USF Libraries
USF Digital Collections

Educational policy analysis archives


Material Information

Educational policy analysis archives
Physical Description:
Arizona State University
University of South Florida
Arizona State University
University of South Florida.
Place of Publication:
Tempe, Ariz
Tampa, Fla
Publication Date:


Subjects / Keywords:
Education -- Research -- Periodicals   ( lcsh )
non-fiction   ( marcgt )
serial   ( sobekcm )

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E11-00154
usfldc handle - e11.154
System ID:

This item is only available as the following downloads:

Full Text
xml version 1.0 encoding UTF-8 standalone no
mods:mods xmlns:mods http:www.loc.govmodsv3 xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govmodsv3mods-3-1.xsd
mods:relatedItem type host
mods:identifier issn 1068-2341mods:part
mods:detail volume mods:number 8issue 10series Year mods:caption 20002000Month January1Day 2727mods:originInfo mods:dateIssued iso8601 2000-01-27


1 of 19 Education Policy Analysis Archives Volume 8 Number 10January 27, 2000ISSN 1068-2341 A peer-reviewed scholarly electronic journal Editor: Gene V Glass, College of Education Arizona State University Copyright 2000, the EDUCATION POLICY ANALYSIS ARCHIVES. Permission is hereby granted to copy any article if EPAA is credited and copies are not sold. Articles appearing in EPAA are abstracted in the Current Index to Journals in Education by the ERIC Clearinghouse on Assessment and Evaluation and are permanently archived in Resources in Education Collegiate Grading Practices and the Gender Pay Gap Alicia C. Dowd Cornell UniversityAbstractExtending research findings by R. Sabot and J. Wake man-Linn (1991), this article presents a theoretical analysis showin g that relatively low grading quantitative fields and high grading verbal fields create a disincentive for college women to invest in quantit ative study. Pressures on grading practices are modeled using higher educa tion production functions. The gender pay gap has narrowed in the Un ited States since the 1970s, but is still of sufficient magnitude to warrant concern about th e equal employment and status of women. The decrease in the size of the gap can be e xplained in part by the increasing numbers of college women who responded to expanded opportunities in the labor market and chose to enter technical and applied fie lds, particularly business (Eide, 1994; Loury, 1997). Women entering fields requiring quant itative skills can expect a greater return on their educational investments, because su ch skills are a relatively scarce human capital input (Paglin & Rufolo, 1990). Numerous stu dies have demonstrated that, all else equal, college graduates with quantitative skills w ill earn more than their counterparts without such skills (Berger, 1992; Eide, 1994; Jame s & Alsalam, 1993; Rumberger &


2 of 19Thomas, 1993; Sharp & Weidman, 1989). However, wome n continue to be disproportionately represented in the humanities an d social sciences and underrepresented in mathematics and the applied and physical sciences (National Center for Education Statistics, 1997). The theoretical an alysis presented in this article shows that one way to increase the participation of colle ge women in quantitative fields, and potentially reduce the pay gap even further, is to institute uniform collegiate grading practices in quantitative and nonquantitative field s. Previous research (Kuh & Hu, 1999; Sabot & Wakeman-Linn, 1991) has provided evidence that grade inflation and compression has o ccurred in collegiate disciplines at different rates, creating non-uniform (or divergent ) grading practices. One factor contributing to the underenrollment of women in qua ntitative fields may be the use of relatively high grading practices in nonquantitativ e, or "verbal," fields and low grading practices in quantitative fields. This article has two purposes. The first is to show that grading disparities between academic disciplines ha ve a significant impact on the curricular and career choices of female students. T he second is to apply the analytical tool of the higher education production function to explain pressures on assessment practices from within and outside the academy that lead to divergent grading practices. This analysis also considers from which quarter pre ssure might come to change such practices. The discussion takes account of the publ ic and private nature of institutions of higher education, noting that human capital formati on is not their only, or necessarily even primary, function.Theoretical Framework Students earn college credits and degrees by investing time, money, and effort. At the majority of institutions of higher education, s tudent performance in classes is evaluated with grades, and students must receive pa ssing grades to receive credit for coursework. Students must also earn a sufficient nu mber of credits in prescribed areas to be granted a degree in any given field of study. Va riation in the effort students must expend to successfully complete coursework in diffe rent fields creates variation also in the costs of earning credits in those fields. The f ull costs of that effort will be tempered by a student's motivation and interest. A student might pay the same tuition to m ajor in mathematics or in English, but if she has strong mathematical skills and weak writing skills, she will have to invest more time to earn passing grades in English than in math ematics. Thus, the cost of earning a degree in a given field depends on the effort a stu dent must expend to complete courses with a passing grade, or, for students with higher standards, to be satisfied with his or her own performance. In addition, some fields have more numerous or rigorous requirements, which raises the cost of study in tha t field relative to other fields for any student. (Note 1) The grades students receive infor m them of their area of comparative advantage in completing coursework in a subject, th e probability of successful completion of a course of study, and the costs (in time and effort) of obtaining a degree (Altonji, 1993). The analysis presented in this article is based on an economic approach (Becker, 1976) to understanding the curricular and career ch oices of college students. Educational choices are treated as investment decisions, influe nced by pecuniary and non-pecuniary costs and benefits. By her curricular choices, a st udent determines the specific type of human capital she will acquire. She thereby influen ces potential future returns to the educational investment and her ability to maximize her "utility," or satisfaction. The economic approach to understanding human behavior m akes a number of assumptions


3 of 19about the way in which individuals conceive of thei r well being. Self-interest is conceived of broadly, beyond the pursuit of materia l concerns, to include a wide range of values and preferences. Individuals are consider ed to be forward-looking, to have consistent preferences over time, and to seek to ma ximize their welfare. There are a number of constraints on a person's capacity to pur sue his or her self-interest and these include time, income, incomplete information, and l apses in judgment (Becker, 1996). Altonji (1993) has highlighted the fact t hat individuals make educational choices under considerable uncertainty regarding their abil ity to complete a course of study in their selected field. His analysis (p. 51) models h ow "new information about preferences and academic performance, and new information about payoffs influence choice of major and the decision to stay in school." Within t his human capital framework, as individuals gain new information, they make their c urricular choices, transferring from one field to another or dropping out of college, ba sed on an estimation of their ability to complete degree requirements. The probability of co mpletion is influenced by their stock of knowledge, academic ability, and by degree requi rements. The utility function indicated by Altonji's analysis also includes educa tional and occupational preferences and the present value of lifetime earnings.Prior Research In a 1991 article published in the Journal of Economic Perspectives Sabot and Wakeman-Linn examined the influence of collegiate g rading practices on student course choice. They documented the existence of grade infl ation and compression (low variation) and also observed grading patterns that characterized high and low grading departments. They concluded that students face a di sincentive to study in low grading fields, which, in their study of a small but varied sample of U.S. colleges, were predominantly quantitative fields. They found that economics, chemistry, and mathematics are consistently low grading fields, wh ile art, English, music, philosophy, psychology, and political science are consistently high grading fields. In a survey administered to a small sample of English majors at a research university (Dowd, 1998), I also found that responding students believed that the average grades in biological sciences, physics, computer science, and chemistry at their institution was a B-; in political science, philosophy, economics, and mathe matics a B; and in foreign languages, English, sociology, and history a B+. Co nsistent with Sabot and Wakemann-Linn's study, the low grading fields inclu ded quantitative subjects and the high grading fields included verbal subjects. Davis (1966) argued that college students assess their areas of comparative advantage (where their skills and aptitudes put the m ahead of their peers) based on the local competition for grades at their institution. Students then shape their career plans based on the feedback grades provide. However, Sabo t and Wakeman-Linn (1991) observed that due to varying rates of grade inflati on and compression among academic departments, "grades as a signal of relative streng ths and weaknesses {are} more difficult for students to interpret." They noted (p 167) that students do not adequately adjust their perception of differentially-scaled gr ades in order to gain a sense of their relative strengths and weaknesses, because "the inc entive effects of absolute grades on course choice are far more powerful" than the indic ators of comparative advantage that are weakened by non-uniform grading. Sabot and Wake man-Linn argued that arbitrary differences in grading policies should be eliminate d, because they provide incentives for some students to move away from academic areas wher e they are comparatively strong. Conversely, the effect of more-uniform grading poli cies would be to encourage greater


4 of 19numbers of students to take courses in the currentl y low grading departments, which are those that place emphasis on quantitative skills. W hile the labor market, through high earnings, provides an incentive to invest in quanti tative study, under divergent grading—where quantitative fields are low grading r elative to others—colleges create a disincentive to investment in quantitative study.Divergent Grading and Labor Market Supply The following simple utility-maximizing m odel extends Sabot and Wakeman-Linn's (1991) analysis to highlight the inf luence of non-uniform grading practices, where they exist, on the supply of colle ge graduates with quantitative skills. The model is intended to facilitate a policy analys is of the implications of divergent grading for gender equity in earnings. Under divergent grading practices, when a student decides in which fields of study to invest her time, she faces greater costs to obta in the valuables associated with college study in a quantitative rather than a verbal field. To obtain a certain number of credits in a quantitative rather than a humanities class with a grade of B would on average require more effort, because quantitative classes have lowe r mean grades. The relative costs of the effort to earn a degree through study in quanti tatively or verbally oriented fields may be represented by the ratio EQ/EV where EQ represents the costs, psychic and otherwise, associated with quantitative study, and EV represents the costs associated with verbal study. I assume that this ratio is fixed for each i ndividual (disregarding the fact that costs would vary as students make marginal investments in either field). We can also represent the ratio of the di fferent compensation packages offered by employers to individuals with strong quantitative a nd strong verbal skills as WQ/WV Again, I assume that this ratio is fixed. A forward -looking student with complete information about her future wage potential could d etermine whether to invest in quantitative or verbal study by comparing WQ/W V and EQ/EV If WQ/WV > EQ/E V, she would choose to invest in quantitative study. If WQ/WV < EQ/EV she would choose to invest in verbal study, and if the two ratios are e qual, she would be indifferent to these two options. For example, if the wage ratio is 2:1 (Q:V), then the student should invest her time pursuing quantitative study as long as ear ning credits in quantitative fields is less than twice as difficult (accounting for all co sts, both psychic and material) as earning credits in verbal fields. The forward-looki ng student in this scenario would need to take into account lifelong earnings and career s atisfaction, as well as the continuing education required to succeed at the occupations pu rsued. The college's assessment systems and grad ing policies affect a student's decision to choose to study in a quantitative or verbal fiel d by the fact that the differential between average grades in these two types of fields is one component (along with ability, motivation, and interest) establishing the ratio EQ/E V. As the differential increases, the value of EQ /EV also increases, and a greater number of students w ill determine it is not a wise investment to study in a quantitative field. In this way, the divergent grading system is a contributing factor d etermining the proportion of the population of college graduates who enter the labor market with quantitative skills. Student perceptions of the relative wages offered f or quantitative and verbal skills also influence the proportion of students who enter diff erent fields of study (as Freeman (1978) has illustrated with his cobweb model of cur ricular and career choice). College graduates with different types of interests and abilities encounter different opportunities in the labor market. As strong quanti tative skills are scarce relative to


5 of 19strong verbal skills, quantitative skills are compe nsated at a higher rate in the labor market than are verbal skills. Recent studies indic ate earnings advantages over comparison groups of humanities and education major s of 23% to 61% for engineers, up to 25% for business majors, 13% to 35% for students of mathematics and the physical sciences, and 8% to 24% for social scientists (Angl e & Wissmann, 1981; Berger, 1992; Bishop, 1994; Daymont & Andrisani, 1984; Eide, 1994 ; Griffin & Alexander, 1978; James & Alsalam, 1993; Rumberger & Thomas, 1993; Sh arp & Weidman, 1989). When students are influenced by divergent grading practi ces to invest in verbal skills rather than in quantitative skills, the supply of verbal s kills provided by college graduates to the labor market increases over the supply of gradu ates who would have made this choice, given their aptitudes and interests, under uniform grading practices. Labor economic theory indicates that the impact of this s upply shift would lead to a decrease in wages paid to graduates offering verbal skills to e mployers (Ehrenberg & Smith, 1993).Influences on the Curricular Choices of Women Divergent grading leads to a greater quan titativeskills deficit among women than among men for several reasons. The first relates to the distribution of quantitative skills among men and women. In the population of college-b ound high school graduates, women are less likely to be among those with the st rongest quantitative skills. In addition, the measured quantitative and verbal skil ls of men show greater variance than that of women (Cole, 1997), and those students at t he tails of the quantitative and verbal skills distribution are least affected by divergent grading. Students who have average skills in both quantitative and verbal fields are t hose who are most likely to receive misinformation about their comparative skills advan tage as a result of low grading in quantitative fields and high grading in verbal fiel ds. On the basis of their abilities, these students should be indifferent regarding choice of field. However, the degree of misinformation they receive is the full difference between average quantitative and verbal grades, and they are then motivated to choos e verbal fields. Students with close to average quantitative and verbal skills are also lik ely to receive erroneous feedback. Students with a quantitative/verbal skills differen tial so large that the grading differential does not change the direction of the signal regardi ng their area of comparative advantage are not affected. Second, women may be more affected by the quantitative/verbal grading differential because they may already face higher c osts of study in quantitative than in verbal fields as a consequence of participating in a learning environment that is oriented toward men. Sandler, Silverberg, and Hall (1996) ha ve described a "chilly classroom climate" for women, which is exacerbated in traditi onally male fields. In such a climate, women would experience psychic costs as they find t heir intellects and class contributions devalued. In particular, the competit iveness of study in quantitative fields relative to verbal fields may create high costs for women who pursue quantitative study (Dowd, 1998; Strenta, Elliott, Adair, Matier, & Sco tt, 1994). Even when women have equal measured abilities and aptitudes as men in qu antitative fields, they have been found to enjoy science courses less than their male counterparts and to choose at greater rates to exit the field (Ware, Steckler, & Leserman 1985). Prior research has shown that women persist in quantitative fields at greater rat es if they attend women's colleges (Jacobs, 1996; Solnick, 1995), which suggests that women find a more welcoming environment in all-female classes, experiencing low er costs than those imposed by a malecentered environment. However, the findings o n the effect of women's colleges on female educational attainments are not conclusive ( Riordan, 1994; Smith, Wolf, &


6 of 19Morrison, 1995). Finally, women may also give greater weig ht in making their curricular choices to their present or "local" status, to use Frank's ter m (1985), in the collegiate environment than to their future economic status. Loury (1997) found that women are less motivated than men by the college wage premium in making the decision to attend college. Frank (1996) and Daymont and Andrisani (1984) found that women place greater value than men on moral and personal dimensions of career sati sfaction. These findings suggest that women are less concerned than men with future monetary returns to education. This disinterest may cause women to spend less time acqu iring information about salaries and to underestimate the relative economic returns to q uantitative and verbal fields of study. Disinterest may also be fostered by greater uncerta inty concerning labor market participation, due to the fact that child-rearing r esponsibilities often interrupt women's careers. As Polachek (1981) observed, the prospect of discontinuous employment may provide an incentive for women to acquire human cap ital that does not depreciate quickly during their time outside the labor force a nd lead them to avoid rapidly changing technological fields. However, England (1982) count ered that available data do not support this hypothesis.The Higher Education Production Function The discussion above has shown that diver gent grading creates a disincentive to study in quantitative fields. Further, it demonstra tes that these disincentives are likely to have a greater influence on the curricular choices of women than of men. At this point, beginning with an overview of relevant aspects of s everal theories of the higher education production function, I evaluate the facto rs creating patterns of low and high grading in quantitative and verbal fields of study. The need for and purposes of grading can be understood as part of a higher education pro duction function, and the existence of divergent grading practices suggests that quantitat ive and verbal fields experience a different kind or degree of pressure to produce gra des. Production functions consider the outcome s of schooling as educational "outputs" resulting from various inputs including faculty, qu ality of students, and physical and financial capital. The demand for these outputs, wh ich include teaching, research, and public service, comes from students, private and pu blic funding agencies, and donors (Garvin, 1980; Hopkins, 1990; Hopkins & Massy, 1981 ; James, 1990). Production functions typically are based on the assumption tha t the goal of a private firm is to maximize profits. It is further assumed that market forces create an imperative that firms produce at the most efficient technological boundar y of production. These assumptions do not apply to higher education, however, and in m odifying the production function model for the higher education context, researchers have proposed several other objectives, including the maximization of administr ative scope, income, and prestige. The role played by grading in the production functi on varies depending on the outcome to be maximized. Niskanen (1971) described universities as "mixed bureaus," non-profit organizations with public and private characteristi cs, due to the fact that they are funded through grants as well as through revenues generate d by selling their output at a per-unit rate. He viewed universities as income-maximizers, whose administrators and faculty gain utility by increasing the size and scope of th eir bureaucracy. Breneman (1976) observed that faculty members seek to optimize depa rtmental prestige, and Garvin (1980) elaborated on this and other research to dev elop a model of institutions as a whole as prestige maximizers. Faculty members gain utility from increasing levels of


7 of 19prestige associated with their departments in the f orm of higher salaries, better quality graduate students who can be attracted at a lower p rice, higher caliber colleagues, and greater success rates in seeking internal or extern al funding. Zemsky and his colleagues drew on element s of the prestigeand bureaucracy-maximizing utility models to argue that faculty members increasingly expend their energies toward individual goals, away from the goals of the institution (Pew Higher Education Research Program, 1990; Zemsk y, Massy, & Oedel, 1993). They attribute this phenomena to misplaced incentive str uctures that motivate faculty to focus on their research at the expense of teaching and ad vising. Faculty members maximize prestige in their disciplinary labor market by publ ishing academic papers. Teaching, the quality and value of which is difficult to present to external observers, carries little reward, they argued.The Demand for Grades The prestigeand bureaucracy-maximizing production models of higher education provide a theoretical basis for examining the chara cteristics of high and low grading departments. In this section, I extend these models to explain the pressures on departments at four-year research institutions to a dopt high or low grading practices. I also use a utility maximization analysis to describ e the interests students have in the prestige of their institutions and the demand they create for grades. As Breneman (1976) and Garvin (1980) have illustrated theoretically and empirically, departments at research universities m aximize prestige through research and scholarly output. They can increase their output by hiring very productive faculty members or by increasing the total number among the faculty. As increasing student enrollments provide a rationale for additional facu lty hiring, there is a derived demand for a larger quantity of students. As faculty membe rs prefer to work with talented students, there is also a demand for higher quality students. When departments attract external research funds from the government, founda tions, or corporations, they can afford to lose a share of university resources allo cated on the basis of student enrollment. The availability of external funding creates pressu re to "weed out" less talented students and reduce enrollments. Departments that attract a lesser share of external research dollars will attempt to maximize enrollment, a goal that would relax pressures for competitive grading practices intended to dissuade the least capable students to leave the field. Under certain conditions, students themse lves create a demand for competitive grading, in a way that the other agents in the high er education output-demand system do not. Funding agencies, such as the government and f oundations, are primarily interested in the outputs of research and teaching, as they ma ke investments in higher education to develop public goods and promote social welfare. Fo r students, higher education is both a consumption and an investment good (Schultz, 1961 ). The immediate value of their consumption is affected by the quality of teaching and learning, including modes of assessment. The value of their investment benefit i s influenced by the status of their college (Heath, 1993). Heath (1993) has illustrated theoreticall y that students value both local and global status, where local status is defined as a student' s academic standing at her institution. As was discussed above, local status informs a stud ent's understanding of the investment costs of completing a degree in any given field of study (Altonji, 1993). Local status also has psychic costs and benefits (Frank, 1985) and co ntributes to determining the consumption value of a student's education. In Heat h's analysis, global status is


8 of 19determined largely by a college's ability to place graduates in high paying occupations and in graduate and professional programs. Global s tatus is influenced by an institution's academic rigor and the quality of enrolled students with greater rigor attracting an academically talented student body. Students value the positive effects of higher standards on their global status, but fear the pote ntially negative effects on their local status and the increased costs of completing their work. Student interest and influence on collegi ate grading practices stem from their investment and consumption decisions. Students can be expected to endorse competitive grading practices, in which performance is graded o n a curve and where average grades are low relative to other fields, if they perceive that such practices enhance their global status and ability to compete for high paying jobs. Students who are competing for scarce places in lucrative professions will have th e greatest concern for global status. Under heavy interest, access to an occupation becom es limited and institutions have a prestige-maximizing incentive to certify only a por tion of their students for entry into that field. The response to this incentive is the a doption of assessment practices that are designed to motivate or require those who are least capable to leave the field of study (Breneman, 1976). Students who are not career oriented and who place a greater value on higher education as a consumption good can be expected to resist competitive grading and to avoid such practices when making their course choic es, because it imposes immediate psychic costs and reduces the consumption value of their classes. If a field of study does not provide a closely articulated link to lucrative and competitive career paths, students will demonstrate a lack of interest in the credenti aling function of grades. These students may value grades intrinsically as a reflection of t heir talents, but they do not create a demand for comparative rankings. In the absence of preprofessional student pressures, the field has an incomeand resourcemaximizing i ncentive to become high grading in order to attract enrollment. In summary, the prestige-maximizing and b ureaucracy-maximizing model of higher education production provides a theoretical basis for understanding the pressures on collegiate grading practices. External research dollars enable departments to maximize prestige and income while "weeding out" th e least successful students from their programs. Student careerism also creates pres sures for competitive grading, as students wish to enhance their global status. The m odel clearly predicts the behavior of departments experiencing a combination of low stude nt careerism and low external funding (high grading practices) and high careerism and high external funding (low grading practices). As quantitative and applied fie lds are influenced much more greatly by research interests and strong links to employers than are arts and letters fields (Becher, 1989; Breneman, 1976), they are more likel y to adopt low grading practices to maximize prestige. Verbal fields, with weak ties to employers and low levels of research funding, are more likely to adopt high grading prac tices to maximize administrative scope and enrollment.Traditions of Scholarship The educational production function ident ifies the utility-maximizing goals of scholars in different disciplines and provides a mo del that predicts grading practices in response to different output-demand systems. Intern al features of departments stemming from disciplinary traditions and epistemologies may also account for different assessment practices. In Academic Tribes and Territ ories, Becher (1989) characterizes modes of scholarship in academic disciplines. His f our-part taxonomy of "hard pure,"


9 of 19"hard applied," "soft applied," and "soft pure" fie lds bears resemblance to the simpler quantitative/verbal dichotomy I have used. Hard fie lds are quantitative and soft fields, which include the humanities, social sciences, and "social professions" (education, social work, law), may or may not employ quantitative anal yses. The applied fields, whether hard or soft, are those whose research practices ar e influenced strongly by practitioners and a search for practical knowledge. Becher's appl ied fields are those I have described as having ties with employers. Whether these employ ment relationships influence grading practices depends on the level of competiti on among students for entry into related occupations and professions. These relation ships can be influential in a hard applied field, such as engineering, as well as in a soft applied field, such as business. As Becher (1989) indicates, the modes of scholarship in the applied fields follow from those of their pure counterparts, but are alte red by the focus of applied fields on generating solutions to practical issues outside ac ademe. For this reason, the epistemological distinctions that Becher observes b etween hard pure fields (natural sciences and mathematics) and soft pure fields (hum anities and social sciences) describe the predominant disciplinary traditions and culture s that may influence grading practices. He offers a set of contrasts that, in sum, indicate that hard pure fields have a more clearly defined body of knowledge than the soft pur e fields. First, Becher (1989, p. 13) observes, knowledge in hard pure fields is cumulati ve through the work of generations of researchers building on each others' findings re levant to clearly defined and bounded questions. In contrast, soft pure fields address is sues that retain their currency over time. Researchers in soft pure fields make contributions, not by generating new knowledge, but by providing insights into familiar topics. Sof t pure fields lack the clear boundaries that specify areas of investigation in hard pure fi elds. Second, while hard pure fields "break down complex ideas into smaller components," in soft pure fields "complexity is regarded as a legitimate aspect of knowledge, to be recognized and appreciated" (p. 14). Third, in hard pure fields, scientists make "strong arguments based on mathematical models, measurement, and observed regularities. In soft pure fields, where explanation revolves around numerous concepts and the absence o f clearly defined variables, scholars make apparently weak arguments and rely mo re heavily on "judgment and persuasion" (p. 14). Finally, soft pure knowledge r ecognizes and admits the "intentionality" of the scholar, while hard pure fi elds convey knowledge as "impersonal" and "value-free" (pp. 14-15). Becher, himself, does not comment on diff erences in assessment practices between fields. This likely stems from the fact tha t participants in his case study at "elite departments" defined their membership in their acad emic professions "in terms of excellence in scholarship and originality in resear ch, and not to any significant degree in terms of teaching capability" (p. 3). For this same reason, grading practices may be given peripheral attention, be little affected by discipl inary norms, and be easily modified by external influences. Or, they may follow closely fr om the research traditions. In the latter case, the openness of soft pure fields to divergent viewpoints combined with the acceptance of unresolved complexities in subject co ntent would be consistent with assessment practices that allow numerous "correct" answers. In contrast, hard pure fields would be expected to rely on assessment practices t hat test students' abilities to convey their understanding of established subject content and to make greater distinctions between right and wrong answers. Low grading practi ces in hard pure (quantitative) fields and high grading practices in soft pure (ver bal) fields may, therefore, have epistemological roots. This explanation is not comp letely persuasive, however, because the soft pure fields awarded lower grades on averag e in earlier times (Kuh & Hu, 1999; Sabot & Wakeman-Linn, 1991). Understanding of the r elative influence of external


10 of 19demands and internal traditions of scholarship on a ssessment practices would require a study of changes in external and internal departmen tal environments in relation to changes in grading over time. To my knowledge, such a study has not yet been conducted.Empirical Tests Though little research has been conducted that tests the predictions of the production function model of grading practices, two recent studies present relevant findings. Freeman (1999) investigated the predicted relationship that departments with graduates entering lucrative professions have low a verage grades. He hypothesized (p. 344) that "given equal money prices per credit hour across disciplines, departments manage their enrollments by ‘pricing' their courses with grading standards commensurate with the market benefits of their cour ses, as measured by expected incomes." Using data from the National Center for E ducation Statistics on 648 U.S. institutions of higher education, he confirmed that fields associated with higher starting salaries had lower GPAs than those associated with greater "income risk" (p. 350). His research provides evidence that departments manage student enrollment through their grading practices. Those experiencing higher studen t demand due to positive salary prospects for graduates are more likely to grade mo re rigorously. Freeman's work did not also estimate the influence of available research d ollars on grading practices. Kuh and Hu (1999) investigated the causes of grade inflation from the mid-1980s to the mid-1990s, providing evidence that average g rades have increased during that time period. However, their models do not include v ariables representing changes in labor market returns to field of study or changes i n availability of externally funded research dollars, so the work does not provide a te st of the production function model of grading practices. Their results do provide some re levant empirical evidence to evaluate the model, however. Using a large national data bas e including students from approximately 600 four-year colleges and universiti es, they find (p. 306) that grades in the humanities increased at a faster rate than grad es in science and mathematics, with the grade increase in the science and mathematics clust er observed to be minimal. This finding supports the idea that quantitative fields, which have greater opportunities to attract research support, are resistant to inflatio nary pressures on grading. Grades in the social sciences and preprofessional fields were on average lower than those in science and mathematics, which, if the included social scie nces were applied fields, supports the aspect of the model that indicates that preprofessi onal students will create a demand for rigorous grading. In addition, Kuh and Hu found (p. 304) th at while "grades increased across the board the increases were greatest at [research univ ersities]," which suggests that some fields at research universities felt the greatest p ressure to increase grades. Under the production function model, these fields are expecte d to be those attracting few external research dollars, though they could only have had t he observed impact on the average grades if they were, indeed, departments with high enrollments. However, disaggregating the broader results, Kuh and Hu find (p. 314) that grades in general liberal arts colleges and in the humanities and soc ial sciences were actually deflated in private institutions during the period under study. These findings may provide evidence contradictory to the production function model. Alt ernatively, they may indicate that humanities and social science fields without a sign ificant preprofessional student body do not assume inflationary practices unless they ar e in a competitive situation with low grading preprofessional and researchoriented fiel ds, which are more likely to be found


11 of 19at public and research universities. The latter int erpretation of their results is appropriate if the sample included a significant number of priv ate liberal arts colleges among the private institutions, but it is not possible to dra w this conclusion from the article.Discussion The existence of divergent grading indica tes that high grading and low grading departments are subject to different output-demand systems for grades. Institutions themselves are not likely to insist on uniform grad ing practices across their departments without a change in that demand system. If we assum e that departments are maximizing their utility under existing practices, from what q uarter might change toward uniform grading come? As discussed above, students, with th eir sometimes conflicting interests in global and local status, and agencies such as co rporations, foundations, and the state, with their interests in the outputs of research and teaching, are the primary consumers of higher education. In this section, I discuss the po tential motivations of the state and of students to create a demand for change. Foundations with an interest in social justice and economic development may play a role analogous to t hat of the state discussed below. Corporate sponsors of research will be most interes ted in private returns to their investments, but corporations too have an interest in an adequate supply of college graduates who have quantitative training. As a matter of social justice, the state has an interest in promoting equal employment opportunities for women. As a matter of economic development, it has an interest in encouraging women to develop human capi tal in quantitative fields if market mechanisms are not providing an adequate incentive. Through research grants and internship programs, in its role as an employer, an d through direct funding of colleges and universities, the state creates a demand for re search and teaching. Through specialized programs, it structures some of that de mand to create opportunities for women. These opportunities do not attract as many w omen in the presence of divergent grading as they would under uniform grading (as som e women continue to choose verbal fields despite the offer of an incentive, due to th e higher cost of earning a degree in a quantitative field). The state could potentially in crease the enrollment of women in quantitative fields by putting regulative pressure on colleges to adopt uniform grading practices. However, as Strike (1997) has argued, whe n state regulatory processes require educational institutions to promote human capital f ormation as the goal of schooling, the resulting regulations promote a particular concepti on of what constitutes a good life. Such an intrusion as defining human capital formati on as the goal of education, at the exclusion or expense of other legitimate schooling goals, is beyond the purview of the state. Colleges and universities do not have an obl igation to motivate female students to plan their educational investments with an eye towa rd future economic success. The traditional liberal arts curriculum has been intend ed to produce people who are "virtuous, of good taste and liberated interests" ( Strike), not people whose educational and life goal is to attain high earnings. Liberal a rts colleges may very legitimately wish to structure the curriculum, including grading prac tices, to require or encourage students to take liberal arts courses. If liberal arts colle ges choose to promote enrollment in liberal arts courses by intentionally lowering the psychic costs of study in those courses, that approach may well be consistent with instituti onal goals. Pressures for uniform grading might therefore come from the state, not in a regulatory mode but in its capacity as a consumer. The state addresses its human capita l concerns by supporting educational programs that provide training in areas it deems va luable, thereby increasing the


12 of 19attractiveness of those areas to prospective studen ts (by reducing associated tuition costs or by providing enhanced instructional facilities, for example). To further increase enrollment of women in quantitative fields, the sta te could attempt to alter aspects of the learning environment in those fields that create gr eater costs for women than for men. As competitive learning environments appear to plac e a particularly onerous burden on women (Dowd, 1998; Sandler et al., 1996; Strenta et al., 1994), the creation of non-competitive workshops, internships, research pr ojects, or other opportunities of this type may serve to attract women to the study of mat hematics and science. Non-graded instructional programs in quantitative fields could rely on other types of assessment to provide students with an incentive to learn the mat erial presented. Such programs would provide certification of the attainment of threshol d levels of knowledge, but would not provide comparative rankings. The instructional pro gram would be structured to allow students multiple opportunities, as needed, to acqu ire the skills and knowledge necessary to capitalize on their investment in the labor mark et. Such an approach may be less efficient than using competitive grading to identif y the most able students, but may be more efficient in fostering occupational gender equ ity. Astin (1990) has advocated a "talent development approach" to assessment in high er education, arguing for noncompetitive assessments on the basis of both equ ity and efficiency. Demand for competitive grading in verbal fields might be created by trends in student enrollment. As the human capital model indi cates, both grades and the present value of lifetime earnings are part of the equation determining the best human capital investment for a particular student. If the earning s associated with verbal fields of study fell so low as to outweigh the benefits of high gra ding, enrollment in verbal fields would fall. In that case, colleges might seek to create b etter links with employers for liberal arts graduates in order to place graduates in higher pay ing positions and to bolster enrollments. One way to establish these links would be to take an active role in supplying the most talented students to those labor markets. Such an approach would lead to comparative grading practices that would be ar more resemblance to grading practices in quantitative fields. Alumni donors might support such developm ents, because the increased success of graduates in the labor market would enhance inst itutional prestige. As Heath (1993) observed, alumni benefit most from increases in an institution's prestige, experiencing positive benefits related to their alma mater's enh anced reputation, without having to pay the costs associated with the academic competition of a higher quality student body. Alternatively, alumni might decry the professionali sm of liberal arts programs and oppose new practices. The effect of their influence would depend on whether alumni donations are of a sufficient amount to motivate in come-maximizing behaviors. Liberal arts colleges and departments do not have an ethical obligation to ensure access to employment information for their students but they may benefit themselves by enabling their students to more efficiently estimat e their future utility and to make investments in course choices that will maximize th eir financial return. If the college's graduates are able to maximize their utility in the labor market at a higher level after having had access to employment information while i n college, the graduates would be able to achieve higher levels of both income and ca reer satisfaction. Such an outcome would increase alumni donations, as well as the dem and from prospective students for a liberal arts education.Conclusion I have presented a theoretical model, bas ed on various explications of a higher


13 of 19education production function, to explain the deman d for college grades. I have described student assessment as part of the process of producing educational outputs. The practice of high grading in verbal fields and l ow grading in quantitative fields was placed in the context of the different levels of de mand placed on those fields for the outputs of teaching and research. Low grading field s are predicted to experience high demand by preprofessional students for entry into o ccupations with scarce positions and/or a high demand for research. The opposite dem and system would affect high grading departments. Students who are concerned wit h entering a lucrative and competitive profession will create a demand for rig orous grading as it contributes to the prestige of the institution and to their own "globa l status," or value in the labor market. Students who are less career-oriented will place gr eater value on the consumption benefit of a college education and be concerned wit h the quality of teaching and learning and the value of their own "local status," or acade mic standing. Evidence from prior research was presented to show that women are more influenced than men in their choice of major by local status concerns, leading t hem to disproportionately choose high grading verbal fields. Thus, divergent grading crea tes an incentive for women to underinvest in quantitative fields of study, and, thereb y, contributes to occupational sex segregation and the gender pay gap.NotesSee Hoenack and Weiler (1975) for a discussion of t he potential impact on university administration of charging different tui tion rates by field of study. 1. While this simple model refers to an either/or inve stment in two different kinds of study, the argument could be extended to evaluate m arginal investments in quantitative and verbal subjects and to take accoun t of the different returns to various subfields. 2. This article is based on the author's dissertation research. 3.ReferencesAltonji, J. G. (1993). The Demand for and Return to Education When Education Outcomes Are Uncertain. Journal of Labor Economics, 11 (1), 48-83. Angle, J., & Wissmann, D. A. (1981). Gender, Colleg e Major, and Earnings. Sociology of Education, 54, 2533. Astin, A. (1990). Educational Assessment and Educat ional Equity. American Journal of Education, (August) 458-478. Becher, T. (1989). Academic Tribes and Territories: Intellectual Enqui ry and the Cultures of Disciplines The Society for Research into Higher Education an d Open University Press.Becker, G. S. (1976). The Economic Approach to Human Behavior Chicago: University of Chicago Press.Becker, G. S. (1996). The Economic Way of Looking at Behavior: The Nobel Lecture Stanford University, CA: Hoover Institution on War, Revolution, and Peace. Berger, M. C. (1992). Private Returns to Specific C ollege Majors. In W. E. Becker & D.


14 of 19R. Lewis (Eds.), The Economics of American Higher Education (pp. 141171). Boston: Kluwer Academic Publishers.Bishop, J. (1994). The Economic Consequences of Sch ooling and Learning (Draft report). Ithaca, NY: New York State School of Indus trial and Labor Relations, Cornell University.Breneman, D. W. (1976). The Ph.D. Production Proces s. In J. T. Froomkin, D. T. Jamison, & R. Radner (Eds.), Education as an Industry (pp. 3-52). Cambridge, Massachusetts: Ballinger Publishing Company.Cole, N. S. (1997). The ETS Gender Study: How Femal es and Males Perform in Educational Settings. Princeton, NJ: Educational Te sting Service. Davis, J. A. (1966). The Campus as a Frog Pond. The American Journal of Sociology (July), 17-31.Daymont, T. N., & Andrisani, P. J. (1984). Job Pref erences, College Major, and the Gender Gap in Earnings. Journal of Human Resources, 19 (3), 408-428. Dowd, A. C. (1998). Influences on Curricular Choice s in College: A Case Study of Female English Majors from a Human Capital Perspect ive. Unpublished Dissertation, Cornell University, Ithaca, New York.Ehrenberg, R. G., & Smith, R. S. (1993). Modern Labor Economics: Theory and Public Policy (Fifth ed.). New York: HarperCollins College Publ ishers. Eide, E. (1994). College Major Choice and Changes i n the Gender Wage Gap. Contemporary Economic Policy, 12 (April). England, P. (1982). The Failure of Human Capital Th eory to Explain Occupational Sex Segregation. Journal of Human Resources, 17 (3), 358-370. Frank, R. H. (1985). Choosing the Right Pond Oxford: Oxford University Press. Frank, R. H. (1996). What Price the Moral High Grou nd? Southern Economic Journal, 63 (1), 1-17. Freeman, R. B. (1978). The Overeducated American New York: Academic Press. Freeman, D. G. (1999). Grade Divergence as a Market Outcome. Journal of Economic Education (Fall), 344351. Garvin, D. A. (1980). The Economics of University Behavior New York: Academic Press.Griffin, L., & Alexander, K. (1978). Schooling and Socioeconomic Attainments: High School and College Influences. American Journal of Sociology, 84 31947. Heath, W. C. (1993). Choosing the Right Pond: Colle ge Choice and the Quest for Status. Economics of Education Review, 12 (1), 81-88. Hoenack, S. A., & Weiler, W. C. (1975). Cost-Relate d Tuition Policies and University


15 of 19Enrollments. Journal of Human Resources, 10 (3), 332-360. Hopkins, D. S. P. (1990). The Higher Education Prod uction Function: Theoretical Foundations and Empirical Findings. In S. A. Hoenac k & E. L. Collins (Eds.), The Economics of American Universities (pp. 11-32). Albany: State University of New York Press.Hopkins, D. S. P., & Massy, W. F. (1981). Planning Models for Colleges and Universities Stanford, CA: Stanford University Press. Jacobs, J. A. (1996). Gender Inequality and Higher Education. Annual Review of Sociology, 22 153-85. James, E. (1990). Decision Processes and Priorities in Higher Education. In S. A. Hoenack & E. L. Collins (Eds.), The Economics of American Universities Albany: State University of New York Press.James, E., & Alsalam, N. (1993). College Choice, Ac ademic Achievement and Future Earnings. In E. P. Hoffman (Ed.), Essays on the Economics of Education Kalamazoo, MI: W. E. Upjohn Institute for Employment Research.Kuh, G. D., & Hu, S. (1999). Unraveling the Complex ity of the Increase in College Grades from the Mid-1980s to the Mid-1990s. Educational Evaluation and Policy Analysis, 21 (3), 297-320. Loury, L. D. (1997). The Gender Earning Gap Among C ollege-Educated Workers. Industrial and Labor Relations Review, 50 (4), 580-593. National Center for Education Statistics. (1997). Women in Mathematics and Science (NCES 97-982). Washington, DC: U.S. Department of E ducation. Niskanen, Jr., William A. (1971). Bureaucracy and Representative Government Chicago: Aldine Atherton.Paglin, M., & Rufolo, A. M. (1990). Heterogeneous H uman Capital, Occupational Choice, and Male-Female Earnings Differences. Journal of Labor Economics, 8 (1), 123-144.Pew Higher Education Research Program. (1990). The Lattice and the Ratchet (Policy Perspectives Vol. 2, No. 4): University of Pennsylv ania. Polachek, S. W. (1981). Occupational Self-Selection : A Human Capital Approach to Sex Differences in Occupational Structure. Review of Economics and Statistics, 63 (1), 60-69.Riordan, C. (1994). The Value of Attending a Women' s College: Education, Occupation, and Income Benefits. Journal of Higher Education, 65 (4), 486-510. Rumberger, R. W., & Thomas, S. L. (1993). The Econo mic Returns to College Major, Quality and Performance: A Multilevel Analysis of R ecent Graduates. Economics of Education Review, 12 (1), 1-19.


16 of 19Sabot, R., & Wakeman-Linn, J. (1991). Grade Inflati on and Course Choice. Journal of Economic Perspectives, 5 (1), 159-170. Sandler, B. R., Silverberg, L. A., & Hall, R. M. (1 996). The Chilly Classroom Climate: A Guide to Improve the Education of Women. Washingt on, D. C.: National Association of Women in Education.Schultz, T. W. (1961). Investment in Human Capital. The American Economic Review, 51 (March), 1-17. Sharp, L. M., & Weidman, J. C. (1989). Early Career s of Undergraduate Humanities Majors. Journal of Higher Education, 60 (5), 543-64. Smith, D. G., Wolf, L. E., & Morrison, D. E. (1995) Paths to Success: Factors Related to the Impact of Women's Colleges. Journal of Higher Education, 66 (3), 245266. Solnick, S. (1995). Changes in Women's Majors from Entrance to Graduation at Women's and Coeducational Colleges. Industrial and Labor Relations Review, 48 (3), 505-514.Strenta, A. C., Elliott, R., Adair, R., Matier, M., & Scott, J. (1994). Choosing and Leaving Science in Highly Selective Institutions. Research in Higher Education, 35 (5), 513-547.Strike, K. A. (1997). Centralized Goal Formation an d Systemic Reform: Reflections on Liberty, Localism and Pluralism. Educational Policy Analysis Archives, 5 (11).[Available online at v5n11.html] Ware, N. C., Steckler, N. A., & Leserman, J. (1985) Undergraduate Women: Who Chooses a Science Major? Journal of Higher Education, 56 (1), 73-84. Zemsky, R., Massy, W. F., & Oedel, P. (1993, May/Ju ne). On Reversing the Ratchet. Change, the Magazine of Higher Learning, 25 56-62.About the AuthorAlicia C. DowdDepartment of Education 415 Kennedy Hall Cornell University Ithaca NY 14853 Email: Voice: 607 255-7755Fax: 607 255-7905Alicia C. Dowd is Senior Lecturer in the Department of Education, Cornell University. Her research and teaching focus on educational equi ty and access. Current research projects include studies of the influence of tuitio n and tuition subsidies on the educational attainments of community college studen ts; student equity in community


17 of 19 college financing in New York State; and the influe nce of school-college partnerships on human capital development and institutional change. She received her Ph.D. in 1998 from Cornell University, where she studied educatio nal administration, with a focus on higher education. In 1998-99, she played a role in creating the Institute for Community College Development at Cornell University, which pr ovides leadership development opportunities to community college administrators a nd faculty members. Prior to completing graduate studies, she was a staff member for over ten years at Cornell's School of Continuing Education and Summer Sessions, where she was involved in developing and marketing academic programs.Copyright 2000 by the Education Policy Analysis ArchivesThe World Wide Web address for the Education Policy Analysis Archives is General questions about appropriateness of topics o r particular articles may be addressed to the Editor, Gene V Glass, or reach him at College of Education, Arizona State University, Tempe, AZ 8 5287-0211. (602-965-9644). The Commentary Editor is Casey D. C obb: .EPAA Editorial Board Michael W. Apple University of Wisconsin Greg Camilli Rutgers University John Covaleskie Northern Michigan University Alan Davis University of Colorado, Denver Sherman Dorn University of South Florida Mark E. Fetler California Commission on Teacher Credentialing Richard Garlikov Thomas F. Green Syracuse University Alison I. Griffith York University Arlen Gullickson Western Michigan University Ernest R. House University of Colorado Aimee Howley Ohio University Craig B. Howley Appalachia Educational Laboratory William Hunter University of Calgary Daniel Kalls Ume University Benjamin Levin University of Manitoba Thomas Mauhs-Pugh Green Mountain College Dewayne Matthews Western Interstate Commission for HigherEducation William McInerney Purdue University Mary McKeown-Moak MGT of America (Austin, TX) Les McLean University of Toronto Susan Bobbitt Nolen University of Washington


18 of 19 Anne L. Pemberton Hugh G. Petrie SUNY Buffalo Richard C. Richardson New York University Anthony G. Rud Jr. Purdue University Dennis Sayers Ann Leavenworth Centerfor Accelerated Learning Jay D. Scribner University of Texas at Austin Michael Scriven Robert E. Stake University of Illinois—UC Robert Stonehill U.S. Department of Education David D. Williams Brigham Young UniversityEPAA Spanish Language Editorial BoardAssociate Editor for Spanish Language Roberto Rodrguez Gmez Universidad Nacional Autnoma de Mxico Adrin Acosta (Mxico) Universidad de J. Flix Angulo Rasco (Spain) Universidad de Teresa Bracho (Mxico) Centro de Investigacin y DocenciaEconmica-CIDEbracho Alejandro Canales (Mxico) Universidad Nacional Autnoma Ursula Casanova (U.S.A.) Arizona State Jos Contreras Domingo Universitat de Barcelona Erwin Epstein (U.S.A.) Loyola University of Josu Gonzlez (U.S.A.) Arizona State Rollin Kent (Mxico)Departamento de InvestigacinEducativa-DIE/ Mara Beatriz Luce (Brazil)Universidad Federal de Rio Grande do Sul-UFRGSlucemb@orion.ufrgs.brJavier Mendoza Rojas (Mxico)Universidad Nacional Autnoma deMxicojaviermr@servidor.unam.mxMarcela Mollis (Argentina)Universidad de Buenos Humberto Muoz Garca (Mxico) Universidad Nacional Autnoma deMxicohumberto@servidor.unam.mxAngel Ignacio Prez Gmez (Spain)Universidad de


19 of 19 Daniel Schugurensky (Argentina-Canad)OISE/UT, Simon Schwartzman (Brazil)Fundao Instituto Brasileiro e Geografiae Estatstica Jurjo Torres Santom (Spain)Universidad de A Carlos Alberto Torres (U.S.A.)University of California, Los

xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam a22 u 4500
controlfield tag 008 c20009999azu 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E11-00154
0 245
Educational policy analysis archives.
n Vol. 8, no. 10 (January 27, 2000).
Tempe, Ariz. :
b Arizona State University ;
Tampa, Fla. :
University of South Florida.
c January 27, 2000
Collegiate grading practices and the gender pay gap / Alicia C. Dowd.
x Research
v Periodicals.
2 710
Arizona State University.
University of South Florida.
1 773
t Education Policy Analysis Archives (EPAA)
4 856