Ethics, Law, and Society Forum – September 19, 2017 – Aliya Saperstein

[ Music ]>>Welcome everybody. Let’s get started. Can I have your attention? So, today’s speaker is
Professor Aliya Saperstein, who is from Stanford University
from their Sociology Department. She has always been a visiting
scholar at Sciences Po in Paris and the Russell Sage Foundation. Her research focuses on how
we conceptualize categories like race and sex and how those
conceptualizations can impact social inequality. And her research
has been published in numerous peer review articles
including the Journal of Soc, I am sorry The American Journal
of Sociology and Demography and for general audiences in
such places as the proceedings for the National Academy
of Sciences and PLoS One. In turn, this work has gained
attention from places like NPR and the Colbert Report. And in 2016 she received
the Early Achievement Award from the Population
Association of America and I am personally a big fan. So I am thrilled to welcome
Professor Saperstein to SSU to speak to us on
this question — Can Racial Inequality be
Fixed by Seeing Race as Fluid? Please join me in welcoming
Professor Aliya Saperstein. [ Applause ]>>Thank you Josh. And thank you all for coming. I really appreciate
you being here today so that I can share
my work with you. But I am also hoping you guys
can help me a little bit. Because I have to admit, I
am not sure I know the answer to the question that is
the title of my talk. So, in true academic fashion, I
am going to kind of hem and haw and talk around [laughter]
the idea of the question. And I am hoping by the end I can
also hear what you guys think the answer is, either what you
already think the answer is or what you think
after I tell you, share some of my
research with you today. So the question in the title
of my talk was inspired by some work, by a body
of work done my colleague in social, in psychology. I am in sociology. And psychologists have found
over and over again looking at lots of different traits
from gender and math ability to sexual orientation
and race that people who believe a particular
trait is innate or stable, are more likely than people who
see the same trait as learned or malleable to exhibit
prejudiced to endorse stereotypes, to
favor policies of segregation, like same gender classrooms or
the North Carolina Bathroom Bill or Trump’s Muslim ban. They are also more likely
to explain behavior based on what psychologists
call dispositional rabid and situational causes. So they are more likely to
say that you do not have a job and you are a drug abuser
because you are lazy and have weak morals, not
because of the economic downturn and the, you know pharmaceutical
companies overprescribing opioids for example. So what this research suggests
is that, what it implies is that if we embarrassed
in the case of race. If we embarrassed
racial fluidity to the idea roughly speaking that race is socially
constructed, that this could be the key
to ending racial inequality. So that if people think that
characteristics are fluid, they are less likely to
do all of these things that support a regime
of race or gender or sexual orientation
inequality. Of course, the catch is race and
inequality has been [inaudible] for hundreds and
hundreds of years. Probably if you are here in
this room, I do not need to go over this history for you. But just to make
sure that we are all on the same page,
the idea of race. People argue about exactly when
it was invented and exactly in what place and
for what purpose. But for the most part,
the ideology that we know of as race was created not
as a neutral recognition of skin color differences or
geographic origin differences. But as a justification
for inequality. So in the 18th and 19th
centuries people looked out on the world. They saw inequality. In many cases some of them were
interested in perpetuating it through slavery and conquest. And they used ideas about
race and who was superior and who was inferior to
justify why it was okay to enslave certain people
and take other people’s land. These two pictures
that I am showing you on the left are the famous
images from [inaudible] skulls, where the Caucasian skull, which he thought was the
most beautiful is on the top. And the image on the right,
which is from the Races of Mankind in 1911 clearly
makes the you know high status, presumably educated white man at
the central of the other races of mankind, at the center of
the other races of mankind. So these kinds of ideas about
race and inequality have been with us for a long,
long, long time. And although perhaps
a couple years ago, this might have seemed
like ancient history. The last couple of years have
give us reason to believe that we are not so far from
this history as we might like. So, Ferguson Black Lives
Matter movement, the backlash against Rachel Dolezal,
the Charleston shooting and most recently the
demonstrations associated with Charlottesville remind us that most Americans are not
ready to think of race as fluid. And perhaps a surprising number
of white people still believe that they deserve
better than other people who live in the same country. Now, most Americans
do not believe that. Most white people
do not believe that. But we still have the challenge of how do we get past this
long history connecting race and inequality. How do we get rid of race
and these old associations and these old stereotypes? They are much harder to get rid
of, even in those of us who want to be allies and those of us who
feel that we are egalitarian, these kinds of connections,
cognitive connections between race and
inequality run deep. And I am going to give
you an example of that. This is. You may be familiar with the idea of
the Welfare Queen. Now, this idea comes to us
at courtesy of Ronald Reagan. In 1976 he introduced the nation to a woman he called the
Welfare Queen in a series of campaign speeches during
his first run for President. He was trying to draw
attention to himself. He was the Governor
of California. He was trying to build
a national reputation. So he complained a lot about
wasteful government spending and this woman was a
great example for him, a political stalking
horse, if you will. Now, Reagan in his speeches
never mentioned her race. He just mentioned her
string of fake ID’s. The thousands of dollars
she allegedly stole. Her many children. Her Cadillac and the
fact that she lived on the South Side of Chicago. Americans imaginations
filled in the rest. So our stereotype between
blackness and welfare is so strong that most people think
of the Welfare Queen as black. It is not clear, however, that
the Welfare Queen is black. Now, we can talk about
well what are the criteria that we would use to
decide if she is black. What does her birth
certificate say? What does her death
certificate say? How was she recorded
on a census? How does she identify? How do other people see her? There are lots of
things that we could use to decide what her race is. What I am interested in is
how she was seen by others and how we came to see
the Welfare Queen as black as just being obvious. Something we did not even have
to give a second thought to. So, the back story
about this woman. And it is hard to
tell you her name because she used
a lot of aliases. The Welfare Queen is known as,
she was known as Linda Taylor at the time that she
was the Welfare Queen. It is believed that she was
born Martha Miller in a family of people who were identified
in the census as white. And 10 years before she became
famous as the Welfare Queen, she was, the story on the
left that says phony heiress of policy king gets six months. So, she was accused of falsely
claiming an inheritance. At that time, she was going by
the name of Constance Wakefield. And her relatives
came to court and, in an episode not unlike the
outing of Rachel Dolezal, announced to the court that she and her family were
in fact white. And so she could not be
this, you know the wife of this particular heiress. And coverage of her
many exploits in the Chicago Tribune reporter
George Blith describes her as — she allegedly has posed as
a Filipino white and a black to obtain Welfare aide. So she kind of made a
habit of allowing people to see what they wanted to see
in her and using race as a way to get what she wanted. The picture on the right is a
picture of her from the 1980’s in Florida shortly
before she died. As journalist, Josh Levin,
put it in a wonderful article in Slate, if you are curious
about more of this history. She was white according to
official records and in a view of certain family members who could not imagine
it any other way. She was black or colored or
Negro when it suited her needs. Or when someone saw a woman they
did not think or did not want to think could possibly
be Caucasian. Sorry, that is her at the
time of her conviction. So, I am telling you this
anecdote because what I am going to tell you now is a bunch of
research that is largely based on quantitative data that
is much less interesting. But, I think the quantitative
data that is at the basis of my research supports
the idea that the story of the Welfare Queen and
how we come to see people as particular races
is more general. So my research is
based on surveys that include multiple
measures of race, both how people self identify
and how other people see them. And often these measures
are recorded multiple times over a period of years. So we can see people’s race
changing in the survey data. The two surveys that I used are
the National Longitudinal Survey of Youth, which started
interviewing people in 1979. And I will also show you some
results from the National Study of Adolescent to Adult Health, which started interviewing
people in 1994 or 95. And if it helps you to kind of
pin how old these people are in your mind, the people in the
National Study of Adolescent to Adult Health are my cohort. So they are people who were in
school, at college, etcetera, around the same time that I was. I am also going to show you
a little bit of evidence from a categorization experiment
that I conducted along with some social psychologists. And this allows us to
control for aspects of physical appearance that are
not recorded in the survey data. So that is where we are headed. By the way, if you guys
have any questions, like short clarifying questions because something I said is
not clear, please feel free to raise your hand and I
would be happy to answer it. So to give you a sense of where
the data that I used comes from, I am going to tell you a little
bit about how the survey works. So the National Longitudinal
Survey of Youth — every year in the beginning
and every other year starting in 1994, someone would,
an interviewer would come to the house of the people
who were part of the survey. And would talk to
them about, you know, how is your life going? Have you had any more kids? Are you still married? Are you still working
at the same job? How much money do you make? Do you smoke? Lots of questions about
how things are going and they check in, you know
year after year after year. Now at the end of the interview, the survey interviewer would
typically go back to their car and fill out this section known
as the interviewer remarks. And what you should
notice is in addition to saying how long the interview
was and what date it occurred on and other things like whether
there were other people in the room or whether
the person was cooperative and understood the
questions, one of the things that the interviewer is
asked to record is the race of the person they talked to. They are given the categories
white, black and other. Those categories sound
kind of strange to us now. In 1979 that was a bit
more typical thinking about what racial
variation looked like in the United States. So, this survey is still
being conducted today. But from 1979 to 1998 the
interviewers were asked to complete this
question at the end of every single interview
they conducted. So what this means is that
each person in the survey has up to 19 years of data on how someone else
perceived their race. So these are some examples of what I would call a racial
classification trajectory. So, the example on the top is
someone whose race is always classified the same
way every year. They were white in
every single year that the survey interviewer
talked to them. The other examples in
the table are examples of people whose racial
classifications changed over time. Now, these are not the only
examples from the data, but these are the
most common examples at each level of
racial fluidity. So, the most common example of having one classification
different was being classified as other in the first year and
white every year after that. But other people who have
been white in all the years, but one year in which they were
black or black in all the years but one year in which
they were white. That one just happens
to be the most common. But so looking at this data,
for one thing 20 percent of the 12 thousand people in the
survey had at least one change in their racial classification
over those 19 years. That suggests race is a lot more
fluid than we typically think. That means 80 percent never
experienced any change. They were always white,
always black, always other. But 20 percent of the people
in the sample fell into one of those other categories where
their race did change over time. And what this raised
for me was — well so if race changes
over time, what predicts why you are other
in one year or black or white? Is there is something going
on in that particular year? Is there something else that
is changing about your life that might help us understand
why someone suddenly sees you different? So we asked the question — if we take all the people
who were not seen as white in the previous year, but are
seen as white in the next year, what other things did
they share in common? What predicted who might
have been seen as white in the next year if they
were not white according to the survey interviewer
in the previous year? And what we find is that a lot
of things that we associate with success or you know the
good life were associated with being classified as
white in the subsequent year. So, if someone moved
to the suburbs or got their Bachelor’s
degree or got married, they were more likely to be seen
as white in the subsequent year than someone who did
not have a change in their status along
those same dimensions. And if we took something
and experienced that maybe society does
not think is so positive, like being an unmarried parent. This was particularly
stigmatized in the 1980’s and early 90’s when
this data comes from. So if we take something
that people do not think of as being positive, you can
see that it made it less likely that someone would switch
from not white to white in the subsequent year. So we concluded that departures. Oh I forgot to tell you. The dash line there is the
average level of fluidity. So, on average about nine
percent of people change from not white to
white in a given year. So what we are interested
in is you know if you think of average fluidity as being like yeah we are just not
sure what to classify you as, so there is some wiggle room. You can think of the average
level of fluidity like that. So what we are interested in is
when do these kinds of changes, when do they differ from the
average level of fluidity? And when they differ,
what it indicates is that upward mobility, being
successful, doing the things that society thinks is right,
increases the odds of being seen as white by the interviewer
in the subsequent year. Now, I want you to pay
attention to the Y axis. On this chart it goes
from zero to 18 percent. On the next chart, it goes from
zero to one point eight percent. So I am not trying
to mess with you. That is a change in magnitude. So what you should take
away from this is changes from not white to white are an
order of magnitude more common than changes from
not black to black. However, the few changes from
not black to black that do occur in the survey suggests
the opposite pattern of what we just saw. So if you become an unmarried
parent between one survey and the next, the survey
interviewer is more likely to see you as black in
the subsequent year. If you lost your job or your
income fell below the poverty line, the interviewer was
significantly more likely to see you as black in
the subsequent year. But if you got your college
degree, they were less likely to see you as black
in a subsequent year. So again, the dash line
indicates average levels of fluidity. Point seven percent of the
sample changed from not black to black in any given year. But changes were more likely when the person experienced
downward mobility. When something bad
happened to them. When they had some
kind of economic shock. They were more likely
to be seen as black by the survey interviewer
in the subsequent year. So you might be wondering
who are these people? What do they look like? How could this possible happen? Race is so obvious. I mean the woman talking
to me is clearly white. So we were interested
in this too because the surveys
do not say things like what hair color they had
or what skin color they had. So we cannot control for whether
people looked different using the survey data. But we could run a
categorization experiment where we controlled for physical
appearance, but also tried to vary the status
information that somebody had about the person that
they were categorizing. So we did this. This is exactly the
screen that the subjects of the experiment saw and
so they were shown a face that was dressed in
this case in a suit. And they were asked to
categorize the faces that they saw as quickly as they
could by moving the mouse up and to the left to
categorize someone as black. And up and to the right to
categorize someone as white. So we were interested
in two things. One — what category
did they pick? And two — we could track
the mouse trajectory as they made their decision. So that even if they
picked the wrong category for the status information,
we could see if they weighed both
the physical appearance of the person and the status
queue implied by their clothing. So we used both male
and female faces. These are examples
of the male faces. So you can see that at each
level it was actually a 13 point spectrum of phenotype from
the most stereotypically white on the left to the most
stereotypically black on the right. So we varied appearance
gradationally across the spectrum and then
showed them some faces that were in the suit and some faces
that were in, it is supposed to be janitorial
coveralls, I am not sure that totally comes clear. But you can see the
person is wearing a T-shirt and kind of overalls. So we were trying to signal
high status and low status. So what we found. These are the exact same faces. They are not the exact middle. But in the middle
of the spectrum. And what we found
is that as expected if the face was wearing
the business suit, they were more likely to
classify that face as white. And that held true
across the entire physical appearance spectrum. But was most significant
in the middle when people’s appearance
was ambiguous. But when we looked at the mouse
trajectories, what we saw is that even when they
took the same, even when they were
classifying the same face in the janitor coveralls, even
when they classified that face as white, you can see
the trajectory veers ever so slightly. But statistically significantly
closer to the black category. So we interpret this to mean
that their brains are sort of, anybody’s brain who
is doing this, is weighing okay I know
white people look like this and black people look like this. But I also know that
white people do this and that black people
do other things. And that they are
weighing status information and appearance information
as they make categorizations. So you can think
of the trajectory as like them fighting
the impulse to classify the face differently
because of the clothing that it is attached to it. Now you might be thinking why
has she been talking about black and white, black and
white, black and white. We are so past that. Unfortunately, a lot of
the data that we can use to answer these questions
are limited in the number of categories that we have. But the other survey that I told
you about that looks at people in my cohort did use
more detailed categories. So they used white and black. But also Asian and
Pacific Islander. And American Indian
and Alaska Natives. Now the drawback of this survey
is instead of 19 years of data on classification where the
categories are exactly the same, we only have two. So, we are a little bit
less sure on these results, but as you see they
line up pretty strongly with the previous ones
from the other survey. The other catch is because
these people are younger, we cannot look at the same
status outcomes in both surveys. But in this survey, they are
in their 20’s and early 30’s and so we can look at contact with the criminal
justice system. This is a status outcome that
unfortunately is most common in people in this age group. So we were interested
in looking at their race in the first survey wave, looking at whether they were
arrested or not in between and then looking
at their race again after they experienced an
arrest to see if it changed. So what this chart shows is that people were less
likely below the one line, they are less likely to be
classified as Asian or white after they have experienced
an arrest. And they are more likely to be
classified as American Indian or black after they have
experienced an arrest. This survey is setup exactly the
same way as the one I told you about before where the survey
interviewer marks their race at the end of the interview,
after they have heard about all this other
stuff, like whether or not they have been arrested. So, let me show you in a little
bit more detail what I mean by they were less likely
to be classified as Asian and more likely to be classified
as American Indian or black. So let’s take someone who
was classified as Asian in the first wave of the survey. Then we are going to look at whether they were
arrested or not. And we are going to look at the different possible
racial destinations that the survey interviewer
could have put them in. They could stay Asian. They could move to white. They could move to
American Indian. And they can move to black. So what you can see here is that
someone who starts out Asian in the first wave of the survey
is less likely to stay Asian if they were arrested in
between the two survey waves. They are ironically more likely
to be classified as white, which tells us maybe
something about the spectrum of our stereotypes
about race and crime. So Asians are perhaps the least
likely to participate in crime. Whites, American Indians and
blacks perhaps are more likely to participate in crime
in our stereotypes. So then the person
who was classified as Asian is more likely to be
classified not only as white, but also as American Indian. Now if we take someone
who is American Indian in the first survey
wave and look at how they were
classified depending on their arrest record, we
see that they are less likely to be classified as Asian. They are less likely to
be classified as white. They are more likely to
stay American Indian. And they are more likely to have
the survey interviewer change and list them as black. So what do we take
from all of this? Other than the fact that
race is clearly more fluid than we typically think. I want to take you
back to inequality. What did this do
about the association between race and inequality? So you need to think about
three different directions of stability and change
in racial classification. You need to think about who
is entering the category from one time point to another. So, let’s take someone who is previously not
classified as black. They were anything but black
in the first survey wave. If they were arrested — those are the darker
bars in the chart. If they were arrested,
they are more likely to be newly classified as
black in the next survey wave. We take someone who was
not classified as Asian — what category did they enter? They are more likely to
enter the Asian category if they were not arrested. So you are more likely to
enter the black category if you were arrested. You are more likely to
enter the Asian category if you were not arrested. What about exiting? Who left a category from
one time point to the next? So if you were classified as
Asian, you are less likely to be classified as Asian
again if you were arrested. That association between
Asianess and the lack of crime is being
reinforced by who is entering and who is leaving the category. Similarly, if you were
previously classified as black. Again, there are many fewer
changes to and from black. But you are statistically
significantly less likely to be classified as black
again if you were not arrested. So you leave the black category
when you were not arrested. And then lastly, you
have the people who stay in the same racial category from
the one time point to the next. So this is not only
about fluidity. Most of what happens here
is actually stability. But there are patterns
in the stability as well. So you are more likely to
stay in the black category if you were arrested and
you are more likely to stay in the Asian category if
you were not arrested. So we have a vicious cycle here between existing
racial inequality in criminal justice systems. And the inequality that
we create through the way that we perceive people. Now, the last category you might
be wondering about is Hispanic. In most of this data and in
most data in the United States up until now, we do not
categorize Hispanic as a race. We ask about it in a separate
question called ethnic origin. That is potentially changing. The census bureau has been
considering adding Hispanic as a racial category. So that you could
pick white, black, Hispanic, Asian, etcetera. But we can look at
how people identi. And so. Blah. I forgot [laughter]. So because it is not
classified as a race, the survey interviewers
were never asked — does that person look Hispanic? But we can look at whether
people identify themselves as Hispanic and whether
that changes over time. And what research suggests from
other scholars and my own is that something similar
happens in terms of how people identify
themselves, not only by race, but also by Hispanic origin. So then not only does
it change over time, people do not identify
necessarily consistently as Hispanic. Either over time or across
generations, so from parents to children, parents
might identify as Hispanic, but
children do not. The other thing, what I find
in my research is that people with ancestors who were born in
Spanish speaking countries — so that is our definition
of Hispanic origin. People whose ancestors were born
in Spanish speaking countries, are less likely to identify as
Hispanic the more successful or assimilated they are. So, they will tell you
that they have ancestors from Spanish speaking countries. But if you ask them
do you identify as Hispanic, they will say no. And those people are more
likely to be employed, to have higher incomes, are
less likely to have been in jail and are less likely to
have received welfare. So that means the
opposite is true. People who are consistently
identifying as Hispanic over time are more likely to have experienced
negative things. So someone who was
unemployed at some point in their life is more likely
to continue to identifying as Hispanic over and over
and over again later in life. So we see a similar pattern
with how other people see you and potentially how we
identify ourselves as well. So let’s go back to that
question that we started with. Can we fix racial inequality
by seeing race as fluid? If we weigh that data,
uhh that is depressing. Seems like if race if fluid, we
are only reinforcing inequality. That people are moving in
and out of categories in ways that cement and maintain the
inequality that we already have. That suggests that
fluidity is not going to help us eliminate inequality. But I do think the psychologists
are on to something because we cannot
see those patterns that I just showed you unless
we think that race is not fixed. You would not even think
to look at that data. You would not even think
to look for that data. If you thought race was
stable at birth and fixed, you would only need
one point in time in which you saw how
someone identified or how someone else classified
them to know their race. So you cannot even see what I
just showed you without taking that first step and
thinking that race and ethnicity might be fluid. I also think that often we
associate seeing race as fluid with distracting from the
real problems that matter — from the people being
shot in the streets, from the Nazis marching
on our campuses. I would argue paying attention
to the social construction of race does not side
track us from the things that matter and here is why. Because when we think of
race as fixed and stable, we are more likely to look for
explanations for inequality in the essence of individuals. So you are just a bad
weak, immoral, lazy person. And that is why you are in
the position that you are. If we think of race as fluid, I
think it forces us, it forces us to look at who we categorize as
what and why and it forces us, it exposes the social
and political mechanisms that maintain inequality. We can no longer
blame individuals for the inequality
that they experience. We have to look at the whole
system including our own perceptions that
helps maintain it. Thank you. [ Applause ] So what did you think? Did you buy it? [Laughter]. How are we going to do this? How are we going to fix it? Can we fix it? Any question is fair game
including are you serious? Is that really what you found? Yes.>>Is there any evidence
that suggests that gender affects how
people self identify?>>That is a good question. Can you tell me a little bit
more what you are thinking?>>Just like if someone
was I guess ambiguous that they might chose
to self identify. Like if you are a woman, you
might chose to self identify in a way that you thought
might benefit you or something. I do not know. I am not sure if that is
a weird question to ask. But.>>So it is not a weird
question at all to ask if the patterns are
different by gender. Not a weird question at all. And it is. The answer is complicated. But it always is. So what I can tell you is. First I will tell you about
how other people see you because of my data
is related to that. So in terms of how other people
see you, the amount of fluidity, that 20 percent is the
same for men and women. So the amount of fluidity
is exactly the same. But the, what we find
is that the experiences or characteristics that seem to
be linked to when you change, do appear to be different. And they are different in
the way of our stereotypes. So the Welfare Queen is a woman. And we associated women
and welfare and black women and welfare in particular. So, having received welfare
benefits is more likely to affect the classification
of a woman in that stereotypical
direction, whereas going to jail is more likely to affect
the classification of a man. So, the overall level
of fluidity is the same, but the things that
people use as cues if you will, do appear
to differ. Now, in terms of
self identification, we do not have as
good data on this. In part because we do not think
race changes, so we do not tend to ask people how they
identify over time. We only ask them when we change
the question, when we are going to give them more categories. And to them it is really hard
to tell well did you change because we gave you a
different category or is it because something
really changed. But there is some
evidence for example that women are more
likely identify as multiracial than men are. And there is lots of arguments
over why that might be, women are more likely to keep
in touch with their relatives and know their ancestry
and be into genealogy. So maybe they just know more and
are more willing to report it. But mostly we need
to do more research because we have not been
thinking about race this way. So we have not even been
asking those questions. Yes.>>With the experiment that you
did, with the picking the white or black person, I was wondering like did you use all white
people to do the experiment? Or did you keep track of
like if you know [inaudible].>>It is a great question. So the experiment was
run at Tufts University with using Tufts
University college students. And so they were not all white,
but they were mostly white. And there were not
enough, you know. So the results did not. There are non-white
people in the sample, but there are not enough
non-white people to be able to say oh white people do this, black people do this,
Asian people do this. If I can assume why you are
asking, which maybe not, so you can correct me. But, there is. A lot of people think like
oh you know different races, like you are more likely
to classify someone as black if you are black. You are more likely
to classify someone as Latino if you are Latino. And there is lots of
speculation about why this is. For the most part in
psychological experiments, we do not have good data
on whether that is true. And in the surveys where
we do have interviewers who are of different races. Unfortunately, we cannot
disentangle whether black people are more likely to,
people who identify as black are more likely to
classify other people as black because they are more likely
to be assigned in neighborhoods that are predominantly black. So, we have not. This is another example
of if we start thinking about race difficulty, we would
setup our studies differently to account for these
kinds of factors. But that may not have been
where you were heading, so correct me if I am wrong.>>One thing I was wondering was
if like black people stereotyped as much as white people did
on that specific subject?>>I do not know of any evidence
that suggests that they do. There is some. There is reason to believe
that what all of our studies, psychology and mine, show you
is what white people think because they. I mean, survey interviewers,
the majority of survey interviewers
are middle aged, middle class, white women. So we can generalize the
most easily to that group. We are pretty sure that that, that this is how
they classify people. and given that middle aged,
middle class, white women tend to be teachers and human
resource managers and the kinds of people who are gate
keepers in the institutions that we all navigate how they
classify people is actually really important. But we do not know. We do not have a
lot of good data on whether those
patterns generalize. Yes. [ Inaudible ] So let me, I am going to tell
you what I think you said and you can tell
me if I am wrong. So, I think what
you are asking is if you know how someone
identifies themselves does that change these results?>>Yeah kind of.>>Okay. I will tell
you the answer to that and you can follow-up
if you need to. So what is interesting. So in the survey where the
interviewers classified them every year for 19 years — only
in the first year did they hear, did they directly ask the
person who they identified. So only in the first year would
they have first heard oh I identify as Hispanic or I
identify as African American or I identify as Chinese
before they classified them. In all the other years, they would not have heard
explicitly what the person said. Now of course there
is lots of ways that you might guess
someone’s race. The other people
in their household. What their appearance
looks like, right? So it is very likely that the
interviewer was using other cues other than how they
self identified in all those other years. And we did not find that the
patterns were different in terms of linking race and status when the interviewer
also heard what the, how the person identified. So, we do not have
great data on it. But the data that
we have suggests that interviewers
are perfectly happy to ignore what you
say about yourself. Like, they have an idea of
what category you belong in. And their instructions
say go by observation. Do not go by what the
other person says. So they are perfectly
willing, you know even if you said I identify
as Latino, they might be like nope black. Nope white. And so there is a
lot of evidence that suggests that yeah we do. I mean overall we tend to take
other people’s word for it. But if we think that someone — Rachel Dolezal is perhaps
a good example of this — that if we think someone is
claiming something they should not, we are happy to fix it for
them [inaudible] I should say. Yes.>>Does geographical origin of
like a person play a role in or determine their like
classification [inaudible]?>>So, like the survey
interviewer or the person they
are classifying?>>Both.>>That is a good question. So like do you mean
if I am in the South, do I classify people
differently. So we are in California, do
we classify people differently than someone in Arkansas? I am going to say this a lot
and it is going to sound boring. But we do not have good
data on that [laughter]. So, we can do a little bit
with the survey that I have. But it is 12 thousand people. And once you start then trying
to break it down by geography, you get fewer and fewer people
with fewer and fewer changes. So you do not have a lot of
statistical leverage on it. But one of the things that we
did notice is that in the South, in the [inaudible]
interviewers were more likely to just use white and black. They were much less
likely to classify people as other just in general. So there are, there do seem
to be regional differences in that sense, like what
categories are salient to people. I think that is definitely true. More nuance than that,
we need better studies and you guys can do them. That is why you are here. Well it is one of the
reasons you are here.>>I think that is actually
the last question [inaudible] a great note to finish on. So thank you very much.>>Thank you all for coming. [ Applause ] [ Music ]

Leave a Reply

Your email address will not be published. Required fields are marked *