A Crisis of Authority in Predominantly Black Schools?
by Sean P. Kelly - 2010
Background/Context: Black students are no less engaged or more disruptive than other students of similar achievement levels and socioeconomic status. However, because Black students are more likely to have disadvantaged family backgrounds and lower levels of achievement, segregation concentrates the risk factors for problem behavior in predominantly Black schools. As a result of the behavioral climate in predominantly Black schools, teachers may rely on instructional methods that facilitate an orderly classroom and minimize the negative effects of disruptions, possibly resulting in an instructional approach that is less engaging for students in those classrooms. I rely on Metz’s typology of developmental versus incorporative instruction and research on classroom discourse to identify instruction that may negatively impact student engagement.
Purpose/Objective/Research Question/Focus of Study: First, how do reports of problem behavior from teachers and administrators in predominantly Black schools differ from those in integrated and non-Black schools? Second, how does the prevalence of developmental instruction vary across schools with different racial compositions?
Research Design: This study uses the 2003–2004 Schools and Staffing Survey data to analyze reports of problem behavior from teachers and administrators. Logistic regression models are used to provide estimates of the prevalence of behavioral problems, adjusting for academic performance and socioeconomic status of the student body. The Chicago School Study (CSS) and the Partnership for Literacy Study (Partnership) data are then used to investigate the prevalence of developmental instruction. The CSS data contain student and teacher reports of the incorporation of student ideas into instruction. In the Partnership data, time summary statistics of observational records of teachers’ instructional activities (e.g., lecture, discussion, seatwork) and question property statistics from a coding of classroom discourse are presented.
Conclusions/Recommendations: Consistent with prior research, teachers are much more likely to report incidences of problem behavior in predominantly Black schools. Consequently, the instructional environment in predominantly Black schools and classrooms is tailored somewhat to reduce classroom disruptions and maintain an orderly environment. Specifically, the result is less interactive discourse and more seatwork. However, the differences in teachers’ instructional approach are relatively modest; there is no “crisis” of authority. Further research is needed on the effects of segregation on social relations in schools and how classroom instruction is affected.
In understanding the effects of school racial composition, educational researchers have not given adequate attention to the nature of classroom instruction in predominantly Black classrooms and schools. Despite decades of research on the effects of school racial composition on student outcomes, we know relatively little about the instruction that contributes to these outcomes. Perhaps the two most robust findings associated with instruction are research demonstrating the difficulty of staffing predominantly Black schools in underfunded urban areas (Guarino, Santibanez, & Daley, 2006), and studies demonstrating that course-taking in predominantly Black schools is quite rigorous given student achievement levels (Kelly, 2009; Lucas & Gamoran, 2002). However, these studies do not show how racial composition affects actual classroom instruction.
One dimension of instruction that may differ across schools and classrooms with different racial compositions is teachers approach to maintaining classroom order. A teachers perspective on authority is a central element of his or her pedagogy (Metz, 1978). Is there a crisis of authority in predominantly Black schools? If so, it may impact the nature of classroom instruction. Teachers in predominantly Black schools may become preoccupied with maintaining classroom order and minimizing classroom disruptions, which could lead inadvertently to less engaging instruction. Importantly, these social processes may have little to do with race per se, stemming instead from differences associated with race, such as students levels of achievement or socioeconomic status. However, that would not diminish the relevance of a crisis in authority for those who teach at predominantly Black schools, or for the students who attend them. Moreover, the initial problem, the concentration of students with problem behaviors in certain schools, is a process driven by residential and school segregation, and that social process is heavily impacted by race (Quillian, 2002). As other research shows, and will be reiterated here, predominantly Black schools do have poor behavioral climates. My foremost concern in this analysis is not what generates problem behavior, but what effect it has on teachers approach to instruction.
The concern that a crisis of authority in predominantly Black schools may lead to less engaging instruction is examined here using teacher and principal reports from the 20032004 Schools and Staffing Survey, observational data from the Partnership for Literacy Study, and student and teacher reports of instruction from the Chicago School Study. These studies were selected for this analysis because they each have unique strengths, and unfortunately, no single database exists that can adequately investigate this issue at the level of individual teachers. Although it is clear that many predominantly Black schools have poor behavioral climates, teachers responses to authority problems appear to be relatively modest, not a wholesale change in instructional approach.
AUTHORITY IN THE CLASSROOM AND TEACHERS APPROACH TO INSTRUCTION
In this analysis, I draw on Metzs (1978) conceptual framework of teachers perspectives on authority relations in the classroom, which are described as being primarily incorporative on the one hand, and developmental on the other. Metzs ideal types are closely related to Cubans (1993) investigation of the prevalence of student-centered versus teacher-centered instruction over the 20th century. For incorporative or teacher-centered teachers, the task of teaching entails transmitting an existing body of knowledge and skills to students. There are two important and interrelated dimensions to this perspective. First, the incorporative teacher believes that the authority to make decisions on both what a student is to learn and how it is to be learned resides solely with the teacher. Second, in selecting a method and style of instruction, incorporative teachers emphasize classroom order as an important instrumental goal because an orderly classroom is a necessary prerequisite to achieving coverage of material. Incorporative instruction is often seen as traditional, evoking images of orderly rows of students completing worksheets or taking notes on a lecture.
For teachers who adopt a developmental or student-centered perspective, fostering student engagement is the fundamental challenge in a classroom. Rather than focusing on content coverage and maintaining order, developmental teachers direct their energies toward cultivating interest, concentration, and effort under the assumption that students must be engaged for achievement growth to occur (Finn, 1989; Fredericks, Blumenfeld, & Paris, 2004; Newmann, 1992; Shernoff, Csikszentmihalyi, Schneider, & Shernoff, 2003). For teachers who strive to connect children to school and learning, their goals in the classroom are not strictly cognitive in nature; what they find exciting is the prospect of inducing positive attitudes among their students toward school or toward a particular branch of learning (Lortie, 1975, p. 114).
For the developmental teacher, teaching is anything but straightforward. Students are presumed to be eager to learn, but teachers have to compete for the attention of young children and adolescents, among whom engagement can never be taken for granted. The developmental teacher cedes some authority to students, allowing them to have input into what will be learned and tailoring material to students in hopes that they will engage the material. An orderly classroom is good, but not if it is achieved at the expense of engagement. Ideally, the developmental teacher sees moral development and its expression in classroom behavior as being self-directed; you can force a student to be quiet, but you cant force him or her to be quiet and engage the material.
How do we know developmental or incorporative instruction when we see it? The typology of instruction itself refers to abstract ideas about authority and classroom instruction, but specific instructional techniques can often be identified as developmental. For example, Cuban (1993) looked for several indicators of student-centered instruction that were readily observable, including the ratio of teacher talk to student talk, the use of varied instructional materials as opposed to reliance on textbooks, freedom of movement, and arrangement of classroom furniture. In a classroom with a higher ratio of student-to-teacher talk, for example, student ideas are likely to have a greater influence on classroom instruction.
INCORPORATIVE INSTRUCTION AS A RESPONSE TO A CRISIS OF AUTHORITY
In Metzs (1978) study, a relationship existed between the student body composition of the school during the teachers formative early years in the classroom, and to a greater extent, between the track level of a teachers students, and his or her instruction in the classroom. Low-track students are often inattentive, withdrawn, and disruptive (Carbonaro, 2005; Eder, 1981; Felmlee & Eder, 1983; Gamoran, Nystrand, Berends, & Lepore, 1995; Oakes, 1985; Smerdon, 1999). Thus, observation of instruction in differently tracked classrooms provides a good example of the relationship between crises of authority in the classroom and adoption of incorporative teaching strategies. A parallel process may unfold in predominantly Black schools where teachers encounter low-achieving and/or disruptive students.
The teachers whom Metz observed stressed the element of control in the design of classroom instruction (Metz, 1978). Teachers avoided whole-class forms of instruction with their low-track studentssuch as discussion, during which disruptive students could quickly get out of controlfavoring worksheets and other individual forms of seatwork. The teachers emphasized the desirable aspects of such an approach; the emphasis on individual written assignments allowed those students who wanted to work the chance to do so, isolating the negative effects of disruptive students. Metzs findings were particularly striking because these accommodations for misbehavior were made not only by incorporative teachers but also by teachers whose philosophy and instructional approach in other classes was developmental. Although it does not speak directly to the context of predominantly Black schools, Metzs analysis, and studies of instruction in tracked classrooms (Caughlan & Kelly, 2004; Gamoran et al., 1995; Oakes, 1985), illustrates the dramatic effect that students have on teachers instructional approach.
NEGATIVE EFFECTS OF OVERLY INCORPORATIVE INSTRUCTION
Setting aside the rare permissive or authoritarian teacher, there is no conclusive evidence demonstrating that an incorporative or developmental approach is associated with higher levels of achievement growth. To the contrary, research on teaching stresses the diversity of approaches taken by successful teachers. However, research does suggest two possible benefits of developmental instruction: fostering intellectual initiative (Silberman, 1970), and increasing student engagement (Metz, 1978). If developmental instruction is entirely absent, engagement and students long-term intellectual development may suffer.
If the children do not embrace a given educative goal, simple insistence is often a poor strategy of persuasion. In this circumstance, the teacher must work first upon their desire to learn. It is helpful ifin developmental fashionhe makes his goals more general and more various, while spending much of his energy in experimenting with strategies to lure the students into interest in the undertaking. (Metz, 1978, pp. 251252)
In this passage from the concluding paragraph of Classrooms and Corridors, Metz articulated the hypothesis that developmental instruction can foster student engagement. Perhaps the best evidence on the link between developmental instruction and student engagement comes from research on dialogic instruction, a form of developmental instruction that emphasizes student-centered classroom discourse. Gamoran and Nystrand (1992) distinguished between forms of classroom discourse that elicit substantive, as opposed to merely procedural, engagement. Their research has since been replicated in two major studies involving hundreds of English and language arts classrooms (Applebee, Langer, Nystrand, & Gamoran, 2003; Langer, Applebee, & Nystrand, 2005), and the link between a teachers approach to classroom discourse and substantive engagement among students is well established.
Although research on dialogic instruction has focused primarily on English and language arts classrooms, studies using the experience sampling method (ESM) have investigated the link between instruction and engagement across a full range of classroom subject matters. The ESM approach measures students affective engagement (interest, concentration, enjoyment, and so on) during classroom instruction by randomly prompting students for feedback throughout the school day (Csikszentmihalyi & Schneider, 2000).1 ESM research confirms that classroom instruction that is interactive in nature, including classroom discussion and exchanges between teachers and students and among students in the context of groupwork, is more engaging than noninteractive forms of instruction (Shernoff et al., 2003; Uekawa, Borman, & Lee, 2007). Unfortunately, researchers have found that engaging, student-centered instruction occurs only infrequently. In math and science classrooms, Borman and associates (2005) found that less than 20% of instructional activities were student centered. In middle school English classrooms, Nystrand (1997) found that less than a minute of genuine discussion occurred during the average class session.
Even when students are not at risk of being disengaged, developmental instruction might be related to an important educational goal: fostering intellectual initiative. Intellectual initiative is the positive mirror-image of docility, a prime player in Silbermans (1970) influential Crisis in the Classroom. According to Silberman, American schools are overly preoccupied with order and control, which leads to intellectual docility in students. Strict adherence to lessons and timetables; an emphasis on procedure and order rather than ideas, interest, and learning; and a lack of opportunity for independence combine to produce students who are good at following orders and completing assignments but who lack independent thinking skills and the drive to produce knowledge on their own. Connecting docility to teachers incorporative instructional approach, Silberman stated, Whatever rhetoric they may subscribe to, most schools in practice define education as something teachers do to or for students, not something students do to and for themselves, with a teachers assistance (p. 135).2 Empirical research on this topic is lacking, perhaps because of the abstract nature of the causal argument. However, similar connections between teachers instructional approach and docility appear in work by Goodman (1956/1983) and Sizer (1984).
THE BEHAVIORAL CLIMATE IN PREDOMINANTLY BLACK SCHOOLS
Early research (Fordham & Ogbu, 1986; Gibson, 1987; Ogbu, 1990) found that race/ethnic identities can be important determinants of students attachment to school and adoption of proschool behavior. Black students levels of attachment and engagement in particular have been the subject of numerous studies, both qualitative and quantitative. The overwhelming majority of the evidence shows no association between Black racial identity per se and negative alignments to school (Ainsworth-Darnell & Downey, 1998; Carter, 2005; Horvat & Lewis, 2003; Kelly, 2008; MacLeod, 1987; Morgan & Mehta, 2004; OConner, 1997; Ogbu & Simons, 1994; Tyson, 2003; Tyson, Darity, & Castellino, 2005), although teachers do perceive Black students as putting forth less effort (Ainsworth-Darnell & Downey). However, even if there is no direct effect of minority status in and of itself on student behavior, low levels of achievement (National Center for Education Statistics [NCES], 2003) may lead Black students to hold antischool attitudes and to be more likely to exhibit disruptive or inattentive behavior. Social-psychological theories of achievement motivation predict that students who enter school with weak academic skills, regardless of their racial or ethnic identity, are at risk of becoming disengaged in the competitive atmosphere of schools.
Explanations of achievement motivation based on self-worth (Covington & Berry, 1976) and self-efficacy (Bandura, 1977) stress that perceptions of ability are an important determinant of student engagement. Students with low self-worth or self-efficacy on a task are more likely to engage in failure-avoiding behaviorsin particular, avoiding effort. When little effort is exerted, students can maintain the belief that they could potentially succeed at a task, or at least no evidence to the contrary is publicly presented. Students who enter class with weaker skills than their peers have an incentive, in the short-term, not to participate in socially risky tasks that would further solidify their identity as poor students. Over time, a reciprocal process of disengagement, further deterioration in levels of self-efficacy and self-worth, and reduced achievement growth unfolds. Low-achieving students are less likely to be engaged than high-achieving students (Ainsworth-Darnell & Downey, 1998; Finn, Pannozzo, & Voelkl, 1995; Roeser, Strobel, & Quihuis, 2002; Shernoff et al., 2003; Valeski & Stipek, 2001; Voelkl, 1997). These well-documented social-psychological processes put minority students, who enter school with lower achievement levels than Whites, at a disadvantage, setting the stage for behavioral problems as students progress through school.
Beyond competency-based issues, an additional problem for many low-achieving students, especially in high school, is the uncertain connection between success in school and success in the world of work. Most American schools lack a clear articulation between school and work, and low-track students level of effort suggests that they perceive few negative consequences of low achievement in high school (Rosenbaum, 2001). It is difficult to reach conclusions about students motives, but Schwartz (1981) believed that she observed the link between extrinsic rewards, or absence thereof, and the motivation to be well behaved in the middle school grades. Schwartz found that high-achieving students in high-track classes were better behaved than low-track students, but only in their academic courses for which grades were important in maintaining their elite position.
A PREOCCUPATION WITH BLACK STUDENTS BEHAVIOR?
Even if Black students are frequently disengaged, inattentive, or even disruptive, it is also possible that teachers perceptions of Black students behavior are biased, or that teachers are especially preoccupied with classroom order in predominantly Black schools. Tyson (2003) investigated the behavioral climate of two all-Black schools, one public school and one Black independent school. Black independent schools are schools founded by Black educators specifically to meet the educational needs of Black students. She observed a strong emphasis on good behavior and deportment, with relatively mundane infractions (e.g., students talking out of turn, moving about the class, and so on) taking on a high level of significance among the staff. Tyson traced the preoccupation with orderly classrooms to more than the common incorporative stance that classroom order is an essential part of the instructional process. Staff at the schools she observed articulated, both to her and to the students involved in her classroom observations, that because of racial inequality beyond the walls of the school, behavioral conformity is especially important to the success of Black students (see also Ferguson, 1995). Findings in the present analysis are consistent with Tysons hypothesis, yet, without a large body of research comparing observational data on student behavior with teacher perceptions of behavior, it is difficult to say how widespread or powerful the phenomenon may be that Tyson and others observed. My primary concern in this analysis, however, is not whether teachers concerns with student behavior in predominantly Black schools are justified, but what effect their concerns might have on classroom instruction.
DATA AND METHODS
In this study, I analyzed three sources of data: the 20032004 Schools and Staffing Survey (SASS), the Chicago School Study (CSS), and the Partnership for Literacy Study (Partnership).3 In the first part of the analysis, I used the SASS data to investigate teachers perceptions of student behavior in predominantly Black and non-Black/integrated schools. In the second part of the analysis, I used the CSS and Partnership data to compare instructional practices across schools with differing racial composition. The SASS data were useful for addressing teachers perceptions of student behavior because (1) the SASS questionnaires contain a number of items pertaining to this issue, and (2) the data are nationally representative.
The CSS data contain a set of items concerning developmental instruction from both teachers and studentsin the school as a whole and in specific subjectswhich correspond well with Metzs conception of instruction. Because the data were drawn from Chicago, where segregation rates are quite high (Lewis Mumford Center, 2002), the data are also well suited to investigating differences associated with racial composition. A weakness of the CSS data is the relatively low response rates at both the school level and among teachers and students within schools, which may lead to biased estimates.
The Partnership data constitute a unique body of observational data that measure two dimensions of classroom instruction related to authority relations within the classroom: (1) teachers use of different instructional activities, and (2) patterns of classroom discourse. The Partnership data are not nationally representative, however, and because they were part of a professional development program, the classroom instruction was more dialogic than that observed in previous National Research Center on English Learning and Achievement (CELA) studies. However, we found that teachers in both the urban predominantly Black schools and other schools responded similarly to the professional development (Langer et al., 2005), and I believe that the comparisons provided here are representative of the way in which teachers respond to teaching in schools with different racial compositions. In other words, the specific values of the statistics on classroom discourse and instructional activities are not representative, but the comparisons among teachers in different schools are nevertheless informative. The Partnership data contain a larger number of classrooms in a diverse array of schools than in previous CELA research and are well suited to investigating the effects of racial composition.4
SCHOOLS AND STAFFING SURVEY
I present descriptive findings from the teacher and principal questionnaires in the 20032004 SASS public school data files. Although the NCES compiles many useful reports using the SASS data, no report is available that summarizes school behavioral climates or instructional practices associated with racial composition.5
CHICAGO SCHOOL STUDY
The CSS administered questionnaires in the spring of 2001 to all participating public school students and teachers in the Chicago Public Schools. Individual principals determined participation, and the study included about 75% of all schools in 2001. Within schools in 2001, about 59% of elementary students and 49% of high school students in the public school district responded. The data reported here summarize reports of developmental instruction from more than 90,000 students and 9,000 teachers in 580 elementary, middle, and high schools.
Four items on the teacher survey indicate a developmental approach to instruction, taken from a set of questions concerning teachers general instructional strategy. Dev. A is the unstandardized scale of the following items (" = .67):
[How often do you]
Relate the subject matter to students experience and interests.
Have students brainstorm ideas for written work.
Have students discuss and debate ideas for more than half a period.
Each on a scale of 15 (1= never 5 = almost every day)
[For about what percent of the lessons you have taught is the following statement true?]
The lessons had students explaining to you or to their classmates how the topic relates to their personal experiences or to a problem in the contemporary world.
On a scale of 16 (1 = none 6 = 76 to 100 percent)
Indicators of developmental instruction from the student survey come from three sections: questions about the students teachers in general, and questions about instruction in English and language arts, and mathematics.
Dev. B is an unstandardized scale that captures the students report of developmental instruction in the school as a whole (" = .65). It consists of four items, two concerning the teaching in the school as a whole, and the average of two identical questions asked in both the English and mathematics sections. In the second half of the survey, half of the students completed the English and language arts section, and half completed the mathematics section. A small number completed both sections accidentally, in which case Dev. B uses items from the first section (English). The questions are:
[How much do you agree with the following?]
My teachers dont care what I think. (reverse coded)
My teachers will always listen to students ideas.
On a scale of 14 (1 = strongly disagree, 4 = strongly agree)
And the average response in English and math to the question:
Really listens to what I have to say.
Relates this subject to my personal interests.
On a scale of 14 (1 = strongly disagree, 4 = strongly agree)
Dev. C measures developmental instruction with a set of items specific to instruction in English and language arts (" = .73):
[In your English class this year, how often do you do the following?]
Apply what you have learned in English class to situations outside of school.
Explain your ideas to the teacher or other students.
Participate in a debate in your class.
Review or edit another students writing.
On a scale of 15 (1 = never, 5 = every day)
Dev. D measures developmental instruction with a set of items specific to instruction in mathematics (" = .71):
[In your math class this year, how often do you do the following?]
Explain how you solved a problem to the class.
Write math problems for other students to solve.
Discuss possible solutions to problems with other students.
Discuss your ideas about math with the teacher or other students.
On a scale of 15 (1 = never, 5 = every day)
PARTNERSHIP FOR LITERACY STUDY
The Partnership collected data on middle school English and language arts teachers in Wisconsin and New York state over a 2-year period (20012003). These analyses relied on 398 observations of classroom instruction in 21 predominantly Black and 76 non-Black/integrated Partnership classrooms for which observational data were available. Only regular-track and untracked (heterogeneous ability group) classrooms were sampled in the Partnership. The classrooms were evenly split between seventh and eighth grades, and many of the classrooms were racially homogenous; a large number had predominantly White or Black student populations.
Partnership classrooms were observed twice each in the fall and spring on adjacent days when teachers were conducting literature lessons. Researchers videotaped each classroom observation and recorded instructional properties using the CLASS software (CLASS 4.24), a real-time laptop-based data collection system. Files were later edited in the laboratory using the videotapes to improve coding consistency.
The Partnership data are analyzed here at two levels of analysis. Time-summary statistics on use of different instructional activities (seatwork, lecture, Q&A, discussion, and so on) are presented at the level of the observation. Properties of classroom discourse are summarized at the level of the individual question. The CLASS program recorded all substantive instructional questions during Q&A, for a total of 25,573 questions in this analysis. The properties of classroom discourse reported in this analysis are briefly summarized as follows:
Source of question (teacher or student)
Authenticity (Was the question a test question? In other words, did it have a predetermined answer, or was it authentic?)
Uptake (Did the question incorporate a students previous response? If so, was the uptake genuine uptake, or was it test uptake, with the students previous response entirely predictable?)
Cognitive level (Reporting versus thinking. Cognitive level distinguishes recitations or reports of what happened, with unknown information such as an analysis, generalization, or speculation based on the text.)
These properties reflect differences in the extent of dialogic instruction within classrooms, when teachers facilitate substantive engagement by relinquishing authority to students during Q&A sessions (Applebee et al., 2003; Nystrand, 1997). Further information on the coding of instructional activities, the typology of classroom instructional activities used in the Partnership, and the sample of schools and classrooms used are available in Langer et al. (2005).
In this analysis, I present descriptive findings on the prevalence of developmental instruction in predominantly Black as opposed to integrated/non-Black schools. Because of the large amount of data compared in many of the tables, tests of statistical significance (t tests) usually reveal significant differences. To evaluate the magnitude of differences across schools with different racial compositions, decomposition of variance statistics are presented where necessary. In the analysis of teacher reports of problem behavior, two sets of statistics are presented: unadjusted teacher reports, and adjusted statistics generated from regression models that include controls for academic climate and students socioeconomic status.
Adjusted values are predicted probabilities from a logistic regression model (except for # of students tardy, which are predicted values from an OLS regression). These predicted probabilities are useful because they can be used to compare differences across schools associated with racial composition in the original metric of a given variable (the level of misbehavior) while controlling for potentially confounding variables. Regression models include variables related to the academic climate of the school and socioeconomic status of students: percent of students graduating (if high school), percent of students attending 4-year colleges after graduation (if high school), percent eligible for free/reduced lunch, and a dummy variable for whether the school meets district performance criteria. Discrete change calculations (Long, 1997) for the predicted probabilities use mean values of graduation, 4-year college attendance, and meeting proficiency, and refer to high schools having 37% free/reduced lunch students.
I chose a simple measure of racial composition: whether the school or classroom is greater than or equal to 50% Black. In the SASS data, this cut point groups together schools that have quite different racial compositions; 815 (10.5%) schools are 50% or more Black, whereas 362 (4.7%) are 85% or more Black. The differences associated with school racial composition in the estimates of teacher reports of behavioral climate in Table 1 are more pronounced when a cut point of 85% is used, but only slightly so: The cut point does not substantially affect the overall conclusion.6 In fact, preliminary analyses revealed that the behavioral climate in schools that are 50%85% Black is quite similar to that of a school that is more than 85% Black. For this reason, I selected an inclusive cut point of 50%. The CSS and Partnership data used in the remainder of the analysis came from the Midwest and Northeast and are highly bimodal on racial composition; therefore, whether a cut point of 50%, 60%, or even 85% is chosen is of little consequence.
Note. Restricted-use data. Unweighted descriptive statistics. Ns rounded to the nearest 10.
a Predominantly Black schools defined as schools serving 50% or more Black students.
The appendix presents histograms of racial composition among the CSS data, Partnership schools and classrooms, and the SASS schools. Estimates of BlackWhite segregation in the CSS data were produced using Reardons Seg command for STATA (Reardon & Firebaugh, 2002). The level of segregation in the CSS data (dissimilarity index of .85) approximates estimates reported elsewhere for Chicago for a similar time period (.84 in 1999; Lewis Mumford Center, 2002). Because many of the schools in the Partnership data were sampled from urban centers or the urban fringe of midsized cities, these data are also quite racially imbalanced; the BlackWhite dissimilarity index is .72, equivalent to a highly segregated urban area.
TEACHER AND PRINCIPAL REPORTS FROM PREDOMINANTLY BLACK SCHOOLS
Table 1 summarizes teacher reports of behavioral problems and faculty support for discipline from the 20032004 SASS in predominantly Black and integrated/non-Black public schools. The findings are consistent with student reports of behavior in NELS:88 and the Add Health data, with Black students having higher levels of misbehavior than non-Black students (Ainsworth-Darnell & Downey, 1998; Crosnoe, Johnson, & Elder, 2004). Teachers are more likely to report a range of behavioral problems, including tardiness, absenteeism, disrespect, lack of control in the classroom, and threatening behavior in predominantly Black schools than in other schools. Almost 3 times the proportion of teachers in predominantly Black schools strongly agree that the level of misbehavior interferes with teaching, as compared with teachers in integrated and non-Black schools. Likewise, nearly twice the proportion indicate that disrespect for teachers happens dailyalmost 40% of the teachers sampled in predominantly Black schools. School racial composition alone accounts for a substantial proportion of the variance in student behavioral problems at the school level. For example, 18.7% of the variance in teacher reports of misbehavior (T0333) lies between schools in the SASS data, about 8.2% of which is accounted for by school racial composition.
White teachers are more likely to perceive behavioral problems among Black students than are Black teachers (Alexander, Entwisle, & Herman, 1999; Downey & Pribesh, 2004; Farkas, Grobe, Sheehan, & Shuan, 1990). For example, on question T0362, concerning disrespect for teachers, 33.9% of Black teachers in predominantly Black schools reported that disrespect occurred daily, compared with 43.2% of non-Black teachersa statistically significant difference (p ³ .0001). Thus, the statistics in Table 1 hide variation in Black and White teachers reports of problem behavior within schools. However, differences associated with the race of the teacher are small compared with the differences associated with the racial composition of the student body.
Table 1 also provides adjusted estimates of teachers reports of problem behavior, controlling for several measures of academic climate and a simple measure of socioeconomic status. The magnitude of the differences in the adjusted estimates is generally less than the raw differences. This reflects the fact that predominantly Black schools are more likely to have a student body of lower socioeconomic status and with lower levels of academic achievement. However, in all cases, teachers in predominantly Black schools still report substantially more problem behavior. This finding is consistent with Tysons hypothesis that teachers in predominantly Black schools are preoccupied with student behavior. However, without detailed observation of actual behavior, this conclusion is only speculative. The most important point is simply that teachers in predominantly Black schools, for whatever reason, perceive substantial problems associated with student behavior.
The findings in Table 1 are reinforced by reports from principals. Table 2 reveals that principals in predominantly Black schools are about twice as likely to report that student absenteeism is a serious problem and that disrespect for teachers is a daily occurrence. However, this does not seem to translate into a preoccupation with order and control among principals. Table 3 shows that principals of both integrated/non-Black schools and predominantly Black schools are concerned first and foremost with academic matters. Principals do list good work habits and self-discipline as an important secondary goal, but those leading predominantly Black schools are no more likely to do so than other principals.7
Note. Restricted-use data. Unweighted descriptive statistics. Ns rounded to the nearest 10.
a Predominantly Black schools defined as schools serving 50% or more Black students
Note. Restricted-use data. Unweighted descriptive statistics.
PB= Predominantly Black; Int/Non-B=Integrated/non-Black public schools.
THE NATURE OF INSTRUCTION IN PREDOMINANTLY BLACK CLASSROOMS
The CSS collected useful data on the prevalence of developmental instruction, which are reported in Table 4. The differences between predominantly Black and integrated/non-Black schools are not particularly striking. On three of the four measures, predominantly Black schools reported more, not less, developmental instruction. If teachers tailor their instruction to maintain classroom order, it does not surface in student or teacher reports of developmental instruction. Note, however, that although teachers in predominantly Black schools are just as likely to place student ideas at the center of instruction as teachers in other schools, students did not report high levels of developmental instruction in either case. For example, the mean values of Dev A (3.65, 3.68) indicate that the average teacher engages in such practices as relating the subject matter to students experience and interests somewhere between once or twice a month and once a week. Proponents of developmental instruction will find the statistics reported in Table 4 disheartening.
Note. I also examined elementary/middle and high schools separately. Racial composition differences were the same across school levels.
a t test: Diff, -.031, p > |t| = 0.0843.
b Because of the large number of cases, all differences in student-reported scales are sig. at the .0000 level.
The data from the CSS are relatively abstract, with students and teachers reporting their overall impression of developmental instruction. The Partnership data provide a closer look at what happens in classrooms. Table 5 provides time summary statistics for the most common instructional activities within English and language arts classrooms. Consistent with the data on students behavior in Tables 13, predominantly Black classrooms had higher rates of classroom interruptions, and more time was spent disciplining students. Over the 317 observations in predominantly White and integrated classrooms, instances of discipline occurred in only 34 of those observations. In contrast, discipline occurred in almost half (35 of 81) of the observations in predominantly Black classrooms. The racial composition dichotomy accounts for 23% of the variance in time spent in disciplinary activities.8
Note. Columns do not total to 3,600 (60-minute class) because of other activity types (student presentation, guest speakers, and so on). In all cases, data collections were supposed to be scheduled for nontest days. Thus, these estimates of time use do not provide accurate information on the prevalence of that activity.
*p < .05. **p < .01. ***p < .001. (t test)
Beyond the differences in discipline, three other important differences emerged: Predominantly Black classrooms were more likely to engage in seatwork and reading aloud, and less likely to engage in question-and-answer sessions. In middle school English and language arts classrooms, Q&A sessions constituted a substantial element of instruction. The CELA coding scheme distinguished between substantive discourse, and procedures and directions about classroom scheduling, events, and so on. Substantive instructional discourse can take three forms: lecture, Q&A, and discussion. Q&A represents the most prevalent form of interaction between teacher and students. Thus, in predominantly Black classrooms, lower levels of Q&A indicate less studentteacher interaction. Put another way, there is simply less substantive talk going on. Instead, students are engaged in reading aloud or seatwork. These forms of instruction share two common features: They are noninteractive and highly controlled by the teacher.9
These differences are robust in tests of statistical significance because of the large number of observations. They also represent important substantive differences: 25% less Q&A and almost twice as much seatwork and reading aloud, for example. But the difference is still one of degree in most cases. Although observed infrequently, predominantly Black classrooms were more likely to have class sessions with no time or almost no time devoted to Q&A (7 of 19 sessions with less than 2 minutes of Q&A occurred in predominantly Black classrooms). But the modal class session, regardless of classroom racial composition, contained a balance of interactive discourse in a whole-class setting, as well as less interactive forms of instruction.
In addition to the time-summary statistics, the CLASS program tracks classroom discourse at the level of individual questions. Table 6 reports differences in the nature of classroom discourse across classrooms of differing racial compositions in the Partnership data. Because of the large sample size, many of the differences in Table 6 are technically statistically significant differences. However, with the possible exception of the proportion of questions that are of a high cognitive level, I see few important differences across classroom types. At the level of individual utterances, the nature of classroom discourse is mostly independent of classroom racial composition. Q&A sessions were quite dialogic in the Partnership data as a whole. Given the strong link between properties of discourse during Q&A sessions and the likelihood of discussion occurring (Nystrand, Wu, Gamoran, Zeiser, & Long, 2003), these results help make sense of the prevalence of discussion reported in Table 5. Predominantly Black classrooms have as much discussion as other classrooms do because the teachers in the Partnership frequently used authentic questions and uptake.
Table 6. Question Asking and Answering in Predominantly Black and Integrated/Non-Black Classrooms. Question-Level Data (N= 26,573), Proportions
In Tables 4, 5, and 6 the main difference appears to be simply the amount of talk, or opportunity to speak, rather than the nature of classroom discourse. Thus, although it appears that teachers may have chosen activities that minimize disruptions and allow them to maintain control of the classroom, I cannot characterize the predominantly Black classrooms we observed as wholly less developmental. Again, the difference seems to be more one of degree than of kind.
In this analysis, I used three large-scale databases to investigate behavioral climates and classroom instruction in predominantly Black schools. The Schools and Staffing Survey, a nationally representative database with reports of student behavior from teachers and principals, provides convincing evidence that many predominantly Black schools have poor behavioral climates. This finding is consistent with prior research on student behavior in predominantly Black schools. Authority relations are a central element of a teachers pedagogy, and one possible solution to crises of authority in the classroom is an overly incorporative approach that disengages students and leads to docility.
To investigate teachers instruction in predominantly Black schools, I examined data from the Chicago School Study and the Partnership for Literacy Study. The observational data from the Partnership study is consistent with the teacher and student reports of instruction from the CSS. Teachers in predominantly Black schools do modify their instruction somewhat in response to their students behavior, but the difference is modest. In the Partnership data, there is less talk and less give and take between teachers and students in predominantly Black classrooms. Importantly, though, this did not represent a wholesale change in teachers approach, but a change in emphasis away from activities in which disruptive behavior might quickly spread. Likewise, in the CSS data, teachers and students did not report substantial differences in the prevalence of developmental instruction in schools with differing racial composition. Although many teachers in predominantly Black schools struggle with the behavioral climate in those schools, at least with respect to teachers instructional approach, the behavioral climate does not yet constitute a crisis.
Given the nature of the data presented here, my conclusions about teachers instruction in predominantly Black schools are still somewhat speculative. Data from the CSS are limited by the teachers and students frame of reference; more stark differences might appear with observational data. The students perception that their ideas have a place in the classroom is important, but the CSS student measures might be influenced by some wishful thinking that hides true differences in instruction. The observational data from the Partnership circumvent these problems but may create Hawthorne effects, where the observed instruction is influenced by the presence of the observer and is not representative of actual instruction. These are difficult problems to overcome, although in these data, the results are similar across research method, and that lends support to my conclusion. An alternate approach is to collect data on classroom instruction directly from students, using, for example, the experience sampling method (Csikszentmihalyi & Schneider, 2000). A limitation of that approach however, is that students may be unable to distinguish among methods of instruction that are of the same basic format but that have a much different effect on student engagement and learning (Q&A vs. discussion).
A second limitation concerns the causal link between behavior and instruction. Perhaps students behavior can be traced to teachers instructional approach more so than the other way around. The relationship between instruction and student behavior seems to be a reciprocal process that unfolds iteratively. Thus, it is difficult to infer teachers motives in the classroom, such as why a teacher adopts a particular instructional approach. On the other hand, when asked, teachers will readily explain why they do what they do in the classroom, and the relationship between their instructional approach and student behavior seems quite clear (Metz, 1978). Finally, an additional shortcoming of the Partnership data is that low-track classrooms are omitted from the analysis, and it is almost certainly these classrooms where teachers struggle the most to overcome behavioral problems.
Research suggests that school segregation is on the rise in Americas public schools (Lewis Mumford Center, 2002). Moreover, between-district segregation accounts for a substantial portion of total segregation, and rising segregation, in recent decades (Clotfelter, 2004). White avoidance of many urban areas and legal decisions thwarting desegregation efforts across district lines make this between-district segregation an intractable barrier to school integration. As we seek to understand the effects of segregated schooling on student outcomes, we must certainly consider the most basic indicators of access to educational opportunity, such as per pupil expenditures, teacher quality, Advanced Placement courses, and so on. In addition, researchers must look more closely at the way in which segregation affects the social relationships between teachers and students within schools.
In this analysis, I have investigated the effects of behavioral climate on teachers perspectives on authority, asking if teachers use of developmental instruction is constrained. But further research is needed on how teachers respond to students behavior in predominantly Black schools. Even if the behavioral climate in these schools does not decrease the possibility of engaging instruction, research on teacher attrition suggests that the behavioral climate in some predominantly Black schools places teachers under duress (Kelly, 2004). Rates of attrition are higher in schools serving low-income, low-performing students in urban settings (Guarino et al., 2006). Unfortunately, these risk factors often occur jointly in predominantly Black schools, leading to higher rates of attrition. It would be easy, and a mistake, to minimize the potential problems associated with poor behavioral climates in predominantly Black schools by focusing too narrowly on a particular outcome.
1. The methodology of an ESM study is described in detail and a sample ESM instrument is provided in Csikszentmihalyi and Schneider (2000), pp. 2728, appendix B.
2. Classroom discourse appears as one of the elements of instruction that can reinforce docility in Silbermans (1970) analysis, where classrooms dominated by teacher talk and scripted discourse foster docility.
3. SASS was administered by the National Center for Education Statistics. The CSS was administered by the Consortium on Chicago School Research and the Center for Research on Educational Opportunity. The Partnership was administered by the National Research Center on English Learning and Achievement (CELA).
4. Classrooms are the unit of analysis reported here, but the predominantly Black classrooms were all from predominantly Black schools.
5. But see What Happens in Classrooms? Instructional Practices in Elementary and Secondary Schools (NCES, 1999) for an overview of the SASS findings on this topic without reference to racial composition.
6. For example, 24.5% of teachers in schools that are 85% or more Black strongly agree that the level of misbehavior interferes with teaching, whereas 23.3% strongly agree in schools that are 50% or more Black.
7. Work habits and self-discipline is a more frequently listed third most important goal. In general, teachers place somewhat more emphasis on work habits and discipline than do principals (NCES, 1996).
8. The CLASS program prompts the observer to record the average number of students off-task during different instructional activities. Off-task estimates associated with racial composition are consistent with the time summary statistics on discipline; predominantly Black classrooms had a greater proportion of students coded as off-task in a number of different instructional activities. However, with so much activity to take note of, classroom observers indicated that the off-task prompt was often ignored. Thus, this variable is not coded with the same level of reliability as instructional activities and classroom discourse. 9. The CELA coding scheme differentiated between multiple forms of seatwork on such dimensions as the cognitive level of the seatwork and who organized the task (teacher or student directed). However, almost all the seatwork observed in any classroom was highly structured by the teacher and thus unlikely to be very developmental.
Ainsworth-Darnell, J. W., & Downey, D. B. (1998). Assessing racial/ethnic differences in school performance. American Sociological Review, 63, 536553.
Alexander, K. L., Entwisle, D. R., & Herman, R. (1999). In the eye of the beholder: Parents and teachers ratings of childrens behavioral style. In C. L. Shehan (Ed.), Contemporary perspectives on family research: Vol. 1. Revisioning children as active agents of family life (pp. 225245). Greenwich, CT: JAI Press.
Applebee, A. N., Langer, J. A., Nystrand, M., & Gamoran, A. (2003). Testing a framework for effective English instruction: A study of relationships between achievement gain and instructional approaches in middle and high school classrooms. American Educational Research Journal, 40, 685730.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191215.
Borman, K. M., & Associates. (2005). Meaningful urban education reform: Confronting the learning crisis in mathematics and science. Albany: State University of New York Press.
Carbonaro. W. (2005). Tracking, students effort, and academic achievement. Sociology of Education, 78, 2749.
Carter, P. L. (2005). Keepin it real: School success beyond Black and White. New York: Oxford University Press.
Caughlan, S., & Kelly, S. (2004). Bridging methodological gaps: Instructional and institutional effects of tracking in two English classes. Research in the Teaching of English, 39, 2062.
Clotfelter, C. (2004). After Brown: The rise and retreat of school desegregation. Princeton, NJ: Princeton University Press.
Covington, M., & Berry, R. (1976). Self-worth and school learning. New York: Holt, Rinehart and Winston.
Crosnoe, R., Johnson, M. K., & Elder, G. (2004). Intergenerational bonding in school: The behavioral and contextual correlates of student-teacher relationships. Sociology of Education, 77, 6081.
Csikszentmihalyi, M., & Schneider, B. (2000). Becoming adult: How teenagers prepare for the world of work. New York: Basic Books.
Cuban, L. (1993). How teachers taught: Constancy and change in American classrooms 18901990. New York: Teachers College Press.
Downey, D. B., & Pribesh, S. (2004). When race matters: Teachers evaluations of students classroom behaviors. Sociology of Education, 77, 267282.
Eder, D. (1981). Ability grouping as a self-fulfilling prophecy: A micro analysis of teacher-student interaction. Sociology of Education, 54, 151162.
Farkas, G., Grobe, R. P., Sheehan, D., & Shuan, Y. (1990). Cultural resources and school success: Gender, ethnicity, and poverty groups within an urban school district. American Sociological Review, 55, 127142.
Felmlee, D., & Eder, D. (1983). Contextual effects in the classroom: The impact of ability groups on student attention. Sociology of Education, 56, 7787.
Ferguson. A. (1995). Bad boys: Schools and the social construction of Black masculinity. Unpublished doctoral dissertation, University of California, Berkeley.
Finn, J. D. (1989). Withdrawing from school. Review of Educational Research, 59, 117142.
Finn, J. D., Pannozzo, G. M., & Voelkl, K. E. (1995). Disruptive and inattentive-withdrawn behavior and achievement among fourth graders. Elementary School Journal, 95, 421434.
Fordham, S., & Ogbu, J. (1986). Black students school success: Coping with the burden of acting White. The Urban Review, 18, 176206.
Fredericks, J. A., Blumenfeld, P. C., & Paris, P. H. (2004). School engagement: Potential of the concept, state of the evidence. Review of Educational Research, 74, 59109.
Gamoran A., & Nystrand, M. (1992). Taking students seriously. In F. Newmann (Ed.), Student engagement and achievement in American secondary schools (pp. 4061). New York: Teachers College Press.
Gamoran, A., Nystrand, M., Berends, M., & Lepore, P. C. (1995). An organizational analysis of the effects of ability grouping. American Educational Research Journal, 32, 687715.
Gibson, M. A. (1987). The school performance of immigrant minorities: A comparative view. Anthropology & Education Quarterly, 18, 262275.
Goodman, P. (1983). Growing up absurd. New York: Random House. (Original work published 1956)
Guarino, C. M., Santibanez, L., & Daley, G. A. (2006). Teacher recruitment and retention: A review of the recent empirical literature. Review of Educational Research, 76, 173208.
Horvat, E. M., & Lewis, K. S. (2003). Reassessing the burden of acting White: The importance of peer groups in managing academic success. Sociology of Education, 76, 265280.
Kelly, S. (2009). The Black-White Gap in Mathematics Course Taking. Sociology of Education, 82, 47-69.
Kelly, S. (2004). An event history analysis of teacher attrition: Salary, teacher tracking, and socially disadvantaged schools. Journal of Experimental Education, 72, 195220.
Kelly, S. (2008). Race, social class, and student engagement in middle school English classrooms. Social Science Research, 37, 434448.
Langer, J. A., Applebee, A. N., & Nystrand, M. (2005). Final report: National Research Center on English Learning and Achievement. Albany: State University of New York.
Lewis Mumford Center. (2002). Choosing segregation: Racial imbalance in American public schools, 1990-2000. Albany: State University of New York, Lewis Mumford Center for Comparative Urban and Regional Research.
Long, J. S. (1997). Regression models for categorical and limited dependent variables. Thousand Oaks, CA: Sage.
Lortie, D. C. (1975). Schoolteacher: A sociological study. Chicago: University of Chicago Press.
Lucas, S. R., & Gamoran, A. (2002). Track assignment and the Black-White test score gap: Divergent and convergent evidence from 1980 and 1990 sophomores. In T. Loveless (Ed.), Closing the gap: Promising strategies for reducing the achievement gap (pp. 171198). Washington, DC: Brookings Institution.
MacLeod, J. (1987). Aint no makin it: Leveled aspirations in a low-income neighborhood. Boulder, CO: Westview Press.
Metz, M. H. (1978). Classrooms and corridors: The crisis of authority in desegregated secondary schools. Berkeley: University of California Press.
Morgan, S. L., & Mehta, J. D. (2004). Beyond the laboratory: Evaluating the survey evidence for the disidentification explanation of Black-White differences in achievement. Sociology of Education, 77, 82101.
National Center for Education Statistics. (1996). Schools and staffing in the United States: A statistical Profile, 199091. Washington, DC: U.S. Department of Education.
National Center for Education Statistics. (1999). What happens in classrooms? Instructional practices in elementary and secondary schools, 199495. Washington, DC: U.S. Department of Education.
National Center for Education Statistics. (2003). Status and trends in the education of Blacks. Washington, DC: U.S. Department of Education.
Newmann, F. M. (1992). Student engagement and achievement in American secondary schools. New York: Teachers College Press.
Nystrand, M. (1997). Opening dialogue: Understanding the dynamics of language and learning in the English classroom. New York: Teachers College Press.
Nystrand, M., Wu, L. L., Gamoran, A., Zeiser, S., & Long, D. (2003). Questions in time: Investigating the structure and dynamics of unfolding classroom discourse. Discourse Processes, 35, 135196.
Oakes, J. (1985). Keeping track: How schools structure inequality. New Haven, CT: Yale University Press.
OConner, C. (1997). Dispositions toward (collective) struggle and educational resilience in the inner city: A case analysis of six African-American high school students. American Educational Research Journal, 34, 593629.
Ogbu, J. U. (1990). Minority education in comparative perspective. Journal of Negro Education, 59, 4557.
Ogbu, J. U., & Simons, H. D. (1994). Cultural models of school achievement: A quantitative test of Ogbus theory. Washington, DC: Office of Educational Research and Improvement.
Quillian, L. (2002). Why is Black-White residential segregation so persistent? Evidence on three theories from migration data. Social Science Research, 31, 197229.
Reardon, S. F., & Firebaugh, G. (2002). Measures of multigroup segregation. Sociological Methodology, 32, 3367.
Roeser, R. W., Strobel, K. R., & Quihuis, G. (2002). Studying early adolescents academic motivation, social-emotional functioning, and engagement functioning, and engagement in learning: Variable and person-centered approaches. Anxiety, Stress, and Coping, 15, 345368.
Rosenbaum, J. E. (2001). Beyond college for all: Career paths for the forgotten half. New York: Russell Sage.
Schwartz, F. (1981). Supporting or subverting learning: Peer group patterns in four tracked schools. Anthropology and Education Quarterly, 12, 99119.
Shernoff, D. J., Csikszentmihalyi, M., Schneider, B., & Shernoff, E. S. (2003). Student engagement in high school classrooms from the perspective of flow theory. School Psychology Quarterly, 18, 158176.
Silberman, C. E. (1970). Crisis in the classroom. New York: Random House.
Sizer, T. R. (1984). Horaces compromise: The dilemma of the American high school. Boston: Houghton Mifflin.
Smerdon, B. (1999). Engagement and achievement: Differences between African-American and White high school students. Research in Sociology of Education and Socialization, 12, 103134.
Tyson, K. (2003). Notes from the back of the room: Problems and paradoxes in the schooling of young Black students. Sociology of Education, 76, 326343.
Tyson, K., Darity, W., & Castellino, D. (2005). Black adolescents and the dilemmas of high achievement. American Sociological Review, 70, 582605.
Uekawa, K., Borman, K., & Lee, R. (2007). Student engagement in U.S. urban high school mathematics and science classrooms: Findings on social organization, race, and ethnicity. Urban Review, 39, 143.
Valeski, T. N., & Stipek, D. J. (2001). Young childrens feelings about school. Child Development, 72, 11981213.
Voelkl, K. E. (1997). Identification with school. American Journal of Education, 105, 204319.