Higher Education Is Not A Game

[NOTE: The following is not a crisis call nor is it meant as an indictment of the integrity of the profession or any particular individual with in it. Instead, it is observation meant to raise consciousness and present PSA’s correction of the gaming problem in higher education systems around the world.]

The Nature of Higher Education Gaming

Gaming the higher education (HE) system is a form of cheating and cheating is a problem for all education systems, at all levels. Along with other known and yet to be conceived means, gaming occurs when: 1) Students exchange homework, use crib notes during exams, and engage in contract cheating such as hiring ghost writers for essays or “gunmen” (as they are called in China) to impersonate and take tests for them or even take the entire course or degree. 2) Teachers artificially inflate class averages, create inappropriately easy assignments or tests, take bribes to inflate individual grades, take bribes to accept students to athletic programs, and engage in acts of fraud and plagiarism in research publication. 3) Higher education institutions (HEIs) accept bribes to admit un(der)qualified students, lie about key stats that affect institutional placement on university ranking systems, admit un(der)qualified students who might contribute to collegiate sports, put pressure on faculty to (for instance) artificially reduce class failure rates.

This happens in public and private education systems across the world, and doesn’t even include the gaming undertaken by recruitment and application agents or those that produce counterfeit documents or sell accreditation – the latter of which my own alma mater was shut down for in 2011. For a more complete survey of gaming in HE, see this article from WENR.

The entire HE system is undermined by gaming. It is a question of reliable honesty, quality and equity. That is, public trust in the quality and equity of the public service of education provided by HEIs - in short, system integrity. Considerable effort has been expended to eliminate or reduce gaming, with mixed results. The corruptible are incorrigible.

PSA must confront the gaming problem as well. I estimate that the model offers a better response than the institutional model. To explain, let’s begin by briefly characterizing the gaming dynamics that form the focus of this post.

First, as the public face of HE, universities and colleges must not be seen to engage in (too much) egregious gaming. The public can tolerate the occasional unqualified student who gains admission because the family has donated money to build a new library or sports stadium; or the admission of unremarkable students because they are remarkable athletes that contribute to inter-collegiate sport competition. Arguably, this behaviour by HEIs is now so entrenched it has become standard operating procedure in the public eye – and perhaps just one among many necessary evils in light of the longstanding fiscal troubles of the HEI model.

But it seems when individuals within the system start colluding in a systematic and organized fashion with students, parents, and recruitment or application services in order to gain admission for unqualified students, the public cries scam. One of the loudest voices in the chorus of condemnation is the HEI. Look at these American and Chinese cases: Stanford University and Renmin University.

However, just as in the corporate world, the distinction between employer and employee action is not always clear. An individual employee might engage in behaviour of which the employer is genuinely unaware or perhaps the employer is aware and even (tacitly) encourages or requires the offensive behaviour of the employee - while the employer publicly denies collusion or complicity. Across these cases, there is neither meaningful distinction in kind nor a clear demarcation of culpability. 

A prime example of this is grade inflation. In an era when higher education is commercialized as a consumer product (diplomas) or service (skills-based training for the job market), the philosophy of customer service has steadily replaced the philosophy of public service. This is coupled with a combination of decreased public funding and increased tuition and contingent labour (adjunct instructors). HEIs have entered a corporate mindset, selling a commercial credence good, with its accompanying emphasis on advertisement, competition and customer satisfaction. These factors have contributed to the steady inflation of grades.

By the mid-to-late 1990s, A was the most common grade at an average four-year college campus (and at a typical community college as well).  By 2013, the average college student had about a 3.15 GPA [see Figure 1] and forty-five percent of all A-F letter grades were A’s [see Figure 2].  If you pay more for a college education in the consumer era, then you of course get a higher grade.  By 2013, GPA’s at private colleges in our database were on average over 0.2 points higher than those found at public schools. (gradeinflationdotcom)

 

Figure 1. gradeinflationdotcom

 Figure 2. gradeinflationdotcom

Education has become a consumer good whose purveyors – HEIs and their faculty employees – operate like Gold’s Gym outlets and their staff of personal trainers. There are, however, fundamental differences that in this context commend the pursuit of exercise over education. In the case of exercise, the customer must be committed to personal improvement, must work hard to achieve it, with the results being objectively measurable and unique to each individual based on natural ability and level of effort. This is evidence of receipt of a real consumer good. In the case of education, apparently the results are spectacular for many, with little effort required, and if all else fails, the desired results can be bought with a bribe. In fact, there are a growing number that feel simply being admitted into college is enough to legitimize customer demands for a desired good: “I qualified and paid my money. Now give me my degree.”

A former university chancellor from the University of Wisconsin, David Ward, summed up this change well in 2010:

…Today, our attitude is we do our screening of students at the time of admission. Once students have been admitted, we have said to them, ‘You have what it takes to succeed.’ Then it’s our job to help them succeed. (gradeinflationdotcom)

I guess it all depends on what one means by "help" and “succeed.” If getting high grades and GPAs are a reliable measure of success, then HEIs are succeeding in the service of education. Of course, they could be doing better – 70-80% of students could be getting As!

The administrators, faculty and students of the Chinese HE system enthusiastically embrace this philosophy of education. The prevailing attitude is that once students pass the National College Entrance Examination (“gaokao”) and are admitted to an HEI, the rest is mere formality. Consider that students in China have an on-time degree completion rate of 97% - except in those subjects where life and death hang in the balance, such as medicine and some types of engineering. Such a figure is only possible where institutions and faculty have decided academic standards need not apply and everyone is "helped" to pass regardless of performance.

Similar corrupt gaming dynamics are found throughout the world.

PSA Response to Gaming

PSA is not a panacea for the problems faced by the current HEI model, but it does offer viable, vast improvement. With respect to gaming, four key components of PSA help to control its occurrence: 1) HEIs are assigned an elective and altered role in HE; 2) Evaluation is done by anonymous external crowd-sourcing; 3) Academics across the system are self-employed; and 4) Academic income is respectable. Ultimately, the PSA response is a combination of prevention and penalty.

With HEIs repositioned as mere vendors for core professional academic practices of the system, their current gaming habits are neutralized. They would be unable to admit un(der)qualified students in return for bribes or gains in collegiate sports. They would no longer be able to influence the teaching and evaluation practices of faculty to artificially inflate their institutional reputations in the competitive HE consumer market. Professional academics are not employees of HEIs, as they are now. Instead, they are self-employed professionals, operating independent academic practices as members of a professional society. Consequently, in the PSA model HEIs hold no sway over professional academic practice. They are, as they should be, mere facilitators of the core teacher-student relationship in HE.

It might be argued that in the absence of institutional employers the potential for gaming in PSA is even greater. As with other professions such as psychiatry or dentistry, professional academics are paid directly by clients (students) for their services. A wealthy family, with an underachieving child, could simply bribe the academic to issue stellar grades to the student – even when poor or no work has been done by the student. Bribery could be used to secure the services of a high profile academic, when normally the academic would not accept such an unimpressive student. The gaming now done by HEIs and their faculty employees simply falls to the professional academic in private practice, with no HEI employer to offer possible policing or penalty.

However, this is mistaken because it overlooks certain features of the PSA model.

Academics provide their service within the protection and direction of their professional society. Part of the function of a professional society is to police and penalize its members for failing to conduct themselves according to the law or their professional codes of ethics and conduct - as occurs in the professions of medicine and law. So, there is oversight.

Of course, many are understandably distrustful of professional oversight, especially in law and medicine. However, that’s insufficient reason to dismiss the proposal to legislatively form a professional society of academics, with appropriate legal stipulations to protect societal interest in the public good of education. We might be justifiably cynical about the legal and medical societies, but that cynicism can offer insight in the construction of a new professional society of academics. And anyway, PSA offers other more tangible mechanisms to control gaming by academics.

A key component of the PSA model is external marking. That is, all student work produced for formal evaluation that contributes directly to the final grade is neither marked nor solely constructed by the academics providing the instruction service – whom we might simply call, Instructors. This is accomplished through the following mechanisms: 1) All evaluation material (e.g., assignments, tests, exams, presentations, projects, models, designs, etc.) are constructed by the Instructor, but are subject to crowd-source academic review and approval by what we might call, Evaluators. Or, using the same sort of crowd-sourcing, the Evaluators could rate the standard of the evaluation material at the same time as they mark student work. This would result in two grades – one for the evaluation material and another for the student’s performance. In either case, all of this is conducted with anonymity for all parties. 2) Along with the commonplace student-completed Instructor/Course evaluations, each Instructor has their student and evaluation material grades made public, with anonymity for the students ensured. 3) Student full academic profiles are made public, with anonymity for the students ensured. 4) Accessibility to all this information is aided by an informative statistical analysis tool and database.

This is the sort of information that should be made available to all parties within and without the HE system so that they can make decisions in pursuit of their respective interests. For instance, academics, students and employers can better decide if the education and employment relationships they enter into are suitable: Should I take on this student? Do I want this academic to provide me instruction? Is this a qualified applicant for the position?

 Consider kindred legislation recently introduced in the state of Texas and the existing self-reporting practices at universities such as Columbia, Dartmouth, Indiana University, and Eastern Kentucky University. The aim is to contextualize grades by providing on transcripts information such as the number of students, grade distribution, the average grade, grade point average (GPA), and average student GPA in each class.

This transparency requirement would yield big dividends. Employers, as well as graduate admissions committees, would welcome the better information that comes with grade contextualization. Students would get a truer sense of what their abilities are and where they truly stand. Transcripts would again be the indicators of accomplishment they were intended to be. (Forbesdotcom)

Though these practices demonstrate a will to correct grade inflation gaming, they seem ineffectual - though they can help to expose the scale of such gaming, where for example 40-50-60% of students in a course receive a grade of A.

External evaluation of the sort offered by PSA does more to reduce the possibility of grade inflation and helps establish measurable student performance standards without the need for standardized evaluation materials (e.g., tests or exams). With external crowd-sourced evaluation, individual academics are unable to hand out grades, since they are not solely responsible for the marking. Evaluation materials designed to make their completion an easy task for students are identified and appropriately graded or ultimately eliminated. Thanks to the publicity of student and evaluation material grades, anyone wishing to assess the performance of both students and academics can do so. A student who receives a grade of 90% on an assignment which is reviewed as having a difficulty level of (say) 3 or 4 out of 10, is not that impressive, when compared to a 90% on an assignment reviewed at 8 or 9. In this way, academics have no incentive to game the evaluation materials, since this would be publicly exposed by professional peer review. At the same time, should the monetary incentive offered by a wealthy family prove too tempting, then Instructors are free to take on the un(der)qualified student and work (perhaps in tandem with a qualified TA hired by the practice) to ensure the student improves enough to perform as desired when reviewed by the external Evaluators. This is providing good, quality-controlled education service.

When you combine this with the publication of the student’s full academic profile, there is now rich contextual information for decision-making and less room for inflation that dilutes or destroys the value of evaluation and credentials. And none of this requires exclusive use of numerical or alphabetical grades at the expense of written evaluations, which are preferred and recommended by some HEIs (e.g., Evergreen State College, Sarah Lawrence College, and Antioch University).

As an improvement, PSA-style evaluation at least blocks the ubiquitous secretive, subtle, often self-deceptive road to evaluation inflation that is motivated by the customer satisfaction goals of their institutional employers and the desire to keep their job, secure promotions or simply make their job less thorny (e.g., vetting fewer complaints from students). The externality and anonymity of the Evaluators makes them immune to these concerns. They cannot be fired, passed over for promotion or inundated with student complaints or inquires because of their professional judgement. Under such circumstances, a crowd-sourced cross-section of Evaluators is far less likely to inflate alpha-numerical grades and can temper any written evaluation that an Instructor might be tempted to inflate.

The technological and administrative systems necessary to facilitate this sort of external evaluation are widely available and, in some cases, already in place. Consider the existing bureaucratic, administration, accreditation and national examination (e.g., gaokao in China) bodies found throughout the world - not to mention the increasing quality and use of AI in evaluating student work, which has none of the human flaws associated with gaming. As needed under the PSA model, these human and technological resources, which are now committed to the failing and corrupt HEI model, could be enlisted by the professional society of academics. In fact, accreditation bodies are populated by academics who are or have been employed by the very HEIs they accredit. PSA simply suggests relying on the experience and expertise of these academics in professional society to not only license themselves to practice HE (rather than HEI middlemen), but to set and apply evaluation standards across the system.

This is not to say that the PSA-style of evaluation is flawless where gaming is concerned. The possibility remains that a critical mass of professional society members might engage in a quid pro quo expectation where giving generous grades encourages reciprocity across the system. This tacit collusion could inflate grade and narrative evaluations. After all, in order to earn a living, each member has an interest in building an attractive academic practice, in much the same way as HEIs want to attract students.

This is possible, but the concern is diluted by the following salient points. First, since the Evaluators are anonymous and not subject to the pressures that faculty experience as institutional employees, they are freer to act according to their professional conscience. It is important not to underestimate the persuasive power that senior administrators and department faculty members can have over the general faculty body. In the competitive, corporate atmosphere of the current HEI model, there is opportunity and incentive for each university and college to remain willfully ignorant of or knowingly complicit in gaming violations in the interest of promoting or protecting institutional reputation and revenue. In the HEI model, even if there exist faculty who would prefer to see such violations exposed, their institutional employers can resist them, using various subtle and overt means to silence them. Though the freedom to resist or contradict an institutional employer is purportedly provided by tenure and union representation in the HEI model, the reality is that tenure is a very rare status, while unions impose their own unnecessary detrimental costs on the HEI model.  

Academics in PSA are self-employed and so can voice their opinions and direct their profession without the threat of repercussions from an institutional employer. From within a professional society of peers who have anonymous control over a key component of HE – evaluation - there is true academic freedom to protect and direct the integrity of the HE system against gaming, and thereby uphold the reputation of the profession. It is true that the reputations of individual academic members will also play a role in the integrity of the PSA model. There will be those members who have good and those who have tarnished reputations, those who provide good and those who provide poor service. But as we shall see in a moment, under PSA, this will be public record.

Second, generating a respectable income in PSA is unproblematic for academics. In the US, calculations indicate that the model can support a 48% increase in the number of academics providing service, each with an annual practice revenue ranging from $120,000-$200,000 USD and income range of $88,800-$137,600 USD. This is well above the average income for faculty in the current HEI model and does not include other sources of revenue common to HEIs (e.g., research grants, tax exemptions, etc.). To earn such revenue the baseline teaching load is a full-time equivalent teacher-student ratio of 1:16 or 80 full (160 half) credit course registrations – or 80 to 160 students per annum. This is already respectable pay for a moderate workload that is well below that typically found in the current HEI model and includes exclusive access to a TA working 20 hours per week (earning between $19,200-$38,400 USD per annum). All of this is possible for at least 50% less than the total cost to run the current HEI model, leaving plenty of public financial flexibility in the PSA model. Similar numbers are true for the Chinese HE system.

The fact that this sort of respectable income is available without the need to attract large numbers of students diminishes the level of competition among professional academic practices, as compared to what one finds in the HEI model among universities and colleges with shockingly high operational costs. Even so, because academics have control over their work, some might like to teach more students and so make more money. This could cause practice competition to increase. In such a circumstance, though the PSA model doesn’t explicitly require it, a salary cap could be introduced. Professional sports have introduced salary caps, so surely the same could be done for professional academics who provide a service in which the state has a far greater social interest in stability.

Third, PSA is meant to provide stable, sustainable employment for academics – unlike the contingent employment the HEI model offers 50-75% of faculty. This sort of security removes the pressure to please institutional employers in order to keep their jobs. But it also means that with this sort of security and control over their work, academics can teach without the common research requirements imposed by HEIs, allowing them to spend time improving teaching service or grooming a qualified TA to help provide good education. Of course, they could choose a practice balance where research (as a source of income) dominates revenue and teaching is subordinate. The point is that the range of professional prerogative offered by PSA relieves the pressures associated with part-time, contingent employment.

There is nothing in these three features of PSA that guarantees academics will refrain from collective evaluation inflation. But these features and the very nature of professional society do provide framework within which to better detect and address it, should it take place. In the HEI model there is no effective unifying body that binds universities and colleges to either systemic detection of or response to evaluation inflation. It also lacks the anonymity needed for honest, individual expression of judgements related to quality. And its current financial stresses incline HEIs toward a customer satisfaction philosophy that encourages gaming violations of various sorts. Under these conditions system integrity becomes a matter of every institution for itself.

There remains one further feature PSA that improves system integrity beyond that offered by the HEI model. Though under PSA-style external evaluation academics gain nothing by taking the road of creating easy materials or inflation, they can still engage in more blatant acts of gaming. They might collude with students to ghost write, provide answers to tests and exams, or other yet to be conceived and known gaming. And, of course, students who engage in contract and other forms of cheating might continue to do so independently of Instructor collusion.

However, as a professional society established through an act of legislation designed to safeguard provision of the public interest in education, PSA would have the authority to investigate and penalize its members for violations of law, conduct, and ethics codes. Under such circumstances the self-policing powers entrusted to the professional society can be defined and overseen by the state, and so long as violations are diligently investigated and strictly punished – e.g., the suspension or revocation of the license to practice in academia – the inclination to blatantly game the system is diminished. Where such violations involve (collusion with) students, PSA has the system-wide capacity to apply penalties to students – e.g., probation, temporary or permanent expulsion from the higher education system (not merely one HEI, as one finds under the current model). Any academic who faces the loss of license to legally make a living in the HE system or student who faces the loss of access to HE credentials will think twice before playing any games.

To maintain system integrity, this function of the professional society is done with full transparency, including the publication of relevant information related to violations, violators and penalties (as is done in law and medicine).

With these features of the PSA model in play, it seems premature to suggest that academics in professional society would, as a whole, degenerate to the lowest common denominator and engage in system-wide evaluation inflation in the self-interest of generating income for as little work as possible. Instead, PSA creates an environment where system integrity can more readily be championed against all forms of gaming.

But in the end, even if the internal, highly gameable evaluation practices of the HEI model were maintained, the PSA model is still preferable, offering numerous improvements to the provision of HE – as identified in the posts on this blog and a summary document called, A New Tender for HE. As always, I welcome comment and collaboration.