Part I:
Executive Report and Recommendations

D. Major Findings and Conclusions

From the extensive data sets obtained by the research team, there were several key findings that are summarized below. These findings provide the opportunity to better understand means of increasing recruitment and extending retention of URM faculty.

The results described below were observed in at least one of the three key modes of inquiry used in the study, which will be designated in the text with a parenthetic italicized letter at the end of the finding statement (survey (S), cohort analysis (C), qualitative interviews (I)), and in many cases were substantiated by two or more of these methods, often in conjunction with information obtained from minority faculty forums (MFF). Ultimately, these major findings have helped to inform and shape the recommendations of this committee. Full statistical analysis and details obtained from the study are presented in the research report in unabridged form. The research report also contains additional research findings, including findings on the salary analysis, details on the survey results and additional findings from the qualitative interviews.

Definitions

For this report, URM faculty were defined based on records of the provost, which included all faculty of African, Hispanic or Native American ethnicity who fit under the federal employment definition of all U.S. citizens and permanent residents regardless of country of origin, plus those holding permanent visas in the U.S. The list included citizens, permanent residents and newly arrived faculty on temporary visas who either identified as an underrepresented minority or who were identified as such if he/she had not identified at all. It did not include faculty who meet the guidelines for minority status, but who self-identified as White. As described in the Introduction section, it is recognized that this sample represents the broader definition of underrepresented minority faculty and includes many international faculty with African and Hispanic ethnicity.

Recruiting

MIT recruits heavily from its own and a few peer institutions: Data from the minority faculty who were interviewed (80% response rate) indicate that 36% have an MIT degree (UG or grad) and 60% received their doctoral degrees from three key universities (MIT, Stanford, Harvard - see Table 2) (I). Institute numbers are consistent with these data, indicating that 55% of all URM faculty have their Ph.D. from these schools, with similar, though slightly lower numbers from White (50%) and Asian (43%) faculty from the same three key universities.

To determine means of increasing the number of underrepresented groups on the MIT faculty, it is important to appreciate the universities that serve as primary sources of URM faculty for MIT. A large number of MIT minority faculty have acquired an MIT degree of some kind in the past, indicating a strong tendency of MIT departments to recruit from
the Institute's own alumni. These statistics also indicate that URM candidates with
previous exposure to MIT may be more likely to consider a faculty position at MIT. On the other hand, the narrowness of the sources of URM faculty - essentially more than half with Ph.D. degrees from only three top-tier institutions - indicates a significant lost opportunity to gain faculty from other schools. The fact that these schools also do not have a large number of minority candidates in their pools can exacerbate a problem presented from narrow recruitment sources.

Table 2. Feeder school data for MIT URM faculty from interview sample

URM faculty with an MIT degree (n=47)
Degree	Frequency	Percentage
No MIT degree	30	64%
Undergraduate	6	13%
Master’s	4	9%
Doctorate	10	21%
Post-Doctorate	3	6%
MIT Lifer	3	6%

Note: MIT Lifer denotes faculty receiving B.S., Ph.D. and faculty appointment at MIT

Graduate schools feeding MIT’s URM faculty (n=47)
Institution	Frequency	Percentage
MIT	10	21%
Harvard	9	19%
Stanford	9	19%
Yale	2	4%
Univ. of Chicago	2	4%
UC-Berkeley	3	4%
International	2	4%
Other	11	23%

Note: From URM interview data (n=47). Applies to schools from which faculty received a Ph.D.
Other includes Caltech, Cornell, CUNY, Julliard, Northwestern, NYU, Princeton, UCSB,
Univ. of Michigan, Univ. of Pennsylvania and Univ. of Virginia.

Conclusion: On the one hand, MIT has made good use of itself as a resource for faculty hiring. The Institute's ability to attract its own students can be used as a great opportunity to influence its future URM faculty numbers by cultivating positive student experiences and long-term relationships with its former students. On the other hand, an increase in the breadth of search could yield much larger numbers of URM faculty.

URM faculty hires reported more active recruitment than non-minority counterparts (I). Approximately 79% of non-URM faculty report directly applying to MIT over being specifically recruited, compared to just 37% for URM faculty. These numbers indicate that the dominant route for non-minority faculty to a position at MIT begins with a decision to apply, generally unsolicited. In the case of minority faculty, however, it is more often the case that
a member of the department approached the prospective faculty member and actively encouraged/recruited his/her application.

Conclusion: Proactive measures initiated by MIT have been key to recruitment of URM faculty. Direct engagement by department heads and deans in these cases has often made the difference in a successful hire.

There is some ambiguity and misinformation on the nature of Provost Opportunity Hires among both URM and non-URM faculty (I). Such perceptions may be damaging to faculty who are thought to result from such hires. The provost retains access to a small number of faculty slots that can be made available to departments that, upon completion of a search, find an excellent faculty candidate who will increase diversity and whom the department wishes to hire. The faculty slot is provided by the provost, and resources such as start-up funds and laboratory space are provided by the department, as is the case for other departmental hires. General understanding among some faculty about the Provost Opportunity Hire process was often incomplete or incorrect and, in many cases, such misunderstandings influenced perceptions about the program and those hired using the provost slots. In some cases, it was incorrectly believed that such hires take place outside of the usual departmental search and hire process, which is not true for most schools and departments. Often, URM or women candidates hired using a Provost Opportunity could be negatively perceived (by fellow faculty and/or self-perception) to be a second-choice or lower-ranked candidate or, in some cases, to have been hired without the same qualifications. In other cases, understanding around how the provost hire slot is made available and how it relates to the departmental role of providing new faculty resources (e.g., startup package or lab space) was unclear to faculty at large. These uncertainties seemed to primarily exist because of non-uniform information about the purpose, use and process surrounding such hires. The result of such perceptions could also influence those not hired using a provost's slot, due to the sometime presumption that a URM was hired under a "different" circumstance.

Figure 3. URM hiring by school, 1991-2009, from cohort analysis.

 

Note: Data counts dual hires as 50% in each department.

Figure 4. URM hiring by department from cohort analysis
(includes minorities of U.S. and international origin).

Conclusion: Lack of clarity about the use and purpose of Provost Opportunity Hires can lead to an undesired negative perception that could be alleviated with more open communication about the program and its process.

Hiring by school and department shows patterns in which minorities are consistently not hired in certain departments. There are also positive hiring patterns that are apparent in certain other departments/disciplines (C). The cohort analysis included the examination of incoming hiring of all faculty from 1991 to 2009 and determined the percentage of URM hires that took place during this time period. The number of URM hires is shown by school and by departmental unit in Figures 3 and 4, respectively. There are definite and consistent trends among the different schools, as seen in Figure 3, with the percentage of hires over this time period varying across a range: from the Whitaker School (22%) to MIT Sloan (13.3 %), SHASS(12.5%) to Engineering(9.3%), Architecture and Planning (6.3%) to the School of Science (3.4%).

The numbers provided per department indicate significant differences even within schools and also point to some departments in which there has been no minority hiring in the past two decades. On the contrary, there are certain departments that seem to have achieved relatively significant hiring of URM faculty.

It is clear that the hiring patterns reflect, in some part, the relative pools available within a given field. Successes within some of these more challenging fields in the recent past, however, indicate the potential to experience gains in faculty even given these kinds of challenges. A careful analysis of such departments within sets of fields or disciplines can lead to the learning and sharing of new approaches at MIT for increasing diversity in departments in similar disciplinary areas. Discussion and analysis with units that have had some difficulty in this area may also yield additional ideas about both increasing the pipeline and addressing the search and recruitment process.

Table 3. Current numbers of URM faculty by school and departmental unit
(2009 – 2010 academic year)

	URM	Asian	White	Grand Total	% URM
Architecture & Planning
Architecture	2	5	28	35	5.7%
Program in Media Arts & Sciences		5	15	20	0.0%
Urban Studies & Planning	3	3	23	29	10.3%
Total	5	13	66	84	6.0%
Engineering
School of Engineering			2	2	0.0%
Aeronautics and Astronautics	3	3	26	32	9.4%
Chemical Engineering	2	4	25	31	6.5%
Civil & Environmental Engineering	5	1	31	37	13.5%
Biological Engineering	1	2	15	18	5.6%
Electrical Engineering/ Computer Science	9	23	92	124	7.3%
Engineering Systems Division	1		6	7	14.3%
Material Sciences and Engineering	2	2	32	36	5.5%
Mechanical Engineering	3	18	47	68	4.4%
Nuclear Science and Engineering			15	15	0.0%
Total	26	53	291	370	7.0%
Humanities, Arts, & Social Sciences
Anthropology Program	1	1	6	8	12.5%
Economics	2	2	30	34	5.9%
Foreign Languages & Literature Section		3	5	8	0.0%
History Section	1	1	12	14	7.1%
Linguistics & Philosophy	2		24	26	7.7%
Literature Section	2	1	12	15	13.3%
Music & Theater Arts Section	2	3	8	13	15.4%
Political Science	1	2	21	24	4.2%
Program in Science, Technology & Society	1		12	13	7.7%
Program in Writing & Humanistic Studies	4	1	4	9	44.4%
Total	16	14	134	164	9.8%
Sloan School of Management
Total	10	15	81	106	9.4%
Science
Biology		5	49	54	0.0%
Brain & Cognitive Sciences	2	7	29	38	5.3%
Chemistry		3	26	29	0.0%
Earth, Atmospheric & Planetary Sciences	1	2	34	37	2.7%
Mathematics	1	6	43	50	2.0%
Physics	4	11	59	74	5.4%
Total	8	34	240	282	2.8%
Whitaker
Harvard/MIT Division of HST		1	6	7	0.0%
Office of Provost Area
Office of the Provost		1		1	0.0%
Dean for Student Life – DAPER
Total	1	1	10	12	8.3%
Grand Total	66	132	828	1026	6.4%

*Data from Provost Office of Institutional Research. Note: Dual hires are only counted once for the primary department or division. All data from 2009-2010 academic year, as reported November 2009.

These hiring numbers are not to be confused with the total number of minority faculty per department, which would include all current faculty members regardless of date of hire, and would also take into account losses of minority faculty during the cohort time frame. The total numbers of URM faculty per school and department for the 2009-2010 academic year at press time are provided in Table 3. These numbers can be compared to national university averages for the top 100 science and engineering research universities based on the 2007 Nelson Report,⁹ summarized for several STEM disciplines in Table 4. To also provide an idea of the immediately available pool, the percentage of URM Ph.D.s produced in each of these fields is also included. In many cases, even when viewed on the highest education level - namely the number of Ph.D. graduates - URMs remain underrepresented. MIT is approximately at, or in some cases, exceeds, the national average for certain fields; however, there are also several fields for which MIT is below the average. Given these data, it must be noted that the national Ph.D. numbers are low in general compared to the U.S. URM general population, which now exceeds 30%. Furthermore, there is not readily available data on URM postdoctoral candidates by field and discipline.

Conclusion: Over an extended time period, there are some units within MIT that had consistently low or zero hiring patterns with respect to minority faculty, indicating areas where focus, added resources, support and new strategies - for both pipeline and recruiting - could increase numbers. There are also units that have had relative success in URM hiring in past years, indicating the potential to examine and learn more about recruiting strategies within sets of fields or disciplines.

Table 4. 2007 URM data from top 100 research universities

Field	Discipline	URM % Ph.D. 96-05	% URM in top 100 department faculty*	% URM faculty at MIT in 2009/2010
Physical Sciences	Chemistry	7.5	3.9	0
Physical Sciences	Mathematics	6.1	3.3	2.0
Physical Sciences	Physics	5.2	2.5	5.4
Physical Sciences	Earth Sciences	5.5	3.7	2.7
Biological Sciences	Biology	7.8	4.1	0
Engineering	Chemical Engineering	7.7	5.6	6.5
Engineering	Civil Engineering	8.2	6.1	13.5
Engineering	Computer Science	6.6	2.8	{ 7.3
Engineering	Electrical Engineering	7.9	3.3
Engineering	Mechanical Engineering	3.7	4.1	4.4
Social Sciences	Economics	8.4	5.8	5.9
Social Sciences	Political Science	12.7	7.3	4.2
Social Sciences	Sociology	16.4	13.5	N/A
Social Sciences	Psychology	12.9	6.9	N/A

*At time of survey, URMs represented 28% of the 2006 U.S. population.

Data taken from "A National Analysis of Minorities in Science and Engineering Faculties at Research Universities," Donna J. Nelson, 2007.

Retention

A significant number of minority vs. non-minority faculty leave before or at the associate professor without tenure (AWOT) case. The first three to five years appear more critical to the retention of URM faculty than the majority group (C). Junior faculty at MIT undergo a two-step process to tenure that includes promotion to AWOT and promotion to associate with tenure. Cohort analysis data (see Table 5) indicate that a disproportionate number of URM faculty leave MIT prior to AWOT or after the AWOT case (i.e. without going up for a tenure case), when compared to the non-minority faculty group. For example, 74% of entering White assistant professors were promoted to AWOT, whereas only 55% of URM faculty were promoted, and 79% of Asian faculty. These numbers were statistically significant and provided a meaningful contrast in terms of expected outcomes for URM versus non-URM junior faculty at MIT. Once beyond the AWOT promotion, differences in URM versus non-URM tenure rates still indicate a difference (63% vs. 53%) but it is significantly lower and not statistically significant. The findings indicate that a disproportionately large number of minority faculty are lost within the early stages - generally the first three to five years that precede the first promotion. Reasons for early departure can range from other opportunities offered elsewhere to direct indications about the improbability of tenure, but it is clear that many faculty do not make it through these first critical years and end up leaving the Institute. This phenomenon constitutes a significant loss in the number of URM faculty retained at MIT.

Conclusion: These findings suggest that earlier intervention, more consistent mentoring and oversight, and a strong support structure during this time period could make a significant difference.

Table 5. Promotion rate data for AWOT and tenure taken from cohort analysis

Promotion from AWOT to Tenure (Asst. Professors hired from 1991 to 2000) Promoted to tenure Not promoted N URM 53% 47% 17 Black 58% 42% 12 Hispanic 40% 60% 5 White 63% 37% 230 Asian 60% 40% 42 Overall 62% 38% 289 Promotion to AWOT (Asst. Professors hired from 1991-2004)
	Promoted to AWOT	Left without promotion	N
URM*	55%	45%	38
Black	61%	39%	23
Hispanic	50%	50%	14
White	75%	23%	436
Asian	79%	19%	80
Overall	74%	24%	554

* Includes 1 Native American

Mentoring across the Institute lacks consistency, including level of commitment and a defined role for mentors (I). The interview data indicated there was a broad range of mentoring experiences reported by URM faculty. Among the most positive experiences were those in which mentors were accountable at the departmental or higher levels for taking an active role in mentoring the junior faculty member. Formal programs with such accountability and personal investment from the faculty were most successful. In these cases, mentors were reported to take on an advocacy role rather than a departmental evaluatory role, indicating a difference between the perceived roles of a formal mentor versus a tenure committee member; other studies have found that this kind of role is extremely beneficial to the mentee.¹⁰ More negative experiences included those in which mentors were non-existent, or were not engaged or active, or in which the junior faculty received ill-conceived or overly-directive advice. Interviews with non-minority and minority faculty indicated that poor or negative mentoring experiences are more frequent for URM than non-URM faculty, and they are particularly high among URM women. It should be noted that it can be beneficial for URM junior faculty to have access to at least one mentor from a non-URM group; and in particular, cross-racial or cross-gender mentoring experiences tend to be positive and helpful experiences.^11,12 Finally, some junior faculty expressed a lack of knowledge of how mentors might best be utilized to support their careers.

Conclusion: A consistent mentoring approach across the Institute with accountability, a defined role of a mentor, as well as periodic and timely assessments of progress can contribute to success of junior faculty in the years preceding promotion.

The potential for subjectivity in tenure/promotion decisions, as well as communication about expectations, is more of a concern for URM faculty (I, S). Interview data indicate there is a greater concern among URM faculty about having an objective review process compared to the non-URM sample. This data is complemented by survey data that indicate URM faculty feel requirements for tenure are less clearly communicated with them than their non-URM counterparts.

Conclusion: Concerns exist among some URM faculty regarding a less-than-objective tenure review in general, or a tenure review that is influenced by aspects of race, ethnicity or gender. Aspects of the tenure process that are less defined or less clearly communicated can create increased concerns around subjectivity with regard to these matters.

Many URM faculty, particularly, though not exclusively, in SHASS, SAP and Sloan, work in research areas that are different from the majority of their peers (C). In these cases, there was often concern expressed about the appropriate choice of referees for promotion. There was also concern regarding the level of respect or understanding afforded these different aspects of the chosen research problem by their departmental peers.

Conclusion: Attention and additional effort is required and should be applied toward the support and development of faculty who work in new, frontier areas of the field not well-represented in a departmental unit, or in areas less widely recognized but with potential socio-cultural, national or global impact.

Data from the survey indicate that there is more dissatisfaction among tenured URM faculty compared to their White counterparts (S, MFF), with Asian faculty in the middle. There also is more dissatisfaction among Asian and URM tenured faculty compared to their untenured counterparts. These trends are not statistically significant, but are supported by the interviews and by the discussions heard in the faculty forums. Although it is clear that faculty generally agree they are satisfied with being a faculty member at MIT, when we compare the extremes reported by tenured White to tenured URM faculty (and see whether they are very dissatisfied or very satisfied with being a faculty member at MIT), the difference approaches the p=.10 level of significance. Tenured URM faculty (and, to some extent, Asian faculty as well) are less likely to be highly satisfied with their MIT lives and are more likely to be dissatisfied.

Some possible reasons for this difference culled from qualitative URM faculty interviews and the minority faculty forums with senior URM faculty include accumulation of micro-inequities and stressors such as:

• Lack of peer recognition and acknowledgement;
• Ceilings or barriers at high levels (lab director, senior appointments, chairs);
• Perception of MIT as an equitable place;
• Fatigue, anger or frustration from past efforts to improve diversity;
• Accumulation of micro-inequities.

Ironically, these data is accompanied by the fact that it is the URM non-tenured faculty, particularly the Black faculty, who are most likely to be very satisfied with their lives at MIT (67% Black vs. 47% White) (S). Untenured URM and Asian faculty are more satisfied than White untenured faculty, and more satisfied than their tenured counterparts. This may indicate that recent efforts to provide a supportive environment for junior faculty have met some level of success, at least in terms of overall satisfaction with life at MIT. It is difficult to separate cohort factors - such as changes in administrative practice or departmental climates at MIT - from differences in attitude that may occur over the course of a faculty career, as URM faculty begin to face some of the challenges described by the senior URM faculty. (The latter was also the case for women.)

Conclusion: There is an inverse vector with regard to overall satisfaction in moving from junior to senior faculty rank for URM versus non-URM faculty that is disconcerting, and if addressed could improve the long-time retention of tenured URM faculty at MIT.

Climate

One of the overall issues that impacts the careers and the quality of life of URM faculty is the climate around race and inclusion present within the schools and departments, within classrooms, labs and other localized work environments at MIT. The MIT culture is unique and promotes the scientific standard of objectivity, but it also tends to place less emphasis on humanistic aspects of the academic enterprise. Within this culture, which seeks to view the individual with respect to his/her contributions to a field and levels of productivity, it can often be difficult to address the larger social culture in which MIT is embedded, which includes inherent value placements on aspects such as cultural differences, race and
diversity. Below are findings that address the climate at MIT and its potential impact on URM faculty.

MIT non-URM faculty view diversity as less critical to the Institute's core value of excellence (S). Based on responses from the quality of life survey to the question "I feel a diversified faculty is important for MIT's academic excellence," URM faculty and women both indicate diversity to be a more critical component of MIT's core value of excellence than non-URM males. This difference indicates a deeper dissimilarity in the appreciation of why participation at the highest levels of all groups is needed for future technological and research developments. The idea that MIT's long-term success depends on recruitment of the top talent throughout the U.S. as well as the world is a message that has not yet reached a large part of the faculty. Furthermore, it is clear that the value placed on gaining a diverse faculty is not high.

Conclusion: This low level of valuation speaks to the climate to which minority faculty are recruited. It also indicates the expectation that these faculty exhibit MIT's high ideals of excellence in spite of race or gender differences, rather than demonstrating excellence as a part of the continuum of diverse backgrounds and perspectives gained from a broad spectrum of diverse faculty.

Discussion of race-related issues is avoided at MIT, to the detriment of many URM faculty who may face but cannot confront such issues directly (I). Based on URM and non-URM faculty interviews, there is great awkwardness in openly addressing race and racial differences at MIT, leading to a sense of silence regarding race. URM faculty indicated this difficulty can lead to issues in communicating concerns from minority faculty regarding race, and can also impede the ability of faculty, in general, to move beyond unexpressed concerns or cultural misunderstandings. In some cases for example, URM faculty may feel that speaking on diversity as a topic in any way can potentially "brand" them as someone who focuses only on this concern at the expense of other issues.

Examples of situations in which this kind of "silence" can be inhibiting include the discussion around a minority faculty candidate or a promotion case in which comments from a referee, or a negative interaction with specific members of the field, might bring about a relevant concern impacted by race or gender. These issues are often extended to faculty discussions around other members of the MIT community, such as URM students and staff, in which presumption or misunderstanding would be more readily addressed without the unspoken expectation of silence on these issues.

Conclusion: The social and political issues that surround the ability to discuss race openly, when necessary, create barriers that can impact and inhibit discussion of key issues for faculty. This ultimately impacts the climate at MIT.

Meritocracy is a concept that is key to the ideals at MIT. Although it is important to strive for this ideal, there is tension created by the outward presumption that true meritocracy is already essentially achieved at MIT (I). Such presumptions preempt the potential for hidden bias or preferential behavior and do not acknowledge the use of relatively monolithic criteria of excellence (which often works against those who are minorities by race, gender or field). As a community focused on scientific and technological advances, MIT holds a great deal of pride in the concept of a merit-based society in which those who excel are rewarded proportionately. On the other hand, the presence of bias remains a possibility even among those who are most well-intentioned. For that reason, it is not possible to guarantee that racial, gender and other cultural biases do not impact the way in which faculty are evaluated. In short, it is not possible to proclaim a fully meritocratic process when our society presents innate biases to which all can be susceptible on some level. An excellent example involves the differences found in patient evaluations of physicians that resulted in less favorable ratings of minority and women physicians in comparison to the objectively measured physician performance metrics devised to correlate with patient satisfaction and well-being.¹³ Other studies also confirm the presence of unconscious bias in many evaluative settings and contexts.^14-16

Furthermore, although the ideal of a meritocracy is, in general, one that can be appreciated by many, there are flaws in the belief that merit is equitably assigned to different kinds of contributions. In particular, the tendency to use two or three highly defined metrics as a means of evaluating quality can lead to a more myopic view of excellence. It may also lead to an inability to quantify, value and recognize other types of achievements that also enrich and contribute to the academic excellence of the Institute. On the other hand, the ability to recognize and reward a broader range of merit can lead to creative and significant advances in new areas. A quote from a young URM faculty member describes this concept: "To insist on orthodoxy [i.e. narrow, singular definition of excellence] would stifle one of the pillars of MIT which is to encourage innovation and entrepreneurship of ideas."

Conclusion: Although the meritocracy concept presents an appropriate ideal, tension is created by the presumption that true meritocracy is already achieved at MIT. This view does not acknowledge the potential for hidden bias nor address the fairly narrow means of assessing excellence.

There is tension at MIT around the concepts of inclusion vs. excellence (I, MFF). The promotion of excellence at the highest levels (national and worldwide recognition, significant and high-impact advances) is a key feature of MIT's strength. One of the greatest tensions associated with achieving a diverse faculty is the idea that by being more inclusive, one sacrifices excellence or dilutes quality. This concept and the tension generated by it was an underlying theme in both URM and non-URM interviews. The anticipation from some members of the community that the intentional inclusion or recruitment of a minority faculty member might, in some cases, represent a lowering of standards is one that can yield negative experiences for URM faculty even before their career has begun. On the other hand, this same tension is sometimes used as a reason for the lack of progress in increasing URM faculty numbers (i.e. "no good URM candidates can be found"). The presence of faculty with diverse racial and cultural viewpoints can bring a great deal of perspective to work, as well as a broader and more extensive choice of problems, frequently with high impact. Examples include engineering, scientific or architectural solutions that impact those from lower income groups in the rural U.S., in the country's urban centers, or in developing nations that have not been touched by technological advances.

Conclusion: In general, the belief that inclusion must equal dilution of excellence is one that has not been effectively discussed and countered within MIT's culture, although inclusion of the top scientists and engineers across a broad range of experiences can lead to innovation. It can also lead to the foundation of new research areas that have high impact in many parts of the country and world.

Previous:
C. Engagement with MIT Faculty,
Administrative Leadership
and Advisory Board

Next:
E. Recommendations