Sign up

Factors related to faculty research productivity and implications for academic planners: planners must align the emphasis on research and scholarly products with the overall institutional mission.
Article Type:
Report
Subject:
Academic achievement (Research)
Educational evaluation (Research)
Author:
Webber, Karen L.
Pub Date:
07/01/2011
Publication:
Name: Planning for Higher Education Publisher: Society for College and University Planning Audience: Academic Format: Magazine/Journal Subject: Education Copyright: COPYRIGHT 2011 Society for College and University Planning ISSN: 0736-0983
Issue:
Date: July-Sept, 2011 Source Volume: 39 Source Issue: 4
Topic:
Event Code: 310 Science & research
Product:
Product Code: 9105111 Educational Quality Assessment NAICS Code: 92311 Administration of Education Programs
Geographic:
Geographic Scope: United States Geographic Code: 1USA United States

Accession Number:
299759757
Full Text:
Introduction

Scrutiny over faculty allocations of time and subsequent outcomes is frequent in postsecondary education, especially at state research universities (those that receive appropriations from the state). Particularly during economic downturns, college officials and legislators must make difficult choices in fund allocation. For this reason, faculty workload and productivity continue to be highly debated issues, both within higher education institutions and legislative circles. The focus on research is particularly strong at research universities, but it is also growing in emphasis at other four-year colleges. Toma (2008) cautions that officials at some institutions may see an increased emphasis on research productivity as a way to increase rankings or other indicators of prestige. Because of the continued or increasing emphasis on both research production and institutional policies that encourage faculty research, a greater understanding of research productivity is important for academic planners today.

Because faculty productivity is a complex mix of individual and environmental characteristics, the study described in this article includes a broad set of contributor individual and institutional variables and also examines interactions among some of the variables. To date, only a few empirical studies have examined the contributors to faculty research productivity using the most recent national set of data available on postsecondary faculty in the United States, the 2004 National Study of Postsecondary Faculty (NSOPF:04). The study described herein is significant because it accounts for the nested nature of the data using a hierarchical generalized linear model and examines six separate factors of research productivity. Using the generalized model allows us to address the non-normal distribution of count data, which then enables us to examine the contribution of individual characteristics nested within the context of institutional characteristics. Thus, this examination enables us to expand our understanding of the factors that contribute to faculty research productivity in four-year colleges and universities and consider the implications for academic planning and practice.

Purpose of this Study

Although some previous attention has been given to faculty productivity, we do not yet have a complete understanding of the differences that exist across discipline and institutional type for faculty today. Data available from NSOPF:04 allow us to extend our understanding of six different measures of faculty productivity that vary in emphasis across discipline based on a nationally representative sample of college faculty. Specifically, the following four questions are addressed in this study:

1. What individual factors contribute to research productivity; namely, how do rank, gender, race, marital status, having children, percentage of time spent on research, and credit hours generated from teaching contribute?

2. Are there interactions between gender, marital status, dependent children, and discipline?

3. What institutional factors influence research productivity; namely, does the proportion of time allocated to teaching versus research influence productivity? and

4. Do these differences occur by discipline and type of institution?

Method

The data used for this study include over 9,300 records from the NSOPF:04 database. Non-tenure-track and part-time faculty are excluded, and the weighting procedure available in the data set for all faculty is used. The weighted sample represents over 300,000 faculty members in nearly 1,200 four-year colleges and universities. Institutions are grouped into three categories: doctoral-extensive; doctoral-intensive; and bachelor's and master's (grouped in one category). Data elements for ratio of FTE students to faculty, public/private status, research expenditures, and total expenditures are obtained from the NSOPF:04 Institutional Questionnaire.

In this analysis, the dependent variables include the seven NSOPF:04 variables that asked respondents to indicate the number of academic products completed in the past two years. These items include the number of

* articles in refereed journals

* articles in non-refereed journals

* book reviews, chapters, and creative works

* books, textbooks, and reports

* presentations

* exhibitions and performances

* patents and computer software

These are measures of scholarly work across most disciplines in postsecondary education, and when these measures are examined in light of the individual and institutional characteristics noted above, they can provide much-needed updates on the factors that account for faculty productivity today.

Initial exploratory analyses found that many faculty members had zero scholarly products, thus showing a strong skew toward zero. To examine the research questions and accommodate for the non-normal distribution, a two-level hierarchical generalized linear model (HGLM) is used, HLM version 6.06. Using the counts available in the data set, the Poisson distribution model represents the number of publication events occurring during a two-year time period. Six separate HGLM models are developed for the outcome variables; figure 1 shows descriptive statistics for the outcome variables by disciplinary field. A model could not be developed for the seventh measure, patents and computer software, as originally planned because of its very low count rate. Figure 2 shows the independent variables at the individual (level 1) and institutional levels (level 2) used in the full HGLM model. One can see in figure 2 that the standard deviation for some scores is larger than the mean, indicating the high skew for many of these count variables.

Results

Mean scores show that, on the whole, faculty members produce the greatest number of research products in the areas of presentations, refereed journal articles, and non-refereed journal articles. Faculty members in physical and life sciences produce about one-third more presentations and about twice as many refereed articles as faculty in other disciplines. Faculty members in arts and humanities produce the vast majority of exhibitions and performances, while faculty in physical and life sciences produce the greatest number of patents. These findings emphasize the fact that the type of research product produced differs by discipline.

Mean scores show that 85 percent of the faculty in the sample are white, 31 percent are female, 76 percent are married, and 52 percent have one or more children under the age of 18. Overall, these faculty indicate relatively high satisfaction--92 percent said they would choose an academic career again. The mean scores show that the number of faculty who have research as their principal activity and the percentage of time spent on research are low (17 percent and 25.4 percent, respectively); this may be because baccalaureate through doctoral-granting institutions are included and a relatively small number of faculty report research as their primary activity. Figure 3 lists all NSOPF disciplines within the five groups.

Figure 4 presents results from the HGLM model and shows that a number of individual and institutional variables have a significant effect on faculty research productivity. The 3 coefficients shown in figure 4 are log of the event rates (thus making comparative interpretation more difficult), but the exponential numbers (Exp 3) for each coefficient enable the reader to more easily see the effect or impact of each variable. For example, for refereed journal articles, the Exp 3 value is 1.02. This can be interpreted to say that holding all other variables constant, the faculty member who apportions more than 50 percent of his or her academic time to research has .02 or two percent more refereed journal articles published within the past two years than one who spends less than 50 percent time on research. Conversely, a negative 3 coefficient indicates a negative effect; for example, at level 2, the 3 coefficient for private institutions is -.28, with an Exp 3 of 0.76. These figures can be interpreted to say that (holding all other factors constant), faculty from private institutions publish 24 percent fewer refereed journal articles than faculty from public institutions.

Using these guidelines for interpretation, the results presented in figure 4 show many important findings. Overall, reported time spent on research activities positively affects research productivity in four of the six items shown (recent refereed journal articles; book reviews, chapters, and creative works; books, textbooks, and reports; and presentations). Coefficients show that when faculty spend more time on teaching, they produce up to 10 percent fewer scholarly research products.

Results by teaching, rank, and satisfaction. HGLM results in figure 4 show a significant and negative effect for credit hours generated from teaching on five variables of productivity (recent non-refereed journal articles; book reviews, chapters, and creative works; books, textbooks, and reports; presentations; and exhibitions and performances). Findings indicate that faculty who spend more time on teaching are less productive with their research. Examining productivity by rank, full professors are more productive across all measures. For example, full professors are 38 percent more productive in producing refereed journal articles than associate and assistant professors. The finding that one would choose an academic career again is significantly related to one measure of productivity--recent books, textbooks, and reports. Faculty members who said they would choose an academic career again are nine percent more productive in writing a book, textbook, or report within the past two years (but the variable for overall satisfaction did not have a significant effect on any other measure of research productivity).

Results by race, marital status, and gender. Results in figure 4 show that race (whites vs. all others) and marital status have no significant effect on the six measures of research productivity. These findings indicate that for all respondents in this data set, race and marital status do not contribute significantly to research productivity. Being female has no significant effect on the production of refereed articles, books, textbooks, or presentations. This indicates that females have production rates similar to their male peers for these items. However, female faculty produce significantly fewer non-refereed articles, book reviews and chapters, and exhibitions and performances than their male peers. Some interaction effects are also noted; female faculty in the physical and life sciences produce 25 percent fewer refereed journal articles than faculty in other fields or male faculty in the physical and life sciences, and female faculty with one or more dependent children produce eight percent fewer books and textbooks than female faculty without dependent children or male faculty. Overall, however, having dependent children has a significant and positive effect on refereed (13 percent more) and non-refereed journal articles (18 percent more) compared to faculty with no dependent children.

Results by academic discipline. Results in figure 4 also show significant differences in research productivity by academic discipline. Faculty members in physical and life sciences produce 32 percent more recent refereed articles but 25 percent fewer non-refereed articles than faculty in arts and humanities. Faculty in business and legal fields produce 62 percent fewer book reviews, chapters, and creative works compared to faculty in arts and humanities and social sciences and education. In addition, faculty in arts and humanities produce significantly more exhibitions and performances than faculty in all other discipline groups. These significant differences in production rates by discipline group show that different forms of scholarship are emphasized across academic fields.

Results by research expenditures, size, and type of institution. Not surprisingly, those institutions that apportion more funds to research saw a positive and significant effect on recent refereed and non-refereed journal articles, books and textbooks, and presentations. The proportion of expenditures allocated for research and the institution type (non-doctoral, doctoral-extensive, or doctoral-intensive) have a significant positive effect on most of the productivity items. For example, for each standard deviation unit increase in the ratio of expenditures for research to total expenditures, there is a two-fold increase in the number of refereed journal articles and a four-fold increase in the number of non-refereed articles. Respondents from public institutions produce 24 percent more articles in refereed journals; 23 percent more book reviews, chapters, creative works, and books; 12 percent more books and textbooks; and 17 percent more presentations than those from private institutions. Size of institution (indicated by student-to-faculty ratio) is significant for only one of the six forms of productivity: book reviews, chapters, and creative works. Findings show that faculty who work in institutions with a higher FTE student-to-faculty ratio produce about two percent fewer book reviews, chapters, and other creative works than those faculty in institutions with a lower student-to-faculty ratio. As might be expected, faculty at doctoral institutions produce more recent refereed journal articles, book reviews, and presentations than peers in non-doctoral institutions. In particular for refereed articles, 98 percent of the variance in level 2 is accounted for by the variables included; results show that there is great variation in the number of refereed articles produced by type of institution. For example, respondents from doctoral-extensive institutions report 62 percent more refereed articles than those from master's and bachelor's institutions, and respondents from doctoral-intensive institutions report 63 percent more refereed articles than those from non-doctoral institutions.

Discussion and Implications

This analysis sought to further examine and update the factors that contribute to research productivity for faculty members in today's four-year colleges and universities. Six multilevel models were developed to examine individual factors nested within characteristics of institutional type. Since the distribution of frequency scores for many of the productivity measures was highly skewed toward zero, an over-dispersion Poisson HGLM technique was used. These six models allowed us to comprehensively examine a variety of scholarly products that are frequent and critical measures of faculty productivity.

Overall, findings from these analyses are consistent with the majority of previous reports on faculty productivity, but also offer additional insights on the interactions among some of the variables. Of the six measures examined, presentations and refereed journal articles are the most frequent outputs. Although different products are emphasized across disciplines, public distribution of research findings is most commonly done through verbal and written communication, so this finding makes sense. Consistent with the findings of Marsh and Hattie (2002) and Porter and Umbach (2001), findings show that faculty members who apportion more of their time to research or who identify research as their primary activity produce the highest number of scholarly products. The negative relationship found between teaching and research productivity is consistent with findings reported by Marsh and Hattie (2002) and Fairweather (2002). In institutions in which knowledge production is a primary focus, officials may wish to consider how to allocate the tasks of teaching, research, and service to best achieve all facets of the institution's mission. This may result in faculty hired specifically for research who have limited or no teaching duties.

Indeed, college and university rankings have prompted some officials to consider (or to have actually changed) their mission emphases as a way to climb the ranks. Institutions with higher visibility achieved through notable patents or other scientific discoveries receive high marks on peer assessments completed by college presidents and capture a significant portion of federal research and development funds. Decreases in state funds for higher education further entice institutional officials to consider a strong emphasis on research. For some institutions, an increased emphasis on research is a reasonable consideration. New scientific discoveries and knowledge are critical to the economic and social improvement of our society, and academic research serves as a vehicle for knowledge production, contributes to economic growth, and can guide public policy.

However, before jumping to add higher expectations for research, institutional officials may wish to consider where and how much research and knowledge production fits within the institution's mission. Unlike the research university, non-doctoral-granting institutions may have a much larger emphasis on teaching, workforce development, or service to the community. In this type of institution, a focus on and reward for success in teaching, student certification, or community service may be more advantageous.

The primary emphases of an institution will not only determine how faculty members spend their time, but will also determine resource allocation priorities, policies for promotion and tenure, and faculty satisfaction. Results reported in this study show that apportioning more funds to research significantly contributes to higher research productivity. Especially in some disciplines, research requires extensive equipment, physical space, and additional personnel to support activities such as grant administration. If institutional officials choose to emphasize research, then expectations related to salary, time for teaching, criteria for promotion and tenure, and general satisfaction may change. Productive faculty members who have a unique skill or cumulative set of skills may consider moving to another institution with a better salary or facility or to industry. Given issues of salary compression and industry market demands, institutional officials are challenged to retain current, valued faculty within limited budgets while also trying to bring in new "star" faculty who hold promise for teaching and research.

The endeavor of knowledge production is not only resource intense, but also time intense. Faculty who engage in high levels of research may not have as much time to devote to instruction. To lighten the teaching load to allow faculty to focus on research, senior officials sometimes employ adjunct faculty or teaching assistants. While this can assist with ensuring credit hour production, it might cause unintended consequences in students not receiving the same quality of instruction as they would from tenured and tenure-track faculty. Indeed, it is a fine balance for senior administrative officials to manage the many needs related to institutional mission.

However, faculty members who are productive researchers may serve as role models for students in the classroom, laboratory, and other campus-based activities. Ideas gleaned from research may be brought into classroom discussions, and this can bring classroom learning to life. In this capacity, faculty as instructors are sharing the most up-to-date thinking as well as demonstrating positive mentoring practices. Such practices may move students to consider additional study or an academic research position as a career option. To ensure a continued effective balance between faculty roles in teaching, research, and service, institutional leaders must consider what is right for their institution, being mindful that movement away from the balance has consequences for student learning and public service.

Along with implications related to the emphasis on teaching and research, these findings may have implications for interactions by gender and other demographic characteristics. Overall, female faculty produce fewer non-refereed articles, book reviews, and chapters than their male peers but similar levels of peer-reviewed journal articles and scholarly presentations. There were some significant interaction effects for women that indicate differences by discipline and for those with young children. Female faculty in the physical and life sciences produce 25 percent fewer refereed articles than their male peers, and female faculty with dependent children produce eight percent fewer books and textbooks than males and females without children. These negative effects for women and women with dependent children echo findings from Creamer (1998) and Perna (2005), but the mixed findings for female faculty overall show the complexities that may exist among gender, rank, and marital status. Typically, junior-ranked faculty members are younger, may have young children, and are also trying to publish to earn tenure.

Additional complexities related to departmental and discipline expectations make success in the academic role even more challenging, especially for women. The "pipeline" metaphor assumes that an enlarged pool of females with doctoral degrees will expand the number of female faculty. Although the proportion of women who have earned doctorates has certainly grown over the past two to three decades, Kulis, Sicotte, and Collins (2002) argue that the number of female faculty has not increased proportionately to the number of women with earned doctorates due to leaks in the pipeline.

It is likely that attitudes about and expectations for women are changing, even if slowly, and that the number of mixed findings for female faculty related to research productivity may indeed result from changes that are occurring at this time in history. To advance our institutions toward greater diversity and equity, we need to find ways to ensure that female and underrepresented faculty receive and perceive strong institutional support. Female faculty in male-majority departments will benefit when workloads and expectations for teaching and research are comparable for all. All junior faculty, regardless of rank or gender, may benefit from mentoring programs.

The finding that more research productivity occurs in institutions that apportion more funds to research is not surprising. By devoting more expenditures to research, institutions are outwardly showing what they value and, subsequently, how faculty will be rewarded. The finding that research institutions are apportioning more funds to research affirms Slaughter and Rhoades' (2004) assertion of an increase in academic capitalism and the likelihood that faculty will be motivated to spend more time on research and those other activities (not teaching) that will be most highly rewarded (for example, on refereed publications rather than on institutional service or substantial attempts to increase skills in the teaching-learning process). Findings of lower research productivity at non-doctoral institutions and those that apportion fewer expenditures to research are also not surprising. The mission of master's and bachelor's institutions is typically focused on teaching more than research, and therefore more faculty time should be apportioned to activities related to instruction. As mentioned above, however, clear caution is urged for institutional leaders who encourage greater research production merely as a way to climb the ladder in external rankings or to seek other forms of prestige. Academic planners must align the emphasis on research and scholarly products with the overall goals and mission of their institution.

Findings of difference by discipline are consistent with previous findings (Baird 1986; Fairweather 2002; Porter and Umbach 2001). The typical physical and life sciences discipline may include traditional laboratories and multiple post-doctoral researchers and graduate students. In this setting, the nature of scientific inquiry is simply different--it facilitates different products and larger numbers of products than the typical arts and humanities department. Differences in the emphases of research may also affect budget needs. Along with more equipment, science faculty often require more undergraduate or graduate student assistance. Students who serve as research assistants receive valuable hands-on training that can increase technical skills and critical thinking. However, they also require a department to ask for more institutional financial support. Research expectations by discipline may also affect expectations for tenure and promotion. Faculty and senior administrative officials who serve on tenure and promotion committees must be aware of such differences by discipline so that faculty members are not evaluated on a one-size-fits-all mentality.

In addition, research methods and techniques that differ by discipline may affect resource allocations and expectations for extramural funding. For example, compared to the equipment-intensive laboratories required in many science fields, research completed by a historian typically involves review of documents done individually with few (if any) pieces of equipment. Few institutions can afford to pay for the total set of equipment required in today's high-tech laboratories; happily, some granting agencies fund the addition of expensive equipment. This kind of extramural funding can be a win-win--the faculty member and the institution produce new knowledge and the granting agency contributes to economic and knowledge production and may also benefit from patent proceeds.

Results reported herein confirm the complex nature of faculty research productivity for many of today's college and university faculty. While individual characteristics and perceptions of satisfaction and career choice greatly contribute to productivity, so do interactions with other individuals in one's department, across campus, and even across institutions. Interactions among faculty members (by gender, race, and rank) are critical and can help create a sense of community and may produce positive results for research productivity and for organizational effectiveness more broadly. In a detailed study on faculty workload, Lee (2004) discusses the need for, and the high levels of, collaboration found among postsecondary scientists. Further, Bozeman and Boardman (2004) believe that collaboration among researchers can bring together a diverse set of faculty who may have different experiences in teaching, grant acquisition, and tenure. Work with peers is also important for participation in faculty governance. Collaboration may help some faculty members feel confident about getting involved in departmental or campus committees, and this may affect policy decisions made through governance procedures.

Increased expenditures apportioned to research are likely to shift the areas of focus and reward for faculty. This may mean that as tenure and tenure-track faculty devote more time and cognitive effort to research, adjunct faculty will take on a larger role in instruction. Academic planners must continue to maintain a careful balance in addressing institutional fiscal needs, both through research as well as instruction. A lack of faculty interaction and greater use of adjunct faculty could potentially have a negative impact on students' level of satisfaction and success.

Limitations

Findings for women's productivity may be limited since we focus exclusively on full-time, tenured or tenure-track faculty. This subgroup was chosen because the nature of research responsibilities is more consistent for this group; however, effects for women may be underestimated since a high proportion of female faculty are placed in part-time and non-tenure-track positions. Due to the nature of multiple interactions for factors related to productivity, this study chose to examine a more limited set of interactions. A broader set of interactions may add insight; yet, a large set of interactions may also yield more findings that overlap and become unwieldy to interpret. Findings presented herein are for scholarly products created within a two-year period by faculty in four-year institutions. Because expectations for research may be less at non-doctoral institutions, additional analyses for only doctoral-granting institutions may be beneficial. Analyses were also conducted to examine research productivity (the same seven measures) over a faculty member's career, and the findings were remarkably similar. Additional insight into factors related to research productivity may be gained through the use of a multiple multivariate structural equation model in which all or some of the seven measures of productivity are included in one latent construct.

Acknowledgement

I gratefully acknowledge Cigdem Alagoz for her assistance in analyses for this article.

References

Baird, L. L. 1986. What Characterizes a Productive Research Department? Research in Higher Education 25 (3): 211-25.

Bozeman, B., and C. Boardman. 2004. The NSF Engineering Research Centers and the University-Industry Research Revolution: A Brief History Featuring an Interview with Erich Bloch. Journal of Technology Transfer 29 (3-4): 365-75.

Creamer, E. G. 1998. Assessing Faculty Publication Productivity: Issues of Equity. ASHE-ERIC Higher Education Reports, vol. 26, no. 2. San Francisco: Jossey-Bass.

Fairweather, J. S. 2002. The Mythologies of Faculty Productivity: Implications for Institutional Policy and Decision Making. Journal of Higher Education 73 (1): 26-48.

Kulis, S., D. Sicotte, and S. Collins. 2002. More than a Pipeline Problem: Labor Supply Constraints and Gender Stratification across Academic Science Disciplines. Research in Higher Education 43 (6): 657-91.

Lee, S. 2004. Foreign-Born Scientists in the United States: Do They Perform Differently than Native-Born Scientists? Ph.D. diss. Georgia Institute of Technology. Retrieved April 23, 2011, from the World Wide Web: www.rvm.gatech.edu/paperfiles /04/lee_sooho_200412_phd.pdf.

Marsh, H. W., and J. Hattie. 2002. The Relation between Research Productivity and Teaching Effectiveness: Complementary, Antagonistic, or Independent Constructs? Journal of Higher Education 73 (5): 603-41.

Perna, L. W. 2005. Sex Differences in Faculty Tenure and Promotion: The Contribution of Family Ties. Research in Higher Education 46 (3): 277-307.

Porter, S. R., and P. D. Umbach. 2001. Analyzing Faculty Workload Data Using Multilevel Modeling. Research in Higher Education 42 (2): 171-96.

Slaughter, S., and G. Rhoades. 2004. Academic Capitalism and the New Economy: Markets, State, and Higher Education. Baltimore: Johns Hopkins University Press.

Toma, J. D. 2008. Positioning for Prestige in American Higher Education: Case Studies of Strategies at Four Public Institutions toward "Getting to the Next Level" Summary paper. Los Angeles: Center for Enrollment Research, Policy, and Practice (CERPP), University of Southern California. Retrieved April 23, 2011, from the World Wide Web: www.usc.edu /programs/cerpp/docs/CERPPSummaryPaperToma.doc.

Karen L. Webber, Ph.D., is associate professor in the Institute of Higher Education at the University of Georgia. She previously held positions in institutional research at the University of Delaware and the University of Georgia and served as interim associate provost at the University of Georgia.
Figure 1 Descriptive Statistics for Outcome Variables

                          Disciplinary Field Category

                                   Arts and     Physical and
                                  Humanities    Life Sciences
                                   (weighted      (weighted
                                  N = 33,940)    N = 88,010)

Recent articles,     Mean               1.78            4.67
refereed journals    Standard           3.34            5.77
                     Deviation

Recent articles,     Mean               1.26            1.50
non-refereed         Standard           3.10            3.33
journals             Deviation

Recent book          Mean               1.25             .87
reviews, chapters,   Standard           2.27            1.93
creative works       Deviation

Recent books,        Mean                .46             .55
textbooks, reports   Standard           1.20            1.67
                     Deviation

Recent               Mean               3.96            6.03
presentations        Standard            4.9            7.62
                     Deviation

Recent               Mean               4.16             .15
exhibitions,         Standard          10.94            2.14
performances         Deviation

Recent patents,      Mean                .03             .22
computer software    Standard            .40             .80
                     Deviation

                                  Disciplinary Field Category

                                  Social Sciences,    Business
                                   Education, and     and Legal
                                    Agriculture       (weighted
                                     (weighted       N = 17,310)
                                    N = 60,590)

Recent articles,     Mean                    2.45          2.21
refereed journals    Standard                3.48          2.62
                     Deviation

Recent articles,     Mean                    1.48          1.06
non-refereed         Standard                3.26          2.01
journals             Deviation

Recent book          Mean                    1.20           .59
reviews, chapters,   Standard                2.08          1.44
creative works       Deviation

Recent books,        Mean                     .66           .53
textbooks, reports   Standard                1.47          1.29
                     Deviation

Recent               Mean                    5.45          4.04
presentations        Standard                6.67          4.78
                     Deviation

Recent               Mean                     .27           .03
exhibitions,         Standard                2.40           .41
performances         Deviation

Recent patents,      Mean                     .07           .05
computer software    Standard                 .41           .31
                     Deviation

                                     Total
                                   (weighted
                                  N = 206,410)

Recent articles,     Mean                3.05
refereed journals    Standard            4.98
                     Deviation

Recent articles,     Mean                1.40
non-refereed         Standard            3.12
journals             Deviation

Recent book          Mean                1.03
reviews, chapters,   Standard            2.03
creative works       Deviation

Recent books,        Mean                 .57
textbooks, reports   Standard            1.47
                     Deviation

Recent               Mean                5.20
presentations        Standard            6.59
                     Deviation

Recent               Mean                1.00
exhibitions,         Standard            5.52
performances         Deviation

Recent patents,      Mean                 .11
computer software    Standard             .58
                     Deviation

Note: All values of mean, standard deviation, and frequency are
weighted and rounded. Statistics do not include the "Other" category,
weighted N = 6,570. Ns are weighted and rounded.

Figure 2 Descriptive Statistics for Independent Variables

                                           Weighted   Minimum   Maximum
                                              N

Level 1:          Principal activity       348,900         0         1
Individual          is research
Characteristics   Percent time spent on    348,900         0       100
                    research activities
                  Race, white              348,900         0         1
                  Married = 1              348,900         0         1
                  Having one or more       348,900         0         1
                    dependent children,
                    yes = 1
                  Would choose an          348,900         0         1
                    academic career
                    again, yes = 1
                  Arts and humanities      348,900       .00      1.00
                    discipline
                  Physical and life        348,900       .00      1.00
                    sciences discipline
                  Social sciences,         348,900       .00      1.00
                    education, and
                    agriculture
                    discipline
                  Business and legal       348,900       .00      1.00
                    discipline
                  Female = 1               348,900       .00      1.00
                  Full professor = 1       348,900       .00      1.00
                  Total credit hours       348,900       .00     60.00
                    generated in
                    teaching

Level 2:          Bachelor's and               500         0         1
Institutional       master's
Characteristics   Doctoral-extensive           500         0         1
                  Doctoral-intensive           500         0         1
                  Ratio of FTE                 500         2        35
                    enrollment to FTE
                    faculty
                  Ratio of expenditures        500       .00      0.47
                    for research to
                    total expenditures

                                            Mean    Standard
                                                    Deviation

Level 1:          Principal activity         .17        .379
Individual          is research
Characteristics   Percent time spent on     24.8       22.18
                    research activities
                  Race, white                .85         .36
                  Married = 1                .76         .43
                  Having one or more         .52         .50
                    dependent children,
                    yes = 1
                  Would choose an            .92         .28
                    academic career
                    again, yes = 1
                  Arts and humanities        .20         .40
                    discipline
                  Physical and life          .36         .48
                    sciences discipline
                  Social sciences,           .32         .47
                    education, and
                    agriculture
                    discipline
                  Business and legal         .09         .28
                    discipline
                  Female = 1                 .31         .46
                  Full professor = 1         .40         .49
                  Total credit hours        6.88        4.42
                    generated in
                    teaching

Level 2:          Bachelor's and            0.53        4.59
Institutional       master's
Characteristics   Doctoral-extensive        0.28        0.44
                  Doctoral-intensive        0.19        0.38
                  Ratio of FTE             14.22        4.68
                    enrollment to FTE
                    faculty
                  Ratio of expenditures    0.0443      0.085
                    for research to
                    total expenditures

Note: Ns are weighted and rounded.

Figure 3 Academic Fields in Each Discipline Group

Arts and Humanities    Arts--visual and performing
                       Communication/journalism/communication tech
                       English language and literature/letters
                       Foreign languages/literature/linguistics
                       Philosophy, religion, and theology

Physical and           Biological and biomedical sciences
Life Sciences          Engineering technologies/technicians
                       Health professions/clinical sciences
                       Mathematics and statistics
                       Physical sciences
                       Science technologies/technicians

Social Sciences,       Agriculture/natural resources/related
Education, and         Computer/information sciences/support tech
Agriculture            Education
                       Family/consumer, human sciences
                       Library science
                       Psychology
                       Public administration/social services
                       Social sciences and history

Business and Legal     Business/management/marketing/related
                       Legal professions and studies

Others                 Architecture and related services
                       Area/ethnic/cultural/gender studies
                       Multi-/interdisciplinary studies
                       Parks/recreation/leisure/fitness studies
                       Security and
                       Other

Figure 4 HGLM Results of Model for Faculty Productivity

                                           Recent refereed journal
                                                   articles

                                           Coeff.     Exp ([beta]) Sig

Intercept                                   0.268         1.31 **

Level One
Research is principal activity
  (1 = yes, 0 = no)
Percent time spent on research              0.016         1.02 **
Female (1 = yes, 0 = no)
Race (1 = white 0 = other)
Married (1 = yes, 0 = no)
Full professor (1 = yes, 0 = other)         0.322         1.38 **
Have dependent children                     0.124         1.13 **
  (1 = yes, 0 = no)
Would choose academic career again
Physical and life sciences disciplines      0.280         1.32 **
Social sciences, education,
  and agriculture disciplines
Business and legal disciplines
Other disciplines
Total credit hours in teaching
Female x Physical and life sciences        -0.290         0.75 **
  disciplines
Research is principal x Physical and        0.250         1.28 **
  life sciences
Time on research x credit hours
  generated
Female x Children
Female x Social Science, education,
  and agriculture discipline
Level Two
Ratio of FTE enrollment/FTE faculty
Ratio of expenditures for research/         1.110         3.03 **
  total expenditures
Control: Private not-for-profit            -0.280         0.76 **
  (1 = yes, 0 = no)
Level: Doctoral-extensive                   0.480         1.62 **
  (1 = yes, 0 = no)
Level: Doctoral-intensive                   0.490         1.63 **
  (1 = yes, 0 = no)

Variance Components
Variance between institutions                0.48
Proportion of variance explained at          0.99
  level 2
Variance within institutions                 0.04
Proportion of variance explained at          0.05
  level 1
Intraclass correlation                       0.11

                                           Recent non-refereed journal
                                                   articles

                                           Coeff.    Exp ([beta]) Sig

Intercept                                  -0.092          0.91

Level One
Research is principal activity
  (1 = yes, 0 = no)
Percent time spent on research
Female (1 = yes, 0 = no)                   -0.240         0.79 *
Race (1 = white 0 = other)
Married (1 = yes, 0 = no)
Full professor (1 = yes, 0 = other)         0.311         1.36 **
Have dependent children                     0.164         1.18 *
  (1 = yes, 0 = no)
Would choose academic career again
Physical and life sciences disciplines     -0.287         0.75 **
Social sciences, education,
  and agriculture disciplines
Business and legal disciplines
Other disciplines                          -0.273         0.76 *
Total credit hours in teaching             -0.040         0.96 **
Female x Physical and life sciences
  disciplines
Research is principal x Physical and
  life sciences
Time on research x credit hours             0.001         1.00 **
  generated
Female x Children
Female x Social Science, education,
  and agriculture discipline
Level Two
Ratio of FTE enrollment/FTE faculty
Ratio of expenditures for research/         1.650         5.21 **
  total expenditures
Control: Private not-for-profit
  (1 = yes, 0 = no)
Level: Doctoral-extensive
  (1 = yes, 0 = no)
Level: Doctoral-intensive
  (1 = yes, 0 = no)

Variance Components
Variance between institutions                0.18
Proportion of variance explained at          0.26
  level 2
Variance within institutions                 0.05
Proportion of variance explained at          0.03
  level 1
Intraclass correlation                       0.04

                                              Recent book reviews,
                                            chapters, creative works

                                           Coeff.    Exp ([beta]) Sig

Intercept                                  -0.269         0.76 **

Level One
Research is principal activity             -0.285         0.75 **
  (1 = yes, 0 = no)
Percent time spent on research              0.007         1.01 **
Female (1 = yes, 0 = no)                   -0.174         0.84 **
Race (1 = white 0 = other)
Married (1 = yes, 0 = no)
Full professor (1 = yes, 0 = other)         0.322         1.38 **
Have dependent children
  (1 = yes, 0 = no)
Would choose academic career again
Physical and life sciences disciplines     -0.732         0.48 **
Social sciences, education,
  and agriculture disciplines
Business and legal disciplines             -0.977         0.38 **
Other disciplines                          -0.454         0.64 **
Total credit hours in teaching             -0.028         0.97 **
Female x Physical and life sciences
  disciplines
Research is principal x Physical
  and life sciences
Time on research x credit hours             0.001         1.00 **
  generated
Female x Children
Female x Social Science, education,
  and agriculture discipline
Level Two
Ratio of FTE enrollment/FTE faculty        -0.025         0.98 *
Ratio of expenditures for research/
  total expenditures
Control: Private not-for-profit            -0.266         0.77 **
  (1 = yes, 0 = no)
Level: Doctoral-extensive                   0.521         1.68 **
  (1 = yes, 0 = no)
Level: Doctoral-intensive                   0.420         1.52 **
  (1 = yes, 0 = no)

Variance Components
Variance between institutions                0.30
Proportion of variance explained             0.63
  at level 2
Variance within institutions                 0.03
Proportion of variance explained at          0.02
  level 1
Intraclass correlation                       0.09

                                             Recent books, textbooks,
                                                    reports

                                           Coeff.    Exp ([beta]) Sig

Intercept                                  -1.192         0.30 **
Level One
Research is principal activity
  (1 = yes, 0 = no)
Percent time spent on research              0.004         1.00 *
Female (1 = yes, 0 = no)
Race (1 = white 0 = other)
Married (1 = yes, 0 = no)
Full professor (1 = yes, 0 = other)         0.090         1.09 *
Have dependent children
  (1 = yes, 0 = no)
Would choose academic career again          0.090         1.09 *
Physical and life sciences disciplines
Social sciences, education,                 0.245         1.28 *
  and agriculture disciplines
Business and legal disciplines
Other disciplines
Total credit hours in teaching             -0.045         0.96 **
Female x Physical and life sciences
  disciplines
Research is principal x Physical
  and life sciences
Time on research x credit hours             0.001         1.00 **
  generated
Female x Children                          -0.252         0.78 *
Female x Social Science, education,
  and agriculture discipline
Level Two
Ratio of FTE enrollment/FTE faculty
Ratio of expenditures for research/         0.878
  total expenditures
Control: Private not-for-profit            -0.245         0.78 **
  (1 = yes, 0 = no)
Level: Doctoral-extensive
  (1 = yes, 0 = no)
Level: Doctoral-intensive                   0.225         1.25 *
  (1 = yes, 0 = no)

Variance Components
Variance between institutions                0.18
Proportion of variance explained            <1.0%
  at level 2
Variance within institutions                 0.03
Proportion of variance explained at          0.47
  level 1
Intraclass correlation                       0.06

                                             Recent presentations

                                           Coeff.    Exp ([beta]) Sig

Intercept                                   1.446         4.25 **

Level One
Research is principal activity
  (1 = yes, 0 = no)
Percent time spent on research              0.009         1.01 **
Female (1 = yes, 0 = no)
Race (1 = white 0 = other)
Married (1 = yes, 0 = no)
Full professor (1 = yes, 0 = other)         0.113         1.12 **
Have dependent children
  (1 = yes, 0 = no)
Would choose academic career again
Physical and life sciences disciplines
Social sciences, education,
  and agriculture disciplines
Business and legal disciplines
Other disciplines
Total credit hours in teaching             -0.021         0.98 **
Female x Physical and life sciences
  disciplines
Research is principal x Physical
  and life sciences
Time on research x credit hours
  generated
Female x Children
Female x Social Science, education,
  and agriculture discipline
Level Two
Ratio of FTE enrollment/FTE faculty
Ratio of expenditures for research/         1.709         5.52 **
  total expenditures
Control: Private not-for-profit            -0.191         0.83 **
  (1 = yes, 0 = no)
Level: Doctoral-extensive
  (1 = yes, 0 = no)
Level: Doctoral-intensive
  (1 = yes, 0 = no)

Variance Components
Variance between institutions                0.15
Proportion of variance explained at          0.91
  level 2
Variance within institutions                 0.05
Proportion of variance explained at         <1.0%
Intraclass correlation                       0.03

                                              Recent exhibitions
                                                 performances

                                           Coeff.    Exp ([beta]) Sig

Intercept                                   1.569        4.802 **

Level One
Research is principal activity
  (1 = yes, 0 = no)
Percent time spent on research             -0.012        0.989 *
Female (1 = yes, 0 = no)                   -0.570        0.566 **
Race (1 = white 0 = other)
Married (1 = yes, 0 = no)
Full professor (1 = yes, 0 = other)
Have dependent children
  (1 = yes, 0 = no)
Would choose academic career again
Physical and life sciences disciplines     -3.784        0.023 **
Social sciences, education,                -2.812        0.060 **
  and agriculture disciplines
Business and legal disciplines             -5.220        0.005 **
Other disciplines                          -2.918        0.054 **
Total credit hours in teaching             -0.096        0.909 **
Female x Physical and life sciences
  disciplines
Research is principal x Physical
  and life sciences
Time on research x credit hours
  generated
Female x Children
Female x Social Science, education,
  and agriculture discipline
Level Two
Ratio of FTE enrollment/FTE faculty
Ratio of expenditures for research/
  total expenditures
Control: Private not-for-profit
  (1 = yes, 0 = no)
Level: Doctoral-extensive
  (1 = yes, 0 = no)
Level: Doctoral-intensive
  (1 = yes, 0 = no)

Variance Components
Variance between institutions                0.47
Proportion of variance explained at          0.04
  level 2
Variance within institutions                 0.19
Proportion of variance explained at          0.03
Intraclass correlation                       0.02
Gale Copyright:
Copyright 2011 Gale, Cengage Learning. All rights reserved.