1 1 UNITED STATES DISTRICT COURT 2 FOR THE EASTERN DISTRICT OF MICHIGAN 3 SOUTHERN DIVISION 4 5 BARBARA GRUTTER, 6 For herself and all others 7 Similarly situated -- 8 Plaintiff, 9 -v- Case Number: 97-CV-75928 10 LEE BOLLINGER, JEFFREY LEHMAN, 11 DENNIS SHIELDS, and REGENTS OF 12 THE UNIVERSITY OF MICHIGAN, 13 Defendants, 14 And 15 KIMBERLY JAMES, et al., 16 Intervening Defendants. 17 ------------------------------------/ VOLUME 10 18 BENCH TRIAL 19 BEFORE THE HONORABLE BERNARD A. FRIEDMAN 20 United States District Judge 21 238 U.S. Courthouse & Federal Building 22 231 Lafayette Boulevard West 23 Detroit, Michigan 24 THURSDAY, FEBRUARY 8, 2001 25 2 1 APPEARANCES: 2 3 FOR PLAINTIFF: Kirk O. Kolbo, Esq. 4 R. Lawrence Purdy, Esq. 5 6 FOR DEFENDANTS: John Payton, Esq. 7 Craig Goldblatt, Esq. 8 On behalf of Defendants. 9 10 George B. Washington, Esq. 11 Miranda K.S. Massie, Esq. 12 On behalf of Intervening Defendants. 13 14 COURT REPORTER: Joan L. Morgan, CSR 15 Official Court Reporter 16 17 18 19 20 21 22 23 24 25 3 1 I N D E X 2 3 WITNESS: PAGE: 4 WALTER ALLEN 5 Cross Examination by Mr. Kolbo 5 6 Re-Direct Examination by Ms. Massie 110 7 8 EUGENE GARCIA 9 Direct Examination by Mr. Washington 125 10 11 ERIC FONER 12 Direct Examination by Ms. Massie 13 Cross Examination by Mr. Richter 14 Cross Examination by Mr. Payton 15 16 E X H I B I T S 17 MARKED RECEIVED 18 19 20 21 22 23 24 25 4 1 Detroit, Michigan 2 Thursday, February 8, 2001 3 (At or about 9:10 a.m.) 4 -- --- -- 5 THE COURT: You may be seated. Thank you. Good 6 morning. 7 MS. MASSIE: Judge, a rescheduling thing. I think 8 particularly with Professor Allen's cross exam starting now, 9 we should plan on starting David White tomorrow morning, 10 go from Professor Allen into Dean Garcia of the Berkley 11 Graduate School of Education, and then into Professor Eric 12 Foner today, and I think that will pretty much take up the 13 day. 14 THE COURT: And tomorrow is who? 15 MS. MASSIE: Tomorrow we start with David White, 16 who will testify about race and the LSAT. 17 We also have Rick Lempert who has done a study of 18 minority and white grads of the U of M Law School, and 19 Frank Woo, who is an expert on Asian Americans and 20 affirmative action. 21 THE COURT: Great. Okay. 22 Good morning, Mr. Kolbo. 23 MR. KOLBO: Good morning, Your Honor. 24 Good morning, Dr. Allen. 25 THE WITNESS: Good morning, Attorney Kolbo. 5 1 CROSS EXAMINATION 2 BY MR. KOLBO: 3 Q We have met before? 4 A Yes. 5 Q I want to start out by asking you some questions about 6 the methodology of the focus groups that you assembled at 7 Michigan and elsewhere. 8 First of all, there are -- among the schools that 9 you looked at, three of them are quite large institutions; 10 correct? 11 A That's true. 12 Q University of Michigan, Michigan State, and University 13 of California Berkley are quite large undergraduate 14 institutions, is that correct? 15 A All the institutions are sizable, yes. 16 Q And correct me if I'm wrong, but my understanding is 17 that Michigan, the undergraduate school, the principal 18 undergraduate school has about 20,000 students or so, 19 does that sound about right? 20 A I would have to check the figures, but I can take your 21 word for it. 22 Q Okay. And Berkley has about just a few more, between 23 20,000 and 23,000 students, roughly, undergraduate, would 24 that be correct? 25 A Probably seems appropriate. 6 1 Q Michigan State a little larger, around 30,000-some 2 students at their undergrad school, seem fair? 3 A I haven't studied the figures, so I'll have to trust 4 your figures. 5 Q Harvard, around 6,000 or 7,000 students, undergraduate 6 school. If I'm right, it looks like you were conducting 7 focus groups at institutions that had, at the undergrad, 8 these undergraduate institutions, between 75,000 and 80,000 9 students, would that be -- does that sound reasonable? 10 A As I said, I have not looked at those figures, so 11 I'll have to trust your numbers. 12 THE COURT: Your mike may not be on. Everybody 13 came, they might as well be able to hear us. 14 BY MR. KOLBO: 15 Q Okay. About 75,000 to 80,000 students, if I'm right, 16 in terms of the number of students at these undergraduate 17 institutions, and am I correct that the number of students 18 in your focus groups all together, for all of the focus 19 groups together at the undergraduate level, you looked -- 20 you had 68 students; is that right? 21 A We had 68 students in the focus groups and we had 22 another 200 surveys from those campuses, and as I had 23 mentioned yesterday, in the past I have done research that 24 included 5,000 African American students randomly sampled 25 nationally, and in the upper midwest another study of 7 1 2,500 students sampled nationally. 2 Q I was asking about your focus group work. 3 A Well, in terms of the focus group numbers, that's a 4 correct number. 5 Q That's correct, right, 68? 6 A But I was trying to give you some context. 7 Q Sure. I appreciate that, but I want to focus on some 8 of these questions and then we will talk about some of these 9 other issues perhaps later, but 68 students in the focus 10 group, and you said 200 surveys, right? 11 A That's correct. 12 Q And did that include the 68 students? 13 A Yes, it did. 14 Q So another 130 or so students who filled out surveys 15 and did not attend focus groups? 16 A That's correct. 17 Q And about half of the focus groups at the undergrad 18 level were University of Michigan campus, right? 19 A Purposely so, yes. 20 Q So about 30-some students out of 125,000 were included 21 in the focus groups at the University of Michigan undergrad 22 school, true? 23 A That's true. 24 Q And of the remaining, another 30 or 40 students were 25 in the focus groups out of the approximately 40,000 to 8 1 50,000 at these other undergraduate institutions; correct? 2 A Yes. 3 Q Two focus groups at Harvard? 4 A That's correct. 5 Q About a dozen students or so? 6 A Yes. 7 Q Out of 6,000 or 7,000, true? 8 A The 6,000 or 7,000, as I said, I'll trust you for that 9 number. I'll have to again check it in order to, you know, 10 assert that it's a correct figure, but the figure for the 11 focus groups sounds like it's on target. 12 Q About a dozen out of the entire Harvard undergrad 13 campus at the focus groups? 14 A That's probably correct. 15 Q And you had three focus groups at Berkley, so that's 16 about twelve to fifteen students at that campus? 17 A I would have to look at the numbers, because those 18 numbers sound a bit lower for me. Do you want me to give 19 you an accurate count on the focus groups at Berkley? 20 Q I think we have -- - is it report page seven, am I 21 right, following page seven? I guess we can be specific 22 about this. 23 A I'll have you help me. 24 THE COURT: Page 7 in Exhibit 157? 25 MR. KOLBO: It's actually right behind figure two. 9 1 It's an unnumbered page. 2 BY MR. KOLBO: 3 Q Now, these we don't have the identification of the 4 school here, do we? 5 A No. 6 Q But maybe you know what they are. 7 A I do. 8 Q And so from Michigan State, we have one focus group 9 from Michigan State, and which one is that? 10 A Correct. The one focus group from Michigan State -- 11 just a second, I'll check my files. 12 The Michigan State focus groups are twelve. 13 Q Number twelve? 14 A Yes. 15 Q So that's two people? 16 A Yes. 17 Q Two people out of whatever the size of that campus is, 18 consisted of the entire focus group at the Michigan State 19 University? 20 A And I should add we ended up collecting a larger 21 number of surveys from Michigan State. 22 Q Okay. They are part of the 200? 23 A Yes, they are. 24 Q But for the focus group, just two students? 25 A Correct. 10 1 Q One man and one woman? 2 A Correct. 3 Q And then at Berkley you had three focus groups. 4 Which three are those? 5 A The three focus groups at Berkley, the group was 6 number ten, number six and number four. 7 Q Okay. So number four has five people in it? 8 A Correct. 9 Q All African American? 10 A Yes. 11 Q Number six has seven people in it, correct, all 12 Latino? 13 A All Latino. 14 And by the way, they were purposely selected to be 15 restricted to those groups. 16 Q Sure. And I'll ask you about that, but so far, twelve 17 students so far; correct? 18 A That's right. 19 Q And then the third one has got six in it, so we have 20 got -- and I hate doing math up here publicly, but 18? 21 A Sounds correct. 22 Q So 18 students were in the focus groups at the 23 University of California at Berkley? 24 A Sounds correct. 25 Q Out of the entire undergrad campus there? 11 1 A That's correct. 2 Q And two were at Harvard. Which two were those? 3 A The two Harvard focus groups are number three and 4 number seven. 5 Q Okay. Number three had six people in it, all African 6 American, and number seven had five in it? 7 A Number three was an African American focus group and 8 similarly with number seven it was a Latino focus group. 9 Q So a total of 11 students were involved in the Harvard 10 focus group; correct? 11 A Correct. 12 Q The average size of these groups, it looks like it's 13 five to six students; true? 14 A Average size of a focus group that's effective has 15 to be between five and eight, no more than ten students, 16 otherwise, you don't have the discussion. 17 Q And at the -- I'm going to try to talk about these 18 kind of separately, but while we're talking about numbers, 19 the total number in addition to those undergraduate focus 20 groups, you had some law school focus groups, as well? 21 A That's correct. 22 Q Those were all at the University of Michigan Law 23 School? 24 A Exactly. 25 Q There were a total, as I understand it, of 31 students 12 1 on the figure behind page 60, I think, of your report? 2 A That is a correct number, I think. It sounds correct. 3 Q And that's 30 out of, I don't know, what is it, about 4 1,000 students all together at the University of Michigan 5 Law School? 6 A Again, that's a figure I don't necessarily remember. 7 Q Now, and you mentioned, you just testified that these 8 were intentionally by and large racially segregated focus 9 groups, correct, that was by design? 10 A I'm pausing because of the term of segregated focus 11 group. These were targeted focus groups, that's the term 12 that I would prefer, because segregation for me just 13 suggests the old problematic heritage of legal exclusion, 14 and in fact, we did focus groups across each of the racial 15 groups and each of the ethnic groups and each of the gender 16 groups and in fact were trying to standardize the groups, 17 which is a statistical procedure; that is, you try to 18 control for variation in terms of other external 19 characteristics, so thus we targeted the groups and 20 organized them in terms of, in some instances, in most 21 of the instances, in terms of shared racial identity. 22 But you are aware, I'm sure, that there were several 23 of the groups that were purposely mixed race and mixed 24 ethnicity. 25 Q But they weren't -- you mentioned the word legally 13 1 segregated after, and of course I didn't mean that, but 2 they were intentionally or purposively segregated by race, 3 correct, as part of the design of this case? 4 A And again, I object to the term, segregated. 5 They were a targeted group. They were a group that was 6 standardized and they were a group that was assembled with 7 the idea and goal of assembling them around those common 8 identities. 9 Q Okay. Well, I do understand you're in favor of, 10 you believe we ought to have integrated racial campuses, 11 correct? 12 A Absolutely. 13 Q Would you agree with me? 14 A And an integrated society, and the two are linked, if 15 I might. 16 Q And would you agree with me that most of these focus 17 groups were not racially integrated, were they? 18 A And as I have spent some time explaining, anticipating 19 that perhaps such an incorrect assessment might be made, 20 the groups were, as previously mentioned in several cases, 21 specifically targeted and organized around common or shared 22 racial or ethnic identity, in one case, in another case or 23 category around shared gender identity, but as you do recall 24 from looking at the data and as I said a moment ago, some of 25 those groups were purposely mixed, but they were mixed in 14 1 terms of race and ethnicity, but with the members, for 2 example, sharing common gender status. 3 Q Okay. I won't use the word in my next question, 4 integrated or segregated, but only two of the twelve 5 groups were racially mixed; correct? 6 A That's probably correct. 7 Q And only one of the groups was all white; correct? 8 A That's correct. Well, no, no, that's not correct. 9 You mean -- which are you talking about, are you talking 10 about the law school or the undergrad? 11 Q I'm sorry, I shouldn't go back and forth. I was 12 looking at the figure for the undergrad schools, only one 13 of those twelve was -- 14 A One of the twelve for the undergrad. 15 Q And while we are on the subject, I guess -- 16 A And one for the law schools. 17 Q Okay. So one of each? 18 A Yes. 19 Q Each group was all white. Do you think there would 20 have been some benefit to have seen what the interaction 21 might have been with more mixing of the races in these focus 22 groups? 23 A Obviously, the more information you get the better, 24 but given as I mentioned yesterday, limitations in terms of 25 resources, limitations in terms of time and the specific 15 1 focus of this research, that goal was one that was of lesser 2 priority than the advantages that would accrue from having 3 students in a group of other students who are like 4 identified and then benefitting from their feeling free to 5 discuss the kinds of issues that are of importance and note 6 to them. 7 And I would -- as an aside to that, of course, I had 8 taught at the University of Michigan for some ten years, I 9 have been teaching at UCLA for another twelve, and I have 10 been teaching for twenty-five years plus, and needless to 11 say, my classes have had exactly the kind of mix that you 12 describe, and so I have learned some lessons from that 13 mixture, as well, and from the interaction of the sort 14 you're referring to. 15 Q One of the things you just mentioned was that, as I 16 understand, that one of the reasons you chose to organize 17 these groups the way you did was that the students would 18 feel free to say what they are thinking? 19 A Precisely. 20 Q And is your opinion then that the students would feel 21 free to say what they are thinking in groups that were not 22 racially mixed? 23 A I definitely think that's true. 24 Q How about if you had a racial group that was half 25 and half, half minority, half white, say, would, in your 16 1 opinion, would students of all, both races, feel free to 2 express what's on their mind in that situation? 3 A I think as I have pointed out yesterday, if you 4 move to a situation where you have closer to equal 5 representation, then indeed you would likely get some 6 discourse and exchange between equals, but I don't know 7 that I'm following the point that you're making. 8 Q Well, I'm not -- you know, I'm not asking you to 9 accept any point that I'm making, but I'm just wondering 10 whether students would, in your opinion, would feel free 11 to speak their minds as you don't think they would be in 12 some circumstances, as long as, say, there was equal 13 representation of the group? 14 A I think they would feel free to say some things, but 15 there are some utterances that no matter what the mix in the 16 room, I mean, in terms of the proportionality in terms of 17 the two racial groups, for example, black and white, that 18 students would not feel free to say, compared to the sense 19 of liberty that they would experience in, say, an all white 20 group, and it's, I think, a fairly straightforward reality, 21 because if we think about our lives, if you think about your 22 life, I mean, we talk differently in -- depending on the 23 racial group that we are a member of. 24 I mean, if we are in a homogeneous racial group, 25 then some of our comments about people of other races are 17 1 different than what they might be, than what they would 2 likely be in the presence of people of that race, and 3 similarly, the point holds with gender. 4 Q You mentioned something about an all white group. 5 What about if you had an all black group, does that make 6 the discussion freer or less? 7 A Absolutely. That's exactly the point I'm trying to 8 make. 9 Q How about a black group that has a few white students, 10 would black students feel, in your opinion, free to express 11 themselves then? 12 A I would have to think about that a bit and I would 13 have to look at some research that has looked at questions 14 of proportionality and how that influences, either inhibits 15 or enhances interaction, so I'm not quite sure how to answer 16 that question. 17 Q What it sounds like is your opinion, but correct me if 18 I'm wrong, that mixing the races to any extent, there's some 19 inhibiting of people of different races feeling free to 20 express what is on their mind; is that fair summary? 21 A That's very fair in this society, because as you know, 22 race talk is very difficult for us as a society, and it 23 becomes even more difficult when we have other racial group 24 members in the room, but needless to say, those are lessons 25 we need to learn, and to the extent that you look at work 18 1 of someone like Silveri Atado, and you see the sort of 2 variations in exchanges that occur as a result of the mix 3 of racial proportions in a room, then you come away feeling 4 encouraged, if you will, that as students do have a way of 5 working through these issues, but those tensions are there, 6 because race talk, race relations in this society are very 7 difficult. And challenging is the word I should use. 8 Q Let me move on a little bit here. 9 You testified yesterday about the differences 10 between qualitative and quantitative analysis; correct? 11 A Methodology analysis, yes, I did. 12 Q There are differences between those two, there are 13 strengths and weaknesses of both? 14 A Correct. 15 Q What you did with respect to your focus group work is 16 qualitative research; correct? 17 A That's correct. 18 Q Not quantitative; correct? 19 A Correct. 20 Q And I just want to make sure I understand some of the 21 differences, and maybe some of the strengths and weaknesses, 22 as well. 23 Do I understand that with a quantitative -- well, 24 first of all, one type of quantitative analysis is what is 25 called, and I think you referred to this term before, is 19 1 probabilistic studies, right, if you're familiar with that 2 term? 3 A I am. 4 Q Okay. I mean, I think in your deposition I think you 5 mention it a few times. 6 A Okay. I responded to your mention of it. It's 7 actually not a term that we use, but I understood what 8 you were referring to. 9 Q In terms of probabilistic, a probabilistic study is a 10 form of a quantitative assessment; correct? 11 A Studies that are probabilistic and based on 12 probability are normally referred to as inferential 13 statistics. 14 Q Okay. 15 A And so just very quickly, when you start talking about 16 using an inferential methodology, essentially what you are 17 posing is that the sample on which you base your analysis 18 is a representative sample, it's a randomly selected 19 sample, it's a sample that can be said to represent the 20 characteristics accurately or within some degree of error 21 of the total population, and so thus, this notion of using 22 an inferential technique and inferential statistics. 23 Q You said that a lot better than I could have asked 24 the question, so I appreciate that. 25 A Okay. Just trying to be helpful. 20 1 Q What you have described is not what you would do with 2 respect to focus groups, your focus groups, correct? 3 A What I describe is one of the many methodologies that 4 I have used in my research for years, and in fact, if I 5 can elaborate and give you a sense of anticipating your 6 question about strengths and weaknesses of areas -- 7 Q Can I just -- I would like to sort of do this, 8 otherwise I lose track of my thoughts here -- 9 A Okay. 10 Q -- and get the answers to the questions that I have 11 got, first of all, here, and your -- 12 A I apologize. 13 Q And Ms. Massie will have a chance to be asking you 14 some of these questions, as well. 15 A I'm trying to be responsive. 16 Q You described, I think, very -- at least to my 17 perfect understanding what a probabilistic study is and your 18 description of that contains some of the strengths that are 19 associated with quantitative research; correct? 20 A Correct. 21 Q And the characteristics that you have described 22 with respect to this, to a probabilistic or inferential 23 statistics study, those aren't the characteristics that are 24 applicable to the focus group that you did, correct, just 25 the focus group work? 21 1 A That is a correct comment. 2 Q One of the things that a probabilistic study does or 3 an inferential statistics study does is you get -- you look 4 for and it's important to have a random sample of the group 5 that you're studying; correct? 6 A This is correct. 7 Q You don't have that in your focus group work; correct? 8 I think you have acknowledged that. 9 A Focus groups never have random selection in terms of 10 an inferential statistics approach, because it's simply not 11 an appropriate selection procedure to match with focus group 12 methodology. 13 Focus group methodology by its very nature and 14 purpose is intended to focus on a specific small defined 15 population, and then to, if you will, exchange the sort 16 of broad generality you can get from surveys for the rich, 17 detailed content that only comes forward from smaller group 18 discussions or from intensive individual interviews. 19 Q So focus groups don't depend upon randomness; correct? 20 A Not at all. 21 Q They don't depend on a representative sample; correct? 22 A The representativeness of a focus group comes in 23 exactly the point you were asking about earlier; that is, 24 you want focus groups that reflect or incorporate key 25 dimensions of variation within the population, so thus the 22 1 rationale for an African American or a Latina/Latino or a 2 white focus group. 3 Q But representativeness has a -- that's a term of art 4 in inferential statistics; correct? 5 A I'm not sure I'm following that question. 6 Q Well, don't you understand that representativeness 7 of a sample is something that's important in inferential 8 statistics? 9 A For inferential statistics, yes. 10 Q And as that term is used and as it's important in 11 inferential statistics, you don't need that, you don't 12 look for that, in focus group work; correct? 13 A No, you have to have -- well, maybe I'm not 14 understanding your question, but in focus group research, 15 in qualitative research, you have sort of planned and 16 programmed variability, so you do have to have some 17 representativeness, but in terms of it being mathematically 18 calculable, no. 19 Q And I won't belabor this, but just so I understand it, 20 an example of a quantitative inferential statistics analysis 21 that everybody is kind of familiar with is polling data, for 22 example, right? 23 A That's correct. 24 Q We just came through a long presidential election, we 25 all saw a lot of polls, and it's pretty commonly understood 23 1 that one can take a sample of, say, 1,000 American voters 2 and from that 1,000 draw, through inferential statistics, 3 some valid conclusions about how people want to vote with 4 some margin of error that is accurate with respect to the 5 entire population; correct? 6 A It's accurate in the most general sense, correct, and 7 as you said, depending on the degree of -- with some degree 8 of error. 9 Q And just to be -- I want to make sure I'm 10 understanding this, one can draw statistically valid and 11 reliable conclusions from that group of 1,000 to the 12 larger population of, say, 80 million or 100 million, within 13 a margin of error? 14 A But only certain kinds of conclusions, so I think 15 that is a correct point. 16 Maybe it would help if I just made a fairly simple 17 distinction between quantitative and qualitative statistics, 18 which is a discussion that we're having now, and the 19 quantitative approach is more of a snapshot, so it's a 20 snapshot of a larger whole, if you will, but what you get 21 with qualitative methodology is something that's akin to a 22 video. You get more of the process, so that's why, in fact, 23 the field is moving toward using the two methodologies and 24 perspectives in sort of congruence or in concert. 25 Q Okay. But I also understand, am I right, that while 24 1 with quantitative analysis you can draw these statistically 2 valid and reliable conclusions from the sample to the larger 3 population, you can't draw those statistically valid and 4 reliable conclusions from the small sample of a focus group 5 to the larger population, correct, that's the weakness of 6 focus groups? 7 A The weakness of focus groups is that you don't have 8 an estimate of the probability to which the conclusions 9 represent the whole. 10 Q Right. You can't draw -- 11 A And thus, you rely on survey data for that purpose. 12 Q Right. In focus group work you can't draw and rely 13 upon statistically valid and reliable conclusions from the 14 small focus group sessions to the larger population being 15 studied, right, in terms of the statistical reliability 16 of validity, is that a fair statement? 17 A I'm pausing, because really the distinction is a 18 mathematical one, and so the lessons that you learn from 19 inferential statistics or the lessons that you learn from 20 what you referred to as probabilistic studies, even though 21 they may be wrong or incorrect, you can say the degree to 22 which that particular finding can be presumed to represent 23 what you would find -- 24 Q Sure. 25 A -- within the larger population, and I say, even 25 1 though it may be incorrect, because often you're using 2 the wrong questions or you are using faulty instruments, 3 so I'm saying that it's not a substantive determination so 4 much as it is a mathematical or a statistical one, so. 5 Q Right. 6 A It's a matter of it not necessarily being truth, 7 that's what I'm trying to make sure we don't get -- 8 Q There could be flaws in a question, I suppose, but as 9 to statistics and mathematics, one can draw these valid and 10 reliable conclusions to the larger population? 11 A You can, and being cautious that you may in fact be 12 using the wrong instruments and asking the wrong questions 13 and not understanding the process. 14 Q Okay. And you don't do that with focus groups, that's 15 just one of the features of focus groups? 16 A Right. 17 Q Okay. And you have mentioned that your focus group 18 work was purposive; right? 19 A The sampling was purposive. 20 Q And purposive meaning you had certain predetermined 21 criteria that you were relying upon in the selection of your 22 sample, of your focus group sample? 23 A And I have actually shared those criteria. The point 24 is that what we wanted to do is to select from the pool of 25 African American students who said they were interested in 26 1 serving in a focus group, and by the way, that selection was 2 random, but the original pool was not a random pool, but 3 having gotten that pool, then, we randomly selected and 4 assigned people to the two African American focus groups, 5 let's say. 6 Q And my understanding is -- 7 A Excuse me for interrupting. 8 So the purposive dimension is really quite simple. 9 It was in terms of wanting to have a sample of a number of 10 focus groups that were homogeneous in terms of race, and 11 in the second case a number of focus groups that were 12 homogenous in terms of gender. 13 Q But within that, then, you were actually looking for 14 students from particular races? 15 A Yes. 16 Q That's part of the purposive element of this study? 17 A Yes, particular races and particular gender. 18 Q You didn't consider doing a probabilistic study here, 19 did you, or an inferential statistics focus group? 20 A At the University of Michigan? 21 Q Right. 22 A I have done ten years of probabilistic research at the 23 University of Michigan and found it inadequate for reasons 24 we have been discussing. I kept finding these correlations, 25 but I had no handle on the underlying dynamics that would 27 1 produce a correlation, for example, that translated into 2 students of color being less satisfied with their 3 educational experience and students of color having lower 4 GPA's, students of color having higher rates of alienation 5 and feelings of isolation, so thus the need to match up with 6 that that data from qualitative research, because as you 7 know, the inferential research can only go so far, it can 8 just -- so for those ten, twenty years, I have been looking 9 at and examining correlations, and so, thus came the need to 10 try to dig deeper into the relationships and understand the 11 movie part, if you will, what the dynamics are and what the 12 process is. 13 Q And the quantitative work that you have done in past 14 years doesn't answer any of those explanatory questions 15 about what's going on with respect to minorities and grades 16 and discrimination; correct? 17 A It answers those questions, but only to the level of 18 general patterns and relationships between variables, so 19 again, I could do all the analysis and indeed did the 20 analysis using sophisticated quantitative techniques so that 21 you can look at basic correlations between major variables 22 and then break those down, but then ultimately what is 23 missing to a sizable extent is the human voice and the 24 human experience and the sort of dynamic dimension. 25 So I'm simply answering a qualified yes to your 28 1 question. I mean, the surveys could answer certain kinds 2 of questions, but they could not answer a question, for 3 example, about the actual human toll of individuals and how 4 they feel on a daily basis and it could not extract the 5 specific instances of racial discrimination, if you will, 6 or of gender harassment. 7 Q And for that information, we have your focus groups? 8 A Correct. 9 Q Now, you're familiar with the concept, and I think 10 this might have been touched upon yesterday, but you're 11 familiar with the concept of selection bias? 12 A Yes, I am. 13 Q Selection bias is a bad thing? 14 A In inferential statistics, and as you say, 15 probabilistic studies, yes. 16 Q Isn't selection bias also bad for focus group work? 17 A Absolutely not, because it's purposive. 18 Q So you wouldn't mind having a biased focus group 19 designed -- a bias in the design of your focus group? 20 A Let me back up. In terms of selection bias, I should 21 have asked you what you meant by the term, but as I heard 22 the question, it seemed to me to be relinking back to this 23 concern with kind of representativeness, so I answered in 24 that frame. Representativeness is not the same goal with 25 qualitative research as it is with quantitative research. 29 1 Now, obviously, bias itself is problematic or a 2 sort of selection bias that would, for example, fill a 3 focus group with people who are biased on the question. 4 Q Okay. 5 A It could be problematic. It may well be that if 6 you're trying to understand prejudice, you want to have a 7 focus group filled with people who are very prejudiced. 8 Q Okay. And I think I understand where we're diverging 9 here. If you use the word, bias, in a very broad sense, as 10 sort of a neutral sense, your focus group work is biased 11 because you're biased toward finding certain ethnic groups, 12 you know, you want certain ethnic groups, that's the only 13 sense in which that bias occurs; correct? 14 A See, I hesitate to accept that, that's why I tried 15 to back off and reframe, because again, the selection 16 criteria are not biased, they are built into the focus 17 group structure, and I imagine it would become biased if, 18 for example, you left out an important or a major racial 19 group, but even there it just depends on the purpose of 20 the research. 21 If you are trying to understand how women experience 22 gender hostility, then a very reasonable case can be made 23 for studying women and only studying women. 24 Q And you have anticipated one of my questions, which 25 is, if you're studying an issue, how people feel about an 30 1 issue, it would be a bad form of selection bias to design 2 the selection of the group or the focus group so that the 3 individuals were skewed in their views one way or the 4 other on the issue being studied; is that a fair statement? 5 A I think it could be potentially problematic. 6 Q Wouldn't that be -- 7 A It wouldn't be necessarily fatal. It depends on the 8 skill of the person who is facilitating the focus group, and 9 again, it depends on your goals. 10 Q Wouldn't you agree that it would be a perversion of 11 the process to try to skew the sample on the issues before 12 it in a focus group? 13 A I'll try to repeat a point I have just made. There 14 are times when your goal is to include in a focus group, 15 if you will, a certain point of view, a certain set of 16 experiences, and so in that sense it's not at all a 17 perversion, but it's consistent with what you're trying to, 18 if you will, standardize, so if you're doing a study of 19 individuals who have experienced racial profiling in terms 20 of unwarranted stops by traffic cops, then it may well be 21 that you want to include in that focus group structure, 22 instead of students, people who have had that experience. 23 Q Let me ask it this way: If you're studying the issue 24 of whether affirmative action or the taking of race into 25 account in the admissions process is something that should 31 1 be maintained, would it skew the selection of the focus 2 group to do anything in the design to try to get students 3 who have expressed a particular point of view on that 4 subject? 5 A I think that our judgment was that, certainly -- 6 Q It's just a general question, not specifically about 7 this case. 8 MS. MASSIE: I would just ask you to let the witness 9 finish his answer. 10 THE COURT: Well, he can finish the answer, but I 11 think he has limited it generally as opposed to this case. 12 THE WITNESS: Our judgment -- 13 MS. MASSIE: I'm just asking -- excuse me for 14 interrupting, Professor Allen. I'm just asking that the 15 witness not be interrupted by Mr. Kolbo. This happened 16 repeatedly at his deposition. 17 THE COURT: Well, first of all, make your 18 objections. I don't know what happened at the deposition. 19 Make your objections. We're at trial today, and if you have 20 an objection, make it. 21 MS. MASSIE: I'm objecting to the interruption 22 of the witness. 23 THE COURT: Overruled. 24 Go on, but phrase your question the way you want it 25 answered. 32 1 BY MR. KOLBO: 2 Q Would it be fair in a general sense, Dr. Allen, if 3 you're studying, if you want to study whether or not 4 affirmative action is something that should be used in 5 college or law school admissions, it would be a perversion 6 of the process to try to skew the sample, focus group 7 sample, in one direction or the other on that, on how 8 students feel about that issue, fair enough? 9 A I think that's a fair enough statement. I find the 10 reference to a perversion of the process, skewing -- but 11 generally, I guess that's a reasonable position. 12 Q Well, you just mentioned you gave a deposition in this 13 case; correct? 14 A I did. 15 Q You have a copy. I think we placed a copy of your 16 deposition up there. 17 A I see it. 18 Q Could you go to page 68 of the transcript, line three? 19 A Page 68, line three? 20 Q Yes. 21 A Yes, sir, I have it. 22 Q And feel free to read the context here, but I'm going 23 to ask -- I'm going to read a question. You, I think, were 24 clear enough for myself in the answer, but if you want to 25 take a look at the question, the question on page 68, line 33 1 number three: "And would you agree it would be a 2 perversion of the process to try to 3 skew that sample or that group of 4 students in one direction or the 5 other; correct?" 6 Answer: "Exactly." 7 I mean, was that your answer? 8 A That's my answer, and as you pointed out, it's in 9 a context. We had had a discussion or you had asked me 10 several questions about the recruitment procedures prior 11 to that on page 67 and I indicated that there was a very 12 broad recruitment strategy where we used e-mails, we used 13 newspaper articles, we used petitions on the street, visits 14 to classes, and so in that context of the discussion, as we 15 came to this question, most certainly I agreed with you, 16 because I had sort of gone to great pains to demonstrate 17 that, in fact, we took a number of explicit approaches and 18 strategies to avoid exactly such skewing and distortion. 19 Q Would it, in the design of this study, these focus 20 groups, would it have been improper in selecting the 21 students for the focus groups to try to get students who 22 have a view that affirmative action ought to be maintained 23 and to try to skew the sample in favor of that direction, 24 would that be proper? 25 A I think it would not have -- basically, each one of 34 1 our communications indicated that we wanted to have a broad 2 representation of perspectives, and so in that sense it 3 would be no more improper to include students who were 4 supportive of affirmative action than it would be to include 5 those who objected to affirmative action, and indeed, if you 6 look at the transcripts, you see we have clear evidence that 7 we had balance in that respect; that is, people voicing 8 support and others voicing either reservations, if not 9 strong objections to it. 10 Q But my question is, would it be improper in the 11 design, in actually going out and trying to assemble these 12 focus groups, to go out and try to skew the collection of 13 students with students who have a view in one direction; 14 that is, that they favor the use of affirmative action, 15 would that be improper if it were done? 16 A If, in fact, you're asking me would it be improper 17 to just stack the deck and only include students who were 18 supportive of affirmative action, I would say yes, I agree 19 with that. 20 Q Well, that wasn't exactly the question, but would it 21 be improper as part of the design of this group to go out 22 there and to try to assemble a group of students primarily 23 who were proponents of affirmative action? I'm not asking 24 you to admit that you did it here, but would that be 25 improper, if it was done that way? 35 1 A If we're still in the general case, it would not be 2 at all improper if you are trying to study students who are 3 supporters of affirmative action and you want to understand 4 the dynamic, the sort of life course that brought them to 5 the point of supporting affirmative action. 6 Similarly, if you wanted to understand how it is 7 that students came to object to or to not favor affirmative 8 action, then again, it would make sense to assemble a group 9 of students who did not support affirmative action, if 10 that's the focus group. 11 Now, if you're doing survey research, of course, 12 you want a broader representation. 13 Q Backing up a little bit, you gave me an answer a 14 question or two ago that suggested that -- suggested and I 15 want to clarify this -- that in the groups that you actually 16 assembled you got a wide difference, you got widely 17 different views on whether affirmative action is something 18 that ought to be continued; is that -- 19 A Absolutely did. 20 Q So you got a -- was there a significant number of 21 students who participated in these focus groups that thought 22 that affirmative action ought to be discontinued on college 23 campuses and law schools? 24 A No, there were not, but as my survey data demonstrated 25 both in terms of the surveys for this study, in conjunction 36 1 with Grutter, as well as national survey data of my own and 2 national survey data collected by others, that there is a 3 majority support for affirmative action among college 4 students. 5 Q What percentage? Do you have some idea what 6 percentage? 7 A I'd have to look up those statistics, to tell you 8 the truth. 9 Q Well, I'm talking now about focus groups. 10 A Oh, I'm sorry. 11 Q So what percentage if your focus group respondents, 12 roughly speaking, were opposed to the use of race as a basis 13 for admission, affirmative action? 14 A I actually can't tell you that, for reasons related 15 to the discussion we have had to this point. We're talking 16 about qualitative data, and so those kinds of percentages, 17 but indeed, the surveys and the percentage of students who 18 reported that they did not support affirmative action, 19 that's summarized in survey data. That's part of this 20 report. 21 Q Okay. And were there minority students in these focus 22 groups and the surveys who indicated that they were opposed 23 to the use of affirmative action in college admissions and 24 law schools? 25 A In the surveys, a few. 37 1 Q And how about in focus groups? 2 A Some, again. 3 Q If I could ask you to turn to Exhibit 176. Do you 4 have that book up there, that volume? 5 A I don't see it. I have Exhibit 158 through 160. 6 MR. KOLBO: May I approach, Your Honor? 7 THE COURT: You may. 8 THE WITNESS: Thank you. 9 BY MR. KOLBO: 10 Q Is Exhibit 176 one of the e-mails that was sent out to 11 solicit students for the focus groups that you used at the 12 undergraduate level? 13 A Yes. 14 Q Now, you were involved in the design of some of the 15 solicitations, the e-mails and so forth that went out; 16 correct? 17 A I provided the core of the solicitation and literally 18 did not solicit the students. I purposely did not want to 19 solicit the students myself, so what I did was to provide a 20 core to just a group of students, a group of organizations, 21 that would do the recruitment. 22 Q Okay. You were involved -- 23 A Building the pool that we had talked about. Excuse me 24 for interrupting. 25 Q I'm sorry. 38 1 You were involved in the -- you had some hands-on 2 involvement in reading or at least approving some of the 3 solicitations that went out; correct? 4 A No, what I did was to -- and maybe I'm -- well, 5 I wrote and sort of agreed to a core recruitment and 6 description of the studies, so in other words, if you look 7 at each of the solicitations, you'll see that the study 8 purpose is described, the study structure, our research team 9 is identified, and the dates are there. So that general 10 information we provided, yes. 11 Q But isn't it true that you actually approved the text 12 of this particular e-mail exhibit? 13 A Which one. 14 Q Exhibit 176. 15 A Yes. 16 Q Okay. So you read it before it went out and you 17 reviewed it and you approved it; correct? 18 A With each of these, what I approved was the core 19 portion of it, so I cannot definitively say that sort of 20 the formal form is -- the final form is something that I 21 approved, but I will say to you quite, quite clearly that 22 the solicitations went out with my approval. 23 Q Well, you have had a chance now to read Exhibit 176. 24 There is nothing in there that you would have disapproved of 25 or did disapprove of? 39 1 A No, there is nothing here that would have been 2 terribly problematic for me. 3 Q Okay. 4 A Mainly because I had factored in a sort of stop gap or 5 a supplemental selection procedure of sort that I described; 6 that is, in other words, I knew that I would get a pool of 7 students from which we would select and assign to focus 8 groups. 9 Q Okay. 10 A So. 11 Q And one of the things that this e-mail informed 12 people was that: "This is a chance for students at 13 Michigan State University, in this case, 14 to contribute to a legal case that will 15 impact the educational opportunities of 16 minority students for generations to come. 17 It is our Brown versus Board of Education, 18 and we must do everything we can to 19 insure victory." 20 You approved that message going out in a selection 21 of these focus groups; correct? 22 A Sure, yes, I did. 23 Q Wouldn't you agree with me that that injects just a 24 little bit of selection bias into the way in which these 25 focus groups were designed? 40 1 A I could, possibly, yes. Possibly, yes. I hesitate, 2 because as you pointed out at the very beginning of your 3 cross examination, we're talking about two students out 4 of 68 for the Michigan State students, so any selection 5 bias that might have occurred certainly was considerably 6 diminished just by that very insignificant fraction of 7 the total. 8 Q Well, this would have gone out to whoever was 9 solicited at Michigan State University; correct? 10 A In terms of focus groups, yes. 11 Q Okay. And do you know whether this form went out to 12 other schools as well or don't you know that? 13 A No, no, I have the forms for the other schools. I 14 think you have those in 177 and 178, and indeed, I think a 15 second form went out at -- to Michigan, only Michigan State 16 campus, but unfortunately, we just didn't get very many 17 responses from Michigan State, period. 18 Q If you were to design this e-mail over again, is it 19 fair to say you would do it differently, you wouldn't have 20 this exhortation to get students who felt that we must do 21 everything we can to insure victory? 22 A I would have to think about that, because the fact of 23 the matter is that you're trying to get students to take 24 very precious time to participate in research, so you're 25 challenged in terms of interesting them, but probably that 41 1 particular phrase we may have left out, but you'll see 2 phrasing in some of the other e-mails that e-mail list 3 statements that simply said this is a crucial and important 4 case and it's their opportunity to make a contribution, and 5 we then go on to talk about the fact that we want the 6 broadest range and perspectives represented. 7 Q Were any e-mails sent out looking for students who 8 felt that the use of race should be discontinued in the use 9 of college and law school admissions? 10 A The general e-mail attracted students like that, along 11 with attracting students who were in favor. 12 Q But did something go out that actually was suggested 13 it was looking for those types of students? 14 A Yes. That's while saying, the general e-mail was 15 looking for those students, that Michigan State -- you asked 16 about the law school; is that correct? 17 Q Well, I was asking generally. 18 A Okay. 19 Q Let's go to Exhibit 177. Is there someplace in there 20 where it is suggested that students are being looked for 21 who have a view in which they believe that the use of race 22 should be discontinued in the use of college and law school 23 admissions? 24 A Several places. 25 First of all, we assure everyone that their 42 1 expressed opinions will be treated confidentially. 2 Q Which is actually not true, is it, they are not 3 confidential, are they? 4 A I don't know the names of any of those students. 5 Q The names -- 6 A That's the notion of confidentiality. And are you 7 aware of names and identities of any of the comments? 8 Q I was asking whether the opinions are going to be 9 held confidential. The opinions are out there. 10 A Well, the opinions are out there before we do the 11 research, and they will be out there after we're done 12 with our research, but in terms of the promise of 13 confidentiality, what you're saying to a student is that 14 anything he or she says will not be associated with their 15 names. 16 Q With their names, okay. 17 A Right. 18 Q If you could then show me where in Exhibit 177 there 19 is a request for students who view -- who believe that 20 affirmative action should be discontinued or erased. 21 A Just a second. I'm trying to -- if you go to the 22 second page, which is page four, at the very bottom of the 23 fax, there is a paragraph. 24 Q I'm sorry, what page are we on? 25 A We're still under Tab 177, the second page, and the 43 1 paragraph that begins: "Please help us recruit 2 as many undergraduates as possible to 3 be participants in this study. We need 4 students of every race and ethnicity, we 5 need slightly more women than men, and 6 we need students with a broad--" and 7 this is the passage that's important -- 8 "and we need students with a broad range 9 of perspectives and experiences, and 10 most of all, we're looking for students 11 who are interested in advancing the 12 debate." 13 So for my purposes, that clearly asks students 14 from all perspectives, and whether they support or oppose 15 affirmative action, to come forward, and indeed, if you look 16 through the transcripts, in one set of responses you see a 17 white male in the law school say that he opposes affirmative 18 action because he feels that it would make black students 19 feel inferior, and it was a very rich discussion that 20 ensued, because then the white female said, oh, is that 21 how you feel about your classmates? And they had an 22 exchange, and it was very interesting and enlightening 23 to hear and experience. 24 Q These e-mails were sent out by United for Equality 25 and Affirmative Action? 44 1 A I think so. 2 Q Do they have a view on this subject, as far as you 3 know? 4 A I don't know the organization very well, but I'm sure 5 they do. I think the designation is very clear. 6 Q Now, who actually was -- who were the -- who was 7 responsible for actually the logistics of actually trying 8 to get these focus groups assembled? 9 A As I said, what we tried to purposely do was to 10 identify students, colleagues on the campus, and then to 11 allow them -- or not allow them, to ask their assistance in 12 distributing the appeals, and so beyond having a central 13 contact person or two on each of the campuses, I, for 14 reasons that we have just been discussing earlier, wanted 15 to stay out of the process of selecting students until 16 after which time I had a pool of students that I had not 17 assembled, and then from that pool we could select students, 18 and where the pool was large enough we actually selected 19 them randomly; that is, we numbered them and did a random 20 selection and assignment of the students from the pool of 21 African American students who expressed an interest. 22 So I'm simply saying on each of the campuses, we had 23 several contacts and I would have to pull the list of some 24 of the people that I used for that purpose. 25 Q But your control, other than approving the -- you 45 1 didn't exercise much control over the selection of the pool 2 other than exercising some approval over the text of some of 3 these messages; correct? 4 A That's correct. And I can't say that anyone could 5 exercise that much control, because, you know, it was a 6 voluntary response. You solicited individuals and then the 7 individuals volunteered to participate, and in some cases 8 students who had volunteered just could not -- we either 9 didn't select them into a focus group or they couldn't make 10 the assigned time and place. 11 THE COURT: If you had more time and resources, 12 somebody gave you a gigantic grant or something of that 13 nature, not necessarily in this case, but in another case 14 of something that you're studying and you wanted to do focus 15 groups, would you -- I have seen ads in newspapers and 16 things like that where it says, we're looking for people and 17 here's what we're going to be studying, something like that, 18 would that be a method that you would use, assuming that you 19 had, number one, a lot of resources and a lot of time? 20 THE WITNESS: Yes, Your Honor. The hypothetical of 21 a lot of research resources and the large grant brought the 22 smile to my face. 23 THE COURT: I'm sure it would. I'm sure if you had 24 a lot more time and a lot more money you would have done a 25 lot of different things. 46 1 THE WITNESS: Yes, sir, I would have, but with the 2 resources that we had, we tried to be sure just for reasons 3 of self-protection to get as many students as possible, so 4 we used a variety of avenues for going into classes, but 5 most definitely, yes, to your question. 6 THE COURT: But that's how you would normally, if 7 you had all the resources and ideal kind of situation, you 8 would do similar -- as I have seen ads in newspapers and 9 journals? 10 THE WITNESS: Yes, sir, right. 11 THE COURT: And things like that? 12 THE WITNESS: Exactly. 13 BY MR. KOLBO: 14 Q Did you rely on help from the lawyers for Intervenors 15 in assembling any of the focus groups? 16 A No, not help from the lawyers, as such. 17 Q Ms. Massie, for example? 18 A No. I basically would spend my considerable network 19 resources. As you know, having been in the business for 20 a while, I have quite a few contacts and colleagues. 21 Q And then you were actually at Michigan for some of 22 these focus groups; correct? 23 A For all of the focus groups. 24 Q For all of them? 25 A Yes. 47 1 Q And who assisted you in sort of the logistics of 2 organizing those groups when you were there? 3 A I would have to look up the names of the students who 4 gave us a hand with the organization, but essentially the 5 process was as follows: After we had a pool, then we would 6 reserve rooms and assign time slots and then our team worked 7 in each of those time slots with students who we contacted, 8 so in other words, I would have a list of students who 9 were in the pool, then I would solicit a set of research 10 assistants to call and schedule people, so that was the 11 process. 12 Does that answer your question? 13 Q I think so. 14 A Okay. 15 Q And did you work at all at the actual focus groups 16 where any of the lawyers for the Intervenors were present? 17 A Oh, absolutely not. 18 Q Ms. Massie wasn't present for any? 19 A No, not for the focus groups. 20 Q I'm sorry, for the actual getting people in the 21 rooms together and so forth. 22 A No, that was -- no, no, no. That's -- 23 Q That would have been -- 24 A Not acceptable. 25 Q That would not have been acceptable? 48 1 A Absolutely not. 2 Q You didn't, in selecting these students for these 3 focus groups, you didn't make any effort to ascertain who 4 was actually interested in applying to law school, did you? 5 A No, but in the course of our discussions and in the 6 data that came forward from the focus group interviews, 7 it turned out that there was a sizable representation of 8 students who were either interested in the law or interested 9 in going to law school. 10 Q That wasn't the qualification, one of the 11 qualifications? 12 A No, no, it wouldn't be, because that would defeat the 13 purpose. 14 Q Now, I just want to talk about some of the things that 15 you did. Again, I'll move on eventually here from these 16 focus groups and the methodology, but in the things that 17 you studied with respect to these focus groups, one of the 18 things you were looking at, I think your report uses the 19 term, to illustrate and elucidate some of the issues with 20 respect to academic performance and discrimination and so 21 forth; correct? 22 A Possibly. Are you quoting from the report? 23 Q Well, the words, illustrate and elucidate are there, 24 but it's not important if they are not, would you agree 25 that that's -- 49 1 A That would be a purpose. 2 Q In trying to look at some issues with respect to 3 correlation between academic performance and racial climate, 4 discrimination, things like that? 5 A So if we're talking -- if we're talking about 6 correlations, then you're referring to the survey portion 7 of the report. 8 Q Well, I shouldn't use the word, correlations. You're 9 looking to sort of eliminate in these focus groups those 10 connections; right? 11 A We were trying to understand them, yes. 12 Q Not to draw mathematical conclusions? 13 A Okay, correct. 14 Q And you did not, as part of that process, actually 15 ask students to get transcripts for you; correct? 16 A No. And I didn't for a specific reason, if you would 17 like for me to elaborate. 18 Q My understanding is that you believe that self-reports 19 of grades are accurate? 20 A I would be a bit stronger. It's not my belief, the 21 literature has demonstrated conclusively that students are 22 pretty accurate in reporting their grades and this has been 23 demonstrated with scientific research of the probabilistic 24 sort for students who were asked to report grades and then 25 those grades were compared to transcripts. 50 1 Q But in any event, you didn't get those? 2 A No, I did not. 3 Q Either as part of the survey or as a part of the focus 4 group work? 5 A The actual transcripts? 6 Q Right. 7 A No. 8 Q You don't have any information on how many of these 9 students at focus groups or in the surveys, how many of them 10 that might have taken the LSAT test, you didn't have any 11 testing data on that, did you? 12 A No, I didn't ask any questions about the LSAT test 13 scores for those students. 14 Q Am I correct that between your focus group analysis 15 and your survey work, you concluded that all the students 16 were doing well academically across all racial lines? 17 A I concluded that the students were making reasonable 18 progress and doing well enough to graduate, yes. 19 Q And basically -- 20 A But you know, actually, there were variation in their 21 levels of performance. Excuse me for interrupting. 22 Q But am I correct that your conclusion was that the 23 participants academically were basically the same across 24 racial lines and that the results, these were generally high 25 performing students, fair enough? 51 1 A The second part I agree with, but not the first part, 2 no. The students were not performing the same across racial 3 groups, but the students were high performing students, 4 for example, in terms of the -- in the cases of the law 5 students, in terms of the profiles that they had as 6 undergraduates, and in terms of the undergraduate students 7 in terms of their profiles as high schoolers, and 8 ultimately, as I have said a moment ago, the point was 9 that the students were performing at an acceptable and 10 reasonable level across the board. 11 Q Could you go to page 34 of your deposition? Do you 12 have it there? 13 A Yes, sir, I do. 14 Q Actually, it starts on page 33. This is a question: 15 "And my question is, and I'm sure 16 I have looked at it, it must be here 17 somewhere, but I'm trying to get a 18 general sense of your understanding 19 as to whether there were significant 20 differences between grade point 21 averages among the racial and gender 22 lines among these students at these 23 feeder institutions." 24 Answer: "What was striking as a 25 whole, the students that we interviewed 52 1 were high performing students academically." 2 Question: "Along all racial lines?" 3 Answer: "Along all racial lines, 4 but there was some difference in terms 5 of their reported experiences in classes 6 and in terms of their experiences -- 7 in terms of their reported experiences 8 with respect to how they were perceived 9 and treated." 10 Did I read that accurately? 11 A You did. 12 Q Is that a true statement? 13 A It is, but once more, it has to be in that larger 14 context, because if you remember the questioning and answers 15 that continued over to page 35, for example, line seven, 16 where I talked about grades being negatively affected by 17 discrimination, and so in the context of what is a very 18 complex situation, that's a correct assessment. 19 Q You didn't actually, as part of your work, you didn't 20 measure the extent to which there were differences in 21 grades, academic performance, among these students in 22 these focus groups or surveys; correct? 23 A I had done that work previously at the University of 24 Michigan and in a couple of instances actually making such 25 measures, measurements for the entire population of all 53 1 students who were enrolled at the University. I had used 2 the University's retention data set and done exactly those 3 correlations of race with the grade point average, so in 4 the focus groups, you know, we didn't do that, because for 5 one thing, as we have discussed, it would not have been 6 representative of the entire community, but more 7 importantly, I had that information from previous research 8 that used the entire collected data on all students who had 9 entered the University and been there for five years, as 10 well as data from probability-based surveys. 11 Q Okay. Let me ask you about some of the survey work. 12 Surveys are the written materials, right, as opposed to the 13 focus groups, written survey questions? 14 A Yes, self-completed questionnaires, yes. 15 Q And 200 out of all these, among these schools, among 16 these four schools? 17 A For this particular study, yes. 18 Q And just to clarify myself on this, is that exclusive 19 of the law school, the 200 or so surveys? 20 A I would have to look at the numbers again, but the law 21 school -- yes, it is, it's exclusive of the law school. 22 Q And how many -- 23 A We had about 40 surveys at the law school. 24 Q Okay. So let me ask you about the -- first of all, 25 the 200 or so undergraduate surveys, am I correct that you 54 1 found -- well, first of all, this was predominantly, among 2 the surveys, like the focus groups, this is predominantly 3 minority students; correct? 4 A Yes. 5 Q I think I read somewhere that you have about 6 17 percent whites in the survey response? 7 A Right. And for reasons that are -- became very clear, 8 and became very clear in, for example, the responses of 9 the white students, the white students said they had not 10 experienced racial harassment or racial discrimination 11 and this focus group study was a focused study of 12 experiences with race discrimination. 13 Q Most of the students were A or B students, correct, 14 about 90 percent of them? 15 A I would have to look at the percentages. Are you 16 referring to the survey? 17 Q I think this is around -- well, it may be in several 18 places. I took my notes here and I am looking at page 49, 19 I think, of your report, where you have it in narrative 20 fashion. 21 A Page 49, you say? 22 Q I can't guarantee that's where it is, but -- 23 A I have tables under tab two which would include a 24 specific grade breakdown. 25 Q Actually, page -- 55 1 A Item five. 2 Q And then page 50, actually, goes into this, as well. 3 A Okay, page 50. 4 Q And 47 percent of the respondents reported A averages, 5 46 percent B averages? 6 A I'm sorry, sir, on page 50? 7 Q Page 50, the second full paragraph. 8 A Yes. 9 Q So that's confirms that these were pretty high 10 scoring, pretty well performing students by and large 11 in these focus groups; correct? 12 A This is true. 13 Q Doing well. 14 Most of them also indicated that they were well 15 prepared, or prepared at least, to enter college? 16 A That's correct. 17 Q And I think the number I counted was about 82 percent 18 answered very -- answered well prepared or prepared, that's 19 on page 50 and 51? 20 A 60 percent answering well prepared and 22 percent 21 prepared, somewhat prepared, yes. 22 Q A third of the parents of these students had a 23 father with a graduate or professional degree; correct? 24 A Right, but -- yes, this is true, but as you know, 25 this is the group as a whole, including the white students. 56 1 Q Most of the group is minority; true? 2 A Yes, that's true. 3 Q And 25 percent of the mothers of these students had 4 graduate or professional degrees? 5 A And as you pointed out, 17 percent of the sample is 6 white, so that's why I made the point. So we're talking 7 about an additional eight percent, if indeed it turns out 8 the case that the white students were the ones who were 9 the 17 percent who -- the 17 percent of white students had 10 mothers with graduate degrees. 11 Q Well, do you report that data broken down by race 12 at all? 13 A No, not here I don't. 14 Q How about, did you find out -- we're talking about 15 graduate professional degree, correct, that's beyond four 16 years of college? 17 A It's a college graduate, yes, with an advanced degree. 18 Q Well, I think of a graduate degree, is that -- do you 19 mean by that a four-year college or is that something past? 20 A It's past. It's past the BA and then, as you said, a 21 Masters degree, professional degree, a Doctorate. 22 Q Did you ascertain how many of these students had 23 parents, one or more parents, that had a four-year degree? 24 A I think we do have that here. 25 Q I guess I didn't see it, so. 57 1 A Let me look. 2 Q Well, if it's here, it's here, right? 3 A It would be in the tables for sure. I'm pretty sure I 4 have it. 5 Q Well, if you could find it quickly, I just -- 6 A Parent education. If you'll give me a moment, I'll 7 dig it out. 8 Q Sure. 9 A You know, actually, I correct myself. I'm looking at 10 the questionnaire, and what we asked the students about was 11 their own -- no, we do have it, just a second. 12 Question six in the questionnaire, in the survey, 13 asks about parents' education, so I will have to find the 14 equivalent table. 15 Table 15 under Tab 2 is father's education and 16 Table 16 is mother's education. 17 Q Yes, I see that. 18 A And so I can answer that question separately for 19 mothers and fathers in terms of graduate professional 20 degrees. For fathers we're talking about a third of 21 the sample and for mothers we're talking about a quarter 22 of the sample, so 33 percent and 25 percent. 23 Q That's for graduate or professional degree; correct? 24 A Yes, that's what you were asking me, is that correct? 25 Q Right. And then there are -- we have already gone 58 1 over that data. 2 And having a BA or a BS? 3 A Yes, in terms of a BA, solely a BA, 14.5 percent 4 of fathers and only a BA, 21 percent of mothers. 5 Q All right. And then there are additional numbers who 6 have some college; correct? 7 A Correct. 8 Q Okay. And again, you just -- that's not broken down 9 by race, we just don't have that? 10 A No, not here. Except, once more, that's research -- 11 those kinds of comparisons by race of parents' education, 12 I have done extensively in the 80, 85 or so papers that I 13 have published and so those were questions that have been 14 answered and they have been answered generally, that is, 15 on a national level, and then answered previously for the 16 University of Michigan from the other analysis I have done. 17 I mean, the bottom line is that you see these 18 dramatic differences by race in terms of parents' education 19 with black and students of color being disadvantaged. 20 Q Just not shown here for these focus groups or surveys? 21 A I didn't do that analysis for the focus groups for the 22 simple fact that, as I said, it's a point I demonstrated 23 with earlier research. 24 Q And I hesitate to mention it, because I didn't write 25 down the page number, but am I correct that you found that 59 1 two-thirds of the students in these 200 member focus -- 2 200 member surveys, two-thirds of them reported that they 3 had A averages in high school, probably around -- 4 A You found that where? Where was it? I'm sorry. 5 Q Well, I'm not sure. I think it's page 51 or 2. 6 A Okay, 51 or 52. 7 Q A substantial number of them reported high performance 8 in high school, is that your recollection, at least, or do 9 you want to look? 10 A It is my recollection. Just if you needed me to 11 attest to the percentage, I needed to look it up. 12 Q Okay. So you're satisfied that that's -- let's 13 just -- 14 A Generally it's true, yes, but if you want me to 15 look up the percentage, the exact percentage, I can do so. 16 Q We will just -- if it's there, it's there. If it's 17 not, it's not. We will leave it there. 18 Let me ask you about some of the law school survey 19 work that was done. Am I correct that you found the 20 prospective -- there's only 31 of these respondents, 21 correct, first of all? 22 A Yes, that's correct. 23 Q You found that grades were comparable across races and 24 that they were acceptable levels, with only three reporting 25 something less than a B average? 60 1 A Yes, sir, that's correct. 2 Q And most of the students felt extremely well prepared 3 for law school, about two-thirds of them? 4 A Correct again, yes. 5 Q By and large, these students in the -- among the 6 31 students at the law school reported that their college 7 grades had been high; true? 8 A Yes, sir. 9 Q And can you remind me, what was -- the 31 law schools 10 respondents, were they predominantly minority students? 11 A Once again, we had one focus group that was white 12 students, so they were predominantly students of color and 13 predominantly female, actually, because those were the two 14 categories of focus in terms of questions about racial 15 climate and gender climate. 16 Q Let me ask you, I'm going to change subject now a 17 little bit her. 18 Do I fairly understand that your opinion is that 19 racism, race discrimination, is something that is pervasive 20 in this country? 21 A Oh, absolutely, yes. 22 Q And you have focused on these feeder group 23 institutions and the law school itself and you have found 24 persistent patterns of racial hostility, racially hostile 25 climate; correct? 61 1 A That's correct. 2 Q You have found evidence from these studies, from these 3 survey respondents, of patterns of discrimination and racism 4 on these campuses, that's your finding? 5 A That's my finding and it matches up with the research 6 that others have done, as well as my previous research. 7 Q And is it fair to understand your opinion to be 8 that these feeder institutions and the law school are not 9 atypical for primarily white college universities and law 10 schools around the country? 11 A That's a safe assumption, yes, sir. 12 Q Okay. So this is a problem that's pervasive 13 around the country, other colleges and universities? 14 A It's a problem that's pervasive, but it is a problem 15 that varies in its intensity, depending on characteristics 16 and traits of a campus; for example, the history of a 17 campus, history of race relations on the campus, the sorts 18 of policy issues or debates that are under way on a campus, 19 for instance, on a campus where affirmative action is being 20 hotly debated, then often those kinds of patterns will be 21 exacerbated. 22 Q Okay. And patterns of discrimination, racism, are not 23 limited to colleges or universities, either, correct, in 24 your opinion? 25 A Well, no, unfortunately, they are part of our society 62 1 as a whole and as such you see those problems from the 2 larger society being reflected on the campuses. 3 Q I mean, it's fair to say to understand your opinion 4 to be that we have an existing climate in this country of 5 societal discrimination against minority students, fair 6 enough? 7 A A climate that -- I have trouble answering that 8 question yes or no, and if you'll allow me. I mean, we 9 do have discrimination in the country. It varies in terms 10 of its force and shape and the targets depending on the 11 particular regions, and it also varied over time. 12 So if you talk about California and discrimination 13 right around Prop 209, then you have a different kind of 14 picture than, say, California and whichever group you're 15 focusing on pre209. 16 Q And I understand there may be variations and you 17 may -- 18 A Yes. 19 Q -- believe there are variations, but you believe it's 20 a national phenomenon, correct, in society? 21 A Okay. I believe, and those beliefs are validated by 22 empirical research -- 23 Q Okay. 24 A -- of an extensive body, that yes, the country 25 continues to engage in racial discrimination, that 63 1 continues to be a problem for us, and again in terms of 2 gender discrimination, as well. 3 Q Our country has a history of societal discrimination 4 against minorities? 5 A Absolutely. 6 Q And there are continuing effects of that societal 7 discrimination? 8 A Correct. 9 Q And some of the continuing effects of that 10 discrimination relate to the academic performance of 11 minorities like African Americans, Hispanics, Native 12 Americans at America's colleges and universities; true? 13 A This is very true and it's based on just a cumulative 14 disadvantage that I have documented from K through twelve 15 education that then translate into different levels, 16 for example, of educational preparation, educational 17 opportunity, and then translating into differences in 18 terms of representation for the different ethnic groups 19 in higher education. 20 Q And is it fair to understand you believe that race 21 ought to be considered in the admissions process at 22 colleges and universities in order to help respond to 23 these continuing ongoing events of societal national 24 discrimination against minorities, true enough? 25 A I think as an educator you cannot make decisions 64 1 about students and educating and admitting students into an 2 institution of higher learning without looking at the whole 3 person, and race is a sizable and in many instances just 4 definitive component of a person's experiences and 5 opportunities. 6 Q And one of the reasons we should do that, as I 7 understand your opinion, is that we must respond to these 8 ongoing effects of societal national discrimination against 9 minorities; fair? 10 A One of the reasons is that we do need to respond to 11 and make and take account of systematic patterns of racial 12 discrimination. Further, I think we have to prepare 13 ourselves, our students, our work force, for a reality 14 of a racially and culturally diverse world and society. 15 Q And those are reasons that, in your opinion -- and 16 you're an educator, right? 17 A Yes, I am. 18 Q Those are reasons that, in your opinion, we all 19 have to be able to use race as a factor in admissions 20 decision making at colleges, universities and law schools; 21 true? 22 A I think that when we are making decisions about 23 admitting students, we have to look at the whole student 24 and a student's racial identity is part of that whole. 25 So absolutely, yes, it's -- and I would just say, as an 65 1 educator and researcher, it's just about impossible to 2 look at a student apart from their racial experiences, their 3 race, because race structuralizes in this society in very 4 definitive and very clear ways. 5 You look at a city like Detroit, you see that the 6 city is residentially segregated. You look at one of those 7 pockets, you see it systematically continues in terms of 8 educational opportunity, and so all of that ties together, 9 so it's not merely skin color, per se, but it's what skin 10 color determines about life experiences and opportunities in 11 our society. 12 Q Okay. And so housing segregation, for example, in 13 Detroit, is a reason that ought to be considered as a 14 justification for considering race in the admissions 15 process at a law school or undergraduate school, true? 16 A Housing discrimination in Detroit, in Minneapolis, 17 first of all, sets the catchment area for the schools that 18 you attend. It also determines the resources that are 19 available in the schools that you attend. So it sets you 20 on a trajectory either to higher educational accomplishment 21 and achievement or to lower educational accomplishment and 22 achievement, and unfortunately, those kinds of forces are 23 bigger than individuals, and so the most motivated, the most 24 innately talented young kid cannot learn algebra if algebra 25 class is not available in her school, and unfortunately, 66 1 what we see is that algebra and other subject areas are 2 differently available in schools coded by race; that is, 3 the schools that are coded by race tend to be poor and they 4 tend to have fewer offerings, and so, thus the point I'm 5 trying to make. 6 Q And just so I understand, your opinion is that those 7 kind of social demographic factors are the kinds of reasons 8 we have to -- that we should use to justify using race as 9 one factor in the admissions process at colleges and 10 universities, law schools, true? 11 A Yes, and the sociodemographic factors as related 12 to the educational institutions and the educational 13 opportunity structure that they shape. 14 So it's more about educational opportunity, but it's 15 educational opportunity that is differentially experienced 16 and available based on race and based on racial segregation 17 in this society and systematic deprivation by race. 18 Q I want to talk about the phenomenon of poor 19 performance, poor academics, poor grades in the college 20 or law school. 21 You're an educator. You have been an educator for 22 how long? 23 A I have been an educator for 26, 27 years. 24 Q Would you agree with me that there are a lot of 25 reasons why students, any particular student or any group of 67 1 students, might be suffering from poor academic performance, 2 a lot of reasons out there for that? 3 A I would accept that general principle, yes. 4 Q There are a lot of reasons that would explain why a 5 white student might have poor grades; correct? 6 A There are reasons that would explain why a white 7 student might have poor grades, but one of those reasons 8 would not be racial discrimination. 9 Q But there is a lot of -- 10 A Because this society is a society that has been 11 dominated by whites and has been structured in ways that 12 privileges whites and so racial discrimination against 13 whites in this society, I have trouble buying. 14 Q But there is a lot of reasons that have nothing to do 15 with race that can cause a person to perform poorly on their 16 grades and test scores; true? 17 A See, that's the interesting thing about how race 18 affects educational performance in this society, because in 19 fact, as I tried to point out, educational performance at 20 the college level, let's say, is a function of cumulative 21 either educational advantages or disadvantages, and so if 22 the original situation was created by racial segregation, 23 then by the time you get, say, to an SAT test, you could 24 in theory say that, well, this student's performance, low 25 performance on, say, the mathematical part of the SAT, is 68 1 because he doesn't know math, but that rips the person from 2 his experience and rips the person from an experience that 3 was dictated, structured, required by racial identity and 4 that put him in schools that set him on an educational track 5 that insured that he wouldn't know algebra, because it would 6 not be available to him. 7 And so that's why I'm hesitant to accept an 8 assertion that they are just random causes, because in fact, 9 the causes of low educational performance of Chicano-Latino 10 students are not at all random. They are very systematic. 11 There is very systematic deprivation in terms of 12 preschool educational-related experiences, systematic 13 deprivation in terms of the elementary years and 14 opportunities to learn, and systematic deprivation in 15 terms of high school, so at the end of the process, the 16 predictable has happened, that kid is not as well prepared, 17 is not as able to compete with his white peers who were set 18 on a course where those resources were available and where 19 the opportunities were there and present in their schools. 20 THE COURT: Define systematic for me. 21 THE WITNESS: Systematic is just a kind of patterned 22 effect, if you will, so just following a young person 23 through school. A poor youngster and many youngsters of 24 color who are disproportionately poor, who does not have the 25 advantage of educated parents to prepare him or her in those 69 1 earliest years just for readiness to learn, if you will, 2 so in those early years, in the preschool years, whether the 3 parents could afford to put that kid in a quality preschool, 4 then in terms of the K through twelve years, the quality of 5 schools that those youngsters have available to them. 6 And we educated parents and the parents who are of 7 middle class work very hard to get their kids into certain 8 schools, and that's because we understand that high quality 9 schools will produce kids who are better educated and who 10 are better equipped to compete, and once you get into the 11 high school years, those differences become even more 12 pronounced. 13 The research I have been doing in California shows 14 that the availability of advanced placement classes -- 15 THE COURT: I understand. So if you don't start at 16 the bottom, you can't -- 17 THE WITNESS: Exactly. 18 THE COURT: You used the word, systematic, so that's 19 what you mean by systematic? 20 THE WITNESS: Yes, sir. Sort of a pattern, 21 repeating structure -- 22 THE COURT: I see. 23 THE WITNESS: -- of just deprivation, deprivation 24 in the earliest years, all the way through to deprived or 25 blocked opportunity into high school and college. 70 1 BY MR. KOLBO: 2 Q Let's just accept, at least for point of argument, 3 that race is a factor that can explain academic performance 4 in some cases. Let's just put that aside for a minute, for 5 the moment, but my question actually was, isn't it true,