To the Harvard Faculty of Arts and Sciences Research Integrity Officer

(RIO):
I am submitting these materials to recommend that you initiate two related
inquiries.
1. Inquiry concerning research misconduct of Claudine Gay
See the materials below related to the scholarship of Claudine Gay, involving work
listed on her C.V.
Know that it is impossible that your office has already reviewedthe entirety of
these materials as many examples below have not been previously reported or
submitted to Harvard. Harvardreceived allegations from the New York Post in late
October, but these materials involve more articles and passages of text from her
academic work.
As you examine whether the materials below show a pattern of plagiarism, keep in
mind that Harvard has a detailed guide that explains your standards for plagiarism.
That guide reads, “When you fail to cite your sources, or when you cite them
inadequately, you are plagiarizing, which is taken extremely seriously at Harvard.”
See this PDF:
https://usingsources.fas.harvard.edu/sites/projects.iq.harvard.edu/files/sources/files
/avoiding_plagiarism.pdf
The Harvard Corporation already admitted in writing that “a few instances of
inadequate citation” were revealed by its inquiry. Again, Harvard’s guide reads:
“When you fail to cite your sources, or when you cite them inadequately, you are
plagiarizing.” The Harvard Corporation can resolve to amendHarvard’s current
definition of plagiarism, but it has no standing to speak for Harvard’s FAS
Research Integrity Officer (RIO) or the Chair of the Committee on Professional
Conduct (CPC),who are responsible for inquiries into allegations of research
misconduct.
The Harvard FAS “Interim Policy and Procedures for Responding to Allegations of
Research Misconduct” defines research misconduct this way: “Research
misconduct: fabrication, falsification, or plagiarism in proposing, performing, or
reviewing research, or in reporting research results. Research misconduct includes
fabrication, falsification, and plagiarism (as defined in this Policy). Research
misconduct does not include honest error or differences of opinion.”
See this PDF:
https://research.fas.harvard.edu/files/research/files/interim_policy_amended_09_20
21.pdf?m=1635522426
The passages shown below were clearly written in the course of“proposing,
performing, or reviewing research, or in reporting research results.”
Your policy applies to “allegations of research misconduct (as defined by this
Policy) involving any person who, at the time of the alleged research misconduct,
was employed by, was an agent of, or was affiliated by contract or agreement with
FAS, including without limitation officials, tenured and non-tenured faculty,
teaching and support staff, researchers, research coordinators, technicians,
post-doctoral and other fellows, students, volunteers, and agents.”
Dr. Gay has held a faculty position within the Faculty of Arts and Sciences since
2006. She was a graduate student in Arts and Sciences from 1992-1998. Her
research conduct falls within your jurisdiction. The publications produced while
Dr. Gay was a professor at Stanford University were submitted by Dr. Gay to
Harvard as part of her review for appointment as professor at Harvard.
2. Inquiry concerning mishandling of earlier allegations
In late October, Claudine Gay and the Fellows of Harvard Corporation received
credible evidence of research misconduct by Dr. Gay. The Harvard FAS policy on
allegations of research misconduct applies to Dr. Gay’s work implicated in the
allegations. Her faculty appointment is and has been in FAS. The policy has no
exclusions for administrators. But based on the Corporation’s letter of December
12, Harvard does not appear to have followed its policy.
The Corporation letter states that “the Fellows [of the Corporation] promptly
initiated an independent review by distinguished political scientists.” But your
policy does not provide for such an ad hoc review at the direction of the Fellows.
At Harvard, the rules adopted by the faculty govern. In this case, the rules are
again the Harvard FAS “Interim Policy and Procedures for Responding to
Allegations of Research Misconduct” cited above.
This policy imposes an obligation to report even “apparent research misconduct” to
the Harvard FAS Research Integrity Officer (RIO) or the Chair of the Committee
on Professional Conduct (CPC). The policy states: “All individuals subject to this
Policy will report observed, suspected, or apparent research misconduct to the RIO
or to the Chair of the CPC.”
When Harvard was contacted with allegations of research misconduct in October,
did the RIO and CPC Chair receive notification of “apparent research misconduct”
as required under this policy? Did any individuals subject to this policy fail to
report observed, suspected, or apparent research misconduct to the RIO and CPC
Chair? Note that Dr. Gay is subject to this policy, as are any other FAS personnel
who reviewed the earlier allegations.
The policy states further, “Upon receiving an allegation of research misconduct,
the RIO and CPC Chair immediately willassess the allegation to determine whether
the allegation … falls within the definition of research misconduct.” Their specific
obligation is to determine whether “potential evidence of research misconduct
may be identified” in the allegationsreceived.
If the CPC Chair and RIO find “potential evidence” of research misconduct, the
policy requires that they open an inquiry. The policy states, “The Committee on
Professional Conductordinarily shall serve as the inquiry committee.” However,
the Harvard Corporation’s letter of December 12 makes no mention of the
Committee on Professional Conduct. Was this faculty body bypassed, and if so,
why? Were they consulted on the decision to bypass them? If not, why not?
The Harvard community and the public have also learned that Harvard hired a top
defamation attorney who confronted a complainant (the New York Post) and
presented testimony purporting to refute the allegations against Dr. Gay.
Your policy states: “Harvard community members may not retaliate in any way
against complainants, witnesses, the RIO, or committee members. Any alleged or
apparent retaliation against complainants, witnesses, the RIO, or committee
members should be reported immediately to the RIO or the CPC Chair, as
applicable, who shall review the matter and, as necessary, make all reasonable and
practical efforts to counter any potential or actual retaliation and protect and restore
the position and reputation of the person against whom the retaliation is directed.”
Any reasonable person would see confrontation by a university-paid defamation
lawyer as “apparent retaliation” against the complainant. Any Harvard personnel
involved in this decision should be investigated for potential violation of the policy
against retaliation in these proceedings.
According to the New York Post, Harvard was sent allegations of Dr. Gay’s
research misconduct on October 24. On October 27, the New York Post received
“a 15-page letter by Thomas Clare, a high-powered Virginia-based attorney with
the firm Clare-Locke who identified himself as defamation counsel for Harvard
University and Gay,” which “contained comments from academics whose work
Gay was alleged to have improperly cited.” But the Harvard Corporation admitted
in its public statement that they only received the “results” of their “review” of Dr.
Gay’s published work on December 9:
https://www.harvard.edu/blog/2023/12/12/statement-from-the-harvard-corporation-
our-president/.
Why did Harvard authorize its outside counsel to send communications defending
Dr. Gay’s work and implying a threat of legal action against a complainant before it
knew the results of its “review” and before any determination by the RIO and the
CPC Chair?
The Harvard FAS policy on allegations states, “The Office of General Counsel
shall be available to advise the investigation committee and the RIO.” There is no
mention in your policy of university-provided counsel for the accused. Therefore,
I askwhether Dr. Gay’s case was dealt with equitably. This looks like a potentially
serious abuse of power. Has Harvard ever before provided counsel to a faculty
member credibly accused of plagiarism or research misconduct?
Pleased keep me informed of the progress of your inquiries.

RAW DATA FOR CONSIDERATION BY THE RESEARCH INTEGRITY

OFFICE AND CHAIR OF THE COMMITTEE ON PROFESSIONAL
CONDUCT
1. Gay, Claudine. “Between Black and White: The Complexity of Brazilian
Race Relations.” Origins 1(4), 1993, p. 27:
On June 18, 1978, representatives from a number of Afro-Brazilian
organizations, outraged by a series of racially-motivated incidents, joined
forces in Sao Paulo to form the Unified Movement Against Racial
Discrimination (MUCDR).
As their first project, MUCDR organized a July 7, 1978 demonstration to
protest two acts: (a) the April 28 beating death of black worker Robson
Silveira da Luz, by a Sao Paulo policeman; and (b) the May expulsion of four
young black athletes from the volleyball team of the Tiete Yacht Club because
of their color. Two thousand people participated in the protest on the steps of
the Municipal Theater. The organizers read an open letter to the population in
which they outlined their campaign against "racial discrimination, police
oppression, unemployment, underemployment and marginalization."
Covin, David. “Afrocentricity in O Movimento Negro Unificado.” Journal of
Black Studies 21:2, 1990, p. 130:
 The earliest manifestation of the MNU was the Unified Movement Against
Racial Discrimination (MUCDR).Representatives from a number of Black
organizations,entidades, met at the Center of Black Art and Culture in São
Paulo on June 18, 1978. They resolved to create a movement to defend the
Afro-Brazilian community against racial exploitation and human disrespect.
They designated as their first activity a demonstration to be held at 6:30 p.m.
on July 7, 1978, at the Chá viaduct in São Paulo. The demonstration was to
protest two acts: (a) the torture and assassination of a Black worker, Robson
Silveira da Luz, by policemen in São Paulo on April 28, 1978; and (b) the
dismissal of four Black male children from the volleyball team of the Tiete
Yacht Club in May, 1978, because of their color (Gonzalez, 1982, p. 43).
The demonstration was held on the designated date on the steps of the
Municipal Theater of Sao Paulo. Two thousand people were present
(Gonzalez, 1982, p. 48). An open letter to the population was read. Its
principal features were (Gonzalez, 1982,pp. 48-50):
1. A campaign against racial discrimination, police oppression,
unemployment, underemployment, and marginalization.
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” David Covin’s work is not
mentioned there.]
2. Gay, Claudine. “Between Black and White,” p. 28:
The fundamental organizational unit, as approved by the founding National
Assembly in 1978, was to be the Center of Struggle (Centro de Luta). Centers
were to be formed in work areas, villages, prisons, candomble and umbanda
temples, samba schools, churches, and favelas.
Covin, David. “Afrocentricity in O Movimento Negro Unificado,” Journal of
Black Studies 21:2, 1990, p. 131:
In the basic organizational structure which they approved, the Centers of
Struggle were the fundamental organizational units. These were to be formed
 in work areas, villages, prisons, candomblé and umbanda temples, samba
schools, afoxés, churches, and favelas.
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” David Covin’s work is not
mentioned there.]
3. Gay, Claudine. “Between Black and White,” p. 28:
It was a younger generation of Afro-Brazilians, many with one or more years
of university education, that were among the first and most eager respondents
to the MNU 's organizational call. Their eagerness was in large part a measure
of the economic and political exclusion they had suffered under the military
dictatorship.
As their numbers grew, it was their aspirations and rhetoric which came to
define the organization. The movement became an expression of frustration
among upwardly mobile Afro-Brazilians denied admission to the middle-class
status to which their education and qualifications entitled them. To that extent,
the aspirations of these activists were too removed from the lives of the poor
to facilitate cross-class linkages.
Andrews, George Reid. “Black Political Protest in Sao Paulo, 1888-1988.”
Journal of Latin American Studies24(1), 1992, p. 164-65:
a younger generation of Afro-Brazilians, many with one or more years of
university study, were starting to organise a new black movement in response
to the economic and political exclusion which they were experiencing under
the dictatorship. This movement, most vividly symbolised by the Movimento
Negro Unificado, created in Sao Paulo in 1978 …
This new movement of the 1970s and 1980s was to a large degree the
expression of frustration among upwardly mobile Afro-Brazilians denied
admission to the middle-class status to which their education and
qualifications entitled them. Its activists worked hard at recruiting supp in the
slums and favelas of the urban periferia, but their rhetoric and aspirations
 often seemed somewhat removed from the lives of poor and working-class
blacks facing the immediate, grinding problems of poverty, crime and hunger.
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” In that section, there are 5
works mentioned, the first of which is the following book: Andrews, George Reid.
Blacks & Whites in São Paulo, Brazil, 1888-1988. University of Wisconsin Press,
1991.]

4. Gay, Claudine. “A Room For One’s Own? The Partisan Allocation of

Affordable Housing,” Urban Affairs Review53(1), 2017, p. 51:

Theory predicts an interaction between county partisanship and party control,

such that the more Democratic a county, the more LIHTC allocations it should
receive when the state is under Democratic control; if the relationship is
symmetrical, as Hypothesis 1 predicts, a more Republican county should
benefit when the state is under Republican control.

Ansolabehere, Stephen and James M. Snyder, Jr. “Party Control of State

Government and the Distribution of Public Expenditures.” Scand. J. of
Economics 108(4), 2006, p. 551:

Theoretical arguments predict an interaction between partisanship of voters

and party control of state government. Democratic counties are expected to
receive more transfers when the state is under Democratic control than when
the state is under Republican control; and Republican counties should receive
more transfers when the state is under Republican control.

[Gay cites Ansolabehere and Snyder 2006 twice in parentheses on p. 46. She does
not cite the paper anywhere thereafter, including on p. 51.]

5. Gay, Claudine. “A Room For One’s Own?” p. 52:

 The key tests of the partisan allocation argument are whether the coefficient
on Democratic vote share is positive and significant for Democratic
governors; for Republican governors, the coefficient should be either negative
and significant (Hypothesis 1), or indistinguishable from 0 (Hypothesis 2).

Ansolabehere, Stephen and James M. Snyder, Jr. “Party Control of State

Government and the Distribution of Public Expenditures,” Scand. J. of
Economics 108(4), 2006, p. 558:

The key tests of the majoritarian argument are whether the slope on
Democratic Vote times Democratic Control is positive and the slope on
Democratic Vote times Republican Control is negative. The coefficient on
Democratic Vote times Divided Control should lie between these two and
possibly be indistinguishable from 0.

[Gay cites Ansolabehere and Snyder 2006 twice in parentheses on p. 46. She does
not cite the paper anywhere thereafter, including on p. 52.]

6. Gay, Claudine. “A Room For One’s Own?” p. 51:

To measure the direction in which the governing party skews funds, I include
the interaction between which party holds the governor’s office (Dst) and
county partisanship (DemVoteist); this specification allows for different
slopes on Democratic vote share for Republican and Democratic governors.
The estimated coefficient ρ1 represents the average within-county relationship
between Democratic vote share (in the most recent election) and LIHTC
allocations under a Republican governor.
 Ansolabehere, Stephen and James M. Snyder, Jr. “Party Control of State
Government and the Distribution of Public Expenditures,” Scand. J. of
Economics 108(4), 2006, p. 556:

First, to measure the direction in which the governing party skews funds we
include the interaction between which party controls the state government and
the partisanship of the county. Specifically, average Democratic Vote is
interacted with Democratic Control, Republican Control and Divided Control.
(The sum of these three variables is the average Democratic vote share.) This
specification allows for different slopes on Average Democratic Vote for the
three cases of party control. In Table 2 these variables are labeled: Democratic
Vote times Democratic Control, Democratic Vote times Republican Control
and Democratic Vote times Divided Control.

[Gay cites Ansolabehere and Snyder 2006 twice in parentheses on p. 46. She does
not cite the paper anywhere thereafter, including on p. 51.]

7. Claudine, Gay. “Moving to Opportunity,” Urban Affairs Review 48(2),

2012, p. 159-60:

Under this assumption, one knows that the treatment impact for the
noncompliers (i.e., individuals who did not use the voucher) is zero, making
the ITT estimate πITT a weighted average of the impact on compliers and the
zero effect on noncompliers (Bloom 1984); the weights are the portions of the
sample that are compliers and noncompliers. As a result, the TOT impact can
be estimated by dividing πITT by the program compliance rate for the
treatment group (for similar approaches to estimating MTO impacts, see
Katz, Kling, and Liebman 2001; Ludwig et al. 2008; Orr et al. 2003).
 Ludwig, Jens, Jeffrey B. Liebman, Jeffrey R. Kling, Greg J. Duncan,
Lawrence F. Katz, Ronald C. Kessler and Lisa Sanbonmatsu. “What Can We
Learn about Neighborhood Effects from the Moving to Opportunity
Experiment?” American Journal of Sociology 114 (1), 2008, p. 153:

Under these assumptions, we know that the average outcomes of the

noncompliers in the treatment group and of the potential noncompliers in the
control group are the same. Put differently, we know that the experimental
impact for the noncompliers was zero. Thus, under the TOT assumptions, the
ITT estimate is simply a weighted average of the impact on compliers and the
zero effect on noncompliers—the weights are the portion of the sample that
are compliers and the portion that are noncompliers (Bloom 1984). This result
implies that the TOT impact can be calculated by simply rescaling the ITT
estimate by the program compliance rate.

[On p. 160, Gay cites Ludwig et al 2008, among others, writing: “(for similar
approaches to estimating MTO impacts, see Katz, Kling, and Liebman 2001;
Ludwig et al. 2008; Orr et al. 2003).” Gay uses no quotation marks around
verbatim language and does not cite specific page numbers.]

8. Gay, Claudine. “A Room For One’s Own?” p. 54:

I also construct a county-level measure that captures the financial incentives

developers have to build or rehabilitate affordable housing in the most
impoverished places (Hollar and Usowski 2007).

Freedman, Matthew, Emily G. Owens. “Low-income housing development

and crime.” Journal of Urban Economics 70(2-3), 2011, p. 119:
 As an instrument for low-income housing development, we construct a
county-level measure that captures the incentives developers have to build or
rehabilitate affordable housing in certain tracts.

[Gay never cites Freedman and Owens 2011. She thanks the two of them for letting
her use their data (p. 66n.22).]

9. Gay, Claudine. “A Room For One’s Own?” p. 55:

As QCT coverage increases, so does the ability of developers to take

advantage of the larger tax credit by siting new housing in the county.

Freedman, Matthew, Emily G. Owens. “Low-income housing development

and crime.” Journal of Urban Economics 70(2-3), 2011, p. 119:

As QCT coverage increases, however, so does the ability of developers to take

advantage of the larger tax credit.

[Gay never cites Freedman and Owens 2011. She thanks the two of them for letting
her use their data (p. 66n.22).]

10. Gay, Claudine. “A Room For One’s Own?” p. 43:

The LIHTC is widely considered one of the nation’s most successful housing
programs, and accounts for an estimated one-sixth of all multifamily
housing—subsidized or unsubsidized—built in the United Statessince
program inception (Schwartz 2010).

Williamson, A. R. “Can They Afford the Rent? Resident Cost Burden in Low
Income Housing Tax Credit Developments.” Urban Affairs Review 47(6),
2011, p. 776:
 The LIHTC is responsible for an estimated one-sixth of all multifamily rental
housing—both subsidized and unsubsidized—produced in the United States
each year (Schwartz 2010).

[Gay never cites Williamson 2011.]

11. Gay, Claudine. “A Room For One’s Own?” p. 64n2:

Demand for credits declined sharply during the financial crisis, as the most
active investors in the tax credits (large financial institutions bound by the
requirements of the Community Reinvestment Act) pulled out of the market.

Williamson, A. R. “Can They Afford the Rent? Resident Cost Burden in Low
Income Housing Tax Credit Developments,” Urban Affairs Review 47(6),
2011, p. 779-80:

The strength of this interest has declined since the global financial crisis,
however, chiefly because the most active investors in the tax credits were
large financial institutions that purchased the credits not only for investment
purposes but also to fulfill Community Reinvestment Act requirements.

[Gay never cites Williamson 2011.]

12. Gay, Claudine. “A Room For One’s Own?” p. 43:

What began as a modest item in the Internal Revenue Code has evolved over
time into the nation’s single largestsubsidy for affordable housing, replacing
nearly all previous tax incentives for investing in rental housing of any kind.
 Schwartz, Alex F. Housing Policy in the United States. New York: Routledge
(2010), p. 98:

The Low-Income Housing Tax Credit has evolved from an esoteric financial
instrument to the single most importantsource of equity for low-income rental
housing in the United States. Created by the Tax Reform Act of 1986, the tax
credit replaced virtually all previous tax incentives for investing in rental
housing of any kind.

[Gay cites Schwartz twice in this paragraph. She cites Schwartz once in the first
sentence, and again in the last sentence. The second reference to Schwartz is part
of a sentence copied verbatim from Williamson. See #10 above. Gay does not cite
specific page numbers and uses no quotation marks around verbatim language.]

13. Gay, Claudine. “A Room For One’s Own?” p. 43:

What began as a modest item in the Internal Revenue Code has evolved over
time into the nation’s single largest subsidy for affordable housing, replacing
nearly all previous tax incentives for investing in rental housing of any kind.

Schwartz, Alex F. Housing Policy in the United States. New York: Routledge
(2010), p. 83:

The single largest subsidy for low-income rental housingis not a federal
housing program but an item in the Internal Revenue Code.

[See #12 above.]

14. Gay, Claudine. “Between Black and White,” p. 27:
Again, part of this failure can be attributed to the political
environment—particularly, the party reform of 1979, which allowed the
 opposition to return to multi-party competition. Operating in much the same
way as in the Second Republic, party competition and the promise of free
elections encouraged competition for black electoral support. Parties adopted
anti-discrimination platforms, and organized commissions to investigate race
relations. Affiliation with one of these political parties served a critical
function which the MNU could not: it created a sense of attachment to the
mainstream political apparatus, thus defusing the perceived need for racially
defined political activity.
Andrews, George Reid. “Black Political Protest in Sao Paulo, 1888-1988.”
Journal of Latin American Studies24(1), 1992, p. 165-66:
A major part of the explanation for this success may be found in the 'party
reform' of 1979. Prior to that time, the dictatorship had permitted the
existence of only a single opposition party, the Movimento Democrático
Brasileiro. Eventually recognising that such a policy provided the perfect
mechanism for its opponents to join forces, in 1979 the government freed the
opposition to return to multi-party competition. The Movimento Negro
Unificado had been founded just the year before and, inspired by its example,
local-level black organisations were coming into existence throughout Brazil.
As the newly created opposition parties competed among themselves for
electoral support, they directed particular attention to this burgeoning black
movement, inserting anti-racism planks into their platforms, and creating
special working groups and commissions to investigate racial problems in
Brazil.
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” In that section, there are 5
works mentioned, the first of which is the following book: Andrews, George Reid.
Blacks & Whites in São Paulo, Brazil, 1888-1988. University of Wisconsin Press,
1991.]
15. Gay, Claudine. “Between Black and White,” p. 25:
 Later, "racial democracy" was grafted into official state policy and rooted in
the official view that Brazil is a country in which race makes no difference to
opportunity or status. Furthermore, the Brazilian government claims that the
country is free of both institutional and informal forms of discrimination.
Access to public resources—be it education, health care, employment,
political participation—is equally available to all citizens. Therefore, the
potential for upward mobility is perceived as simply a matter of fair
competition and individual merit.
By insisting that Afro-Brazilians enjoy opportunities to better themselves and
freedom to compete
Andrews, George Reid. Blacks & Whites in São Paulo, Brazil, 1888-1988.
University of Wisconsin Press, 1991, p. 129:
The concept of racial democracy asserts that Brazil is a land entirely free of
legal and institutional impediments to racial equality, and largely … free of
informal racial prejudice and discrimination as well. The nation offers all its
citizens, black, brown, or white, virtually complete equality of opportunity in
all areas of public life: education, politics, jobs, housing. Thus
Afro-Brazilians enjoy opportunities to better themselves, and the freedom to
compete
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” In that section, there are 5
works mentioned, the first of which is the following book: Andrews, George Reid.
Blacks & Whites in São Paulo, Brazil, 1888-1988. University of Wisconsin Press,
1991.]
16. Gay, Claudine. “Between Black and White,” p. 27:
At the First National Congress of MUCDR, held in Rio in December 1979,
the organization changed its name to the Unified Black Movement (MNU)
Covin, David. “Afrocentricity in O Movimento Negro Unificado.” Journal of
Black Studies 21:2, 1990, p. 132:
 At the First National Congress of the MNUCDR, held December 14-16, 1979
in Rio, the name was shortened to the Unified Black Movement (MNU)
(Gonzalez, 1986, p. 128), after considerable discussion.
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” David Covin’s work is not
mentioned there.]
17. Gay, Claudine. “Between Black and White,” p. 24:
The Brazilian concept of "whitening," symbolized in the popular saying "we
are becoming one people,"represents an ideology entirely different from white
European and North American phobias about race mixture prevalent at the
turn of this century. In the Brazilian view, the mixing of whites and
non-whites (be they of Indian or African descent) led to the "breeding out" of
non-white racial characteristics, and the propagation of white racial heritage
and attributes. Over many generations and on a wide scale, inter breeding
would lead to the eventual "whitening" of Brazil.
Miscegenation, or racial interbreeding, has had a long history in Brazil, dating
back to the colonial period …
Skidmore, Thomas E. “Toward a Comparative Analysis of Race Relations
Since Abolition in Brazil and the United States.” Journal of Latin American
Studies 4(1), 1972, p. 9-10:
The Brazilian adage that 'we are becoming one people'rests on an implicit
assumption that this final amalgam will be, at worst, a light mulatto
phenotype and at best a moorish Mediterranean physical type. The ideal of
whitening differs so categorically from white European and North American
phobias about race mixture, that the Brazilian willingness to accept mixed
bloods according to a varying scale of racial categories has often led to the
false conclusion that there is no hierarchy of color values. In other words, the
Brazilian familiarity with, and qualified acceptance of, miscegenation has
misled observers into concluding that the outcome of race mixing was a
matter of indifference.
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” In that section, there are 5
works mentioned. Skidmore’s piece is one of them.

18. Gay, Claudine. “A Room For One’s Own?” p. 43:

The LIHTC incentivizes private development of affordable housing by

allowing investors to reduce their federal income taxes by US$1 for every
dollar of tax credit received, with the amount of the tax credit dependent on
the cost, location, and projected occupancy (by low-income households) of
the housing development. When the program was introduced, it represented a
departure from an historical approach that had relied heavily on direct public
funding and management of affordable housing inventory.

Unlike other tax breaks associated with real estate, the housing tax credits are
not awarded automatically.Rather, authority to issue tax credits is allotted to
states on an annual basis; the total dollar amount of credits available is
determined by state population. Developers must apply to designated state
agencies (ordinarily, the state’s housing finance agency) for credits

Schwartz, Alex F. Housing Policy in the United States, p. 83-84:

The LIHTC allows investors to reduce their federal income taxes by $1 for
every dollar of tax credit received. Investors receive the credit for 10 years;
the property must remain occupied by low-income households for at least 15
years. The amount of the credit depends on the cost and location of the
housing development and the proportion of units occupied by low-income
households. Unlike other tax breaks associated with real estate, the LIHTC is
not awarded automatically. Tax credits are assigned to individual housing
developments by designated state agencies (usually state housing finance
agencies, or HFAs). The total dollar amount of credits available is determined
 by state population. As of 2004, states may allocate $1.80 per capita per year
in tax credits, with the amount adjusted for inflation thereafter.Developers
apply to HFAs for tax credits.

[See comments on #12, 13, and #10 above, all of which are drawn from this same
paragraph. Gay does not use quotation marks around language copied verbatim and
does not cite page number.]
19. Gay, Claudine. “Between Black and White,” p. 25:
The concept of racial democracy was already taking shape in Brazil in the
early part of the twentieth century, to be fully developed in the 1930s by
Brazilian Gilberto Freyre. After witnessing the overt racism of the United
States, Freyre extolled the virtues of the Brazilian racial order, linking its
tolerant and democratic nature to the nation's more humane experience of
slavery.
Andrews, George Reid. Blacks & Whites in São Paulo, Brazil, 1888-1988.
Brazil, University of Wisconsin Press, 1991, p. 129:
The concept of racial democracy received its fullest and most coherent
exposition in the writings of Gilberto Freyre, beginning in the 1930s. It was
visibly taking form during the early decades of the 1900s, however, and its
roots run well back into the nineteenth century.
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” In that section, there are 5
works mentioned, the first of which is the following book: Andrews, George Reid.
Blacks & Whites in São Paulo, Brazil, 1888-1988. University of Wisconsin Press,
1991.]

20. Gay, Claudine. “Moving to Opportunity,” p. 154-56:

While the MTO demonstration was not designed to address issues of racial
and ethnic concentration directly, the racial characteristics of the new
 neighborhoods differed modestly from the original locations and between
voucher groups.

Orr, Larry, Judith Feins, Robin Jacob, Erik Beecroft, Lisa Sanbonmatsu,
Lawrence Katz, Jefferey Liebman, and Jeffrey Kling. Moving to Opportunity
for Fair Housing Demonstration Program: Interim impacts
evaluation.Washington, DC: U.S. Department of Housing and Urban
Development, 2003, p. 36:

MTO was not designed to address issues of racial or ethnic concentration

directly.

[Orr et al 2003 is not cited in this paragraph, or any at any later point in the paper.
It is cited early in the previous paragraph on p. 154, and it is cited on p. 155 in the
text accompanying Figure 1. No quotation marks are used around verbatim
language.]

21. Gay, Claudine. “Moving to Opportunity,” p. 159:

Thus the TOT effects, unlike the ITT effects, are nonexperimental, in the
sense that they are not directly observed for whole randomly assigned groups

Orr, Larry, Judith Feins, Robin Jacob, Erik Beecroft, Lisa Sanbonmatsu,
Lawrence Katz, Jefferey Liebman, and Jeffrey Kling. Moving to Opportunity
for Fair Housing Demonstration Program: Interim impacts
evaluation.Washington, DC: U.S. Department of Housing and Urban
Development, 2003, p. 9:

The TOT estimates are non-experimental, in the sense that they are not
directly observed for whole randomly assigned groups
[Orr et al 2003 is not cited in the paragraph where the passage quoted above
occurs. It is cited at the end of the next paragraph, in the same reference discussed
earlier: see comments on #7 above. No quotation marks are used around verbatim
language.]

22. Gay, Claudine. “Moving to Opportunity,” p. 159:

The TOT impact is inferred from the ITT impact πITT based on the weak
assumption that the effect of the treatment occurs entirely through moving
using a program voucher

Orr, Larry, Judith Feins, Robin Jacob, Erik Beecroft, Lisa Sanbonmatsu,
Lawrence Katz, Jefferey Liebman, and Jeffrey Kling. Moving to Opportunity
for Fair Housing Demonstration Program: Interim impacts
evaluation.Washington, DC: U.S. Department of Housing and Urban
Development, 2003, Appendix B, p. 7:

Under the weak assumption that the effect of the treatment occurs entirely
through moving using an MTO program voucher or certificate

[See again comments on #7 above. The parenthetical citation mentioned there

includes Orr et al 2003, among others. Gay’s text here appears in the first sentence
of the paragraph. The citation—“(for similar approaches to estimating MTO
impacts, see Katz, Kling, and Liebman 2001; Ludwig et al. 2008; Orr et al.
2003)”—occurs at the end of the last sentence. There are no quotations used
around the verbatim language.]

23. Gay, Claudine. “Moving to Opportunity,” p. 170:

Poor people who live in high-poverty neighborhoods may be different in

politically relevant ways from poor people who seek out and settle in
 low-poverty neighborhoods. Past research potentially has confused the effects
of neighborhoods with the effects of the (unobserved) characteristics of
individuals who live in different types of neighborhoods.

Orr, Larry, Judith Feins, Robin Jacob, Erik Beecroft, Lisa Sanbonmatsu,
Lawrence Katz, Jefferey Liebman, and Jeffrey Kling. Moving to Opportunity
for Fair Housing Demonstration Program: Interim impacts
evaluation.Washington, DC: U.S. Department of Housing and Urban
Development, 2003, p. 1:

Until recently such effects could only be studied by comparing the behavior
and life outcomes of low income residents of high-poverty areas with those of
poor families in low-poverty neighborhoods. Such comparisonspotentially
confused the effects of neighborhood with the effects of the characteristics of
families who lived in those two types of residential areas.

[Gay does not cite Orr et al 2003 in this paragraph or on this page. She cites Orr et
al 2003 on p. 167 and (in text accompanying a graphic) on p. 168. She does not
cite the piece again after that.]

24. Gay, Claudine. “Moving to Opportunity,” p. 173n13:

The standard errors for the TOT estimates are similarly adjusted. Thus, while
TOT impact estimates are substantially larger than ITT estimates (because
they are not attenuated by zero effects of the intervention on the
noncompliers), they are statistically significant only if the ITT estimate is
significant.

Orr, Larry, Judith Feins, Robin Jacob, Erik Beecroft, Lisa Sanbonmatsu,
Lawrence Katz, Jefferey Liebman, and Jeffrey Kling. Moving to Opportunity
for Fair Housing Demonstration Program: Interim impacts
 evaluation.Washington, DC: U.S. Department of Housing and Urban
Development, 2003, p. 9:

this makes the TOT estimates substantially larger than the ITT estimates.
However, since the standard errors for the TOT estimates are adjusted in the
same way, TOT impacts are statistically significant only if the corresponding
experimental ITT estimates are significant.

[Gay never cites Orr et al 2003 in the footnotes. The passage from Gay here is the
text of an endnote. The endnote does not contain any citation to Orr et al 2003. It
does not use quotation marks around verbatim language.]

25. Gay, Claudine. “Moving to Opportunity,” p. 159:

There are two estimates of interest that follow from MTO’s experimental
design and are reported in the tables: the intent-to-treat (ITT) and the
treatment-on-treated (TOT)effects. The ITT effect, estimated from the
difference in mean outcomes for the treatment and control groups as a whole,
is the effect of being offered the voucher, regardless of subsequent compliance
(i.e., whether or not the family offered the voucher actually used it to lease
up). The TOT effect is the effect of the voucher on the compliers—the MTO
sample adults who actually leased up using the program voucher. Whereas the
offer of a voucher was extended to every member of the experimental and
section 8 treatment groups, not every group member used their assigned
voucher. Thus the TOT effects, unlike the ITT effects, are nonexperimental, in
the sense that they are not directly observed for whole randomly assigned
groups, but only for the subset of compliers within the groups.

Ludwig, Jens, Jeffrey B. Liebman, Jeffrey R. Kling, Greg J. Duncan,

Lawrence F. Katz, Ronald C. Kessler and Lisa Sanbonmatsu. “What Can We
 Learn about Neighborhood Effects from the Moving to Opportunity
Experiment?” American Journal of Sociology 114 (1), 2008, p. 146:

We focus on two types of estimates that follow from MTO’s experimental

design—termed “intent to treat” and “treatment on the treated” in the
experimental literature. Roughly speaking, the MTO intent-to-treat (ITT)
effect on a given outcome is the simple difference between the outcome for all
individuals assigned at random to MTO’s experimental condition, regardless
of whether they “complied” by actually moving through MTO to a
low-poverty neighborhood, and the outcome for all individuals assigned to the
control group. In contrast, the treatment-on-the-treated (TOT) estimates are of
outcome differences for families actually moving in conjunction with the
program.

[See comments on #7 above. There, the parenthetical reference was included in the
relevant paragraph. Here, the parenthetical reference is in the next paragraph. There
are no quotations used around the verbatim language.]
26. Gay, Claudine. “Between Black and White,” p. 24:
The distinctions of black and white—or even white and non-white—that exist
in other societies such as the United States, are not used by Brazilians.
Instead, they refer to a multi-category system, based not on origin, but on
phenotype (i.e. appearance). Essentially, Brazilians are "color-conscious"
whereas as Americans and Western Europeans tend to be "race-conscious."
Brazilians evaluate an individual based on certain physical characteristics:
namely, amount of pigmentation, hair type, nose, and lips. From this
assessment, an individual is placed along a color continuum, anchored at the
two extremes by those with either no (visible) African traits or no (visible)
European attributes.
 Skidmore, Thomas E. “Toward a Comparative Analysis of Race Relations
Since Abolition in Brazil and the United States.” Journal of Latin American
Studies 4(1), 1972, p. 10-11:
Definitions of racial categories: The greatest single difference in race
relations between the United States and Brazil is the practical definition an
individual's race. The United States has developed a bi-racial system: one is
either 'white' or 'black' (the latter category formerly termed 'Negro or 'colored
'). The individual case is resolved not by the person's physical appearance, but
by his ancestry. State law in most states (including the North) had developed
a functional definition (e.g. one 'Negro' grandparents made the offspring
'Negro') either by statute or practice. The only escape from the ancestrally
defined 'Negro' caste was by 'passing', i.e. by being able to appear white in
one's physical characteristics and thus conceal one’s ancestry.
In Brazil, on the other hand, race has been primarily defined by physical
appearance, thereby creating a multi-racial system. In place of two rigidly
defined castes, there has been a sliding spectrum, with three principal
categories: white, mulatto, and black. In practice Brazilians have used a wide
variety of racial sub-categories, which shade into one another. Individual
judgements are based on an evaluation of the physical characteristics (hair,
skin color, lips, nose, general physical bearing), as well as the person's
apparent social status.
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” In that section, there are 5
works mentioned. Skidmore’s piece is one of them.]
27. Gay, Claudine. “Between Black and White,” p. 27:
In this new political environment, Brazilians of color came to question
publicly the reality of "racial democracy."
Skidmore, Thomas. “Race and Class in Brazil: Historical Perspectives.”
Luso-Brazilian Review 20(1), 1983, p. 110:
 Brazilians of color began to question publicly the myth of racial democracy.
[Gay’s article has no citations. At the end of the article (p. 28), there is a separate
section entitled, “Suggestions for Further Reading.” In that section, there are 5
works mentioned. Skidmore’s piece is one of them.]
28. Gay, Claudine. Taking Charge: Black Electoral Success and the
Redefinition of American Politics. Dissertation submitted to the
Department of Government, Harvard University, 1997, p. 32:

The average turnout rate seems to increase linearly as African-Americans

become a larger proportion of the population. This is one sign that the data
contain little aggregation bias. (If racial turnout rates changed depending upon
a precinct’s racial mix, which is one way to think about bias, a linear form
would be unlikely in a simple scatterplot. A linear form would only result if
the changes in one race’s turnout were compensated by changes in the turnout
of the other race across the graph.

Palmquist, Bradley and Stephen Voss. “Racial Polarization and Turnout in

Louisiana: New Insights from Aggregate Data Analysis.” Paper prepared for
the 54 Annual Meeting of the Midwest Political Science Association,
th

Chicago, Illinois, April 18-20, 1996, p. 10:

… the average turnout rate seems to decrease linearly as African Americans

become a larger proportion of the population. This is one sign that the data
contain little aggregation bias. If racial turnout rates changed depending upon
a precinct’s racial mix, which is one description of bias, a linear form would
be unlikely in a simple scatter plot (resulting only when changes in one race’s
turnout rate somehow compensated for changes in the other’s across the
graph).
[Palmquist and Voss 1996 is never cited.]
29. Gay, Claudine. Taking Charge: Black Electoral Success and the
Redefinition of American Politics. Dissertation submitted to the
Department of Government, Harvard University, 1997, p. 34:

The idea behind the “method of bounds” is that the beginning point for any
ecological inference should be the knowledge a researcher has for certain.
This knowledge includes the fact that any proportion is by definition bound
by 0 and 1. Furthermore, the marginals of a table, Xi (black population
density) and Ti (total turnout) dictate the minimum and maximum possible
values of the cells in the table. King’s method makes direct use of this
information to establish absolute bounds on the values of the quantities of
interest.

Palmquist, Bradley and Stephen Voss. “Racial Polarization and Turnout in

Louisiana: New Insights from Aggregate Data Analysis.” Paper prepared for
the 54th Annual Meeting of the Midwest Political Science Association,
Chicago, Illinois, April 18-20, 1996, p. 12-13:

The beginning point for any ecological inference should be with the
knowledge we have for certain. Almost from the beginning of methodological
work in this area, researchers have used the fact that proportions must by
definition be between 0 and 1 (Duncan, Cuzzort, and Duncan 1961; Achen
and Shively 1995). Recently, Gary King (n.d.) has emphasized the particular
advantages of using the precinct-by-precinct constraints. Making direct use of
this information to establish absolute (i.e. not probabilistic) bounds on the
percentages of the internal cells is entirely straightforward. For any single
table (either of a precinct or of a the state as a whole), the marginals dictate a
minimum and maximum possible value for each of the cells.
[Palmquist and Voss 1996 is never cited.]
30. Gay, Claudine. Taking Charge: Black Electoral Success and the
Redefinition of American Politics. Dissertation submitted to the
Department of Government, Harvard University, 1997, p. ii:

I am also grateful to Gary: as a methodologist, he reminded me of the

importance of getting the data right and following where they lead without
fear or favor; as an advisor, he gave me the attention and the opportunities I
needed to do my best work.

Finally, I want to thank my family, two wonderful parents and an older

brother. From kindergarten through graduate school, they celebrated my every
accomplishment, forced me to laugh when I’d lost my sense of humor, drove
me harder than I sometimes wanted to be driven, and gave me the confidence
that I could achieve.

Hochschild, Jennifer L. Facing Up to the American Dream: Race, Class, and

the Soul of the Nation. Princeton University Press, 1996, p. xx:

Bill Wilson taught me how to think about the relationship between race and
class, gave me confidence that I could write a book on the subject, and
provides me and many others with a model of how to express the courage of
one’s convictions with dignity, evidence, and toughness. Sandy Jencks
showed me the importance of getting the data right and of following where
they lead without fear or favor. His example of iconoclasm about what the
right answer is combined with passion for finding the right answer drove me
much harder than I sometimes wanted to be driven.
[Hochschild 1996 is never cited.]

31. Gay, Claudine. Taking Charge: Black Electoral Success and the
Redefinition of American Politics. Dissertation submitted to the
Department of Government, Harvard University, 1997, p. 12-13:

Bobo and Gilliam (1990) stands as the most complete research to date in this
area, focusing on black elected officials at the city-level. Using 1987 survey
data, Bobo and Gilliam found that African-Americans in “high
black-empowerment” areas—as indicated by control of the mayor’s
office—are more active than either African-Americans in low empowerment
areas or their white counterparts of comparable socioeconomic status.
Empowerment, they conclude, influences black participation by contributing
to a more trusting and efficacious orientation towards politics and by greatly
increasing black attentiveness to political affairs.

As for whites, Bobo and Gilliam (1990) found that they pay less attention to
local politics when blacks control local offices; however, they do not become
generally less trusting and efficacious as a result.

Gay, Claudine. “The Effect of Black Congressional Representation on

Political Participation.” American Political Science Review 95(3), 2001, p.
590:

In the most complete research to date, Bobo and Gilliam (1990) find that
African Americans in areas of high black empowerment—as indicated by
control of the mayor’s office—are more active than either African Americans
in low empowerment areas or whites of comparable socioeconomic status.
 Empowerment, they conclude, influences black participation by contributing
to a more trusting and efficacious orientation toward politics (see also Abney
and Hutcheson 1981; Howell and Fagan 1988) and by greatly increasing black
attentiveness to political affairs. As for whites, the findings suggest that they
pay less attention to local politics when blacks control local offices but do not
become generally less trusting and efficacious as a result.

Bobo, Lawrence, and Franklin Gilliam. 1990. “Race, Sociopolitical

Participation, and Black Empowerment.” American Political Science Review
84 (June): 377–93, p. 377:

The results show that blacks in high-black-empowerment areas—as indicated

by control of the mayor's office—are more active than either blacks living in
low-empowerment areas or their white counterparts of comparable
socioeconomic status. Furthermore, the results show that empowerment
influences black participation by contributing to a more trusting and
efficacious orientation to politics and by greatly increasing black attentiveness
to political affairs

Bobo and Gilliam 1990, p. 383:

In sum, whites tend to pay less attention to local politics when blacks control
local offices but do not become generally less trusting or efficacious as a
result.

[Gay uses no quotation marks around verbatim language.]

32. Gay, Claudine. “Spirals of Trust? The Effect of Descriptive
Representation on the Relationship between Citizens and Their
Government.” American Journal of Political Science, vol. 46, no. 4 (2002),
p. 720n3:

King et al. (2001) identify four conditions, all of which must hold, in order for
listwise deletion to be preferable to multiple imputation: (1) The analysis
model is conditional on X (i.e., explanatory variables containing missing
data), and the functional form is known to be correctly specified; (2) There is
nonignorable (NI) missingness (i.e., the probability that a cell is missing
depends on the unobserved value of the missing response) in X, and there are
no other variables available that could be used in the imputation model to
predict X; (3)Missingness in X is not a function of Y(i.e., the dependent
variable containing missing data) and unobserved omitted variables that affect
Y do not exist; (4) The number of observations left after listwise deletion
should be so large that the efficiency loss from listwise deletion does not
counterbalance the biases induced by the other conditions.

King, Gary, James Honaker, Anne Joseph, and Kenneth Scheve. “Analyzing
Incomplete Political Science Data: An Alternative Algorithm for Multiple
Imputation.” American Political Science Review vol. 95, no. 1 (2001), p. 58:

For listwise deletion to be preferable to EMis, all four of the following

(sufficient) conditions must hold. (1) The analysis model is conditional on X
(such as a regression model), and the functional form is known to be correctly
specified (so that listwise deletion is consistent, and the characteristic
robustness of regression is not lost when applied to data with measurement
error, endogeneity, nonlinearity, and so on). (2) There is NI missingness in X,
so that EMis can give incorrect answers, and no Z variables are available that
could be used in an imputation stage to fix the problem. (3) Missingness in X
 is not a function of Y, and unobserved omitted variables that affect Y do not
exist. This ensures that the normally substantial advantages of our approach in
this instance do not apply. (4) The number of observations left after listwise
deletion should be so large that the efficiency loss from listwise deletion does
not counterbalance (e.g., in a mean square error sense) the biases induced by
the other conditions.

[Gay uses no quotation marks around verbatim language.]

33. Gay, Claudine. “Spirals of Trust? The Effect of Descriptive

Representation on the Relationship between Citizens and Their
Government.” p. 720:

Multiple imputation, which assumes that information in the observed data

provides indirect evidence about the likely values of the unobserved data, can
correct for the inefficiency and bias that result from listwise deletion.

Gay, Claudine. “Moving to Opportunity,” p. 158:

Multiple imputation, which assumes that information in the observed data

provides indirect evidence about the likely values of the unobserved data (and
that, after controlling for the observables, missingness is independent of the
unobserved data), is known to outperform listwise deletion (i.e., removing all
out-migrants from the analysis) by correcting for the inefficiency and bias that
result from the latter approach (Schafer and Olsen 1998).

Joseph L. Schafer & Maren K. Olsen (1998). “Multiple Imputation for

Multivariate Missing-Data Problems: A Data Analyst's Perspective,”
Multivariate Behavioral Research, 33:4, p. 546:
 In any incomplete dataset, the observed values provide indirect evidence
about the likely values of the unobserved ones. This evidence, when
combined with certain assumptions (described in the following section),
implies a predictive probability distribution for the missing

[In “Spirals of Trust?” Gay cites Schafer and Olsen 1998 only in the previous
sentence. She uses no quotation marks around verbatim language.]

34. Gay, Claudine. Taking Charge: Black Electoral Success and the
Redefinition of American Politics. Dissertation submitted to the
Department of Government, Harvard University, 1997, p. 146:

Preston (1978) argued that descriptive representation is “not only desirable

but necessary for Black Americans,”because, in part, it fulfills a host of
psychological needs.

Swain, Carol M. Black Faces, Black Interests: The Representation of African

Americans in Congress.Cambridge: Harvard University Press, 1995, p. 217:

The presence of black representatives in Congress, regardless of their political

party, fulfills a host of psychological needs that are no less important for being
intangible. One need only attend an annual Black Caucus legislative weekend
to see the pride that the hundreds of blacks who attend the affair have in the
group of congressional black representatives. Black representatives are
celebrities-icons for their group. Michael Preston writes: "Symbolic
representation is not only desirable but necessary for black Americans.”

[Swain 1995 cited elsewhere (p. 144 and 148, etc.) but not here.]
 35. Gay, Claudine. “Fighting Poverty, Mobilizing Voters: Housing Investment
and Political Participation,” Working Paper, April 15, 2014, p. 5n4:
For a project to be eligible for tax credits one of two income criteria for
occupants must be met, 20–50 or 40–60: Twenty [40] percent of the units
must be rent restricted and occupied by households with incomes at or below
50 [60] percent of area median income.
Khadduri, Jill, Carissa Climaco Kimberly Burnett, Laurie Gould, Louise
Elving. What Happens to Low Income Housing Tax Credit Properties at Year
15 and Beyond?Washington, DC: U.S. Department of Housing and Urban
Development, 2012, p.1:
Properties must meet one of two criteria to qualify for tax credits: either a
minimum of 20 percent of the units must be occupied by tenants with incomes
less than 50 percent of Area Median Income (AMI), or 40 percent of units
must be occupied by tenants with incomes less than 60 percent of AMI.
[Khadduri et al cited in the next note, p. 5n5, not in this one.]
36. Gay, Claudine. “Fighting Poverty, Mobilizing Voters: Housing Investment
and Political Participation,” Working Paper, April 15, 2014, p. 12n17:
LIHTC program guidelines provide higher tax credit amounts to projects
developed in areas designated by HUD as ‘qualified census tracts,’ defined as
tracts where at least 50 percent of the households have incomes below 60
percent of their metropolitan area’s median family income or where the
poverty rate exceeds 25 percent.
Keren Horn and Katherine O'Regan. “The Low Income Housing Tax Credit
and Racial Segregation,” Housing Policy Debate, 21:3, 2011, p. 446:
Projects that are built in qualified Census tracts (QCTs), defined as
neighborhoods where at least 50 percent of the households have incomes
below 60 percent of their metropolitan area’s median family income, receive a
30 percent bonus in their qualified basis.^13
^13 In 2000, the definition of QCT was modified so that localities could use a
25 percent poverty threshold instead, which on average is equivalent.
[Horn and O’Regan 2011 is cited by Gay only once, on p. 3.]
37. Gay, Claudine. Taking Charge: Black Electoral Success and the
Redefinition of American Politics. Dissertation submitted to the
Department of Government, Harvard University, 1997, p. 2:

To date, social scientists have concentrated their analytical efforts on the

ambiguous link between minority office-holding and minority public policy
agendas, between descriptive representation (the statistical correspondence of
demographic characteristics) and substantive representation (the
correspondence of legislative goals and priorities)

Swain, Carol M. Black Faces, Black Interests, p. 217:

Pitkin distinguishes between "descriptive representation," the statistical

correspondence of the demographic characteristics of representatives with
those of their constituents, and more "substantive representation," the
correspondence between representatives' goals and those of their constituents.

[Swain 1995 is cited elsewhere, but not here.]

38. Gay, Claudine. Taking Charge: Black Electoral Success and the
Redefinition of American Politics. Dissertation submitted to the
Department of Government, Harvard University, 1997, p. 187:
For the last 40 years, the most widely used statistical approach to ecological
inference (m 2 x 2 tables) has been Goodman's Regression. Briefly, the
method involves a regression of total vote on the proportion of the population
that is black and the proportion that is white, with no constant term. The
coefficients generated from this least squares regression are assumed to be
estimates of the district aggregates, Bb (the proportion of black residents who
vote in the district) and Bw (the proportion of white residents who vote in the
district). There are a number of problems with this method. Among them:
[…]

-Goodman's procedure incorrectly assumes that the precinct-level

parameters, Bb and Bw are constant over the precincts.

-Goodman's model does not take into account information from the
"method of bounds,"

King, Gary. A Solution to the Ecological Inference Problem : Reconstructing

Individual Behavior from Aggregate Data. Princeton University Press, 1997,
p. 37, 38, 39:

“The most widely used statistical approach to ecological inference,

Goodman's regression”

[…]

“In terms of the notation in Table 2.3, Goodman's method involves a

regression of T1 (proportion of the voting-age population turning out to vote)
on X1 (proportion of the voting-age population who are black) and (1 — X1)
(proportion who are white), with no constant term. The coefficients from this
least squares regression, Bb and Bw, are intended to be estimates of the
district aggregates Bb (the fraction of blacks who vote) and Bw (the fraction
of whites who vote) since the precinct-level parameters β1 and βw are
assumed constant over precincts.”

[…]
 Goodman's model also does not include information from the method of
bounds.

[Gay does not cite King here and uses no quotation marks around verbatim
language.]

39. Gay, Claudine. Taking Charge: Black Electoral Success and the
Redefinition of American Politics. Dissertation submitted to the
Department of Government, Harvard University, 1997, p. 92:

Since the 1950s, the reelection rate for incumbent House members has rarely
dipped below 90%. In 1994 it was 92.3% (Swain 1997).

Swain, Carol M. Black Faces, Black Interests, p. 31:

Since the 1950s the reelection rate for House members has rarely dipped
below 90 percent.

[Gay uses no quotation marks around verbatim language and miscites Swain in the
next sentence.]