics is, at root, the study of incentives: how people get what they want, or need, especially when other people want or need the same thing….An incentive is a bullet, a lever, a key: an often tiny object with astonishing power to change a situation.”[17] The question, then, is how to structure them in a way that is rewarding, yet, fair and equitable. I do not suppose to solve such a complex problem, but I do wish to demonstrate how misaligned incentives can lead to moral hazard[18] at all levels of the grading process. More importantly, I wish to show how incentives enter into the student’s frame of reference. The linchpin to this idea is that these incentives are economic incentives—those that classical economists have assigned to guiding market transactions—precisely because grades are purchased via a market transaction. No matter the grade that is earned, which undoubtedly depends on the student’s performance, students purchase the opportunity to earn that grade. These are things students and scholars know all too well, but I feel I must state them explicitly because such realizations have become so engrained into our daily activities that we hardly take explicit account of them.

Institutions, it seems, are accountable to the market-type hysteria now infesting American education. The primary role of collegiate institutions is to provide critical information to society regarding which students possess the right skills and attributes that society needs in order to advance economically and socially. However, as grade inflation continues, transforming from phenomenon to trend, it “reflect[s] an erosion in the credibility of higher education to accurately evaluate students’ achievements.”[19] But why has such behavior come to pass? One will see below that misaligned incentives such as those among many financial firms in Wall Street present themselves in similar form to colleges and universities.

The causes of credibility erosion are imbued in the economic incentives that are common amongst all organizations. The incentives that contribute to artificially high grades are a combination of the need to keep enrollments high and the concern over donations from future alumni. Abbot’s case of Fairfield’s declining collegiate standards lends additional clarity. He points to a correlation between grading standards and course enrollment at the department level, establishing a trend of higher enrollment rates in those courses that tend to grade higher.[20] Although in this case Abbott is concerned with faculty, the idea that lower grading standards will increase class enrollment is not unlike a decline in collegiate standards at the institutional level. Rather, the former is simply a micro-trend of the latter. A direct impact is that “grading standards can also affect the number of majors in a department and hence whether it receives funding increases and new hiring lines.”[21] Thus, an institution’s financial stability is directly dependent not upon its role as educator, but upon its grading habits. Furthermore, it is shear folly to deny that the institution feels the same economic pressures as individual departments. The practical significance is that higher education has succumbed to moral hazard, as it seems unable to divorce current practices from future monetary concerns. Cushman points to this staggering dilemma:

No institution, especially in difficult economic times, can be expected to pursue a policy that would jeopardize the ability of its students to succeed, especially since such success is so fundamentally related to the future ability of the institution to ask for donations from its graduates.[22]

As economic concerns bear more weight in current decisions, colleges and universities sacrifice quality and prudence in grading for relaxed standards under the rationale of maximizing enrollment and funding. Regarding the end goal, a paradox presents itself, whereby as resources are gathered, the credibility directly associated with the output of quality students diminishes.

When examining similar incentives faced by faculty members, one sees a similar result. Certain observations lead one to see grade inflation not only as a trend, but as a new status quo. Faculty are perhaps in the most difficult position because they act as middlemen and women in the provision of grades. Cushman is quite straightforward in his criticism of higher education, particularly his peers. He posits a “socialization experience,” whereby a “critical mass of [his] peers who create, by assigning grades that students don’t really deserve, the expectation of higher grades across the board.” He continues that once students are reconciled to a belief of higher grades with lower performance, the professor is now bound to reply accordingly.[23] Such affects of the status quo are reinforced by the observation that “grading leniency is associated with superior course evaluations.”[24] In addition, a preference by teachers to use a grading curve only when it limits the number of low grades indicates a double standard.[25] One might observe from such behavior, then, that a perverse incentive to grade artificially high exists among faculty. However, for such a statement to hold validity, one must ask what is at stake for the faculty.

By providing an intimate link between grades and the teacher, Cushman illuminates the full impact of grade inflation on faculty. When interpreting grades as a product of a teacher’s labor, they become “a measure of their professorial authority in the workplace.”[26] The conclusion from here is quite simple—inflating grades not only provides more job security via shining course evaluations, but also elevates a teacher’s status among his or her peers. Ironically however, as grade inflation tightens its grip on the expectations of college students, in effect creating a new status quo, many professors see a diminished ability to grade objectively and independently. The implications are frightening, and lead to what Cushman classifies as a new system of “grades for dollars.”[27]

What emerges is a quid pro quo system: professors give higher grades and students accept them and don’t make waves. The professor satisfies the student consumer’s demand for an acceptable product and, in return, avoids the taxing and unpleasant duties of assigning and defending lower grades.[28]

Some may argue, then, that not only are faculty resigned to similar external pressures that plague the university, but that these pressures have established new demands from students that are counter to their and society’s best interest. Furthermore, as faculty continue to indulged students with artificially high grades, the practice becomes the new norm, and is consequently harder to combat. As shown below, when students view their education through the lens of investments and returns, education’s bottom line changes from that of knowledge, prudence, and ability to little more than maximizing one’s return on an investment.

In addition to universities and faculty, one may argue that students are the most effected by grade inflation. First however, in order for the effects on students to bear full weight, one distinction not yet discussed is important. Such importance lies in an additional, yet consequent, phenomenon known as grade compression. With grade inflation, a “similar quality of academic performance in a given course is awarded higher grades at the present time than before.” Consequently, if grade inflation is allowed to occur over time, “variations in student course grades can no longer differentiate student performance.” [29] This latter concept is grade compression. As average grades rise, more students receive top marks, in effect, compressing performance at the top of the grading scale. As Barriga notes, “This condition does not allow for adequate identification of differences toward the higher end of the grading scale, and thus excellent performance is no longer differentiated from good or possibly even mediocre performance.”[30] Princeton’s Committee on Grading Policy noted the following effect of compression:

An A grade should signify superior work and a B grade should signify work that exceeds minimal expectations. With compression, the A grade had come to cover a spectrum from work that marginally exceeds expectations to truly superior work; the B grade had come to signify work that was barely acceptable.[31]

What this means for grades as a communicative tool is that they are no longer able to signal student achievement and other strongly associated characteristics such as intelligence and a strong work ethic.

Given that students view grades as intellectual capital, and universities and employers hold the complementary view of grades as indicators of a potential employee’s future value to the institution or firm, the primary implication of grade inflation is that it diminishes the functional value of grades as signals of merit and ability. Consequently, “grade inflation could undermine the…social function of grades.”[32] However, the educational function is that grades serve a particular motivational purpose for students. Hu states the motivational function of grades as follows: “Grading affect how students study, what they focus on, how much time they spend, and how involved they become in the course. Thus grading is a powerful part of the motivational structure of the course, for better or worse.”[33] But at second glance, one can see that this motivational function carries over to directly affect students as well. Barriga brings another viewpoint into the issue. Agreeing that, “Some students may become less motivated because the excellence of their work is not being recognized,” the idea that “Other students may become less motivated because their mediocrity has already earned them the highest grade possible” speaks louder to the those students who still have room to improve.[34] An additional impact of grade inflation, then, is that it alienates those students who need improvement in certain areas, but whose grades reflect a level of competency that is artificially high.

The placement of incentives here is that grade inflation, by awarding more A’s to students while holding their level of performance constant, indulges their interests by guaranteeing them an inflated return on their investment. However, in reality this practice alienates them. The adverse effects remain hidden from students and take the form of diminishing returns. “Grade compression diminishes the function of differentiating students’ academic performance and effort…[and] academic performance is rewarded relatively less because the rate of return of performance in grades diminishes in the compression process.”[35] Diminishing returns is a concept well known by economists and has a potentially devastating impact on how students perceive grades as a motivational tool to achieve the best quality work they can and as a marketing tool that fairly and objectively markets such achievement. Important to note, then, is that the decline in the relative rate of performance is not only a phenomenon posited by scholars and administrators, but one that is far reaching in its implication that choices and behavior, more often than not, have unintended consequences.

On a more minute level, however, one more associated with how the individual student values his or her educational achievement, is grade inflation’s affect on the student’s perception of his or her potential payoff. The role of incentives is crucial here. To borrow from William Easterly, incentives are important in understanding human behavior because “people respond to incentives.” Primarily, Easterly states that people “invest in the future when they get a high return to their investment. They do not invest in the future when they do not get a high return to their investments.”[36] While this lends additional support to my discussion above of the moral hazard facing institutions, Easterly’s statement also provides a crucial link between incentives and student behavior. Assuming that students take Easterly’s view, their incentives to attain high grades are a function of their expected payoff—a higher salary upon entering the workforce—and the price they paid—their rate of tuition. Economic incentives, then, must factor into the how grades are administered.

The fact that grade inflation exists and my suggestion that market pressures are acting on this phenomenon I have discussed. The observation that grade inflation is more prevalent among the more prominent private institutions where tuition rates are typically much higher support this. Indeed, some college professors go so far as to assert, “the unfortunate fact is that grades have become the stock-in-trade of contemporary higher education, the product that is being sold to students in exchange for exorbitant tuitions.”[37] The fact that many private universities are also Ivy League institutions suggests that the prestige of a college education is not only viewed by students as a commodity, but that college grades in general are viewed as a commodity of degrees, whereby an Ivy League institutions may add more relative intellectual capital to a student’s coffers than the average university. For example, a firm is more likely to hire a graduate from Harvard with an equivalent GPA than a student from Eastern Oregon University, ceteris paribus. Such a realization explains the appeal of the Ivy League, lending credence to the stronger market pressures that factor into maintaining a higher level of performance at these institutions. Moreover, the burden of moral hazard would seem to fall more on these elite institutions. Consequently, one can see that Cushman’s narrative of “Grades for Dollars” applies more strongly to those more costly institutions. In addition, because costs and benefits typically move in the same direction, as the costs to the university’s credibility rises, the benefits of addressing grade inflation become greater. It is no surprise, then, that the media are targeting universities such as Harvard and Princeton. Nor is it surprising that these institutions have taken the first step in eradicating the grade inflation phenomenon.

Through incentives, certain behaviors society deems inappropriate are discouraged.[38] In the case of grade inflation, Princeton claims artificially high grades are discouraged through implantation of new grading policies. But what if this new policy alters a student’s behavior, ultimately leading to unintended consequences? Levitt demonstrates that the forceful implication of incentives on a group of people does not necessarily change their behaviors in the intended way:

Such is the strange and powerful nature of incentives. A slight tweak can produce drastic and often unforeseen results. Thomas Jefferson noted this while reflecting on the tiny incentive that led to the Boston Tea Party and, in turn, the American Revolution: ‘So inscrutable is the arrangement of causes and consequences in this world that a two-penny duty on tea, unjustly imposed in a sequestered part of it, changes the condition of all its inhabitants.[39]

It ought to be noted then, that incentivizing behavior does not necessarily create the desired outcome. That said, assuming the student’s incentives to receive the highest grade possible has not changed in the face of a now limited supply of A’s, one can posit that their behavior will indeed. Moreover, this is not irrational because the new policy has effectively altered the potential payoff. That is, students, now in a more highly competitive atmosphere, will have a higher incentive to cheat. The connection between incentives and cheating is that one sees an increase in the former to engage in the latter whenever the payoff is altered in a way that diminishes the players’ potential gain. Levitt offers some clarity on this point as well. “Cheating may or may not be human nature, but it is certainly a prominent feature in just about every human endeavor. Cheating is a primordial economic act: getting more for less.”[40] One may suppose that if the payoff is now less, then such a ‘primordial’ act may seem the rational choice. Moreover, as Levitt points out, “[C]heating is more common in the face of a bright-line incentive (the line between winning or losing, for instance) than with a murky incentive.” The difference between winning and losing alters the potential payoff structure to resemble a zero-sum outcome, whereby the sum of the winner’s gains and the loser’s losses equal zero.[41] By altering the structure of the game from non-zero-sum to zero-sum, the policy imposes an additional hurdle on the players that is likely to alter their behavior.

I will now use two models of game theory to illustrate a before-and-after picture, if you will, of a student’s incentivized behavior. Game theory is useful here because it is designed to analyze strategic action between two or more opponents.[42] More specifically, game theory deals with the rational choices of players who are in situations of conflict or cooperation.[43] This distinction will be important later when discussing the primary differences between non-zero and zero-sum games. First however, we must acknowledge four basic characteristics of game theory.

1) There must be at least two players. Players can be individuals or groups such as business firms or political organizations.[44]

2) Each player chooses among a set of strategies, typically referred to as pure or mixed strategies. A player following pure strategies will always play a given strategy at a fixed probability regardless of his or her opponent’s strategy. A player will choose a mixed strategy by randomly mixing his or her play at a fixed proportion.[45]

3) The strategies chosen by each player lead to a specific outcome.

4) Each outcome results in a specific payoff to each player, and a numerical value associated with each payoff represents the value ( or level of utility) of that payoff to each player.[46]

Given these characteristics above, some additional assumptions concerning the players are needed. They are as follows:

1) People respond to incentives.

2) Each player’s incentive is to maximize the return on their investment.

3) Education is an investment.

4) Grades represent degrees of maximization on that investment.

5) Therefore, students want the highest grades possible to maximize their return.

First of all, each player is assumed to be rational. According to game theory, rational players have an incentive to choose the best strategy, namely the strategy that maximizes their average payoff given the other player’s choice. Such a strategy is known as the player’s optimal strategy.[47] This is important because the degree to which each player is compelled to play a given strategy depends upon a mathematically determinate value.[48] The higher the value, the higher the payoff, and we will treat the payoff values as linear to each player’s utility. That said, a player may play either a maximin strategy, whereby each seeks to maximize their utility, given the other player’s choice, or they may play a minimax strategy, whereby each seeks to minimize their opponent’s utility. This is the minimax principle, and reflects each player’s rational choice in that they always pick the most conservative strategy. [49] However, we will simplify here by assuming each player will use the mixed strategy of randomly alternating between cheating and not cheating with a 50 percent probability assigned to each (½ C, ½ DC). After all, it is not each player’s optimal strategy we are after, but to examine the primary conditions of each game, and to find which condition changes when the player’s payoffs are not directly opposed.

That distinction lies in the fact that under a non-zero-sum game scenario one player’s success is not diametrically opposed to his or her opponent’s. That is, player one’s victory does not entail player two’s loss. Although both may not receive equal payoffs, each may better his or her position without affecting their opponent’s payoff. When such a condition is reached, the game is pareto optimal. An additional distinction between non-zero-sum and zero-sum games is that in the latter, every outcome is pareto optimal.[50] This makes sense because zero-sum games are termed as such because the sum of the winner’s gains and the loser’s losses equal zero. This is perhaps the most important distinction for my argument, for it is this fact that I wish to demonstrate. Namely, that each player in a zero-sum game has a higher incentive to play a certain way regardless of his or her opponent’s choice. In other words, the pareto optimal outcome in a zero-sum game is always to maximize each player’s payoff.

To illustrate this argument and its implications, let us engage in a thought experiment. Suppose two students, Tony and Matt, take an exam. Tony, for reasons we may classify as random chance, must take the exam first. Now, suppose the instructor announces to the students that only one of them will receive an A. If both score above a 90 percent, then the higher score will receive the A and the lower score will have to make due with a B. Tony’s decision-making process must acknowledge that his payoff is dependent on Matt’s payoff. Moreover, the game is now a zero-sum game because the alteration of payoffs places their interests in direct contrast. In other words, the sum of Tony’s gain’s and Matt’s losses, or vice versa, will equal zero.

|Two-Person-Zero-Sum | |

| | |Matt | |

| | |Cheat |Doesn't Cheat|

|Tony |Cheat |3 |2 |

| |Doesn't Cheat |2 |1 |

| | | | |

Looking at the payoff matrix above, one can see that this particular game favors Tony. But what we are concerned with is that given both Tony and Matt playing mixed strategies of ½ C, ½ DC, is to demonstrate that each player’s choice of strategy is made with regard to his opponent’s mixed strategy. Tony’s payoff for cheating is ½ (3) + ½ (2) = 5/2, and his payoff for not cheating is ½ (2) + ½ (1) = 1 + ½ = 3/2. As Tony seeks a maximin strategy, we see that his rational play is to cheat regardless of Matt’s decision, as he will win two-and-a-half games for every one that Matt wins. Thus, we say that Tony’s dominant solution is to cheat. Now, Matt’s payoff for cheating, represented by the negative values in the payoff matrix, as he is seeking a minimax strategy, is ½ (3) + ½ (2) = 3/2 + 1 = 5/2, and his payoff for not cheating is ½ (2) + ½ (1) = 3/2. We can look at Matt’s payoff in a couple of ways. We can say he maximizes his own payoff by cheating because 5/2 > 3/2, or what is the same, the only way for him to minimize Tony’s payoff is to cheat because -5/2 < -3/2. That said, Matt’s dominant solution is also to cheat. According to the expected value theorem, Tony and Matt are surely better off cheating because this strategy yields them the payoff with the largest weighted average.[51]

Given the same dilemma as above, but absent any artificial constrictions on grades awarded—grades are simply awarded by merit—Tony may still choose to cheat. But his incentives have changed because this scenario resembles a non-zero-sum game where the player’s interests are not directly opposed. Non-zero-sum games are unique in that they may be either non-strictly competitive or strictly competitive. This means that the former case may allow a degree of mutual benefit via collusion, while the latter will not.[52] I will assume the latter scenario because it more closely resembles both the behavioral norms that dictate proper student etiquette and the general sense of competition that underlies collegiate academia. What is important, however, is that Tony will make his decision on whether to cheat certain that his ability to receive an A is not in direct contrast to Matt’s ability. That is, with no artificial restriction on the number of A’s awarded, Tony will not base his decision to cheat on the notion of a zero-sum outcome.

|Two-Person-Non-Zero-Sum | |

| | |Matt | |

| | |Cheat |Doesn't Cheat |

|Tony |Cheat |(3,3) |(2, 0) |

| |Doesn't Cheat |(0, 2 ) |(1,1) |

| | | | |

This particular non-zero-sum game is well known to many. The prisoner’s dilemma lends particular support to my argument because, although it appears that both maximize their payoff by cheating, it more importantly demonstrates the theoretical possibility of collusion, whereby both players gain some degree of utility by agreeing not to cheat. The two assumptions stated above both hold, namely that no collaboration between Tony and Matt is allowed, and also that each will play the same mixed strategy as above: ½ C, ½ DC. Tony’s expected payoff for cheating is ½ (3) + ½ (2) = 3/2 + 1 = 5/2. And his expected payoff from not cheating is ½ (0) + ½ (1) = ½. Matt’s expected payoff from cheating is ½ (3) + ½ (2) = 5/2, and for not cheating his payoff is ½ (0) + ½ (1) = ½. Therefore, both are still better off cheating.

But looking at the possibility of collusion, a situation theoretically incompatible under the zero-sum scenario, we see that Tony and Matt can agree not to cheat, and more importantly, both may gain by doing so. Luce points out that collusion “can never be achieved if each player randomizes his strategies independently—which is exactly what must occur in the non-cooperative context.”[53] This is the linchpin, because if play must be randomized independently in order to achieve an optimal strategy for each player to maximize their respective payoffs, and if under our zero-sum scenario collusion is not allowed, then players are clearly better off to cheat. Given the assumption of no collusion under the non-zero-sum game above, the same is true. Each rational player will likely cheat absent any agreement from his or her opponent. However, if collusion is allowed, one can see that both players’ interests are not in opposition.

By each player playing their optimal strategy, they receive a higher payoff by cheating in both a non-zero-sum and zero-sum game. However, if collusion is allowed, incentives to not cheat are changed because a mutual benefit may be realized. Although the problem of defecting is still very real, the point I wish to demonstrate is that scenarios absent extremes yield a middle ground where two players may reach agreement, and this fact alters the incentives facing both players. As Levitt points out, “Any incentive is inherently a trade-off; the trick is to balance the extremes.”[54] But with grade ceilings, one can see that with two players competing in a zero-sum match, only the extremes exist. There can only be a winner and a loser.

After such a theoretical exercise, the next logical step is to ask whether the empirical data support the above conclusion. Whitley provides empirical evidence for this response. His mountainous review of 107 studies regarding predictors of student behavior towards cheating found that “students…who see themselves as being in competition with others are more likely to cheat than…those who perceive the environment as less competitive.” Whitley noted that a student’s perception of the atmosphere as more competitive was a moderate predictor of a student’s decision to cheat, [55] and noted further that teachers may be able to limit cheating by reducing competition in the classroom.[56] More precisely, Taylor offers a psychological snapshot of students today. Competitive pressures increase a student’s perception of the atmosphere as strictly competitive, and it is this perception that provides the student with his or her rationalization for cheating.[57] In other words, students who perceive their environment as geared more toward competition than learning, are morel likely to cheat. Other related correlates found by Whitley include a student’s academic beliefs, particularly his or her orientation toward college.

Student who have a learning orientation to college are motivated by a desire to acquire new information and ideas that will enhance their personal and professional development, whereas students who have a grade orientation are motivated by a desire to achieve high grades regardless of whether they learn anything of significance. The correlations found are thus consistent with the theory of the trait: students who want only a grade are likely to cheat for it, whereas those who desire knowledge avoid cheating.[58]

Moreover, one may argue that those grade oriented students, by assigning degrees of success solely to the grades they earn, perceive more benefits from cheating, and they more readily “perceive that academic norms favor cheating.”[59] Indeed, Harding reported cheating to be lower among those who funded most of their college expenses with scholarships versus out-of-pocket funding sources such as parents or jobs.[60] This suggests that market pressures alter students’ behavior. The rationale, one may argue, is that those absorbing higher costs are more willing to cheat because they, like economists and rational consumers, associate higher tuition costs with higher benefits to cheating. For a student with higher costs, then, one can posit he or she also sees a higher payoff to cheating. If true, then the size of the payoff from cheating will surely matter, as grade oriented students would naturally place more emphasis on those big-ticket items such as exams.

If changing the size of the payoff increases a student’s perception of the benefits of cheating, how might altering the structure of the payoff affect competitive pressures to cheat? To examine this question, one need only to look back at the original effects of grade inflation on student incentives. Dowling makes a causal claim in stating that by grade inflation undermining the grading process, it “has brought about the climate in which cheating is perceived as beating the system.” [61] A complex dilemma exists here. If grade inflation contributes to students’ positive perceptions of cheating, one may deduce that the student’s rationale was one of passive disconnection, the feeling being that if he or she is not going to be graded according to traditional academic standards, then why should they hold themselves to those standards. It is presumptuous to assume this affected the majority of students, but one might ask how this contributes to the broader perception toward behavioral norms of collegiate conduct. Whitley provides a clue. “Students who perceive that social norms permit cheating cheat to a greater extent than students who perceive a non-supportive norm.”[62] Indulging both Dowling and Whitley, then, one can see the potential for a very big storm brewing. Namely, that grade inflation has affected student perceptions in a way that favors cheating, which in turn contributes to a broader recognition of cheating as acceptable behavior among students.

However, the story gets worse. Add competitive pressures into this mix, and what one may see is a paradigmatic shift in academic behavior, whereby the behavioral climate changes from viewing cheating as unacceptable to seeing it distorted into a necessary condition through the lens of competitive pressures. It is my claim that grade inflation policies will do just that. In addition to increased perceptions of competition as a correlate of student incentives to cheat we see the potential for a snowball effect, whereby grade inflation policies increase perceptions of higher competition that lead to even higher rates of cheating. Consequently, an increasing number of students begin to view cheating as a necessary condition within an academic environment defined not by learning, but by cut-throat competition.

Furthermore, grade ceilings can only compound such an effect, for they resemble what Whitley calls norm-referenced grading practices. Norm-referenced grading—for example, grading on a curve—means that students are graded relative to each other’s performance. This is not unlike imposing Princeton’s grade ceiling. Whitley makes the comparison valid when he states the implications of this methodology. “Norm-referenced grading could increase competition because it limits the number of high grades and requires that a proportion of students receive low grades regardless of their absolute levels of performance.”[63] Grade ceilings, then, alter the student’s payoff by substituting relative measures for absolute measures, thereby increasing levels of competition within the classroom. And as Taylor points out, “if such school environments are perceived to be structured on a competitive scheme, where students are pitted against each other for rewards like grades and academic recognition, then individual student pressure to succeed within this structure places total responsibility for achievement on the student.”[64] At this point, the student obtains total control over setting the standards of academic conduct. Finally, we come back to Dowling: “It’s the cheaters who are in control of the moral climate in which exams are given and papers assigned, and students who act honorably are very nearly paralyzed by a diffidence or timorousness about calling public action to what is going on.”[65] So we see a kind of Gresham’s law, whereby bad grades are driving out the good, and academic misconduct is becoming the new collegiate norm.

That said, one might ask whether grade inflation policies such as those orchestrated by Princeton and Harvard Universities are prudent in the long term. Looking back to the primary concern of grade inflation, namely that grade compression yields decreasing returns on a student’s investment and diminishes the ability of grading to objectively and fairly signal a student’s capabilities and merit, policy analysts ought to look into the potential consequences of implementing any sort of policy that imposes a grade ceiling on its students. Such policies have the potential to backfire. People respond to incentives, and therefore, explicitly acknowledging them is crucial when examining both the known effects of any new policy as well as conjecturing on that policy’s potentially unknown effects. Because they encourage no preparation, unintended consequences are often drastic by nature. It is for this reason that they can often lead to problems that are not recognized until extensive damage has been done. While many universities are attempting to recover lost credibility as socially responsible institutions, they may be creating a new and further reaching problem, as students react and adjust their own perceptions not regarding what is acceptable behavior, but regarding what is necessary in the face of higher tuition costs, higher competition, and fewer top grades.

Bibliography

Abbott, William M. “The Politics of Grade Inflation.” Change 40 no. 1 (January/February 2008): 32-37. Wilson Web, EBSCOhost (accessed January 10, 2010).

“An Eye for an A.” The Economist. March 7, 2002. (accessed May 20, 2010).

Aronauer, Rebecca. "Princeton's War on Grade Inflation Drops the Number of A's." Chronicle of Higher Education 52, no. 6 (September 30, 2005): A41. Academic Search Premier, EBSCOhost (accessed March 1, 2010).

Barriga, Alvano Q. et al. “Dialogue and Exchange of Information About Grade Inflation Can Counter Its Effects.” College Teaching 56, no. 4 (Fall 2004): 201-209. Academic Search Premier, EBSCOhost (accessed January 20, 2010).

“College Grade Inflation.” Issues and Controversies (June 11, 2004). , Facts on File News Services (accessed April 8, 2005).

Coplin, Bill. "Grade inflation works for today's students." USA Today. Academic Search Premier, EBSCOhost (accessed March 1, 2010).

Cushman, Thomas. “Who Best to Tame Grade Inflation.” Academic Questions (Fall 2003): 55. Academic Search Premier, EBSCOhost (accessed April 25, 2010).

Dowling, William C. “Meaningless Grades and a New Dishonesty.” Academic Questions 16, no. 4 (Fall 2003): 57-62. Academic Search Elite, EBSCOhost (accessed April 7, 2005).

Easterly, William. The Elusive Quest for Growth: Economics, Adventures, and Misadventures in the Tropics. Cambridge: MIT Press, 2002.

Faculty Committee on Grading 2009-2010. Frequently Asked Questions. Princeton University. (accessed March 29, 2010).

Harding, Trevor S. et al. “The Theory of Planned Behavior as a Model of Academic Dishonesty in Engineering and Humanities Undergraduates,” Ethics & Behavior 17 no. 3 (2007): 255-279. ERIC, EBSCOhost (accessed April 25, 2010).

Hu, S. “Beyond Grade Inflation.” ASHE Higher Education Report 30 no. 6 (2005) 1-90. Academic Search Premier, EBSCOhost (accessed January 20, 2010).

Levitt, Steven D., and Stephen J. Dubner. Freakonomics: A Rogue Economist Explores the Hidden Side of Everything. New York: William Morrow, 2005.

Luce, Duncan R., and Howard Raiffa. Games and Decisions: Introduction and Critical Survey. New York: Dover Publications, 1989.

“Recent Key Events in College Grade Inflation.” Issues and Controversies (11 June 2004). , Facts on File News Services (accessed April 8, 2005).

Straffin, Philip D. Game Theory and Strategy. Washington D.C.: The Mathematical Association of America, 1993.

Taylor, Lyn, Mark Pogrebin, and Mary Dodge. “Advanced Placement-Advanced Pressures: Academic Dishonesty Among Elite High School Students,” Educational Studies 33, no. 4 (Dec 2002): 403-421. Academic Search Premier, EBSCOhost (accessed May 6, 2010).

Varian, Hal R. “Game Theory.” in Intermediate Microeconomics: A Modern Approach, 7th ed., 504-519. New York: W. W. Norton & Company, 2006.

Venttsel’, E. S. An Introduction to the Theory of Games. Translated by Jerome Kristian and Michael B. P. Slater. Boston: D. C. Heath and Company, 1963.

Whitley, Bernard E., Jr. “Factors Associated with Cheating Among College Students: A Review.” Research in Higher Education 39 no. 3 (1998): 235-274. ERIC, EBSCOhost (accessed May 18, 2010).

-----------------------

[1] “An Eye for an A,” The Economist, March 7, 2002, (accessed May 20, 2010).

[2] S. Hu, “Beyond Grade Inflation,” ASHE Higher Education Report 30 no. 6 (2005): 4. Academic Search Premier, EBSCOhost (accessed January 20, 2010).

[3] Faculty Committee on Grading 2009-2010, Frequently Asked Questions, Princeton University, (accessed March 29, 2010).

[4] William M. Abbott, “The Politics of Grade Inflation,” Change 40 no. 1 (January/February 2008): 32. Wilson Web, EBSCOhost (accessed January 10, 2010).

[5] Alvano Q. Barriga et al. “Dialogue and Exchange of Information About Grade Inflation Can Counter Its Effects,” College Teaching 56, no. 4 (Fall 2004): 201. Academic Search Premier, EBSCOhost (accessed January 20, 2010). Also, for a list of studies claiming to both demonstrate and refute the grade inflation phenomenon see Hu, 2.

[6] Hu, 30.

[7] Ibid., 1.

[8] Faculty Committee on Grading, 11.

[9] “Recent Key Events in College Grade Inflation,” Issues and Controversies (11 June 2004): 1. , Facts on File News Services (accessed April 8, 2005).

[10] “College Grade Inflation,” Issues and Controversies (June 11, 2004): 3. , Facts on File News Services (accessed April 8, 2005).

[11] Hu, 4. See C. Adelman, “Putting on the Glitz: How Tales from a Few Elite Institutions Form America’s Impression about Higher Education,” Connection 15 no. 3 (2001): 24-31; C. Adelman, Principal Indicators of Student Academic Histories in Postsecondary Education: 1972-2000. Washington D.C.: Institute for Education Science, U.S. Department of Education.

[12] Ibid.

[13] Abbott, 33-34.

[14] Hu, 12.

[15] Barriga, 203-204.

[16] Abbott, 34.

[17] Steven D. Levitt and Stephen J. Dubner, Freakonomics: A Rogue Economist Explores the Hidden Side of Everything (New York: William Morrow, 2005), 20.

[18] The concept of moral hazard is defined as “The presence of incentives for individuals to act in ways that incur costs that they do not have to bear.” See The Economist: Dictionary of Economics, 4th ed., (Princeton: Bloomberg Press, 2003), s.v. “Moral Hazard.” Let it be known that I use the terms moral hazard and perverse incentives interchangeably throughout the remainder of my discussion.

[19] Barriga 202.

[20] Abbott, 33.

[21] Ibid.

[22] Thomas Cushman, “Who Best to Tame Grade Inflation,” Academic Questions (Fall 2003): 49. Academic Search Premier, EBSCOhost (accessed April 25, 2010).

[23] Ibid., 55.

[24] Barriga, 202.

[25] Abbott, 34.

[26] Cushman, 55.

[27] Ibid., 49.

[28] Ibid., 51.

[29] Hu, 19.

[30] Barriga, 201.

[31] Faculty Committee on Grading, 9.

[32] Hu, 17.

[33] Ibid., 23.

[34] Barriga, 201.

[35] Hu, 18.

[36] William Easterly, The Elusive Quest for Growth: Economics, Adventures, and Misadventures in the Tropics (Cambridge: MIT Press, 2002), 38.

[37] Cushman, 48.

[38] Levitt, 22.

[39] Ibid., 23.

[40] Ibid., 25.

[41] Hal R. Varian, “Game Theory,” in Intermediate Microeconomics: A Modern Approach, 7th ed. (New York: W. W. Norton & Company, 2006), 528.

[42] Ibid., 504.

[43] Philip D. Straffin, Game Theory and Strategy (Washington D.C.: The Mathematical Association of America, 1993), 3.

[44] Ibid.

[45] E. S. Venttsel’, An Introduction to the Theory of Games, trans. Jerome Kristian and Michael B. P. Slater (Boston: D. C. Heath and Company, 1963), 7.

[46] Straffin, 3.

[47] Ventssel, 8.

[48] Ibid., 3.

[49] Ibid., 8.

[50] Straffin, 68.

[51] Specifically, the expected value theorem states, “If you know that your opponent is playing a given mixed strategy, and will continue to play it regardless of what you do, you should play your strategy which has the largest expected value.” See Straffin, 13.

[52] R. Duncan Luce, and Howard Raiffa, Games and Decisions: Introduction and Critical Survey (New York: Dover Publications, 1989), 89.

[53] Luce, 94.

[54] Levitt, 23.

[55] Bernard E. Whitley Jr., “Factors Associated with Cheating Among College Students: A Review,” Research in Higher Education 39 no. 3 (1998): 253. ERIC, EBSCOhost (accessed May 18, 2010).

[56] Ibid., 262.

[57] Lyn Taylor, Mark Pogrebin, and Mary Dodge, “Advanced Placement-Advanced Pressures: Academic Dishonesty Among Elite High School Students,” Educational Studies 33, no. 4 (Dec 2002): 404. Academic Search Premier, EBSCOhost (accessed May 6, 2010).

[58] Whitley., 244.

[59] Ibid., 261-262.

[60] Trevor S. Harding et al., “The Theory of Planned Behavior as a Model of Academic Dishonesty in Engineering and Humanities Undergraduates,” Ethics & Behavior 17 no. 3 (2007): 271. ERIC, EBSCOhost (accessed April 25, 2010).

[61] William C. Dowling, “Meaningless Grades and a New Dishonesty,” Academic Questions 16, no. 4 (Fall 2003): 58. Academic Search Elite, EBSCOhost (accessed April 7, 2005).

[62] Whitley, 247.

[63] Whitley, 262.

[64] Taylor, 408.

[65] Dowling, 58.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download