Revisiting the Question of Equality: Tenure and Promotion Differences Among FullTime Women Faculty a

1
Revisiting the Question of EqualityTenure and
Promotion Differences Among Full-Time Women
Facultyand Faculty of Color at U.S. Community
Colleges

• Janna L. Oakes, PhD
• Janna.oakes_at_arapahoe.edu
• League of Innovations
• March, 2009

2
The Problem

• Despite progress made toward equality in the
  American workplace
• the under representation of women and people
  of color among tenured and high-ranking higher
  education faculty persists.
• (National Center for Education Statistics, 2003
  Perna, 2001 Perna, 2003).

3
Availability Data Suggests

• Females
• Female representation among earned PhDs has seen
  dramatic increases in the past 35 years (NSF,
  2007).
• However, female representation among the upper
  faculty ranks across all types of institutions
  does not reflect the increase in females with
  PhDs (NCES, 2006).
• People of Color (NSF, 2006)
• American Indian/Alaskan Native Over 400
  increase since 1975.
• Asian/Pacific Islander Just below 300 increase
  since 1975.
• Black Just below 50 increase since 1975.
• Hispanic Over 200 since 1975.
• Caucasian Less than 1 since 1975.
• Though faculty of color continue to represent a
  small percentage of full-time faculty, 2006 data
  from the NCES demonstrates that their numbers are
  slowly increasing among the higher faculty ranks
  nationwide.

4
Theoretical Framework

• HUMAN CAPITAL focused on investments
  intentionally made by individuals, such as
  advanced education and training, as well as
  intentional productivity efforts. Economic
  egocentricity, a quest for social elevation, and
  occupational dominance are the basis for most, if
  not all, HC investments (Block, 1990).
• LIMITATION Human capital theory, in its
  individualistic view, fails to recognize the
  broader social context in which work and
  inequality exist.
• STRUCTURAL CAPITAL includes fiscal strength and
  resources collective bargaining/unions
  branding tenure systems procedures, policies
  and processes, and other elements of
  organizational functionality that contribute to
  employee productivity.
• LIMITATION Structural capital theory, in its
  organizational view, fails to consider the
  intentional actions of the actors within the
  system.

5
Theoretical FrameworkHuman Capital and
Structural Capital

• By viewing workplace inequity through the lens of
  both human capital and structural capital, the
  current study endeavored to further isolate
  specific factors contributing to inequality in
  the academic workplace.
• Research regarding gender and racial/ethnic
  differences in tenure and promotion in American
  higher education support both human capital
  (Zellner, 1975 Polacheck Seibert, 1994
  Lindley, Fish, Jackson, 1992 Koch Chizmar,
  1976 Johnson Stafford, 1974 Marschke, 1997
  Perna, 2001 Perna, 2003) and structural capital
  perspectives (Bayer Austin, 1968 Bayer, 1997
  Perna, 2001 Perna, 2003 Smart, 1991) regarding
  academic labor markets.
• Social Capital Individuals opportunities and
  social credentials are verified by social
  connections. Thus, social capital reinforces
  identity, reputation, and recognition (Lin, 2001
  Bourdieu, 1983 Bourdieu, 1986). A very real and
  important factor in the lives of actors within
  any economy, social capital is comprised of
  concepts that are difficult to define or quantify
  (and therefore, to measure). Therefore, social
  capital was not a consideration in the current
  study.

6
The Question

• For full-time, citizen faculty at public, 2-year
  colleges, to what extent are gender and
  race/ethnicity related to tenure status and
  academic rank after controlling for human capital
  and structural considerations?

7
The Study

• Focus on Community Colleges
• Though women faculty and faculty of color are
  underrepresented in nearly all institutional
  types, that underrepresentation is less
  pronounced at Americas 2-year colleges (Perna,
  2003).
• By focusing solely on two-year institutions, 2
  extraneous variables (differences in
  institutional mission and in faculty who are
  attracted to different types of institutions) are
  eliminated from the model.
• Focus on Full Time Faculty
• Higher percentage of women and people of color in
  part-time instructional positions in higher
  education (Valian, 1999).

• Focus on Citizen Faculty
• Both gender and race/ethnicity are nested within
  citizenship.
• Unique Focus on Geographic Region
• CHEA regions used to sort institutions into
  geographic regions.
• Focus on Faculty Employed in Centralized
  Community College Systems
• The centralization of community colleges, with
  centralized policies and standards, is becoming
  increasingly common.

8
Factor Analysis

• Why?
• To ensure that the survey items selected reflect
  the constructs in the model previously described.
  
• Very few studies using NSOPF as a data source
  make mention of using factor analysis prior to
  conducting multivariate analyses.
• Resulted in some important and surprising -
  decisions.

9
Factor Analysis - Results

• Number of office hours per week and hours per
  week spent on unpaid duties were removed from
  further consideration due to their failure to
  load on any of the five factors.
• Percent of time spent on instruction was removed
  from further consideration due to ambiguity
  regarding whether this variable is a measure of
  work role (a structural characteristic) or
  intentionally-directed efforts (a human capital
  characteristic).

10
Dependent Variables

• Tenure (Dichotomous variable)
• Rank (Dichotomous variable)

11
Statistical Analysis

• Logistic regression analyses wasused to isolate
  the effects of gender and race on the dichotomous
  dependent variables after controlling for human
  capital and institutional structural
  characteristics.
• Only those factors determined by the factor
  analysis to be viable were included in the
  logistic regression analyses.
• Logistic regression will be employed because both
  outcome variables (tenured or not and full
  professor or not) are dichotomous variables.

12
Limitations

• There are no longitudinal data available that
  track faculty in American higher education.
• Use of the cross-sectional NSOPF data excludes
  those who leave academe and eliminates the
  capture of changes in the same faculty over time.
  This could lead to inaccurate interpretation of
  results regarding gender and racial/ethnic
  differences in the dependent variables.
• Sufficiency of variables provided by the NSOPF.
• Can the variable(s) underlying each response be
  determined?
• An integrated qualitative side to this pure
  research is lacking.
• Social capital, though difficult to quantify,
  must be considered.

13
H 1 Human capital investments will be the most
significant contributors to tenure attainment.

• FAILED TO REJECT
• Experience and holding a doctorate or first
  professional degree, both measures of human
  capital, emerged as significant predictors of
  holding tenured status for both the centralized
  system sample AND the complete sample.

H 2 Faculty of color will be less likely than
Caucasians to hold tenure, controlling for gender.

• REJECT
• Race/ethnicity was not a significant predictor of
  tenure status prior to, or after, introducing the
  independent variables.

14
H 3 Women faculty will be less likely than men
to hold tenure, controlling for race/ethnicity.

• REJECT
• The male/female difference in tenure attainment
  was evident in the initial analysis but was
  eliminated after introducing the human capital
  and structural variables into the model.
• The likelihood of attaining tenure was not
  significantly different for men than for women in
  either sample after controlling for structural
  and human capital considerations.

15
H 4 Centralized community college systems will
eliminate tenure and promotional differences for
women faculty and faculty of color.

• REJECT
• The results failed to support this hypothesis in
  its entirety.
• Neither gender nor race/ethnicity emerged as
  significant predictors of tenure in the complete
  sample or in the sub sample of faculty from
  centralized community college systems.
• However, race/ethnicity was related to attainment
  of the rank of full professor within both the
  larger sample and the centralized subsample.
• Centralized community college systems, as well
  as the larger population of community colleges in
  America, are failing to eradicate the Caucasian
  advantage in the attainment of full professor
  status within those colleges that employ faculty
  rank systems.

16
H 5 Perceptions of Unfair Treatment of Women
Faculty and Faculty of Color Will Be More
Predominant in Regions Showing Lower Tenure
Attainment and Academic Rank for Women Faculty
and Faculty of Color.

• REJECT
• No significant differences found between regions
  with regard to tenure attainment hypothesis
  rendered irrelevant.
• However, the odds of being tenured decreased
  significantly for all faculty employed at
  institutions in which respondents reported that
  minority faculty were treated fairly. This
  result was common to both the larger sample and
  the centralized system sample, and merits further
  investigation.

H 6 Region will be predictive of both tenure
attainment and academic rank for women faculty
and faculty of color.

• REJECT
• The results failed to support this hypothesis in
  its entirety.
• Region was not a significant predictor of tenure
  status for the larger sample or for the sub
  sample.
• However, region was a significant predictor of
  attainment of the rank of full professor for both
  samples with the NCA and WACS regions showing the
  greatest advantage, albeit to a smaller degree
  within the centralized system sub sample.

17
(No Transcript)
18
H 7 Structural characteristics will be
predictive of both tenure and academic rank for
women faculty and faculty of color, but less so
than human capital variables.

• FAIL TO REJECT
• Significance of the structural variables was less
  pronounced in the centralized system sub sample
  than in the larger sample for all four analyses,
  and was not as strong as the human capital
  variables.
• Union status and perception of fair treatment for
  minority faculty (structural characteristics)
  were predictive of tenure status for both
  samples, but less so for the sub sample than the
  larger sample.
• Hours spent on administrative tasks were
  predictive of tenure status, albeit only in the
  centralized community college system sample.
• The hours spent on administrative tasks
  variable was also a significant predictor of full
  professor rank attainment, but only in the larger
  sample.
• Institutional union status was a significant
  predictor of rank only in the centralized system
  sub sample.
• Region, a structural characteristic, was
  predictive of attaining the rank of full
  professor.

19
H 8 Women will be less likely than men to hold
the rank of full professor, controlling for
race/ethnicity.

• REJECT
• An initial regression revealed this to be true (p
  lt .05), with the odds of holding the rank of full
  professor favoring males by a factor of 1.292.
• Differences in the attainment of the full
  professor rank between men and women faculty were
  eliminated by introducing the human and
  structural capital variables into the model.

H 9 Differences in promotion rates for faculty
of color will be eliminated by controlling for
human capital investments and structural
characteristics.

• REJECT
• After controlling for gender, human and
  structural capital variables, Caucasian faculty
  maintained a significant advantage over faculty
  of color in the likelihood of attaining the rank
  of full professor.
• This result was true for both the larger sample
  and the centralized system sub sample.

20
H 10 Academic discipline will be the least
significant factor in the model for both genders.

• FAIL TO REJECT
• Academic discipline was the least significant
  factor in the model for both genders (consistent
  with Pernas observation that discipline was
  unrelated to tenure in four year institutions).
  
• Biglan discipline classification emerged as a
  significant predictor in only the tenure
  analysis, and only for faculty in hard, applied
  life systems (physical sciences, business,
  engineering). The advantage for faculty in hard,
  applied life systems was more pronounced in the
  centralized system sub sample.

21
Conclusions

• GENDER
• After controlling for race/ethnicity, structural
  capital and human capital variables, the tenure
  and academic rank differences between male and
  female faculty were eliminated.
• This finding is consistent with Pernas (2001,
  2003) conclusions, as described in Chapter Two,
  and suggests that womens productivity and
  experience is beginning to equal that of their
  male counterparts.
• This finding may also be due to the increased
  number of women faculty in academe, which results
  in increased social capital for women faculty.

22
Conclusions

• RACE/ETHNICITY
• After controlling for gender, structural capital
  and human capital, tenure differences between
  Caucasian and non-Caucasian faculty were
  eliminated for both samples. This is evidence of
  consistency within tenure practices in academe.
  
• However, the odds of holding the rank of full
  professor were significantly higher for Caucasian
  faculty than for faculty of color in the
  analyses for the full sample and the sub sample.
  
• Differences in rank for faculty of color
  attributable to social capital variables not
  included in the model, such as cultural
  differences regarding interpretations of
  policies, procedures and expectations, social
  connections within the institution or field of
  study, and differing comfort levels with existing
  cultures and practices that are inherent parts of
  the faculty career (Perna, 2001 Perna, 2003
  Toutkoushian, 1999 Tierney Bensimon, 1996).
• Capturing and measuring social capital is an
  important implication for future research
  regarding promotion differentials between faculty
  of color and their Caucasian colleagues.

23
Conclusions

• STRUCTURAL CAPITAL
• The odds of being tenured were significantly
  lower for faculty (regardless of gender or
  race/ethnicity) employed at institutions in which
  respondents reported perceptions of fair
  treatment for faculty of color. This variable
  has not considered in other studies, and merits
  further investigation.
• Analysis by CHEA region was a unique contribution
  to the literature, and was a significant factor
  in the rank analysis for both samples.
  Institutions in WACS and NCA have a decided
  advantage in attainment of the rank of full
  professor. Perhaps because these two CHEA regions
  are west of the colonial areas it has been easier
  for them to surpass the traditional, sometimes
  discriminatory values of the colonial college and
  arrive at consistency in promotional practices.
  Implication for future research Assessment
  regarding how the regions are similar or
  different in terms of promotion practices.
• Union status was a significant predictor of
  tenure for both samples. - Union status was not
  predictive of academic rank for the complete
  sample however, it was significant for the sub
  sample. - For the sub sample, the odds of
  holding the rank of full professor were 1.587
  times higher for faculty employed in institutions
  at which the faculty was unionized. - It may be
  that faculty working within centralized community
  college systems find less freedom both
  professionally and academically, and are
  therefore more likely to exercise their union
  voices with regard to faculty promotion. This
  is a new finding, and merits further research.

24
Conclusions

• HUMAN CAPITAL
• Education was a significant predictor of tenure
  and academic rank for both samples. Not
  surprising - academia has long valued the
  attainment of higher degrees.
• Years of experience was a significant predictor
  of tenure and academic rank for both samples.
  Consistent with one of the basic premises of the
  human capital theory ones worth increases as a
  result of experience. Sociological roles have
  changed many women enjoy professional careers,
  and some experience little if any time away from
  the workplace. This important change is likely
  part of the reason that the experience of women
  faculty is now closely mirroring that of their
  male counterparts, and may also explain the
  elimination of the tenure and promotion
  differentials between genders.
• Biglan discipline classification, a human capital
  consideration that has been used in other studies
  as a measure of structural capital (Perna, 2001
  2003), was a significant factor. The odds of
  holding tenure was significantly higher for
  faculty in hard, applied life systems for both
  samples. Because professionals in these fields of
  study earn substantial incomes outside of
  academia they may be both harder to recruit and
  therefore more highly valued when compared to
  professionals in other disciplines.
• Hours per week spent on advising did not emerge
  as a significant predictor of tenure or academic
  rank for either sample. Whereas this is a duty
  frequently delegated to graduate assistants and
  junior faculty in four year institutions,
  advising is more likely an inherent expectation
  of the community college faculty member.

25
Conclusions

• SO WHAT?!?
• Gender and racial/ethnic tenure differentials
  among citizen, full time faculty employed at
  public, two year institutions were explained by
  differences in human capital investments and
  structural characteristics.
• The lower representation of people of color among
  citizen full time faculty who hold the rank of
  full professor at public, two year institutions
  was not explained by the IVs, even within
  centralized community college systems.
• Results suggest that differences in attainment of
  the rank of full professor persist for citizen
  faculty of color who are employed within public
  two year colleges, and that these differences are
  more pronounced within centralized community
  college systems.
• Centralized community college systems differ from
  other community colleges in their promotion and
  tenure practices.- Centralized systems place
  greater emphasis on possession of the first
  professional or doctorate degree in tenure
  decisions than non-centralized two year colleges.
  - Faculty of color are significantly less likely
  than their Caucasian peers to attain the rank of
  full professor. The analysis of centralized
  community college systems in comparison to the
  larger population of two year public institutions
  is a new direction in higher education research,
  and a unique contribution to the literature.

26

• It is the responsibility of the United States to
  provide an education that is accessible and
  affordable for a diverse population of learners.
  Only in preparing an educated populace can our
  country maintain competitiveness in an
  increasingly diverse and global marketplace.
• To accomplish this, it is the responsibility of
  every institution of higher learning to develop
  healthy, diverse cultures and recruiting,
  retention, and promotion practices that attract
  and honor women faculty and faculty of color.