Agency and autonomy are two constructs critical to
the formation of self,
which is, in turn, critical to motivation. The classroom environment is, in the
first place, psychological in nature. Social interactions ultimately impact on
the students' sense of self by fostering agency and autonomy, or constricting
agency and autonomy. Teachers' comments and expectations for students
are natural occurrences within that social interaction and profoundly
influence the psychological environment of the classroom.
Craig Janes is a graduate student in our Master of Education programme. He has undertaken a review of the literature on teachers' expectations and written a solid paper on the topic. His summary of the research conducted in this area, when considered in light of the other two papers on motivation, illuminates our understanding of schooling in an important way.
Timothy L. Seifert
In a document prepared by the Newfoundland Government entitled, Adjusting The Course (1994), it was emphasized that a fundamental priority of Newfoundland's educational system be that high levels of expectations and standards be maintained for the success of the schools and students. A shift in our approach to educational achievement was necessary, a shift that would form the basis for establishing high standards and for creating an expectation that these standards can be met . The repetition of the word expectations is not merely for literary purpose but is indicative of the growing emphasis placed on the causal relationship between expectations and student achievement. It will be useful at this point to examine how the literature defines expectations.
Lawler, cited in Saracho (1991), defined expectancy as " the persons' estimate of the probability that he will accomplish his intended performance, given the situation in which he finds himself" (p. 27). Saracho (1991) then went on to state that teacher expectation is the "teachers' estimate of the child's academic performance within the classroom" (p. 27) .
The other concept that we are attempting to understand is attribution theory and more specifically how attribution theory and teacher expectations relate to one another. On a very simplistic level attribution theory undertakes to explain "why" an event occurred when there is an unexpected outcome (Weiner, 1984). On a deeper level, this theory analyzes the perceived causes of an event from a number of causal dimensions:
1. Locus of Control - was control of the cause within or outside of the individual.
2. Causal Stability - does the cause always exist or is it only present for a short period of time.
3. Controllability - whether or not the cause was something they could control (effort vs. illness).
4. Intentionality - Poor effort vs. poor use of a strategy.
Once such an examination takes place, the learner will attribute the unexpected event to a particular cause and this will result in some affective or emotional change (Weiner, 1984). For instance, if a student attributes a good mark on a test to ability, a perceived stable and non-changeable cause, they are likely to experience feelings of pride and a sense of accomplishment. Failure attributed to a stable and non-changeable cause results in feelings of guilt or shame (Tollefson, 1988).
In the realm of motivation, how the student attributes the cause of an event will directly affect his or her level of motivation for future tasks. Tollefson (1988) argued that students who attribute success to a stable factor such as ability increase their expectations for success and are therefore encouraged to greater task persistence. When failures are attributed to ability, the student's expectancy for future success decreases and along with it task persistence. There is a sense of hopelessness and resignation in that the learner feels there is nothing he/she can do about it. Weiner
(1984) echoed this hypothesis when he noted that once success or failure has been attained and as long as the conditions or causes of that outcome are perceived as remaining unchanged, then individuals will anticipate success or failure for future tasks with a certain degree of certainty.
If a learner believes, however, that the causes are a result of unstable and changeable factors, such as luck or effort, then the focus of motivation shifts. If failure is attributed to an unstable factor there is still a high expectancy for future success but, if success is attributed to either of these factors, a low expectancy for future success results (Tapasak, 1990). The basic premise is that a learner will not expect to succeed later if the present success was a result of something that can change from situation to situation. In the same vein, failure as a result of something that can change would not reduce one's possibility for future success. The fact that the factor can change implies that it might, so there is no reason to think that one will always fail.
To summarize thus far then, expectations are the beliefs that a teacher and student hold for that student's future success in learning situations. Attribution theory states that where a student attributes the causes for their success or failure will affect his/her emotional state and expectancy for future performance. The connection to be made at this point is that what one expects to happen in the future with regard to success or failure is inherently linked to what one believes to be the cause of past successes or failures. Therefore, expectations for success or failure can sometimes be linked to how one attributes past successes or failures. But this may be a reciprocal relationship in that one's expectations for success or failure may dictate where one attributes the causes for past events. If one expects to be successful, they may determine that a changeable factor was responsible for past failures and a stable one responsible for past successes.
In either case, the expectations and attributions that a student holds must originate from somewhere or, at the very least, be fostered in some manner by outside influences. It is here that we have to examine the role of the teacher in this relationship. In other words, how does the teacher influence the students' expectations for success or failure and thereby influence the attributions that students make.
Theories on how a teacher's expectations, for the success or failure of a student, influences that student's actual achievement are varied and some times even contradictory. One of the original studies by Robert Rosenthal and Lenore Jacobson (1968) entitled Pygmalion in the classroom showed a definite relationship between the expectations of a teacher for a student and that student's level of achievement. Teachers were told that some students were high achievers while others were low achievers when, in fact, there was no actual measured difference. At the end of the study, those labeled as high achievers had actually done better than those labeled as low achievers. However, later studies done along the same lines often failed to produce the same results (Clairborn, 1969). In fact, Williams (1975) put forth that it is mainly the intellectual capacity, social origin, and structural arrangements that a school provides which affect students' performance. This would echo the findings of 'The Coleman Report ' which claimed that no particular school characteristic had a measurable, positive impact on student achievement (Towers, 1992). The report even went so far as to claim the only factor considered to have any impact on student achievement was the social class of the student body.
More recent studies and literature, however, have reported that there is a relationship between teacher expectations and student achievement. Although the major body of literature agrees the relationship exists, it is the exact nature of the relationship which needs to be examined further.
Early research held the notion that there was a direct cause and effect relationship between teacher expectations and student achievement (Anderson, 1991). It was felt that the simple possession of high expectancies for students would translate into increased achievement levels. The belief was that if students knew what they were expected to do and how they were expected to act, they would behave accordingly (Monhardt, 1995). Hassenpflug (1994) asserted that a teacher with high expectations could raise students' expectations and have a positive effect on students' achievement. She goes on to say that "students actually can and will do better if quality work is expected of them..." (Hassenpflug, 1994, p. 161). This line of thinking may, in fact, need further development, for Anderson (1991) maintained that such an interpretation may be naive and superficial in light of the current research. This basic association between expectancies and achievement will need further refinement and clarification.
Much of the literature reviewed attempted to explain the relationship in a more succinct and detailed manner by attributing the relationship between teacher expectations and students' achievement to one or more of the following concepts: Perceptual Bias, Sustaining Expectation Effect, and Self-fulfilling Prophecy (Anderson, 1991; Kolb & Jussim, 1994; Saracho, 1991; and Weinstein, 1995).
The concept of perceptual bias revolves around a very simple premise. Kolb (1994) stated that perceptual biases result when the expectations of the teacher influence the teacher's evaluation of the student's achievement . In other words, a teacher feels that a student is a high achiever and evaluates them higher than their abilities merit. In this particular case, then, there is no action on the part of the student which affects his or her achievement but rather the action is on the part of the teacher.
Closely related to perceptual bias is the notion that in some teachers there is the tendency to expect students to continue or maintain previously developed behaviour patterns, disregarding the students' abilities. This is the process known as sustaining expectation effect (Saracho, 1991). Anderson (1991) further clarified this effect by stating that "teachers expect students to sustain previously developed behaviour patterns to the point that they take these behaviour patterns for granted and fail to see or capitalize on changes in the students' potential" (p. 22).
A further development in the analysis of teacher expectancy comes in the form of the concept self-fulfilling prophecy. Merton (1948) first coined the phrase to describe how erroneous beliefs about people and situations sometimes create their own fulfillment. Kolb and Jussim (1993) refined the notion when they explained that self-fulfilling prophecies occur when teachers induce students to perform at levels consistent with their (teachers') initially erroneous expectations. In other words, if a teacher believes a student to be bright then the interactions between the two may be such as to ensure that this expectation comes true (Anderson, 1991). This is where we begin to see a deviation from the notion of a direct causal relationship to one that is more detailed and explanatory.
The prominent notion being argued now is that it is not the expectations themselves which influence the students' achievement and behaviour but rather how those expectations cause the teacher to interact with the students thereby affecting achievement levels. In each of the three concepts mentioned the onus is placed on the teacher and his/her interactions with the students as the major factor affecting achievement levels. It is also within an examination of teachers' communicated behaviours towards students that we can understand how a teacher's attributions about student success and failure is delivered to the student, internalized by the student, and subsequently affect expectancies for future success. It is here that the complementary nature of expectations and attributions begins to take form when the role of the teacher and their behaviours are incorporated into the argument.
Much of the research and literature now holds fast to the notion that, although teacher expectations are an integral part of the issue, it is more a matter of how the expectations are communicated in 'differential treatment' that actually influences student achievement (Weinstein, 1995). In fact, Anderson (1991) endorsed the notion that "current analysis of teacher expectations shows that while the expectations teachers hold for students may indeed be influential, the way in which the teacher responds or behaves as a result of these expectations is a more important variable" (p. 22). Numerous studies (Gottfredson, 1995; Hall, 1993; Kolb and Jussim, 1994 ; Lee-Corbin, 1994; Taylor and Reeves, 1993) have been done to determine the validity of the thesis that if a teacher has different expectations about pupils, they then respond differently toward those pupils.
Studies (Gottfredson, 1995; Hall, 1993; Kolb and Jussim, 1994; Lee-Corbin, 1994; Taylor and Reeves, 1993) have concluded that there is a great deal of evidence to support the premise that teachers interact differently with students based on their expectations for those students. Teachers tend to call upon those who they think will know the answers more often than those who they feel will simply provide an incorrect response (Taylor, 1993). Also, when a teacher has high expectations for a student, they often develop an interest in that student and focus on improving his/her (student's) performance (Saracho, 1991). This inherently implies, then, that students about whom negative perceptions are held are not provided the same opportunities for performance improvement. In fact, the following behaviours are used more often with perceived low achievers: insincere praise, less frequent and informative feedback, paying less attention to the student, making less eye contact, and making less use of students ideas (Gottfredson, 1995). What could these behaviours communicate to a student about how the teacher attributes the student's success or failure?
A teacher's perceptions about the causes of students' behaviour is extremely important. Peterson and Banger (1988), as cited in Fennema (1990), maintain that a teachers' causal attributions are vital because their view on why a student succeeded or failed influences the teachers' expectations for future achievement on the part of the student. If a teacher felt failure was a result of ability and therefore unchangeable, they are less likely to react toward that student in the same manner than if such failure was attributed to effort. This is where the communication aspect comes into play.
It is my contention that students will develop their own attributions based, to a certain degree, on how the teacher interacts with them following failure or success on learning activities. Complementary with expectancy theory and the transmission of perceived expectancies comes the notion that attributions are also transmitted to a student. Kurtz and Schneider (1990) contend that...
"Teachers influence cognitive development and
school achievement not
only through explicit strategy instruction but also through overt and subtle
messages about their perceptions of children's' abilities and their
attributional theories about other factors that influence achievement." (p.
An example of this line of thought is provided by Tollefson (1988). If a teacher attributes failure to some uncontrollable cause such as ability, he or she is more likely to help and praise the student. However, when combined with the standard behaviours of teachers toward students with perceived low ability, such praise may simply be gratuitous and given simply to placate the individual (Saracho, 1991). Such behaviour communicates to the student that their failure is a result of something they, the student, cannot control and this, in turn, will cause the student to lower their expectancy for success in the future. They lower their expectancy for future success because they have attributed their failure to a stable and unchangeable factor based on the teacher's behaviour toward them. Anger and frustration toward a student's failure communicates that the attribution is a controllable one such as effort and therefore does not result in a reduced expectancy for future success.
Kurtz and Schneider (1990) maintain a similar argument by claiming that the attributional theories about achievement a teacher possesses combined with their expectancies for various students will affect the amount of praise and/or criticism they provide to the children. It also plays a significant role in the level of intimacy and degree of power sharing a teacher has with certain students (Grant and Rothenberg, 1986 cited in Kurtz and Schneider, 1990). Transferred to expectancy theory, the argument is that students with perceived low ability are given less autonomy when it comes to working on tasks (Saracho, 1991). This would communicate to the student that not as much is expected of them because they are incapable of doing the work. Here, then, low ability is equated with low expectancies for success. What results is a certain degree of influence relating to children's' achievement expectations, effort expenditure and resulting achievement. Tollefson (1988) views the relationship in a cyclical manner...
"A student believes he / she cannot do the work
without help. The
teacher believes that the student could do the work if he/she tried harder
and withholds help. The student develops an attitude of ' what's the use
in trying if I am going to fail'. The teacher maintains his/her attribution and
continues to be angry, critical, and unhelpful, and reinforces the student's
beliefs and subsequent behaviour." (p. 264)
While the teacher in this instance may believe the student capable of doing the work, the critical and unhelpful nature displayed actually communicates the opposite according to expectancy theory. Such behaviour is characteristic of teachers who perceive low ability in and have poor expectations for students. Therefore, a perception of low ability on the part of the teacher and subsequent teacher behaviours communicates to the student that the cause of their failure is uncontrollable therefore the student experiences the emotions of hopelessness and resignation.
A detailed list of teacher behaviours based on high and low expectations is provided in Appendix A and while their existence does not guarantee the theorized effects in all situations, the research evidence does hold that there is a correlation.
Fennema (1990) tied this element to the expectancy of teachers in the following way. The instructional decisions that teachers make which, in turn, transmit to the student their views on what caused the event are mediated by the teachers' beliefs. Teachers have a wide range of preconceived ideas and beliefs about students based on a number of factors and such beliefs vary from student to student.
Such discrepancies in expectations can result from a number of traits: race, gender, age, appearance, handicap, perceived effort, and socio-economic status (Anderson, 1991; Gottfredson, 1995). While a complete analysis of all these topics is beyond the scope of this paper, it is interesting to note a couple for their inclusion is important when examining possible solutions to the effect.
These so-called 'foundations' for expectancy differences can be succinctly summed up using two words: stereotyping and labeling. Stereotyping leads people to have preconceived notions or ideas about someone simply because that person possesses some of the characteristics of a particular group of individuals. Once labeled as part of this group the teachers' behaviour may change accordingly. When combined with expectancies in the school setting, a number of interesting conclusions have been drawn.
First of all, students of a particular race, about which preconceived notions are held, often have expectations about them in line with the 'stereotype'. Minority students are often given subtle messages by their teachers about their ability and worth thereby negatively influencing their achievement (Hall, 1993). Studies such as the one carried out by Fennema (1990) addressed the myriad of contentious issues surrounding the perceived gender differences between boys and girls in the area of math achievement. It was discovered that teachers tended provide more encouragement for boys than for girls to engage in math. Boys' success in math was often attributed to ability and, as already discussed, success attributed to ability results in feelings of pride and a sense of accomplishment (Tollefson, 1988). For the girls, their success was often attributed to effort, a changeable factor, which results in a decreased expectancy for future success (Weiner, 1984).
When Tapasak (1990) examined how males and females, themselves, attribute success and failure he discovered the same pattern. Males tended to attribute success to stable factors such as ability and failure to changeable factors such as effort. Women, on the other hand, made attributions in the exact opposite manner. Therefore, they attributed success to an unstable factor such as effort and failure to a stable factor such as ability; both situations often result in reduced expectancies for future success. These situations would appear to be a result of the preconceived stereotypes that individuals hold which have tended to propagate the idea that boys are better at math than girls.
The discussion thus far has been purely one of providing information about a situation that exists within the school system today. Such a task would be worthless without an examination of suggested solutions or intervention strategies to the problems created by differing expectations in the class.
From the study so far it is quite obvious that the problem in this situation lay not with the student but with the teacher. Therefore any discussion of solutions will naturally have to center upon those actions that teachers can take to correct where necessary their thoughts and behaviours.
On a very simple scale, Metcalf (1995) said that it is important to think differently about what you do in the class - to look beyond the negative behaviours... and focus on the positive. Using labels and stereotyping may help to rationalize the behaviours of students but it does not help to solve the problem. Teachers need to be better educated on the effects of racism and discrimination (Hall, 1993). Grant and Zeichner (1995) incorporated this line of thinking into their discussions on reflective teaching. They stressed that the reflective teacher must be dedicated and committed to the teaching of all students not just certain students. This would imply that teachers must reject the thoughts that restrict them in their teaching practices and develop new ways of viewing the teachability of all students. Weinstein (1995) felt it was vital for teachers to be exposed to situations whereby evidence, which he called 'disconfirming' evidence, would be provided to dispel the previously held notions about certain groups of students. Although this particular solution is tied to the notion of stereotypes, it is linked with far more significance to an overall strategy for success:
"...interventions need to provide an ongoing
context in which negative
beliefs can be disconfirmed and more positive beliefs and actions can be
developed, and which would enable teachers and administrators to play a
reflective and active role in both diagnosis and prevention of low
expectancy practices." (Weinstein, 1995, p. 126)
On a deeper or more sophisticated level, the changes necessary are much more complicated. Weinstein (1995) asserted that in order to create a 'positive expectancy climate' changes must occur to the eight interactive features of the organization of classrooms and school life. These eight features are:
b) grouping for instruction
e) student responsibility structures
f) relationships within the classroom
g) relationships with parents
h) relationships within the school
When concentrating on motivation and attribution theory, the possible solutions are quite similar in that the perceived causes of the events, and in particular failure, need to be changed (Weiner, 1984). Altering teacher behaviours when addressing the notion from the vantage point of expectancies is a deliberate step in the right direction. But like expectations, the underlying reasons for the behaviours must be dealt with if a profound and long lasting change is desired. Tollefson (1988) asserted that the solution resides in the elimination of the negative thoughts and actions and their replacement with positive ones. Teachers must become cognizant of the effects that subtle behaviours have on the attribution patterns of students. Once aware that gratuitous praise and pity communicate to the student that their, the students', ability is the cause of their poor results, the teachers can strive to communicate more productive and positive messages. Such radical changes in the thought processes of both teacher and student will require a commitment to work together to solve the problem.
In the area of expectations, collaboration and cooperation from teachers is paramount to the success of any intervention strategy designed to be used in a positive manner (Hassenpflug, 1994; Metcalf, 1995; Taylor, 1993; & Weinstein, 1995). Taylor (1993) insisted that success can only be achieved if teachers agree to train together and provide mutual support for the implementation of new teaching strategies to encourage active student engagement. Kolb and Jussim (1994) suggested that teachers must be educated about the subtle ways in which they may have created an environment that can depress some students' performance. A couple of studies are so confident of the relationship between expectations and student achievement that they stress the importance of maintaining high expectations as a viable intervention strategy (Kolb and Jussim, 1994 & Taylor and Reeves, 1993). In both cases, they felt that for any substantial rise in the achievement of students to occur, it was paramount to keep expectations high as a motivational factor. This would appear to mirror the message stated at the beginning of this paper concerning the document Adjusting the Course (1994, p. 10); "High expectations and standards are necessary and all students, except those with specific disabilities, should be able to meet those expectations and standards".
From a review of the literature it is apparent that the findings of the Coleman Report are dubious in light of the evidence which asserts a direct link between teacher expectations and student achievement. The research would also support the notion that a teacher can communicate 'why' the success or failure occurred influencing how the students attribute their success or failure. Although the exact nature of this relationship is open to some interpretation as to the degree to which perceptual bias, sustaining expectation effect, or self-fulfilling prophecy operate on those concerned, it has been proven that they do operate in some fashion.
First of all, there is evidence of some correspondence between what a teacher expects from a student and what that student's achievement levels turn out to be. How these expectations manifest themselves in the behaviour of teachers is the key to understanding the relationship. Studies have shown that high and low expectations on the part of the teacher lead to observable differences in achievement. Studies have also shown that the perceived cause of the event will influence the expectations for future success or failure and that these perceived causes can be transmitted from teacher to student. In both cases their interaction is closely linked through the behaviour of teachers toward students. The degree and type of feedback and amount of teacher interaction with the pupil are just two such behaviours where differences have been recorded.
Secondly, such expectations arise from the misguided behaviour of stereotyping and labeling. Many teachers hold certain expectations about a student based on the particular group to which that student belongs. Teacher expectations are influenced by gender, race, and socio-economic background just to name a few. Such expectations, based on an irrelevant factor such as gender, have shown that they influence the attributions of both teachers and students.
Due to their close interaction and reliance on the
communicated behaviours of
teachers', solutions to the expectancy/attribution problem are very similar. Efforts must
be made to dispel the underlying thought processes which guide and dictate
behaviours. The elimination of stereotypes and labels is a positive step in the right
direction. Secondly, teachers must become aware of how their obvious and not so
obvious behaviours communicate poor messages to the students. Once aware, they
can work to moderate their behaviour and direct behaviour which communicates
constructive messages to those who need them.
|1. provide them honest and contingent feedback on their responses.||1. give them less honest and contingent feedback but more gratuitous feedback.|
|2. elaborate on their responses.||2.accept their responses and go on to something else.|
|3.help them to arrive at the correct answer by providing them with clues||3.reject their response and call on someone else.|
|4.encourage them to provide open contributions.||4.call them for very brief and controlled contributions.|
|5.respect them as individuals with diverse needs and interests||5.have less respect for them as individuals with diverse needs and interests.|
|.6.treat them with warmth.||6.treat them with less warmth.|
|7.praise any of their efforts and assist them with their responses.||7.fail to praise their strong efforts but criticize their weak efforts.|
|8.encourage students to initiate interaction.||8.discourage students to initiate interaction.|
|9.give them freedom to express their feelings.||9.control their behaviour.|
|10.provide them with opportunities to achieve during group time.||10.provide them with limited opportunities to achieve during group time. (ignored or criticized)|
|11.permit students to reflect on their responses.||11.provide them with limited opportunities to respond to a question.|
|1. assign students to a high ability group with assignments which require students to use their analytical and comprehensive skills.||1.assign students to a low ability group with assignments which require them to work on meaningless tasks such as drill and practice.|
|2.allow them enough time to complete their tasks.||2.allow them limited time to complete their tasks.|
|1. give them more autonomy such as selecting assignments and hardly interrupt them.||1. limit their freedom such as constantly monitor their work and intrude.|
|2. encourage students to conduct self evaluations.||2. evaluate students or have another responsible person evaluate students.|
Source: Saracho, 1991.
Anderson, A., Vogel, P., & Reuschlein, P. (January/February, 1991). The implications of teacher expectations: A review of research. Capher Journal, 21-27.
Clairborn, W. (1969). Expectancy effects in the classroom: A failure to replicate. Journal of Educational Psychology, 60(5), 377-383.
Department of Education (1994). Adjusting the Course: Improving the conditions for learning. St. John's, NF: Government of Newfoundland and Labrador.
Fennema, E. & Others (1990). Teachers' attributions and beliefs about girls, boys, and mathematics. Educational Studies in Mathematics, 21(1), 55-69.
Gottfredson, D.C. (1995). Increasing teacher expectations for student achievement. Journal of Education Research, 88(3), 155-163.
Grant, C.A., & Zeichner, K.M. (1995). On becoming a reflective teacher. In G. Taylor and R. Runte (Ed.), Thinking About Teaching (pp. 54-67). Toronto, Ontario: Harcourt Brace.
Hall, J.L. (1993). What can we expect from minority students? Contemporary Education, 64(3), 180-182.
Hassenpflug, A. (1994). Notes from an English teacher: In pursuit of great expectations. Clearing House, 67(3), 161-162.
Kolb, K.J., & Jussim, L. (1994). Teacher expectations and underachieving gifted children. Roeper Review, 17(1), 26-30.
Kurtz, B. & others (1990). Strategy instruction and attributional beliefs in West Germany and the United States: Do teachers foster metacognitive development? Contemporary Educational Psychology, 15(3), 268-283.
Lee-Corbin, H. (1994). Teacher expectations and the able child. Early Child Development and Care, 98, 73-78.
Merton, R.K. (1948). The self-fulfilling prophecy. Antioch Review, 8, 193-210.
Metcalf, L. (1995). Great expectations. How changing your thinking can change your students. Learning, 23(5), 93-95.
Monhardt, B.M. (1995). Safe by definition. American School Board Journal, 182(2), 32-34.
Rosenthal, R., & Jacobson, L. (1968).
in the classroom: Teacher expectations and pupils' intellectual
development. New York: Holt, Reinhart & Winston.
Saracho, O.N. (1991). Teacher expectations of students' performance: a review of the research. Early Child Development and Care, 76, 27-41.
Tapasak, R.C. (1990). Differences in expectancy - attribution patterns of cognitive components in male and female math performance. Contemporary EducationalPsychology, 15(3), 284-298.
Taylor, R., & Reeves, J. (1993). More is better: Raising expectations for students at risk. Middle School Journal, 24(5), 13-18.
Tollefson, N. (1988). Consequences of teachers' attributions for student failure.Teaching and Teacher Education, 4(3), 259-265.
Towers, J. (1992). Twenty-five years after the Coleman Report: What should we have learned? Contemporary Education, 63(2), 93-95.
Weinstein, R.S. (1995). Raising expectations in schooling: obstacles and opportunities for change. American Educational Research Journal, 32(1), 121-159.