25, no. 1 (Fall 1993): 15-35 To the Editor Search

---

Table of Contents

Previous Article Next Article

Works Cited

IN MARCH 1990 the Chronicle of Higher Education , as part of its ongoing coverage of projected faculty shortages, reported a significant upturn in the number of potential applicants interested in attending graduate school (Blum). At about the same time, staff members at the MLA began to hear anecdotal reports, primarily from chairs of English departments, that applications to doctoral programs had suddenly increased. To get a systematic picture of recent trends, the MLA distributed a short survey in November 1990 to the approximately 500 programs that grant doctorates in modern language fields. The questionnaire inquired into the number of applications received, the number of applicants enrolled, and the number of degrees granted by these doctoral programs during the previous five years. In addition, the questionnaire investigated plans for expanding graduate programs. The version of the survey sent to foreign language departments also asked about the number of graduate students who were native speakers and the percentage of graduate courses taught in the target language. The following pages report on the survey findings, thereby providing a national picture of recent and anticipated expansion in foreign language doctoral programs. ¹ In addition to response frequencies, the results of multivariate regression analyses undertaken to assess subgroup differences among programs are presented. For comparison purposes, figures for the following modern language fields are presented whenever appropriate: comparative literature, linguistics, and English. ²

Thanks to follow-up letters and telephone calls during 1991, four-fifths of the doctorate-granting departments approached returned completed surveys. Because a single institution can have multiple foreign language departments, calculating response rates for such departments is more complicated than for the other disciplines. Of the 86 institutions with one or more doctoral programs in foreign languages, 78, or 91%, are part of the sample. Since not all doctoral programs at each institution completed surveys, the 215 responses received represent 77% of all foreign language doctoral departments. This response rate is high enough to justify the assumption that the data collected provide an accurate picture of recent trends in doctoral education. The response rates are lower for comparative literature (71%) and linguistics (63%). ³

Characteristics of Modern Language Departments with Doctoral Programs

Trends in applications received and degrees granted, as well as responses to other survey questions, were examined for differences among foreign language doctoral programs with differing institutional and departmental characteristics. The characteristics used are briefly described below.

The first section of table 1 indicates that the majority of modern language doctoral programs, be they in foreign languages, linguistics, comparative literature, or English, are in public institutions. ⁴ Most of the remainder are in private independent institutions; few, if any, are in church-related institutions. Between 50% and 63% of the departments in the four major fields are located in the northeastern and midwestern states. Another 13% to 31% are located in south Atlantic and south central states, and 19% to 29% are in the Rocky Mountain and Pacific Coast states. The geographic distributions of English and foreign language doctoral programs are similar, as the second section of table 1 indicates, though more foreign language than English departments are located in the northeastern states and fewer are located in the south central states. Institutions with foreign language or comparative literature doctoral programs, which tend to be somewhat larger than institutions with English programs, as the third section of table 1 indicates, enrolled on average approximately 24,000 students in fall 1988, with half enrolling between 13,000 and 33,500. Linguistics programs are most often located in large institutions, with one-half in institutions enrolling between 22,000 and 36,000 full- and part-time students on average.

The fourth, fifth, and sixth sections of table 1 deal with characteristics that apply to foreign language departments only. The fourth section indicates that almost two-thirds of the doctoral programs in the sample confine their offerings primarily to commonly taught languages (i.e., Spanish, French, and German), while just under a fourth offer less commonly taught languages (primarily Slavic, Asian, and Near Eastern languages). ⁵ The remainder offer both types of languages. (This group is not considered in more detailed analysis.) The fifth section of table 1 shows that close to 60% of the foreign language doctoral programs in the sample are organized as language groups (e.g., Slavic languages, classics, East Asian languages), while just over a third are single-dual language departments (e.g., German, French and Italian, Spanish and Portuguese). ⁶ In keeping with the predominance of language groups, the sampled departments offer courses in an average of 3 languages, with 50% offering between 2 and 6 languages. Only 28% offer courses in 6 or more languages, as table 1 indicates.

Graduate departments in foreign languages, comparative literature, and linguistics are consistently smaller than those in English, as the remaining sections of table 1 indicate. The number of full-time tenure-track faculty members in responding foreign language departments ranges from 1 to 54; half of those departments have between 8 and 17 full-time tenure-track faculty members, while half of the comparative literature or linguistics departments have between 5 and 15 (see section 7 of table 1). In keeping with their smaller faculties, foreign language, comparative literature, and linguistics departments with doctoral programs tend to have fewer MA and PhD students than English departments, as the eighth and ninth sections of table 1 indicate. On average, the number of MA students and the number of PhD students were approximately equal in English departments in 1989–90, while the other three types of departments tended to have more PhD than MA students. For more detailed analysis, the number of MA students and the number of PhD students in each department were combined to yield a total number of graduate students. The average total is approximately 45 to 55 in linguistics departments and approximately 30 to 35 in foreign language and comparative literature departments. The tenth section of table 1 indicates that approximately 85% of the comparative literature departments had 50 or fewer graduate students in 1989–90, as did 69% of the linguistics departments and 86% of the foreign language departments. In contrast, only 14% of the English departments in the sample had fewer than 50 graduate students.

Foreign language doctoral programs responding to the survey were asked to indicate the approximate percentage of native speakers among their graduate students. The percentages ranged from 0% 98%, with a median of 25%. For half of the foreign language departments in the response sample, the percentage of graduate students who are native speakers ranges from 12% to 50%; it exceeds 60% in only 9% of the departments. The percentage of native speakers varies with the total number of graduate students in a department; the percentage is higher, on average, among departments with 31 or more students than among those with fewer students (37% vs. 26%).

The eleventh section of table 1 indicates that in 1989–90 many more graduate students in English had teaching assistantships than in the other three types of departments, hardly a surprising result in the light of the differences in the size of the graduate programs. However, comparing the percentages of graduate students with teaching assistantships in 1989–90 reveals, as the last section of table 1 indicates, that graduate students in foreign language and comparative literature doctoral programs are more likely to have teaching assistantships than graduate students in English or linguistics are. On average, 60% of the graduate students in foreign language departments have teaching assistantships compared with just under 50% of those in English.

Preceding discussion has indicated that foreign language, comparative literature, linguistics, and English doctoral programs tend to be located in institutions with public funding, institutions in the Northeast or Midwest, and institutions that enroll between 12,500 and 31,000 full- and part-time students. The various types of departments differ in size, however. On average, foreign language, linguistics, and comparative literature departments have one-half to one-third as many graduate students and full-time tenure-track faculty members as English departments have. Despite, or perhaps because of, their smaller size, foreign language and comparative literature departments tend to award teaching assistantships to a somewhat higher percentage of their graduate students than English departments.

Links between the institutional and departmental characteristics considered here affect their importance for growth in foreign language doctoral programs. The interrelations are outlined in more detail in appendix A. Source of funding delimits institutional size, with public institutions enrolling more students than private. Institutional size, source of funding, and type of languages offered help determine faculty size and number of graduate students: foreign language departments offering commonly taught languages and those in large public institutions tend to have larger faculties and more graduate students than departments offering less commonly taught languages and those in small private institutions. The percentage of graduate students with teaching assistantships depends on source of funding and type of languages offered. The type of languages offered also helps define type of language program, with 53% of the departments offering commonly taught languages organized as language groups, compared with 87% of the departments offering less commonly taught languages.

Applicants to Modern Languages Graduate Programs

Figure 1 shows the average number of applications received between 1986–87 and 1990–91 by doctoral programs in foreign languages, comparative literature, linguistics, and English. ⁷ In both foreign languages and English, the averages increased steadily. In linguistics, they fluctuated but remained largely unchanged, while in comparative literature the averages increased between 1987–88 and 1990–91. Between 1986–87 and 1990–91, one-half of the foreign language doctoral programs in the sample received between 12 and 49 applications annually. This range increased from 12–34 in 1986–87 to 16–49 in 1990–91.

The number of applications received by foreign language departments varies with the number of graduate students. Figure 2 shows that departments enrolling the most graduate students consistently receive the most applications, on average. In short, large doctoral programs, as measured by number of students, receive more applications than smaller ones, not a surprising finding.

The degree to which application levels changed through time in individual departments was assessed by calculating the percentage change in number of applications received between 1987 and 1991. ⁸ The resulting figures served as the basis for defining three groups: departments experiencing an increase in the number of applications received, departments experiencing no change, and departments experiencing a decrease. (To be included in the increase or decrease groups, departments had to experience a change of more than 10%. ⁹ ) Table 2 indicates that three-fifths of the foreign language departments in the sample received more applications to their graduate programs in 1991 than in 1987, as did just over two-thirds of comparative literature departments and approximately two-fifths of linguistics departments. The percentage is considerably higher for the sampled English departments. Among foreign language departments experiencing an increase in the number of applications received between 1987 and 1991, the average increase was 59%; for half of those departments the percentage increase was between 30% and 86%.

Thirty-four percent of the foreign language doctoral programs in the sample report that growth in applications during the late 1980s was particularly marked for specific languages, while 52% of the English departments reported unusual growth in particular fields. The following list shows the languages mentioned most frequently.

In assessing the figures above, it is important to bear in mid that the frequency with which degree programs are offered affects the chances that an unusual increase in the number of applications may be reported. Thus, estimates of the frequency with which degree programs are offered in the seven languages mentioned more than once were used to calculate the following percentages of programs in specific languages reporting unusual growth.

The figures suggest that applications have increased most markedly in Chinese and Spanish and substantially in French, Japanese, and Russian.

In addition to providing information on the number of applications received during recent years, survey respondents were asked to indicate how the applicants to their graduate programs for the 1990–91 academic year compared with those who applied five years earlier. Table 3 indicates that close to half of the respondents from foreign language and comparative literature departments said that their most recent applicants were more qualified than those who applied five years earlier, and very few said that the recent applicants were less qualified. The percentage reporting less qualified current applicants is similarly low among linguistics and English respondents, though the percentage reporting more qualified applicants is lower for linguistics and considerably higher for English than for foreign language and comparative literature departments.

Although no data were gathered on the number of applicants accepted annually by modern language graduate programs, respondents were asked to indicate whether the number of applicants accepted into their PhD programs had increased during the last five years. Table 4 indicates that the majority of foreign language departments in the sample accepted more applicants during the late 1980s, as did almost two-thirds of the English and comparative literature departments. In all three types of departments, most increases occurred before the 1989–90 academic year. Linguistics departments are more likely than other types of departments are to have made no change in or to have decreased the number of applicants accepted.

Number of Applicants Enrolled by Modern Language Graduate Programs

Figure 3 shows the average number of applicants enrolled annually by modern language doctoral programs between 1986–87 and 1990–91. In comparative literature and linguistics, the average number of applicants enrolled has tended to decrease. For English and foreign languages, in contrast, the averages have increased steadily, though not as substantially as the average numbers of applications received has (figure 1). Between 1986–87 and 1990–91 the number of applicants enrolled annually ranged from 4 to 12 for half of the foreign language departments in the sample; during that time the outer limits of the range increased slightly (from 4–11 to 4–12).

The number of applicants enrolled by foreign language doctoral programs in any given year during the late 1980s varies with the number of graduate students and the percentage of graduate students with teaching assistantships. Figure 4 shows that during the late 1980s the average number of applicants enrolled was consistently higher in departments with more graduate students, while figure 5 shows that the average number enrolled in any year tends to increase as the percentage of students with teaching assistantships increases. The differences are most clear-cut if one compares departments in which 81% or more of the students have teaching assistantships with departments in which a smaller percentage have teaching assistantships. These findings, along with those for number of applications received, indicate that both the number of applications received and the number of applicants enrolled vary with the number of graduate students in foreign language departments.

A percentage change measure similar to the one for application levels was used to assess the degree to which individual programs increased enrollment levels between 1986–87 and 1990–91. Table 5 indicates that close to half of the foreign language and comparative literature departments in the sample experienced an increase in the number of applicants enrolled in their graduate programs, as did three-fifths of the English departments. Over half of the linguistics departments in the sample experienced a decline in the number of applicants enrolled. On average, foreign language departments experiencing an increase saw their enrollments increase by 66% during the five-year period, with one-half experiencing an increase of 32% to 100%.

Percentage of Applicants Enrolled by Modern Language Graduate Programs

The findings discussed thus far indicate that during the late 1980s many foreign language doctoral programs received noticeably larger numbers of applications and enrolled modestly larger numbers of applicants. The question that remains unanswered is whether the same or different departments experienced increased application and enrollment levels. The following figures for foreign language programs provide a preliminary answer to this question.

The figures indicate that doctorate-granting departments that experienced growth in the number of applications received during the late 1980s also experienced increased enrollment, while departments that experienced no increase in the number of applications received tended to experience decreased enrollment. Figure 6 yields a similar picture. Doctoral programs experiencing a decrease or no change in number of applications received show some fluctuation but no real growth in the average number of applicants enrolled between 1986–87 and 1990–91. Among foreign language departments experiencing growth in the number of applications, in contrast, the average number of applicants enrolled increased fairly steadily. But the increase in the average number of applicants enrolled is considerably lower than the increase in the average number of applications received (26% vs. 59%).

Since averages can mask widely differing situations in individual departments, annual application and enrollment figures were combined in a measure of the percentage of applicants enrolled. Figure 7 shows that between 1986–87 and 1990–91 the percentage of applicants enrolled decreased in foreign languages and English but increased somewhat for comparative literature and linguistics; in all types of departments, the averages fluctuated more from year to year than applications or enrollments did. In half of the foreign language departments in the sample, the percentage of applicants enrolled ranged from 16% to 46%; the endpoints of the range decreased slightly from the beginning to the end of the five-year period (19% to 45% in 1986–87 compared with 16% to 44% in 1990–91).

The percentage of applicants enrolled by foreign language doctoral programs varies with source of funding, as figure 8 indicates. Departments in private institutions consistently enrolled a lower percentage of applicants than departments in public institutions did. The percentage of applicants enrolled also varies with the type of languages taught. As figure 9 shows, the average percentage of applicants enrolled in any given year is consistently lower among departments offering less commonly taught languages than among departments offering commonly taught languages. It may be that foreign language doctoral programs offering commonly taught languages in public institutions enroll the highest percentage of applicants (42% in 1990–91 vs. 23% for other programs) because they are dependent on graduate students to staff their extensive offerings in introductory languages. ¹¹

Once again, change in individual departments was assessed using a percentage change measure similar to the one used for applications and enrollments. Table 6 presents the three change groups for foreign languages, comparative literature, linguistics, and English. Among the last three department types, close to two-thirds experienced a decrease in the percentage of applicants enrolled. Just over half of foreign language departments experienced a decrease. Among foreign language departments with a decreased enrollment rate, the average decrease was 37%, and ranged from 25% to 57% for half of those departments.

Table 6 indicates that during the late 1980s the majority of PhD-granting modern language department did not increase the number of applicants enrolled in proportion to the increased number of applications received. Instead departments became more selective about the applicants they admitted. There are a number of possible explanations for this state of affairs. One is that the quality of applicants declined as their numbers increased, a supposition the survey findings do not support. A more plausible possibility is that limits on financial and faculty resources inhibited growth in graduate programs. In addition, continuing doubts about future faculty demand may be leading some English and foreign language departments to decide that enlarging their graduate programs is premature, possibly even imprudent.

Plans for Enlarging Graduate Programs

At the time of the 1990 surveys, no more than one-fifth of respondents said their departments had definite plans for increasing the number of students accepted into their graduate programs in the next five years. Table 7 indicates that a larger proportion will enlarge their programs if conditions for doing so are favorable. Under such conditions, 67% of the responding foreign language departments, 63% of the linguistics departments, 51% of the English departments, and 46% of the comparative literature departments will enlarge their graduate programs. ¹² Expansion appears to be more frequently anticipated by respondents from doctoral programs in foreign languages and linguistics.

Factors that may encourage modern language departments to enlarge their graduate programs are listed in table 8. Most respondents appear to regard the first three factors listed as important. Somewhat over half of the foreign language respondents consider the availability of federally funded fellowships and the number of applicants important, while just under a third view projections of faculty shortages as important. Among respondents in comparative literature and linguistics, a solid majority consider the availability of federal fellowship funds important, while a third or fewer attach importance to the number of applicants and to projections of faculty shortages. Compared with the respondents from English departments, respondents from the other three language fields attach more importance to the availability of federally funded fellowships but are less concerned about projections of faculty shortages. Furthermore, respondents in linguistics and comparative literature are less concerned than those in English and foreign languages are about the number of applicants. Although relatively few respondents consider institutional constraints important in decisions about enlarging graduate programs, respondents in foreign languages are less likely to do so than those in English, linguistics, or comparative literature are.

Although respondents were given the opportunity to name other factors that would influence decisions about enlarging graduate programs in the next five years, relatively few did so. Among the 8 respondents from foreign language departments who specified additional factors, 2 mentioned the importance of sufficient funds, which presumably would not be in the form of teaching assistantships, to support students, while 3 mentioned the need for more faculty members. In addition, 2 respondents mentioned the job market as a factor affecting decisions about enlarging their programs.

The percentage of respondents attaching importance to several of the factors listed in table 8 varies with a number of institutional and departmental characteristics. The percentage of respondents from foreign language departments who consider number of applicants important to decisions about program enlargement varies with the number of languages offered. The fourth column of table 9 indicates that the percentage increases as the number of languages offered increases. The percentage of respondents from foreign language departments who consider the availability of institutional fellowships important is greater in private institutions than in public ones (see second column of table 9), while the percentage who consider the availability of teaching assistantships or federally funded fellowships important varies with the percentage of students currently holding assistantships. The first and third columns of table 9 indicate that respondents whose programs have the lowest percentage of students with teaching assistantships are least likely to consider the availability of assistantships important to decisions about program expansion but most likely to consider federally funded fellowships important. And finally, the percentage of respondents who consider projections of faculty shortages important to decisions about program expansion varies with institutional size. As institutional size increases, the percentage of respondents who consider projected shortages important also increases (see last column of table 9). Even among groups that attach the most weight to projections of faculty shortages (i.e., foreign language programs in the largest institutions) fewer than half consider that factor important to decisions about enlarging graduate programs during the first half of the 1990s. Moreover, the reports of such shortages, which swept the media in late 1989 and early 1990, appear to have had little effect on the numbers of candidates applying to and enrolling in modern language graduate programs. Preceding discussion has shown that increases in the number of applications received and applicants enrolled during the late 1980s is part of a longer-term process rather than a phenomenon that began in 1990–91 (see especially figures 1 and 3).

The figures presented in tables 8 and 9 suggest that two factors will be of prime importance in determining whether doctorate-granting modern language departments decide to enlarge their graduate programs in the near future: the availability of financial support for students and a large and highly qualified applicant pool. Recent increases in the number of applicants to graduate programs and the growing number of undergraduates receiving bachelor's degrees in language—the number has been increasing since 1983–84 for foreign languages—indicate that the necessary pool seems to be available. The requisite financial support, however, may not be available until the economy improves. Even then, expanding the number of applicants enrolled may not be the preferred way to increase PhD production. Rather, the new financial resources may be used to improve completion rates. As William G. Bowen and Neil L. Rudenstine have argued,

working to improve the effectiveness of current programs is by far the most sensible way to begin to prepare for the faculty staffing problems that are anticipated by the end of the decade. In many situations, a combination of increased external support and strong internal efforts to improve program design and program management should make it possible to achieve substantial increases in the number of doctorates conferred without significant increases in the sizes of entering cohorts. (289)

The importance of economic conditions and financial support stands out in responses to a survey question that invited general comments about the recent and future growth of respondents' graduate programs. Among respondents from foreign language departments, 30% answered this open-ended question, with 8 to 64 reporting that their departments had no plans for expansion, usually because maintenance of program quality was deemed the first priority. Among the factors considered crucial to program growth were adequate financial support for students, mentioned by 11 respondents, and general budget limitations and constraints, mentioned by 8. In most cases such limitations were institutionally imposed and thus particularly onerous. A few respondents also mentioned the job market and the need for more faculty members as important considerations in making decisions about program growth. A good many respondents (n=10) simply described the reasons underlying recent or anticipated program growth, rather than specifying preconditions for such growth, while 6 described key features of their graduate programs and 9 commented on the recent improvement in their applicant pools. ¹³

Average Number of PhDs Granted

Survey respondents were asked to list the number of PhDs their departments granted annually between 1985–86 and 1989–90. Between 14% and 26% of the foreign language departments in the sample granted no degrees in at least one year between 1985–86 and 1989–90. (In English the percentage was much smaller: 5% to 8%.) Because the numbers involved are small and fluctuate from year to year, the annual figures were used to calculate the average number of PhDs granted between 1986 and 1990. ¹⁴ The top rows of table 10 indicate that this average is smallest for foreign language departments, with half of the responding departments granting between 1 and 3, on average. English departments granted the largest average number of PhDs, while comparative literature and linguistics departments fall between the two extremes, granting an average of 3 to 4 degrees a year.

The average number of PhDs granted by foreign language departments varies with aspects of departmental size. As the following numbers indicate, the average number of PhDs granted increases as the number of graduate students in a department increases.

Since faculty size helps determine the number of graduate students enrolled in foreign language programs and thereby indirectly affects the average number of PhDs granted, the average number of degrees granted increases as the student-faculty ratio increases (see table 11). In short, the average number of PhDs granted by modern language graduate departments varies with faculty size and the number of graduate students enrolled.

Although larger departments may grant more PhDs, these departments are not necessarily more successful at granting degrees to their entering students. In an effort to isolate characteristics of departments with the highest completion rates, regardless of size, a crude measure of PhD productivity was developed by dividing the average number of PhDs granted between 1985 and 1990 by the total number of PhD students enrolled in 1989–90. The result is the average number of PhDs granted for every 100 PhD students enrolled. The bottom rows of table 10 indicate that the average number of degrees granted for every 100 students is lower for English departments than for the three other types of departments. Half of the foreign language departments granted between 7 and 14 degrees for every 100 students enrolled, compared with 6 to 11 for half of the English departments. Thus, although foreign language departments grant fewer PhDs on average than English departments, their productivity, as measured by degrees granted for every 100 enrolled students, is higher than that of English departments.

Much of the difference in productivity between English and foreign language departments can be accounted for by differences in program size, as the following figures indicate.

For both English and foreign languages, the number of PhDs granted for every 100 PhD students is considerably higher on average in departments with the fewest graduate students. PhD productivity is lower in the three subgroups with more students. These findings suggest that even though smaller doctoral programs grant fewer degrees overall than larger programs do, smaller programs grant degrees to a larger proportion of their students than larger programs do.

Among the small foreign language departments (those with 20 or fewer graduate students), the number of PhDs granted for every 100 PhD students is higher if faculty size exceeds 8 (27.2 vs. 15.8 for departments with 8 or fewer full-time tenure-track faculty members). Thus, a low student-faculty ratio, in addition to small program size, enhances PhD productivity, as table 11 shows. Taken together, the two columns of table 11 indicate that the features of program size that encourage PhD production appear to discourage productivity as defined here.

The percentage of graduate students with teaching assistantships also affects PhD productivity in foreign language departments, though not as strongly as the number of graduate students does. ¹⁵

The figures above indicate that the average number of PhDs granted for every 100 PhD students is highest in departments in which the greatest percentage of graduate students have teaching assistantships. These findings suggest that the widespread availability of teaching assistantships fosters PhD productivity, presumably for two reasons: assistantships provide necessary financial support, and they provide regular contact with faculty members in a professional setting. Such contact with faculty members may be particularly important to degree completion because intellectual stimulation and mentoring relationships may spill over into nonteaching areas. ¹⁶

Unfortunately, the survey did not gather data on the number of students with fellowships, and thus, the effect of fellowship availability on PhD productivity cannot be assessed. Using longitudinal data on individuals enrolled in a single elite research university in the Northeast during a 25-year period, Bowen and Rudenstine conclude in their recent study of PhD attainment patterns that in cohorts commencing graduate study during the late 1960s humanities students with fellowships had higher completion rates than humanities students with teaching assistantships (185–94). ¹⁷ For cohorts commencing graduate study in the 1970s and the early 1980s the opposite is true, however; those with teaching assistantships had higher completion rates. In all cohorts, humanities students with teaching assistantships have a somewhat longer time-to-degree (approximately half a year longer) than those with fellowships did.

Differences in the success of recipients of fellowships and of teaching assistantships fade to insignificance when students receiving any type of institutional support are compared with those receiving none. Using longitudinal data from 5 elite research universities, Bowen and Rudenstine find that the completion rates of humanities students relying on their own resources (e.g., loans, family funds) are one-half to two-thirds as high as those for students receiving some type of institutional support (i.e., fellowships or teaching assistantships) (187–94). Students without institutional support are less likely to reach the ABD stage, they have a longer average time-to-degree. These findings suggest that some type of institutional support is vital for successful graduate study, although it is not entirely clear whether fellowships are a more effective means of enhancing success than teaching assistantships are. Bowen and Rudenstine suggest that combining the two forms of support may make the most sense (194–95, 274–77). Fellowship support, they argue, should be reserved for the first year of graduate study and the dissertation stage, with teaching assistantships awarded during the years in between.

Completion rates and times-to-degree also vary sharply by field, according to Bowen and Rudenstine, with students in the humanities having lower completion rates and longer times-to-degree than students in the natural sciences do (123–41). ¹⁸ Program size, defined by the average size of entering cohorts, is also significant (142–56). Comparison of smaller programs in four elite research institutions with larger programs in three similar institutions revealed that smaller humanities programs have higher completion rates than larger ones (61% vs. 33%). In particular, about 55% of students in the larger programs fail to reach the ABD stage compared with 18% of those in the smaller programs. ¹⁹ Finally, time-to-degree is approximately one year longer in larger humanities programs than in smaller ones (7.5 years from beginning of graduate study to receipt of degree vs. 6.4 years). Bowen and Rudenstine's findings are in keeping with those reported here: PhD productivity is greater in programs with fewer faculty members and graduate students.

Bowen and Rudenstine also find some differences in productivity between programs in Research I institutions and programs in other types of doctorate-granting institutions: completion rates are higher among NDEA fellows in humanities programs in Research I institutions (55% vs. 34% among humanities students at other doctorate-granting institutions), while time-to-degree is somewhat shorter for all humanities students in Research I institutions (8.2 years from graduate school entry to receipt of degree vs. 8.8 years for students in other doctorate-granting institutions) (149–58). Such differences are not present in the data discussed here: PhD productivity is no different in English or foreign language programs in Research I universities than in programs in other doctorate-granting institutions, though the average number of PhDs granted by English programs does differ. (The average number of degrees granted by modern language programs in Research I and other doctorate-granting institutions does not vary, in part because such programs are so heavily concentrated in Research I universities: 79% for foreign languages and linguistics and 92% for comparative literature, compared with 42% for English.) The lack of differentiation may be due to imperfections in the measure of productivity used here or the lesser significance of federal funding (Research I institutions received at least $33.5 million in the early 1980s) for languages compared with other humanities disciplines. The differences that Bowen and Rudenstine identify could also be a consequence of uncontrolled differences in institutional and program size. On average, English programs in Research I universities are in larger institutions than programs in other doctorate-granting institutions are (26,246 full- and part-time students vs. 16,036), and English programs in Research I institutions have more faculty members and graduate students (43.9 vs. 31.2 full-time tenure-track faculty members on average and 140.1 vs. 98.3 graduate students on average). Foreign language programs in Research I universities are also in larger institutions, but the programs are similar in size to programs in other doctorate-granting institutions.

Percentage of Courses Offered in the Target Language

Foreign language departments participating in the survey were asked to indicate the percentage of graduate courses they offered in target languages. Responses ranged from 0% to 100%, with 26% of the respondents saying that 100% of their graduate courses were offered in the target language. The median was 83%, with half of the sampled departments offering between 50% and 100% of their courses in the target language. One in five departments offered 25% or fewer of their courses in the target language. The percentage of courses offered in the target language varies with three factors, as the following figures illustrate.

The figures indicate that, on average, departments offering less commonly taught languages and departments organized as language groups offer a considerably lower percentage of their courses in the target language than programs offering commonly taught languages and programs organized as single-dual language departments or inclusive programs. Furthermore, departments with 51 or more graduate students offer a higher percentage of their courses in the target language, on average, than departments with fewer graduate students do (87% vs. 64%). These findings suggest that departments that offer a high percentage of courses in the target language are likely to be characterized by a large graduate student body and an exclusive focus on one or two commonly taught languages, presumably because the ability to offer courses in target languages depends on having enough students to participate in such courses.

Variation by Language

Distinctions between departments offering commonly taught languages and departments offering less commonly taught languages in the percentage of graduate courses offered in the target language suggest that this percentage may differ from departments offering different languages. Since responses to other survey questions may also differ along this dimension, a set of departmental types was developed with the aid of department titles. Table 12 shows the major categories and their subgroups. The doctoral programs in the sample most frequently fall into four of the major groupings: French and Italian, Romance languages, classics, and Spanish and Portuguese. Table 13 indicates that the percentage of graduate courses offered in the target language indeed differs by department type. The percentage is significantly above average in Spanish and Portuguese departments, French and Italian departments, Romance languages departments, and inclusive programs, while it is below average in Slavic languages departments and Asian languages departments. ²⁰

Not surprisingly, three of the four types of departments that offer the highest percentage of courses in the target language are the largest. On average, Spanish and Portuguese departments, Romance language departments, and inclusive programs have the largest numbers of full-time tenure-track faculty members and the largest numbers of graduate students (see tables 14 and 15). German departments have unusually small faculties and student bodies, while Slavic language departments have atypically small faculties. In addition, the average number of graduate students in classics departments is small. The percentage of graduate students who are native speakers is above average in three of the four types of departments that offer the highest percentage of courses in the native language: Spanish and Portuguese, Romance languages, and inclusive programs (see table 16). The percentage of native speakers is unusually low in only one type of department: Slavic languages, which is among the departments with the lowest percentages of courses taught in the target language. ²¹ Table 17 shows the average percentage of graduate students with teaching assistantships in different types of departments. Differences are not pronounced; the percentage is somewhat above average in the first four department types listed and below average in departments of Asian languages and Slavic language.

Table 18 shows the average number of PhDs granted annually between 1985–86 and 1989–90 by different types of departments. The number is relatively uniform across departments, hovering between 1.4 and 2.4 in most cases. The number is above average in Romance language departments and Spanish and Portuguese departments. Although these departments grant more degrees, their PhD productivity is average. Table 19 shows that the most degrees granted for every 100 PhD students is in inclusive programs, Germanic language departments, and classics departments, while PhD productivity appears to be somewhat below average in German departments. The most productive department types are of differing sizes—inclusive programs are large, while classics departments and Germanic languages departments have small faculties and graduate student bodies—so size does not explain these department types' distinctiveness. Nor does the percentage of students with teaching assistantships, since the average percentage is similar for all three, but not atypically high or low.

Tables 20 through 25 show trends by department type in applications received and in number and percentage of applicants enrolled. Two of the three largest department types consistently received the largest average number of applications between 1986–87 and 1990–91: Romance language departments and Spanish and Portuguese departments. In addition, as table 20 shows, German departments and Asian language departments received an above-average number of applications in most years. Application levels are not significantly below average for any department type, though they tend to be low in Germanic languages. Table 21 indicates that three of the four department types with the highest average application levels also had an unusually high percentage of departments whose application levels increased through time: Romance languages, Spanish and Portuguese, and German. Classics departments were also unusually likely to experience an increase in application levels. The percentage of departments receiving fewer applications is unusually high only among inclusive programs.

Table 22 indicates that Romance language programs and Spanish and Portuguese departments, which both had high average application levels, enrolled the most applicants, on average, between 1986–87 and 1990–91. Classics departments, in contrast, most consistently enrolled a below-averaged number of applicants. With the exception of Romance languages, an atypical enrollment level is not associated with an unusual percentage change in the number of applicants enrolled between 1986–87 and 1990–91. Table 23 indicates that over half of the German departments in the sample experienced a decline in the number of applicants they enrolled between 1986–87 and 1990–91, while more than 60% of the Slavic language departments and Romance language departments experienced an increase.

Table 24 shows the average percentage of applicants enrolled between 1986–87 and 1990–91 by type of departments. The percentage is most consistently below average for Slavic language programs and Asian language programs, while it is most consistently above average for Romance language programs and inclusive programs. The percentage of departments experiencing an increase through time in the percentage of applicants enrolled is highest for Slavic language programs and inclusive programs, while the percentage experiencing a decrease exceeds 58% for three department types: German, Asian language, and Romance language programs (see table 25). These department types, therefore, are least likely to have increased the number of applicants enrolled between 1986 and 1991 in proportion to the increased numbers of applications received during the same period. Instead, these types of departments became more selective about the applicants they admitted.

Bettina J. Huber is Director of Research for the Modern Language Association.

Appendix A

Interrelations among Institutional and Departmental Characteristics

Examining relations among the institutional and departmental characteristics of foreign language departments requires certain assumptions about the causal ordering of variables. Although source of funding the number of students enrolled are relatively fixed dimensions of institutional structure, the funding source, which is usually fixed at an institution's founding, is generally more invariant. The number of students enrolled, in contrast, can vary considerably during the course of an institution's history. Thus, source of funding is assumed to be causally prior to institutional size, as measured by the number of full- and part-time students enrolled in fall 1988.

Because funding source and institutional size are institutional characteristics, they are assumed to be causally prior to the departmental characteristics under consideration: type of languages offered, program type, number of languages offered, faculty size, number of graduate students, and percentage of graduate students with teaching assistantships. Because the type of languages offered is such a fundamental feature of language programs, it is assumed to be causally prior to the other departmental characteristics, while program type and number of languages offered, which generally remain fixed over a period of years, are assumed to be causally prior to the faculty and graduate student characteristics under consideration. Although both faculty and student body size are subject to change and fluctuation, the number of full-time tenure-track faculty members in a department can be changed only every few years at best, and that number is more subject to institutional constraints than the size of the graduate student body is. Thus, the number of full-time tenure-track faculty members in a department is assumed to be causally prior to the number of graduate students and to the percentage of graduate students with teaching assistantships. Faculty size, therefore, may help determine the number of graduate students in a department, while both faculty and graduate student body size may delimit the percentage of graduate students with teaching assistantships.

The first section of table A-1 shows that public institutions with foreign language doctoral programs are considerably larger than private institutions with such programs; on average, public institutions with foreign language programs enroll 2.3 times as many students as private institutions with similar programs do. Type of languages offered and type of language program do not vary by source of funding or institutional size. The two are related, however, as the second section of table 1 indicates. More than 80% of the departments offering less commonly taught languages are organized as language groups, compared with just over half of department offering commonly taught languages. Departments offering commonly taught languages are more likely to be organized as single-dual language departments (47% vs. 14% for programs offering less commonly taught languages). ²² Both type of languages offered and program type help determine the number of languages offered. The third section of table A-1 indicates that, on average, departments offering less commonly taught languages offer courses in twice as many languages as departments offering commonly taught languages. The few inclusive programs in the sample offer the most languages—9 on average—while single-dual language departments offer the fewest. The number of languages offered also varies with institutional size, though not markedly. On average, foreign language departments in institutions enrolling more than 20,000 students offer more languages than departments in smaller institutions do (4.6 vs. 3.1).

The number of full-time tenure-track faculty members in foreign language doctoral programs varies with source of funding and type of languages offered. The fourth section of table A-1 shows that departments in public institutions have larger faculties, on average, than departments in private institutions do, while departments offering commonly taught languages have larger faculties than departments offering less-commonly taught languages do. Faculty size, in turn, affects the number of graduate students enrolled in doctoral programs, as does institutional size. The effect of the latter is not great, as the fifth section of table A-1 indicates. Programs in institutions with more than 20,000 students have more graduate students, on average, than programs in smaller institutions do (37.3 vs. 24.4). The effect of faculty size is more clear-cut: as the number of full-time tenure-track faculty members in a foreign language department increases, so does the number of graduate students. The percentage of graduate students with teaching assistantships varies with two factors. The last section of table A-1 indicates that the percentage is higher in departments in public institutions than in departments in private institutions. The percentage is also higher in departments offering commonly taught languages than in departments offering less commonly taught languages.

The relations summarized here suggest that among foreign language departments source of funding and type of languages offered help determine faculty size, which, in turn, along with institutional size, delimits the number of graduate students departments enroll. Furthermore, type of languages offered and type of language program, which are unrelated to funding source and institutional size, are largely responsible for determining the number of languages offered. Type of languages offered, along with source of funding, also affects the percentage of graduate students with teaching assistantships. The nature of these interrelations explains why the number of graduate students is the most common explanatory factor in the main text.

Notes

¹ Special thanks are due to John Cross, Phyllis Franklin, and Richard Brod for their helpful comments on an earlier version of this article.

² There are too few linguistics and comparative literature departments to permit separate regression analyses for these program types.

³ The English findings are discussed in detail in an article scheduled to appear in the Winter issue of the ADE Bulletin . The response rate for linguistics is lower probably because the Linguistic Society of America chose to collect the same information a month or two after the MLA survey was distributed. Moreover, I subsequently discovered that the list of linguistics doctoral programs we used was incomplete. Rather than the 30 freestanding programs on our list, there are 56 doctoral programs. Thus, the linguistics programs included in this survey represent only 34% of doctoral programs in that discipline. Readers should bear this fact in mind in evaluating the findings presented here.

⁴ The percentages cited in the text and presented in tables may not sum to 100 percent because of rounding. In addition, the number of respondents varies from table to table because those not answering a question were eliminated from consideration.

⁵ Since departments were categorized by title, some languages were grouped under both commonly and less commonly taught. Thus, French and Italian departments and Spanish and Portuguese departments are considered to offer commonly taught languages, while the 4 Italian departments and the 1 Portuguese department were placed in the less commonly taught group. Romance language and classics departments were placed in the commonly taught group.

⁶ The few inclusive programs are responsible for offering courses in all foreign languages offered by their institutions.

⁷ The academic year used when dealing with application figures is the year for which applicants sought admission rather than the year in which they submitted applications. Thus, the applicants included in the 1986–87 group submitted applications in 1985–86. The academic years used for enrollment figures, in contrast, refer to the year in which accepted applicants enrolled. This procedure facilitates comparison of application and enrollment figures for the same year.

⁸ This measure ignores short-term increases or decreases during the years between 1986–87 and 1990–91.

⁹ When no application number was provided fo 1986–87, the 1987–88 number was used to calculate the percentage change, and if no numbers were available for either of these years, the 1988–89 number was used.

¹⁰ Since no data were gathered on the doctoral degree programs offered, the numbers used here were estimated using the names of the foreign language departments in the sample. Moreover, with the exception of Italian, the estimates are conservative; only those departments whose names clearly indicated that a degree program was offered in the language in question were included. Thus, for example, modern and foreign language departments were not included in any of the estimates, since the languages in which they offer PhDs is not apparent from their names. The estimate for Italian may be inflated because all French and Italian departments in the sample were assumed to offer doctoral degree programs in Italian.

¹¹ Such programs have the highest percentage of students with teaching assistantships (72% for programs offering commonly taught languages in public institutions vs. 46% for all other programs).

¹² These percentages combine the first two categories listed in table 7 (i.e., yes and if conditions favorable).

¹³ The full text of the comments made by foreign language respondents is available on request.

¹⁴ When the number of PhDs granted in 1985–86 was not provided, the figures for the following four years were used to calculate a four-year average. Similarly, when no numbers were provided for both 1985–86 and 1986–87, the remaining figures were used to calculate a three-year average.

¹⁵ Regression analysis reveals that variation in number of graduate students explains 2.4% of the total variance in the average number of PhDs granted for every 100 graduate students in foreign languages, while variation in the percentage of students with teaching assistantships accounts for 1.6%. Overall, variation in the institutional and departmental characteristics considered explains relatively little of the total variance in degrees granted per 100 students.

¹⁶ Bowen and Rudenstine make a similar point (188).

¹⁷ Bowen and Rudenstine examine three humanities fields: English, history, and political science.

¹⁸ Among the three humanities fields examined in detail, political science has the highest completion rate (55% vs. 50% for English and history), while English has the shortest average time-to-degree (6.4 years vs. 7.2 for history and 6.8 for political science).

¹⁹ Bowen and Rudenstine ascribe some of this difference to higher incidence of the terminal MA in larger programs than in smaller ones.

²⁰ The percentage of courses taught in the target language is also below average in Near Eastern language departments, though the number of departments in this group is too small to yield reliable means or percentages. For this reason Near Eastern language departments, along with joint German and Slavic language departments and Altaic and Uralic language departments, are not discussed in the text.

²¹ The percentage of graduate students who are native speakers is also very low in classics departments, but since no one is a native speaker of classical Greek or Latin, this is true by definition. The inapplicability of the question about native speakers to classics departments also explains the low response to this question among those departments.

²² Almost 60% of the minority of departments offering both commonly and less commonly taught languages are organized as inclusive programs.

Works Cited

Blum, Debra E. “‘The Word is Getting Out’: Rise in Applications to Ph.D. Programs Reported at Universities Across Country.” Chronicle of Higher Education 14 Mar. 1990: Al+.

Bowen, William G., and Neil L. Rudenstine. In Pursuit of the PhD. Princeton: Princeton UP, 1992.