“The past is never dead. It is not even past.” (Faulkner 1971)

Systematic school wide tracking programs have been on the decline since the 1970s—replaced by subject-by-subject leveled classes and within-class ability grouping—when researchers began to call attention to their negative effects on students in lower levels and tracks (e.g., Eder 1981; Lucas 1999; Oakes 1985). Many educators and researchers believe that ability grouping has been significantly modified since that time. However, while conducting a study of reading instruction in sixth grade English language arts classes, I began to wonder if much had really changed since Oakes’ 1985 book, Keeping Track, in which middle school and high school teachers and students described distinct differences between lower-level classes and middle to high-level classes.

The following excerpt is from an interview conducted with a teacher who was asked to “describe a typical day in your classes”:

Teacher::

Okay. I have a regular class and an honors class. Which one would you like me to go through?

Author::

Are they that different?

Teacher::

Incredibly.

I was interviewing teachers for a study of instruction and materials in middle school language arts classes in a large urban district, in which classes were divided into two levels—honors and regular. Although the interview protocol did not include a single question about differentiation among students or classes, virtually every teacher spoke of sharp distinctions between students in their two levels of classes, as in these examples:

“The regulars, they don’t even bring in their homework, and I’ve given up on them reading at home. Their parents don’t read; they don’t see it modeled.”

“The honors class is wonderful in every respect. They are every teacher’s dream. The regular class is very, very low. They are my most difficult class, and my biggest discipline problems are in that class.”

The dictionary defines “regular” as “customary, usual, or normal” (The American Heritage Dictionary 1982, p. 1041). Yet the teachers I interviewed spoke about the students and instruction in their regular classes in ways that called to mind research from the 1970s and 1980s concerning low-ability groups and low-level classes (e.g., Allington 1983; Gambrell et al. 1981; Gamoran 1986; Goodlad 1984; Rosenbaum 1976; Schafer and Olexa 1971). For example, these studies had found that teachers described such students as having discipline problems, being unmotivated to learn, and having uninvolved parents. Teachers were found to provide inferior instruction and to have modest instructional goals for students when compared to middle and higher-level groups and classes. It appeared the teachers I interviewed and observed were using the term regular as the term low might have been used in the past.

For this research, I conducted an analysis of interviews and observations from 25 sixth grade teachers who taught honors and regular language arts classes, focusing on how they talked about students and instruction in the different class levels, and interpreting their comments in light of historical research on ability grouping and tracking. The findings prompted ideas for further study of ability grouping and implications for practice concerning how class level information is communicated to teachers and parents, how teachers might be prompted to examine more closely their attitudes and practice in ability grouped classes, and how schools might consider alternatives to ability grouping.

Ability Grouping: Definition, History, Characteristics, and Consequences

In this review, I use both the terms ability grouping and tracking to refer to the practice of evaluating and sorting students into categories for the purpose of providing differential instruction within or across classrooms (Goodlad 1985; Lucas 1999; LeTendre et al. 2003). I provide a brief historical account of research on ability grouping followed by a review of research focusing on the characteristics of ability-grouped instruction and the consequences of ability grouping for students, as well as a discussion about the current state of ability grouping.

According to reviews of the history of ability grouping (Barr and Dreeben 1991; Lucas 1999; Oakes 1985), until the early 1900s, students attending secondary schools were almost exclusively from Anglo-Saxon families. By 1920, however, northeastern cities were experiencing a population explosion, mainly consisting of poor, uneducated, and unskilled immigrants from eastern and southern Europe and job-seeking rural youth, later joined by persons of color as southern African-Americans sought employment in the north and Puerto Ricans migrated to their new country of citizenship (Donelan et al. 1994; Nieto 2000). In response to the sudden need to educate unprecedented numbers of students from diverse backgrounds, most cities formed comprehensive schools, which separated students into college preparation and vocational tracks, presumably because of their distinct needs and abilities. Some educators at the time advanced the rationale that the new schools would be the most efficient way to prepare citizens for the industrialized economy, while others asserted that tracking was a form of equal education opportunity designed to meet students’ needs, abilities, and interests, which were openly seen as being based on race, ethnicity, and socioeconomic background. Not surprisingly, the children of immigrants and the poor were more likely to follow a vocational curriculum, which was called the “basic” track, while their middle class, mostly white peers were targeted for college preparation in either the “regular” or “honor” track (Lucas 1999).

In the early 1970s, when the first research about the negative affects of tracking began to appear, school districts in large urban areas like New York, Chicago, and Boston led the way in dismantling schoolwide tracking systems, in which students’ track placement determined all the courses they took, and replaced them with leveled-course systems (Lucas 1999). However, the old track names—basic, regular, and honors—were applied to the leveled courses; thus, according to Lucas, “the foundational element of tracking, the differentiated curriculum, remained” (p. 6). Theoretically, students were free to enroll in different levels of classes in different subjects, such as AP English and remedial mathematics, according to their proficiency in each area. However, in practice this ideal was rarely realized because, due to factors such as peer pressure, scheduling, and historical issues, students often were placed in the same levels of classes for most subjects, resulting in a “hidden in-school stratification system” (Lucas 1999, p. 137). Thus, the end result in many cases was the same as in school wide tracking.

In elementary schools, homogeneous grouping within classrooms was seen as a way to effectively teach students of different achievement and ability levels while avoiding the social and emotional disadvantages associated with tracking (Barker Lunn 1970; Gregory 1984). By mid-century, instruction in low, middle, and high groups was standard operating procedure for elementary reading instruction.

Research on ability grouping was at its height during the 1970s and 1980s. By that time, savvy educators, perhaps attempting to obfuscate concerns about race, SES, and ethnicity (Esposito 1973; Oakes 1985; Rosenthal and Jacobson 1968), offered seemingly logical rationales for the practice, including helping students learn better, promoting equality and higher self concepts, and making teaching and management easier. Researchers studying elementary to high school ability grouping found little support for these claims but instead found “virtually mountains of evidence indicating that homogenous grouping doesn’t help anyone learn better” (Oakes 1985, p. 7), and that the learning of students in lower level groups, classes and tracks was negatively affected by instruction, classroom interactions, and teacher expectations (Allington 1980; Brophy and Good 1970; Esposito 1973; Oakes 1985).

Characteristics of Tracking and Ability Grouping

Studies of tracking in high schools consistently found distinct differences between high- and middle-track classes, which were designed to prepare students for college, and low-track classes, which were geared toward vocational and other non college-bound students (Dreeben and Gamoran 1986; Esposito 1973; Gamoran 1986; Oakes 1985). Students in higher-track classes were more likely to be taught by skilled and experienced teachers who provided fast-paced, engaging instruction in complex texts. Low-track classes, in contrast, focused on basic literacy skills, test preparation, and rote learning within decontextualized reading and writing tasks, and simple, low-difficulty-level texts (Nystrand and Gamoran 1991). Teachers interviewed in Oakes’ book (1985) had differential expectations for students at lower and higher levels. In higher and middle track classes, students were expected to be sophisticated thinkers, active participants, and independent learners. Teachers’ expectations for lower-track students focused on behavior and conformity rather than learning and thinking, and teachers talked openly about students’ negative home lives and characteristics, which were assumed to be the basis of their learning and motivation problems.

In elementary schools, study after study of within-class ability groups found that instruction in low groups, like that in low-track classes, was characterized by slower pacing and more time on decoding and decontextualized skills as opposed to reading for meaning in connected text (Allington 1980; Gambrell et al. 1981). Teachers were more likely to have negative feelings and lower expectations for lower achieving students, to offer them less support, and to communicate their lowered expectations to students (Good and Brophy 1972).

Academic, Social, and Emotional Consequences of Ability Grouping

Ability grouping was found to have important social, emotional, and academic consequences for students. Learning and achievement were positively affected by instructional contexts that included substantive versus procedural engagement, more reading and writing in connected text, more choice and coherence, and a greater degree of student participation, all characteristics of high level but not low level classes (Carbonaro and Gamoran 2002; Nystrand and Gamoran 1991). The differential content and pace of classes was found to affect future achievement as well (Gamoran et al. 1995; Rowan and Miracle 1983). Because lower level classes moved more slowly and covered less material, for example, the same coursework covered in a year of higher level classes was distributed over two or more years, steadily increasing the gap in content covered (Donelan et al. 1994). Wheelock (1992) concluded that, “students at the lower levels move so much more slowly than those at the higher ones that differences that may have been real but not profound in earlier grades become gigantic gaps in terms of achievement, attitudes, and self-esteem” (p. 6). Research also suggested that simply being in low level classes and ability groups compounded and even generated behavioral, attention, and achievement difficulties (Barker Lunn 1970; Eder 1981; Marsh 1984; Rosenbaum 1976).

The importance of placement decisions cannot be overstated, as research found that students tended to stay in high or low level classes and groups year after year (Eder 1981; Ireson and Hallam 1999). However, the accuracy of placements, and thus the idea of true achievement differences in ability grouping, has been questioned on a number of dimensions. Often thought to accurately reflect a student’s academic potential and past achievement, placements decisions were found by many researchers to be questionable, negligible, or erroneous in terms of achievement differences (Donelan et al. 1994; Esposito 1973; Labov 1972).

The Decline of Ability Grouping?

In response to tracking and ability grouping research, researchers and educators called for nonability-grouped structures (Berghoff and Egawa 1991; Crosby and Owens 1993; Radencich and McKay 1995) and drastic reform or eradication of tracking practices. Detracking, in which students are intentionally placed in heterogeneous classes, has been tried with some success (Alvarez and Mehan 2006; Mehan et al. 1996; Oakes et al. 1997; Rubin 2006; Rubin and Noguera 2004; Oakes 2005; Wheelock 1993) but also with much opposition, particularly from middle to upper middle class parents and proponents of gifted education (Kulik and Kulik 1982; Wells and Serna 1996).

By the 1990s, research about ability grouping and tracking had decreased dramatically. Judging from research and reviews written since that time, many educators and researchers believe the lessons learned from research about the damaging effects of ability grouping have been applied in educational practice and that the differential instruction of current times bears no resemblance to the tracking and ability of the past (Barr 1995; LeTendre et al. 2003; Lucas 1999; Rogers 2004 Oakes 2005).

However LeTendre et al. (2003) suggested that the practice of comparing, sorting, and classifying students has not vanished in the U.S. but has merely become a back-burner issue, and a recent resurgence of research and academic debate suggests that ability grouping is again (or still) thriving (Blanchett 2006; Chorzempa and Graham 2006; Orfield and Lee 2004; Watanabe 2006; Yonezawa and Jones 2006).

The purpose of the current study was to examine current forms of ability grouping in middle school language arts classes in light of research of the past. The content area focus was mainly reading instruction. We addressed these questions: (a) How salient is the issue of ability grouping in teachers’ descriptions of students and instruction in honors and regular language arts classes? (b) What is the nature of teachers’ talk about their students in the two classes and how does it compare with research of the 1970s and 1980s?

Method

Setting and Participants

The data for the research was gathered in eight middle schools in a large urban school district in Texas, with 56% of its students living in poverty. The ethnic makeup was 53% Latino, 14% African American, and 33% European American. Four of the eight schools served predominantly minority, low-income areas of the city; two were in mixed ethnic and economic zones, and two were in suburban, upper middle class areas. The sixth grade language arts classes were taught in “blocks”—back-to-back 45-min classes—in which teachers were expected to teach reading, writing, and English grammar. Students were divided into honors and regular classes. There was no “basic” class. Although I did not have any basis for my speculation, I wondered if the reason for removing the basic level was so no students would be in classes with that label. Some schools had further divisions either in separate classes or within their honors or regular classes (e.g., “gifted and talented,” “ESL,” or “academic support”). Special education students attended resource classes for language arts in some schools while other schools followed an inclusion model.

If a student had previously attended a district school, the fifth grade teacher made the placement in honors or regular classes. Otherwise, the school guidance counselor placed the student. However, parent placement requests were almost always granted. In the two schools serving middle income and higher income areas, due to parent requests (personal communication, district literacy coordinator), there were more honors classes than regular classes despite the fact that the dictionary defines honors as “a program of individual advanced study for exceptional students” (p. 636).

Data Gathering

The original purpose of the research was to examine instruction and materials in language arts classes through teacher interviews. The focus was on reading instruction. Of the 41 language arts teachers in the eight schools, 32 (85%) agreed to participate in the study. Two research assistants and I interviewed teachers in their classrooms after school or during planning times for 45–60 min and recorded our thoughts and impressions immediately after the interview. Even though there were no questions about it, the issue of differentiation among honors and regular classes arose repeatedly as teachers described their classes.

Interviews of the 25 teachers who taught at least one regular and one honors class were the major data source for the current study. In this group, there were 21 women (15 European American, 4 Latina American, and 2 African American) and 4 men (3 European American and 1 African American). The average number of years of experience for the teachers was 4.5, with a range of 1–23 years. We conducted observations in both the regular and honors classrooms of 9 of these 25 teachers. Two different research assistants helped to analyze the data.

Researcher Positionality

I am a European American woman from a middle class background. I formerly taught elementary school and middle school, and I am now a professor of literacy education at a large university in the southwest. My experiences as a teacher and a parent, as well as my participation in this study and my reading of research, have contributed to my views on ability grouping, which have fluctuated through the years but are now negative. I taught in the 1970s and 1980s in mostly low-income, ethnically diverse, rural communities where students were sorted into self-contained classrooms by achievement. First because I was an inexperienced teacher, and later by choice, I usually taught the lower-ranked classes. There were times I wished for what seemed like an easier job and more manageable, motivated students, but I mostly enjoyed teaching my students and resented that they were often looked down upon and that the higher-achieving students were afforded more privileges and positive expectations from others in the schools.

When my own children were in school, I experienced both ends of the ability-grouping spectrum. My son, my older child, liked school primarily for social reasons, and was considered an “average” student, although he had many intellectual gifts not recognized by school. He was placed in a regular language arts class in sixth grade when we moved to a new town. Although I was a former teacher and university professor with plenty of cultural and social capital, I did not understand nor did I investigate the district’s system to know the socially constructed meaning of regular nor that I could request a different placement. Through 3 years of middle school, my son suffered a curriculum that included mainly test preparation, basic skills instruction, and low expectations. In high school, I requested that he be moved to “enriched” classes, the middle level, which he found more challenging and motivating, but the rest of school didn’t work well for him, and I feel guilty about the possible contribution of his middle school years to that. In contrast, my daughter received an outstanding education in honors, Advanced Placement, and magnet classes that helped prepare her for college. My appreciation for my daughter’s education is great, although I often felt guilty because so many students did not have access to what she had. My resentment for what my son did not have is greater.

At the same time, I can identify with the reluctance of some parents who are concerned about heterogeneous environments for their children; Goodlad’s (1984) assertion is still true today, “The practice [of ability grouping] has been reinforced from those outside the school by those who believe that able students are held back by slower ones when all work together in the same class.” (p. 151).

After conducting the research for this paper, especially listening to the teachers’ comments, watching their interactions with their students, and studying research about ability grouping, my personal and professional belief is that ability grouping is a harmful practice, especially to students placed in lower tracks and groups, and I am committed to exploring alternatives that do not compromise the education of any student.

All the research assistants were doctoral students in literacy education. Two were former elementary teachers. None of the four had read much research or had strong feelings about ability grouping or tracking before the study began. After participating in the study all were against it for middle and high school, but one of the elementary teachers was doubtful about primary reading being taught without ability groups.

Analysis

The 25 interviews of teachers who taught both honors and regular classes were the primary data source, while the 18 observations provided context. We used three sources to guide the analysis of interviews and field notes (Erlandson et al. 1993; Emerson et al. 1995; Graue and Walsh 1998). Coding of the interviews took place in several stages. First, I read through the interview transcripts and identified content units. For example, if a teacher talked about three topics in answer to one question, I divided that response into three content units. If responses to several questions focused on the same topic, I combined those into one unit. Next, I marked the content units in which teachers mentioned ability grouping or differentiation among students in different classes and calculated the percentage for each interview.

We divided the transcripts among the three of us and analyzed them using constant comparative analysis. The goal of this phase was to capture recurring and salient patterns in the data consisting of both comparable examples and examples of variation, and to begin developing codes (words or phrases) that described the data (Graue and Walsh 1998). We each compared data from a first transcript with data from a different transcript, paying close attention to data units that were interpreted as relating to the same or a similar code. We went through the observations in the same way. Continuing in this way through the data, each researcher made notes about the characteristics of the codes. When the process was finished, each researcher had a list of 20–40 codes for the data set, as well as thoughts about how these could be combined into categories. After each researcher had completed independent analysis of the set of transcripts and observations, we met together to discuss the codes.

Our joint lists included more than 50 codes, many of which overlapped. Each of us talked through our lists across two research meetings, noting both how they could be grouped into broader categories and how these categories could be defined. Returning to the data, we each looked for examples of the categories, noting questions for clarification. At a third analysis meeting, we discussed examples and questions and agreed on final categories, which were: (a) differences in instruction and environment between honors and regular classrooms; (a) students’ behavior, work habits, and interest in learning; (c) differential teacher expectations; and (d) causes for regular students’ “problems.”

Findings

The close examination of teacher interviews supported initial impressions that ability grouping was a salient topic in the data. Despite no questions or prompts that mentioned levels or differences among students or classes, the majority of teachers drew sharp distinctions between students and instruction in their regular and honors classes, and this talk pervaded the interviews. An average of 35% of the content units focused on student and instructional differences between honors and regular classes, with a range across teachers of 8–65%. A handful of teachers in this study, 4 of the 25, supported the idea that homogeneous grouping does not have to mean deficient instruction for students in lower level classes. Two of these teachers taught in a high-poverty area and two in a mixed-economic area. However, the most negative characteristics of ability grouping found in past research were alive and well in the words and minds of most of the teachers who taught ability grouped language arts classes.

According to whether the students were in regular or honors classes, teachers pointed out what they considered striking differences in work habits, behavior, ability, and interest in learning and attributed these differences partly to factors in students’ homes. They also freely talked about lowered expectations and watered-down instruction for students in regular classes, and clearly considered honors classes the “plum” assignment. Much of the content of this talk was troublingly similar to earlier research on junior high through high school tracking and elementary school ability grouping (Brophy and Good 1970; Dreeben and Barr 1988; Goodlad 1984; Oakes 1985; Rowan and Miracle 1983).

The teachers rarely mentioned individual students; instead, they drew broad generalizations, often in the absence of evidence. They described students as having certain characteristics based on their class placement, and these characteristics seemed to be considered static and immutable, as illustrated by the use of phrases like “these kids,” “my regulars,” and “their parents.” Such talk is reminiscent of trait theory (Gutiérrez and Rogoff 2003; McCarthey 2001), in which individuals are assumed to carry common interests, skills, experiences, and characteristics simply by virtue of their membership in a group. Gutiérrez and Rogoff’s (2003) description of instruction based on assumed traits of cultural groups aptly depicts this aspect of the findings from the teacher interviews:

Approaches that accommodate instructional practice to group styles treat what is “known” about a group as applying to all individuals in the group. This makes it more likely that groups will be treated as homogeneous, with fixed characteristics carried by the collection of individuals that comprise the group. (p. 20)

In teachers’ descriptions, then, students came to be defined in terms of the group in which they were placed, and in the minds of most of the teachers, the negative characteristics of regular students were already firmly established by the ripe old age of 12. In the next sections, I present the findings interpreted through the lens of past research in ability grouping.

Differences in Instruction and Classroom Environment

According to the sixth grade teachers’ descriptions, instructional goals, methods, materials, classroom interaction, discourse, and environment were widely divergent for the regular and honors classes. Generally, as in previous research, teachers provided honors classes with creative instruction around sophisticated materials, with the freedom to work interactively and take responsibility for their own behavior. In regular classes, the focus was on isolated and basic skills, along with reading in decontextualized text, using simple materials within a rigidly controlled setting. The common rationale for these differences echoed reasoning from the earliest instantiations of ability grouping, that teachers were supplying what was best for students and what they needed, as one teacher explained.

It’s actually a good environment for them [students in the regular class] because instead of being in less structured classes where they were perpetually getting into trouble, they are in a very structured environment. We do a lot of seatwork and there’s not a lot of talking, not a lot of discussion. Which is unfortunate, but it’s better for them because now they are getting the work in. And a lot of them really enjoy—I don’t know if they enjoy the class—but they enjoy the fact that they are not spending all day in the principal’s office.

This teachers’ rationale for the differentiated environment she provided for her regular students closely matched descriptions found in research conducted by Oakes in 1985. For example, speaking of her high-level English class, a teacher said she “did not have to be mean to keep them under control” (Oakes 1985, p. 123). About her low-level classes, the same teacher said she had to “keep it structured all period long, every day.”

In the current study, honors level reading instruction was based on what teachers considered sophisticated, challenging literature. Students in these classes read for sustained periods of time, engaged in open discussions, and worked collaboratively with their peers to conceptualize and create innovative projects based on the books they read. In short, their work seemed intellectually engaging and stimulating. At best, students in regular classes read below-grade level fiction or stories from the basal reader and filled out worksheets about them; at worst, they engaged in test preparation, reading short paragraphs and answering questions about them.

The nature of instructional interactions surrounding the less-sophisticated materials used with regular students was also vastly different. Virtually all of the teachers gave students in regular classes a reduced amount of time for reading self-selected materials on their own, instead assigning skills activities such as worksheets or test preparation activities, as this teacher’s comment illustrates.

I have two regular classes and one honors class. And in the beginning I gave all the classes silent reading time. But it’s just with these kids [regular] they don’t read if they don’t have to answer questions. And I’m hoping that maybe in 6 weeks or something we could try to something like that, where I could say, “You can read what you want right now.” And they will be in a routine where they can do that. But not right now.

Other teachers made it clear that being quiet and sitting still was their major expectation for regular students during times when independent reading was expected. One teacher said students in her regular class “think it’s stupid: ‘Why do we have to do this, Miss?’ They just open the books up to the middle page and pretend like they are reading.” Another teacher said that “just sitting there” was fine preparation for the major instructional goal for regular students, which was learning what was needed to pass achievement tests:

In my lower classes 20 min is about what they can deal with. And hopefully I’ll work them up a little further as the year goes on. But whether they are just sitting there, they need to learn how to be sustained silent readers. Because that’s what the [test] requires.

Similarly, in the schools studied by Goodlad (1984) and Oakes (1985) there were vast differences in the nature of instructional content as well as in the classroom environment between high and low-tracked English classes in both junior and senior high schools. Teachers’ goals for lower track students focused more on behavior than on learning, with an emphasis on conformity: “students getting along with one another, working quietly, improving study habits, being punctual, and conforming to classroom rules and expectations” (p. 65). Comparable differences were quantified in studies by Dreeben and Barr (1988) and Gamoran (1986).

As shown in the following teachers’ comments, when talking about books they read together, text talk for regular students often meant answering known-answer questions (Mehan 1979) and studying discrete skills, while honors classes engaged in higher-level discussions.

Of course, when we discussed the book [with the honors class] we got a little more in depth with it. And our questions are different. You know, there are more knowledge-level questions with the regulars. Dictionary definitions, which are the easiest thing for them to do because they are used to it. And there are more analyzing questions with the honors kids.

Another teacher described her work with regular students as remedial:

In that [regular] class the reading levels are really low. So we do a lot of what I call remedial-type work…. We go over vocabulary and spelling words and reading skills like drawing conclusions…. The honors is more independent for one thing. I’ll model part of the story and then they are able of course to read some on their own and understand what they are reading. They have more vocabulary words; they have less repetitive work. It’s more analytical.

The sixth grade teachers required less work and made assignments and academic discourse less intellectually demanding for regular classes. The above teacher’s words are disturbingly familiar in light of Oakes’ 1985 description of the instruction and expectations for students in low ability groups and classes:

Teachers of the high track classes reported far more often than others that they demanded critical thinking, problem solving, drawing conclusions, making generalizations, or evaluating or synthesizing knowledge. The learnings in low-track classes, in nearly all classes, required only simple memory tasks or comprehension. Sometimes low-track students were expected to apply their learning to new situations, but this kind of thinking was required far less frequently…” (pp. 76–77).

Observational studies of elementary reading groups found teachers differentiated instruction in similar ways for their low groups in that they focused on decoding and other basic skills rather than reading for meaning or interpretation (Allington 1983; Eder 1981). Considering the word regular, these classes should logically be expected to have “customary, usual, or normal” instruction like middle-level classes rather than low-level or basic classes (The American Heritage Dictionary 1982, p. 1041). In the days of school wide tracking, and when there were three levels of classes, regular was the middle level and was considered a college preparation course, along with honors (Lucas 1999). However, since there was no low or basic level class, it appeared some teachers in this study substituted the word regular for the concept low or basic, even though some of the students might have been in the middle level rather than the lower level classes had there been three levels. Perhaps when the district moved from three levels to two, the middle level was tacitly subsumed into the honors level and the regular level became the low level. The overpopulation of honors classes bolsters this possibility. If so, the message does not seem to have reached parents whose children were in regular/low classes. Past and recent research suggests that many parents of students in lower-level classes have less education and less knowledge of school systems than parents of children in higher-level classes, as well as less access to informal networks that help them obtain information about school placements, including how to override normal placement channels (Apple 2001; Auerbach 2002; LeTendre et al. 2003; Rosenbaum 1976). Parents who do not speak English, who do not feel comfortable in schools, whose skin color differs from those in charge, and who take class labels at face value have more limited chances of successfully advocating for their children (Auerbach; Fine et al. 2005).

As mentioned earlier, four teachers approached instruction for their regular students from a more positive stance. These teachers described approaches and materials they used to teach and motivate both levels of students, including giving them choices in reading materials and making sure all students had access to materials of interest even when teachers (often) had to buy the materials with their own money. Ignoring naysayers in their school, two teachers worked together to plan a unit on Shakespeare’s Hamlet for both their honors and regular classes. The unit included viewing and responding to a popular movie version, learning background information about Shakespeare and his inspiration for the characters, teacher and peer support for reading and practicing the play, and performances by students in small groups, with all students participating. Two teachers in a different school used reading and writing workshop, approaches that offer students choice. Despite initial resistance from students who were not accustomed to being held accountable for independent work, the teachers stuck to their convictions, as illustrated in the following comment:

I expect all of my students to read independently for 30 min every day and keep a journal. At the beginning of the year I felt that [they were thinking], “Oh we don’t want to do that.” Quite honestly. But they are coming around to my way of thinking slowly but surely. Now generally the honors kids don’t feel that way. They like to read. It’s the reluctant readers, the very low level kids, that as they find books that they can read and enjoy, it’s narrowing the gap.

Differential Teacher Expectations

At the foundation of teachers’ differential instruction for their classes of students were differential expectations. Most of the sixth grade teachers’ had low expectations for the students in their regular classes to do meaningful, complex work. Regular students were described as lacking foundational skills and as unable to handle the amount of work or responsibility that went along with the higher-level curriculum, so teachers avoided assigning challenging work or holding students responsible for completing assignments. They did not assign homework because students grumbled about it, and teachers were sure it would not be done. When students were reluctant to read in class, teachers read aloud instead: “We read as a large group because I feel that a lot of times if you tell them to read the story and give an assignment they are not going to read it.” Referring to low-reading-group first graders who were given less time to read in class than their higher-group peers, Allington could have been speaking to some of the teachers in this study: “Too often teachers argue that their poorer readers cannot read independently, but then confess never having attempted to develop such abilities in these children” (p. 875).

During one interview, a teacher showed me some of the literature response projects done by her students. There were dioramas, posters, and even a powerpoint presentation. “Aren’t they wonderful?” she enthused. Then she explained that only her honors students did projects because she feared her regular students would be unable to handle the responsibility, and she did not want them to feel unsuccessful. Her comments reflected the quandary that several teachers expressed about how they dealt with the students in their regular classes. She mused

My honors do the projects but my others are coming in and going, “Can we do that, too?” And so I’m sitting here debating on letting them do it. But the thing is, these are outside projects, it’s not just in class. I’m afraid they won’t get it back to me. And then I’m setting them up for failure.

Several teachers described times they had tried using the honors-type instructional practices and materials with regular classes, but most who had tried had abandoned their plans in short order. For example, one teacher started the year with a complex book project for all of her classes. When I interviewed her, just after the end of the first 6-week grading period, she was having second thoughts:

I think I’m going to just abandon this project. Because I think it’s way above their heads…. I think they just can’t handle this volume. The mass of materials. They can’t handle, “oh I have five things to do.” So I’m going to go strictly to short materials. Either that or stick with our [state achievement test] objectives. That’s what these kids need.

Despite teachers’ assertions that the differential instruction they provided was “what these kids need,” the majority of teachers spoke as if little could be done to improve the chances of students who were behind in sixth grade. Many said that if students had not learned to read or write well by middle school, their chances of doing so were slim to none, as illustrated in the following teacher’s discouraging comments about students in her regular classes:

Sometimes I have students who can’t read. I have one little boy who’s on first grade level, and it’s hard to know how I can motivate him to read what he can’t read. I feel bad that he’s gotten this far, and I feel bad that he’s probably going to leave me not reading and I’m his reading teacher. But I can’t cure what should have been cured way down there. All I can do is try to teach him to cope with the skills he has.

Other teachers seemed similarly hopeless about their influence on students, as reflected in this teacher’s words:

Because I see, unfortunately, how they get pushed through the system. If you’re in 6th grade and you’re on 3rd grade level, I don’t think the chances of your reading level getting higher are really good if the system stays the same way.

Although the language of ability grouping and curricular differentiation was present in all the interviews, the four teachers described earlier spoke as if they thought they could and were making a difference in students’ learning and attitudes. These teachers, two of whom taught in high-poverty areas, said they did not lower their expectations for students in their regular classes, as illustrated in this comment: “In both my classes, in my honors class and the so-called regular class—I guess the attitudes are pretty much what’s expected, but they are going to do the work because they know I expect it.” Another common theme among these teachers was building personal relationships with students as a necessary component of teaching and learning, as this teacher said:

You’ve got to know them a little bit better. Establish rapport. Get to know who they are. What they want from life. What they want to read. Why they don’t want to read. Once you find that out it’s easier to get them interested.

Ashton and Webb (1986) would consider these teachers to have a high degree of self-efficacy, a concept associated with higher student achievement and effective classroom practice. That is, they believed in their ability to reach students regardless of home environment, motivation, or past experience. These teachers said they saw their expectations rewarded, as illustrated in this comment: “I’ve seen just a remarkable difference in all my classes. They like to read. They ask me, ‘Are we going to have reading time today?’ and that has really impressed me.”

Students’ Behavior, Work Habits, and Interest in Learning

While honors students could be counted on to monitor their own learning and behavior, regular students were described by many teachers as needing the teacher’s constant supervision, as reflected in this teacher’s description:

The honors class doesn’t need monitoring. I can be working at my desk or looking at their day starters or checking their homework, and I can look up and it just has that feeling in the room when everyone’s doing what they’re supposed to be doing. [With] the other [regular] class, it’s more walking around and saying, “Do what you’re supposed to be doing.”

I interviewed one teacher just after observing reading instruction in one of her regular classes. She referred to the visit during her interview, pointing out what she saw as clear differences between her two classes in motivation and work habits:

[D]uring silent reading in class, I don’t have any kids in my honors classes that won’t sit and read. They really do enjoy reading. It’s probably for most of them the most fun time in the class. They get to read what they want to read…. And my slow readers, my regulars in the class you just saw? They are struggling. They are trying, but it’s difficult for them because they’ve not had perhaps the necessary training or just necessary time to sit in class and read. Wait until you see the honors. They’re amazing.

These teachers’ words have a disconcertingly familiar ring. Junior and senior high school teachers studied by Goodlad (1984) and Oakes (1985) described students in higher level classes as being better behaved and more motivated in the classroom and the classes as having a “relaxed, informal atmosphere” (Oakes 1985, p. 123). Students in low classes were also described as being slow underachievers who needed to develop discipline and were not interested in learning.

According to some teachers in the current study, regular students’ problems extended beyond the classroom. They were described as irresponsible in completing assignments, as illustrated in the following comment:

They are supposed to read at home every day and do book reports. When those book reports were due, I ended up having one kid read 12 books, and most of the regulars had nothing to turn in. Nothing done. Some of them didn’t have their books checked out. Despite constant reminders and warnings. I called home on all of them and still no response. I ended up getting stuff from seven of those. It was done late and it was poor quality.

Another teacher generalized that her students were unable to keep up with learning materials:

Some of the regulars are real low-level types of people that have learning disabilities. They have to keep their books in the classroom because they’ll lose them. It’s just a fact of life; they are going to lose everything you give them.

Teachers also described what they saw as clear-cut differences in students’ motivation and interest in reading. Honors students were generalized as having strong, varied interests and as enjoying challenging reading material, including the “classics.” The interests of regular students were described as narrow and related primarily to the difficulty of the material, as illustrated by this teacher’s words: “I’ve noticed that my honors kids don’t narrow themselves down to one author. They are doing a variety. It’s my regular kids that are kind of geared towards the same types of books and the same authors.” Another teacher had similar comments about her regular class:

They’re not real particular; the thing is they get real frustrated if they can’t understand it. I know the book we’re reading now, that I’m reading to them, is the story of Roberto Clemente, and they love it. It reads real easily and there aren’t a lot of vocabulary words that they don’t understand.

Causes: Home Environment, Ability, and Previous Schooling

Teachers in the current study advanced theories about the causes of differences they saw between students in their two levels of classes, again without being asked. Although few of the teachers had visited students’ homes, many made assumptions about students’ home environments and opportunities, apparently based primarily on their placement. Students in honors classes were assumed to have supportive home environments. Parents of students in regular classes were not only described as unsupportive, but also were assumed to be poor academic role models, as illustrated in this teacher’s comments: “They [student in regular classes] don’t see reading modeled at home. It’s not role-modeled for them. And most of my kids find it a chore.” Some teachers assumed, that parents of students in regular classes did not help their children with schoolwork: “I have all honors classes except for one and there is a major difference. You can tell which ones are reading at home and which ones their parents work with them on the reading.” Oakes (1985) also reported that teachers she studied assumed the students, their families, and their socioeconomic circumstances were the source of their learning challenges.

Regular students were also assumed to be economically disadvantaged and thus unable to buy books and other school materials. One teacher noted, “My regulars don’t have access to books outside of school. It’s sad but true.” Another remarked, “The honors kids bring books from home. Because the honors kids go to the [bookstore], and their parents will spend $50 on books for them. The honors kids always have a big stack of books.” Based on district records, and consistent with previous research on ability grouping, students who were eligible for the federally funded free or reduced price lunch program were disproportionately concentrated in the lower of the two levels, so teachers’ assumptions were based in truth. However, some inferences went beyond economic logic. Parents of students in regular classes were also presumed deficient in their provision of the “right kinds” of out-of-school experiences, despite the fact that teachers had, at best, limited evidence for such generalizations. For example, one teacher assumed, without evidence, that none of her students had ever been to a public library: “A lot of them have never seen a library, believe it or not. They don’t have library cards; they don’t know how to act in the library; they have never been to the library. These are the regular kids.” Another made a similar assumption about bookstores:

My regulars don’t have the opportunity to go to an actual bookstore a lot. I hear my honors kids that they’re there all the time with their parents. But these parents [regular] don’t go to the bookstore and make a side trip to go there with the kids. The library is unheard of.

Two of the four “high self-efficacy” teachers also spoke about students’ limited access to reading materials outside of school. Because their schools served high-poverty areas, this was a problem for virtually all students. Both of these teachers spoke of solutions, including sharing classroom library books with other teachers, checking out school library books to keep in their classrooms, and buying paperbacks for students to use and check out.

Some teachers tied students’ motivation and interest in learning to innate potential, as illustrated in this comment about honors students: “It’s all ability. Because they can do it and they can be successful and they can understand.” Other teachers made it clear that their regular students were unwilling and/or incapable of handling grade level work due to social promotion, as in the following comment:

And those kids that have been—dare I say it on tape—pushed through the system? They’re reading on second, third grade level and then we’re asking them to use our textbooks and use the stuff on sixth grade level.

Other teachers felt previous inadequate school experiences were to blame for regular students’ limited skills:

The honors have the capability to read and comprehend. They just need a jumpstart. The attitudes with the regular class are hard. Because they just don’t enjoy reading period. I really feel the reason they don’t is that they have trouble reading. And they missed the skills in the elementary grades so therefore it’s something they want to avoid instead of plunge into.

Several teachers remarked, as in the previous comment, that one reason for the reading difficulties of regular students was the lack of challenging instruction in previous schooling. As one teacher said, “My thoughts are: one, they have not been challenged to read before, or not as frequently as they should; two, they are non-readers or slow readers.” However, instead of providing more challenging experiences, the instruction most teachers described was simple and undemanding, as described in the first section of the findings.

Conclusions, Limitations, and Implications for Research and Practice

Attitudes and practices similar to those that existed when tracking was more openly acknowledged in the United States were evident in the words of most teachers in this study. The term regular concealed what was essentially a basic skills or even remedial environment in most of these classrooms, as evidenced by teachers’ descriptions of their regular students as “slow,” “low level,” and “struggling” and their classroom instruction and environments as “easy,” “structured,” and “repetitive.” Since the data were collected for this study, the district has changed the name of the bottom level of two levels of classes from regular to the more palatable but even less accurate “grade level” (personal communication, district literacy specialist). That label clearly implies students are achieving academically what would be expected at sixth grade level. Yet, these students would likely be the same group of students who were in the regular classes in this study, some of whom were described by their teachers as reading many grade levels below their placement. Thus, the message sent by this label is even more misleading than regular. Labels used in the past (e.g., basic or low-level as opposed to regular), while perhaps less pleasant, more clearly alerted parents to the fact that their children were being educated as if they had academic difficulties. I suggest that, while this may be an unintentional by-product of name changes, ability grouping may be more insidious and damaging than earlier incarnations, when harmful attitudes and practices were not so deeply buried under euphemistic labels.

Limitations of the Study

This study was conducted with 25 teachers in one school district, and the findings cannot be generalized beyond those teachers. I regret I was unable to observe multiple times in the classrooms of all the teachers, enlisting multiple observers with diverse stances toward ability grouping, to get a better sense of how their instruction corresponded with their discourse; this would have made the study stronger. The data would have been richer if I had enlisted multiple observers and interviewers with diverse stances on ability grouping to contribute to data collection and analysis. As I made clear in my researcher statement, my beliefs about ability grouping are negative, and my stance is surely reflected in this paper. I took great pains to conduct balanced data collection and systematic data analysis, but I acknowledge the potential that my beliefs may have influenced either or both, just as this is a possibility for any researcher in any paradigm.

Implications for Research

Given the pervasiveness of talk among the teachers in this study about minimal expectations and marginal instruction for students in regular classes, it is reasonable to expect the negative language of ability grouping, as well as practices that are harmful to students in lower groups and classes, might be widespread. More research in classrooms is needed. Such research could include large scale, cross national studies to help determine the scope of the problem, as well as smaller-scale studies in which interviews are paired with multiple observations to explore nuanced differences between current ability grouping practices and those of the past. Continued research on detracking, as well as instructional approaches that do not include ability grouping, such as workshops, is also needed.

Implications for Practice

Although the findings of this study cannot be generalized, it seems likely the attitudes and practices uncovered might be found among other educators. Thus, there are implications for schools who have obligations to parents who send their children to school every day expecting nothing more, but nothing less, than for their children to receive a “usual,” “customary,” regular education rather than one characterized by low expectations and marginal instruction. In addition, educators would benefit from opportunities to examine their language and practices in light of research on ability grouping in the hope of modifying them. These implications as well as alternatives to ability grouping are discussed in the following sections.

Informing Parents

One set of stakeholders in the school system, parents, has been consistently left in the dark about ability grouping (LeTendre et al. 2003). According to Lucas (1999), while students are no longer locked into tracks for the whole school day, “Ironically, this dismantling of formal programs has probably increased the information gap” (p. 6) between parents of middle and upper middle-class students and parents of low-income and minority students by removing clarity that once existed about the ability grouping system. The district’s web site does not offer clear information for parents. The course listings are designated by the letters “R” and “H.” Pre-AP is written in parentheses next to the H. None of these designations is spelled out, and the term grade level is not mentioned, although it is the accepted term used by teachers (personal communication, district literacy specialist and classroom teacher). The course description refers to both levels of classes, but there is a sentence explaining that Pre-AP classes are more advanced. The web site does not mention this, but in some schools there are no Pre-AP classes because there are no teachers qualified to teach these classes (personal communication, district literacy specialist and classroom teacher). The site explains that placements are based on state achievement tests scores or parent requests (information retrieved July 29, 2009).

LeTendre et al. (2003) and Lucas (1999) asserted that schools should establish standard placement classifications, clarify how students are placed, and give students and parents information about the placement process. I agree with this suggestion. Web sites should be more informative, and each campus should provide many opportunities for parents to receive information, both through printed flyers and in sessions offered at a variety of times and in languages spoken by the families served by the schools. However, I wonder what would happen if parents were truly aware of the kind of instruction and expectations their children would potentially receive in regular or grade level classes? It seems likely most would request honors placements. Would there then be two levels of honors classes with one necessarily being the lower level? While providing parents more information is a necessary step, teacher’s attitudes and practices also must be addressed.

Increasing Teachers’ Awareness

Although much of the teachers’ language in this study was harsh, it would be a mistake to blame teachers. Doing this would overlook the role of schools and fail to “take into account the fact that teachers function within particular structures in society over which they usually have little power” (Nieto 1995, p. 43). School and district policies follow a long tradition of tracking and sorting, and teachers rarely have a say in changing those policies. Within their own classrooms, according to Oakes, “It is likely that teachers intend to adapt instruction to differences in students’ learning styles and learning speeds. What they effect with differentiated curricular content, however, appears to be of another sort altogether.” (p. 92). Setting aside who is to blame, continuing the current situation is unacceptable.

The concept of critical reflection is relevant here (Schön 1982). The words of teachers in this study, paired with past research in ability grouping, could be a first step in raising teachers’ awareness of grouping practices and attitudes. Young teachers might never have been exposed to the historical research on grouping; many may find they are mirroring the language of their more experienced colleagues, whose words in turn are “permeated with the interpretations of others” (Bakhtin 1981, p. 202). They reflect the voices, thoughts, experiences, and intentions of others who have used these same or similar words to refer to marginalized students in the past. Similarly, the actions and practices described by the teachers in this study are similar to those described by teachers and researchers in the past. According to Watanabe (2006), “teachers’ notions of ability and intelligence come through in their talk about classroom practice, and it is important for teachers to become adept at identifying these perspectives in each other’s comments” (Watanabe 2006, p. 29). Beyond just raising awareness, using Shön’s concept of reflection-in-action to set up study groups in which teachers study and reflect on their own practices in collaboration with other educators has the potential to impact actual practice.

Exploring Alternatives to Ability Grouping

Even when differences in achievement are slight or nonexistent, mere placement of a child in a lower-ability group can negatively affect expectations, which in turn can negatively affect achievement (Dweck and Leggett 1988; Rosenthal and Jacobson 1968). In the research of Pallas and colleagues (1994)

Children in higher ranked reading groups were perceived by their parents and teachers as more competent than were similar children in low-ranked groups, often independent of actual performance [emphasis added]. These perceptions may structure the educational opportunities that parents and teachers subsequently make available to children, as well as the social-psychological resources they extend to children. (p. 43).

As long as students are grouped by ability, then, there will always be the potential for discrimination and negative outcomes for students in lower-ranked groups. In this section, I explore alternatives to ability grouping. For a more complete discussion, see Worthy et al. (2009).

Since the 1990s, a number of schools and districts have experimented with placing students in heterogeneous rather than leveled classes. Often called detracking, many of the more visible and successful efforts in high schools and middle schools have focused on providing a rigorous academic curriculum for underserved, mainly low-income students (Alvarez and Mehan 2006; Mehan et al. 1996). Students enrolled in such schools and programs have had higher than expected college acceptance and attendance rates. Detracking efforts in schools serving racially and economically mixed areas have had more tenuous results because middle class parents, whose children are more likely to be in higher tracks, are often resistant to heterogeneous grouping, even though achievement levels of high SES students did not decrease (Wells and Oakes 1996; Wells and Serna 1996). Oakes concluded that detracking in these schools often came down to a “struggle between more- and less-privileged families over scarce resources; and it entailed an ideological struggle over the meaning of race and culture as they are enacted in schools” (p. 294). In a rare example, the Rockville Center School district in suburban New York’s detracking effort, which was started gradually in 1990 and included extra support for lower-achieving students, has resulted in the virtual disappearance of a longstanding achievement gap between racial minority and majority students (Oakes 2005). Despite some successes, however, tracking reform has not taken extensive hold in the United States, and most schools continue to group students within and between classes from elementary through high school. However, lessons learned from more and less successful efforts provide important implications for the future.

A first step in tracking reform is to “become much more aware of how local constituents perceive tracking and its outcomes” (LeTendre et al. 2003.p. 83) by having open meetings and then addressing those concerns and perceptions in designing programs (Welner and Burris 2006). Strategies used in the school Welner and Burris wrote about involved phasing in detracking step-by-step, starting with open enrollment in heterogeneous classes while keeping some traditional classes, and gradually expanding unleveled classes after achievement scores in the classes improved. Another strategy is to work first to win the support of parents of students in upper level classes, who are usually the most vocal supporters of tracking and have the most political clout (Datnow and Hirschberg 1996; Welner and Burris 2006).

Advanced planning involving teachers is an essential component of detracking. Rubin and Noguera (2004) found that when detracking plans were implemented from the top down, without teacher input, support, and professional development, some teachers tracked students within their heterogeneous classrooms by requiring less complex work of the students they knew were lower achievers, and providing more complex work, along with more thoughtful feedback to higher achievers. As mentioned earlier, involving teachers in critical reflection about their beliefs and practice has the potential to help in this regard as well.

It is important to acknowledge that students in high tracks do benefit from their placements because they have access to enriched curriculum and instruction, extra resources, and high expectations (Gamoran 1992). When classes are detracked, all students should have access to these privileges and expectations, along with extra support for students who may need it (Donelan et al. 1994; Rubin and Noguera 2004, Yonezawa and Jones 2006).

Promising instructional strategies for heterogeneous classrooms include establishing relationships with students (as some teachers in the study mentioned), involving students in issues that affect their own lives, such as social justice concerns, differences among people, and discrimination (Haberman 1991), and making use of cultural relevant and culturally responsive pedagogy (Gay 2002, Rubin and Noguera 2004; Ladson-Billings 1995). Non-competitive, student-centered practices such as project-based learning, cooperative learning, and workshop approaches are based on the premise that students have unique interests and abilities upon which they draw for learning (Bartolomé 2003; Cone 2006), including some classrooms in the current study.

Although high schools and middle schools typically have more entrenched systems of tracking, with leveled classes, some elementary schools group students into classes at each grade by achievement, and virtually all use within-class ability grouping. Further, because this is where ability grouping starts, sometimes as early as pre-kindergarten, it is important for educators to explore alternatives to ability-grouped instruction and to educate parents about the potentially harmful effects of ability grouping starting in the primary grades.