If mere photos can have such an impact, we should not be surprised to learn that the way children speak can make a big difference in what teachers expect from them. Between 1890 and 1940—the Nadir of race relations—segregation increased in the North as well as the South. Black English grew apart from white English, because blacks increasingly lived apart from whites. Today most African Americans sound identifiably “black” on the telephone. Their accent, dialect, and even the timbre of their voices differ from those of whites. This difference is not genetic; historian Barbara J. Fields points out that no such thing as “black English” exists in England. There, West Indian immigrants’ children learn the English of their class and region. But in America, as a result of the Nadir, “black English” has intensified.9 In turn, this black English then persuades many teachers to expect less from those who speak it.10
Careful research by David Figlio shows that teachers also make judgments based merely on children’s first names. Figlio studied thousands of Florida birth certificates to identify “black” names like Dwayne and “white” names like Drew. He also identified names like “Damarcus” or “Da’Quan” as more likely to be given to working-class children, based on such socioeconomic characteristics as “no father present” and “mother did not complete high school.” Then he compared the academic performance of children from the same family.11 Even after controlling for variables that might influence performance—such as low birth weight—he found that those with “black” or “lower-class” first names did worse in school than their siblings. He inferred that teacher expectations were largely responsible, because teachers were simultaneously less likely to recommend children with black or “lower-class” names for gifted and talented programs and more likely to recommend them for promotion to the next grade. That is, at every level of academic achievement (measured by achievement test scores), from low to high, teachers nominated proportionally fewer black students than white students to gifted and talented programs. At the same time, teachers were more likely to promote black and poor students than white affluent students to the next grade, at every level of achievement. Figlio explained these seemingly paradoxical results by invoking expectations. Because they expected less from black and poor children, teachers didn’t bother forcing them to repeat a grade and were also unlikely to recommend them to gifted and talented programs.12
Figlio concluded that white teachers were more likely to stereotype based on names; other studies have found white employers much less likely to hire or even interview job applicants with black-sounding names. Other research shows that Hispanic and African American teachers also often participate in expecting more from Euro American children. The black community has long been plagued by expectations that differ owing to color and ethnicity. I wrote my doctoral dissertation on a then little-known group, Chinese Americans in Mississippi. As a third party in a social system built for two, they proved sociologically fascinating, as they tested the norms of racial segregation. The minority within this minority, Chinese African Americans, also proved interesting. Today “everyone knows” that Asian Americans are good in school—most of ’em, anyway—the “model minority.” This stereotype abides not only within white subculture, but also among African Americans. One Chinese African American undergraduate told me how he had benefitted in high school: “The teacher calls on you more often, expecting more from you. So you study harder.” When these mixed-race students graduated from high school, almost all went on to college, and not to the usual choices—the state schools, where African Americans tended to go—but to private colleges or Northern universities.13
Before school desegregation, many teachers in black high schools expected more from light-skinned African Americans than from dark-skinned students. These expectations mirrored social class distinctions deriving from slavery times, when whites treated house slaves, often the product of sexual relations between a male owner and female slave, better, compared to darker field hands. As a graduate student in August 1965, I had an “aha moment” on this point. I was spending the summer directing the Social Science Lab at Tougaloo College, a predominantly black school near Jackson, Mississippi. At that time, Tougaloo offered a two-week “pre-freshman program” to its entering students, designed to get them reading, writing, and thinking closer to a college level. Most had just graduated from all-black high schools in Mississippi, the least-funded schools in America, and some of the worst. Students in graduate and law programs at Harvard and Brown universities had devised the program; some regular Tougaloo faculty members also participated. On the first day of the second session, the young undergraduates-to-be and their teachers milled around in a “mixer” in a large hall, trying to make small talk. So did I.
Twenty minutes into this awkward affair, I looked around the hall and had my “aha moment.” Every single graduate student and faculty member—white or black, myself included—was talking with light-skinned students. Darker-skinned African Americans stood in small groups talking with each other. I spent the rest of the hour talking with the darkest students I could find. Of course, my little gesture—one person, in one setting, for 40 minutes—meant nothing. I could only hope that the darker-skinned African Americans were as oblivious to the pattern as I had been. This was hardly likely. Dark-skinned singer Big Bill Broonzy, born in the Mississippi Delta in 1893, incorporated three famous lines about the pattern in a blues song he wrote long before 1965.
If you’re white, you’re all right.
If you’re brown, stick around.
But if you’re black, get back! Get back! Get back!
Lighter-skinned African Americans were sometimes even admitted to parties, fraternities and sororities, and even certain colleges only if their skin color was lighter than a brown paper bag.14 The word “brighter” in black parlance at that time referred not only to intelligence but also to skin tone.
Within three years of my aha moment, the “Black Is Beautiful” movement arose, partly to combat this strain in black culture. It did alleviate the problem somewhat. The matter is more general, however. Sally Zepeda found that teachers in a Midwestern urban elementary school called on white students more often to answer “higher-order thinking questions,” while asking nonwhites and ESL students easier questions that simply reflected the reading. Other researchers have shown that teachers give students whom they consider “stronger” more than twice as much time to answer questions before restating them or moving on to other students. Rhona Weinstein showed years ago that students are very aware of different teacher expectations for different students.15
A change of teacher behavior can mitigate this problem: teachers can regard silence as their friend. If teachers simply wait two or three seconds for a student to respond, usually s/he will respond, and often in more depth. Waiting is particularly important when the question does not ask for mere recall. Waiting can be hard; there is a lot to cover, and the teacher worries that the rest of the class will grow bored. Usually they won’t, though. Silence focuses class attention, adding drama, especially if the teacher is part of it, waiting breathlessly for the student’s important answer. This simple change in teacher behavior has been shown to produce higher achievement from all students, fewer disciplinary comments, and better and deeper questions from teachers. Silence can also be the teacher’s ally when asking questions of the entire class. Again, absent a quick reply, a natural tendency is to rephrase the question or give a hint as to the answer. Wait. Be still. If the only hands raised belong to “the usual suspects,” the teacher can ask, “Do none of the rest of you have a thought about this matter? This issue has implications for us, today. This is important.” Then wait some more.
How quickly teachers’ initial expectations can become fixed! I experienced this myself in 1st grade in Decatur, Illinois. Before September was halfway through, my teacher Miss Stone had divided us into three reading groups. The best were called the “bluebirds.” They had already learned to read from their parents. Next came “redbirds.”
Slowest in this all-white class of 30 students were, of course, the “blackbirds.” Some of them lived on the “wrong side” of Highway 48. Social class and first impression largely generated the initial placements. Whatever ability students developed or displayed as reading was taught played no role whatsoever, for no child moved from one group to another during the rest of the school year.
RESEARCH ON TEACHER EXPECTATIONS
There is a huge literature on teacher expectations. Among other things, it shows that many teachers not only expect better things from affluent white children, they also communicate that expectation to the children and thus help create the superior performance that they expect. In itself, that process seems unobjectionable—most of us got where we are today because someone believed in us and communicated that belief to us. It’s the flip side that does the damage. Most researchers in this area find, as did Figlio, that low expectations also become a self-fulfilling prophecy. Some students internalize the view that they are not very competent. Others resist this inference but choose to slide by with little effort, after they realize that teachers don’t demand much from them. Over time, both paths lead to the same outcome: students who in fact have little knowledge and ability.
Research by Robert Rosenthal and Lenore Jacobson at an elementary school in San Francisco in the late 1960s sparked the entire expectancy literature. Rosenthal taught social psychology at Harvard, where he devised an experiment to see if he could generate teacher expectations and measure their effect on students. He and his team came to the school in late May, flaunting their Harvard credentials, and gave all K–5 students an IQ test—not the usual Stanford-Binet, but an IQ test nonetheless. However, they did not call it an IQ test. They touted it as a new “Harvard Test of Inflected Acquisition” and claimed it could predict which students were about to spurt—show a considerable increase in school performance. Just before school opened the next fall, they returned and met with the teachers of these students, now in grades 1–6. They talked about their research, reminded the teachers of the test they had given at the end of the previous school year, and told each teacher that a handful of their students—listed by name—had scored exceptionally well. They could expect a burst in performance from those students. (In fact, the students had been chosen at random; no test in existence then or now can predict spurting.) The teachers then proceeded to teach for the school year. From time to time Rosenthal and his staff visited and observed. At the end of the year, they gave the IQ test again, along with other measures of students’ performance, and interviewed each teacher about his or her class.
In the 1st and 2nd grades, the experiment worked spectacularly. The students who were expected to spurt did spurt—an average gain of 28 points in IQ among the 1st graders and 17 among 2nd graders. Was this because teachers spent more time with those who were expected to spurt? Taught them better? The researchers considered this unlikely, because the other students, too, gained, showing that teachers had hardly neglected them. Rather, the large increases came about because the teachers subtly expected more from the named students, and the children met their teachers’ expectations. Since the children had been selected randomly, no characteristic of theirs could possibly have played a role. They had about the same mixture with regard to class background, race, sex, even height, as those not selected to spurt.
This result became so famous that it has been called “the Rosenthal effect,” though it is now usually known as “the expectancy effect.” Even laboratory rats running a maze run it better when researchers expect them to, so psychology students now take care not to know which rat had which background when they put them in mazes for their final runs.16
The experiment also caused substantial gains in the performance of students about whom nothing was said. First graders gained an average of 12 points, second graders 8 points, and students in higher grades showed similar increases in their IQs. According to IQ theory, no large gains should occur, because IQ supposedly stays steady over time and may even be mainly genetic. Furthermore, psychologists norm the test so that the usual gains associated with a year’s maturation and education cause no change in IQ. The children in this study—even those in the control group—showed larger gains, resulting in increased IQ scores. The researchers concluded this was another example of the familiar Hawthorne effect, named after the industrial company where social scientists first systematically observed it. This is the effect on those studied that results simply from doing the study. The children may have felt important, since researchers had traveled clear across the country to study them. Their teachers may have taught extra well, since they were being observed. Harvard’s prestige may somehow have rubbed off.
Whatever caused the control group children to excel, to me Rosenthal’s second most important finding—after the expectancy effect itself—was the teachers’ assessment of those in the control group who spurted the most. Teachers often rated them unfavorably. Remember, these are students who did particularly well. Nevertheless, their teachers did not like them. They were the “wrong” students—those from whom nothing special had been expected. They didn’t get the same subtle rewards for good work as the students who were supposed to do good work received. Other researchers have reported that when African American children excelled, their teachers liked them less, while when white children excelled, their teachers liked them more.17 Again, these teachers were not intentionally racist; some were black. They simply had a subconscious sense for the way things should be and didn’t appreciate students who upset that order. Subconsciously, they acted to restore the order.
“STANDARDIZED” TESTS AFFECT TEACHER EXPECTATIONS
If mere photographs and the mention of student names at the start of the school year can prompt such different teacher expectations, followed by different educational results, it should not surprise us to learn that “standardized” test results—many of which purport to measure aptitude—can influence how teachers view students and how they rate and reward their work. The hopefully apocryphal tale of the substitute elementary teacher makes this point. Rushed into service after the sudden illness of the regular teacher in October, he managed the class well and caused his principal no problems. In June, as he was cleaning out his desk, the principal came to his room to commend his performance. “I’m sorry we never really had time to chat until now,” she said. “I was so busy last fall that I never even had time to give you a proper orientation.” “That’s OK,” the teacher responded. “Mrs. Tyler had left me a sheet with the children’s IQ scores, and I found that most helpful. As I looked back over the year, the children really behaved according to their potential.” “IQ scores?” the principal asked. “We don’t have IQ scores! Let me see that sheet.” He handed it to her. “My God, those are your children’s locker numbers!”
Unlike the expectations given to teachers in San Francisco, “standardized” test scores aren’t random, nor do they match locker numbers. They correlate strongly with student characteristics, especially race, sex (girls scoring worse than boys on math), and social class. This problem with “standardized” tests slapped me upside my head when I left graduate study at Harvard University and took up full-time duties as a sociology professor at Tougaloo College. At Harvard I had enjoyed considerable teaching responsibilities, including running the sophomore seminar in sociology, key to a successful major, for some twenty undergraduates. One of my charges could not do the work. Back then, when “gentlemen” like George W. Bush and John Kerry attended Yale and Harvard, some undergraduates would not do the work. But this lad was trying hard, yet still had difficulty. In those days, faculty members could freely examine student records, so I looked up his. He came from a wealthy background. His father and grandfather had graduated from Harvard; he had attended Phillips Exeter Academy, an elite New England prep school. Surely he had taken at least one coaching course for the SAT,18 and he had taken the SAT and PSAT several times. His best scores were 500 on the verbal and 550 on the math, on the SAT’s 200–800 sc
ale. Scores so low were unusual at Harvard, and I realized his poor performance in my seminar was not his fault.
My first year at Tougaloo, my students impressed me. The best ones equaled my best protégés at Harvard, and the middle group was able, too. Only the lowest quartile would have been unable to do the work at Harvard. Owing to a grant the department head had gotten from the Field Foundation, every senior major took the Graduate Record Examination. Not only is it produced by the same corporation (Educational Testing Service, or ETS) that produces the SAT; its verbal and math tests are drawn from the same bank of items. The results were fascinating. Their GRE scores ranked Tougaloo students in a reasonable order, one reflective of their classroom performance, especially taking into consideration a handful of academic underachievers. But their scores were dramatically lower than those of Harvard students. The best student received maybe a 570. The worst scored 200 (which you get for signing your name) to 220. An average Tougaloo student scored from 340 to 440 or so.
I had to realize that a 500/550 at Harvard was very different from a 500/550 at Tougaloo. Tougaloo graduates with scores around 500 went on to prestigious law schools and graduate programs in sociology and completed them with fine academic records. One Tougaloo student who scored about 500/550 as a senior found my introductory course in sociology too slow-moving for her as a sophomore and made a deal with me: after mastering what I was teaching, she read sociological essays by Karl Marx and then wrote a riveting essay on false class consciousness. The regular course material that was too easy for her included some of the same readings that had caused my Harvard student, with similar scores, to despair of majoring in sociology. Upon reflection, this made sense. My Tougaloo student’s scores put her near the top of her milieu, while my Harvard student’s scores put him near the bottom of his.
Teaching What Really Happened: How to Avoid the Tyranny of Textbooks and Get Students Excited About Doing History Page 8