Closing the Achievement Gap One Teacher at a Time
by Bridget K. Hamre & Robert C. Pianta - October 28, 2005
Of great concern to society is whether experiences in the early grades of school can close the gap between students of varying demographic, experiential, and developmental backgrounds. In this commentary we present the results of a recent study providing evidence that for kindergarten children at risk of problems in first grade the instructional and emotional aspects of interactions in first grade appeared to help close the achievement gap. Yet, it is worrisome, in this age of high-stakes testing of children, that national-level studies report exceptional variability in the nature and quality of everyday instructional and social–emotional supports offered to children in the early grades. It appears unreasonable to expect universal levels of minimal performance for students when the opportunities to achieve such levels are so unevenly distributed. If schools are going to close achievement gaps, then efforts to improve student performance must address this variability in the quality of everyday classroom experiences. Our impression is that achieving that goal will require a reconceptualization, redesign, and refocusing of how teachers are trained and supported to do the very difficult job they have chosen.
Of great concern to society is whether experiences in the early grades of school can close the gap between students of varying demographic, experiential, and developmental backgrounds. If classrooms matter in this way, then strategic efforts related to teacher training and support, curriculum implementation, and assessments of classroom settings could counter the tendency toward poor outcomes for such students. It is worrisome, in this age of high-stakes testing of children, that national-level studies report exceptional variability in the nature and quality of everyday instructional and socialemotional supports offered to children in the early grades. It appears unreasonable to expect universal levels of minimal performance for students when the opportunities to achieve such levels are so unevenly distributed. If schools are going to close achievement gaps, then efforts to improve student performance must address this variability in the quality of everyday classroom experiences.
We conducted a recent study to test precisely this possibility that interactions in classrooms can close gaps using data from a large, national prospective study of children and families. Specifically, we examined whether children at risk of early school failure exposed to high levels of instructional and emotional support in first grade displayed higher achievement and lower levels of conflict with teachers than did similarly at-risk peers in classrooms that did not provide this support. Importantly, we examined naturally occurring variation in everyday classroom interactions rather than an intervention designed to improve classroom interactions. Thus, this research has implications for nearly every elementary school in the country.
Two groups of at-risk children were identified: those whose mothers had less than a 4-year college degree and those who displayed significant behavioral, social, and/or academic problems according to their kindergarten teacher. These at-risk children, on average, were behind their peers in early achievement at age four, fell further behind by first grade, and had more conflict with first grade teachers. Yet not all of these children displayed academic or relational problems in first grade; some were doing quite well, and their success seemed connected to the kind of experiences they had in their first grade classrooms.
If placed in classrooms offering low instructional quality, children whose mothers had lower levels of education had poorer achievement than did their peers with more educated mothers. However, in classrooms offering higher levels of instructional quality and support, children with less-well-educated mothers achieved at the same level as those with mothers with a college degree. Similarly, when children whose kindergarten teachers judged them to be showing considerable adjustment problems in their classrooms were placed in emotionally supportive first grade classrooms, they showed achievement and adjustment levels identical to children who had no history of problems in kindergarten. Thus, for kindergarten children at risk of problems in first grade, the instructional and emotional aspects of interactions in first grade appeared to help close the achievement gap.
Unfortunately, the odds are stacked against students actually getting the classroom supports necessary for their success. Research by the NICHD Early Child Care Research Network has shown that, in general, students are exposed to moderate to high quality social and emotional supports in their elementary classrooms but quite low levels of instructional supportlevels that are not as high as those gap-closing first grades we studied. Most notable in nearly every study that includes a large number of classrooms, however, is the exceptional degree of variability in the qualities of classroom experience and interactions to which children are exposedparticularly in the areas that our study shows matter most for children. The NICHD study, involving several thousand classrooms in elementary school, shows that a typical school day for some students includes spending most of their time engaged in productive instructional activities with caring and responsive adults who consistently provide feedback and challenge students to think critically. Yet for others, even in the same grade and same school, a typical day consists of spending most of the time sitting around, watching the teacher deal with behavioral problems, and participating in boring, rote instructional activities such as completing worksheets and spelling tests. Recent studies that have used statistical procedures to group similar classrooms together show that only about 25% of classrooms in a given grade show high levels of emotional and instructional support.
The problems of inconsistent exposure to high quality classrooms are compounded by clear evidence of inequity. Perhaps not surprisingly, students from disadvantaged backgrounds are more likely to be exposed to poor quality classroom supports. For example, in a recent study by the National Center for Early Development and Learning (NCEDL) of 240 state-funded preschool programs in six states, the proportion of poor children in the classroom was one of the strongest and most consistent predictors of classroom quality.
Further troubling is evidence that even the student lucky enough to experience a high quality classroom one year is very unlikely to be systematically exposed to high quality over a period of years. Observations of the quality of interactions and supports in classrooms show almost no association from year to year, even when students remain in the same school. This suggests that school-level resources such as professional development or structural features such as class size, or teachers extra degrees, coursework, or experience, simply may not be sufficient to ensure high quality classroom environments.
Taken together, these resultsthat experiences with teachers in classrooms can close early achievement gaps, the generally low quality of such classroom experiences, the high degree of variability from classroom to classroom, the disconnect between typical measures of teacher quality and observations, and the extent to which less-advantaged children are more like to be exposed to lower qualityshould sound an alarm. Too few of the students who are in greatest need of high quality school experiences receive them, and the few that do are unlikely to receive them consistently, making it unlikely that the positive effects we found in first grade will be sustained for those at-risk children who need consistent supports.
For policy makers and educational administrators, these findings could spark an interest in finding solutions that focus on raising and leveling the quality of classroom supports available to children in preschool and elementary school. Among the range of potential solutions, there are two basic mechanisms through which policy might effect change on classroom quality. The first is to find structural features of schools and classrooms such as teacher education and certification, class size, and curriculum that may affect classroom quality and enact policies to ensure that these proxies for quality are more uniformly in place. The other option is to find ways to more directly change and improve classroom qualitythe dimensions of instructional and social interactions teachers have with childrenin large numbers of classrooms.
The first option appears to be the path preferred by policy makers and administratorsperhaps for good reason. It is much simpler to change education and certification requirements or to buy new curricula than it is to actually try to change classroom-level interactions between teachers and students and how teachers implement curricula. However, we have yet to find evidence that any structural feature of schools or classrooms ensures high quality classroom experiences for students. Gathering results from studies involving observations of about 5,000 pre-K and elementary classrooms make this point quite clearly. For example, these studies consistently show small, if any, effect of teacher qualifications on observed classroom quality; and, most importantly, we continue to find a high degree of variability in classrooms that meet all structural benchmarks of quality. These proxies for teacher or classroom qualitywhich in fact are the indicators written into most NCLB regulations related to teacher qualityare very poor reflections of the ingredients of schooling that matter most for children.
The alterative to policies that require adherence to proxies for quality is to build professional development and training systems for teachers that actually support them to interact more effectively with their studentsa daunting challenge. But ultimately, such systems, if based on strong and valid metrics, may be a more cost-effective mechanism for effecting real change in classroom quality. This would be in part because rather than focusing personal and financial resources in the pursuit of proxies to quality, such a system may actually produce higher quality.
We and our colleagues at the Center for the Advanced Study of Teaching and Learning at the University of Virginia are working to address the need for research on teaching and teacher quality, with observation of teachers practices as a central focus. In our work in the teacher education program, supported by the Carnegie Foundations Teachers for a New Era program, we are using a validated observational measure called the Classroom Assessment Scoring System (CLASS) as a tool for observing all students throughout their training, for increasing student teachers self-analysis skills, and for engaging in discussions among teacher educators about how best to produce high quality teaching. In a major initiative funded by NICHD, we are testing an innovative, patent-pending model for working with teachers to improve classroom quality, at scale. Teachers in this project, called MyTeachingPartner (www.myteachingpartner.net), work in schools all across the state of Virginia, many in rural, fairly remote areas. These teachers videotape their implementation of instruction in their classrooms twice a month and mail the videotape to a trained consultant. The consultant edits the video into a series of short clips highlighting a specific area of classroom quality and puts it up on a secured web site for the teacher to view and respond to a series of prompts. The teacher and consultant then meet, via web-based video conference software, to further discuss the video.
The idea behind this type of professional development is that teachers will be most likely to change if they receive ongoing, personalized feedback about their everyday interactions with students. Our hypothesis is that focusing professional development resources on teachers actual interactions with children and implementation of activities, guided by validated indicators of those interactions, is in the end a more powerful tool for changing those interactions than is requiring them to take a course or attend a workshop.
We began this piece by describing how teachers interactions with children in classrooms mattermatter enough to close the achievement gap. Ensuring that all children, starting with the least advantaged, have access to gap-closing classrooms should be a top priority in the Unites States. Our impression is that achieving that goal will require a reconceptualization, redesign, and refocusing of how teachers are trained and supported to do the very difficult job they have chosen. We now know enough to at least start down that path.