Cultural sensitivity in the application of behavior principles to education.

Sensitivity to cultural differences is generally agreed to be a highly desirable characteristic of sound practice in all human services. The terms "cultural competence," "cultural sensitivity," "cultural responsiveness," and "cultural appropriateness" are frequently seen in contemporary literatures (see Gollnick & Chinn, 2002; Lee, 2003; Shealey & Callins, 2007; Sleeter, 1999). Definitions of these terms are often not what behaviorists would consider operational. For example, Shealey and Callins (2007) state that "culturally responsive teaching refers to the extent to which educators use students' cultural contributions in transforming their lives and the lives of their families and communities by making education relevant and meaningful" (p. 195). They go on to say that demonstration of cultural sensitivity requires teachers to "learn about the cultures represented in their classrooms and translate this knowledge into instructional practice" (p. 196) and state that culturally mediated instruction is "characterized by the use of culturally mediated cognition, culturally appropriate social situations for learning, and culturally valued knowledge in curriculum content" (p. 196). As good as these ideas are, educators guided by them may be unable to identify a practice that is multicultural or culturally responsive or a practice that it is not. Teachers may also be unable to discriminate cultural sensitivity from cultural insensitivity.

Our purpose, therefore, was to seek evidence of responsiveness to behavioral interventions related to cultural identity. Thus, we reviewed behavioral literature related to three cultural markers--ethnicity, gender, and religion--to see whether we could find data supporting differences in behavioral interventions depending on cultural distinctiveness.

Some writers have suggested that behavioral interventions in education are inappropriate in or insensitive to some cultures. These writers find that the very characteristics that make behavioral interventions a practical tool for working in the scientific tradition may make them unacceptable for use with those who do not share the viewpoint of behavioral scientists. That is, they promote "culturally sensitive" practices that have little or no empirical support or urge teachers not to be "culturally insensitive" by using practices that do have considerable empirical support. For example, in his website on behavior management, McIntyre (2007) stated:

  Behaviorist interventions are often inappropriate for, and sometimes
  even discriminatory against large numbers of culturally different
  youngsters with learning and behavioral disorders. Readers are asked
  to ponder whether behaviorism's practices, despite the empirical
  research demonstrating their effectiveness, should be implemented with
  youngsters who do not adhere to a Western European life view and
  orientation.

The same essay concluded:

  Behaviorists' research may be able to "prove" that behavioral change
  occurs, but at what cost? Is a method, despite its empirical
  validation, appropriate for everyone? Is it morally right to change
  culturally different behavior in culturally insensitive ways? One need
  only ask former students made to attend Bureau of Indian Affairs
  schools in decades past, or those youngsters who are willing to comply
  and love to learn, but refuse to do so in ways that they perceive to
  be "acting white" (identifying with a group they believe to be
  oppressors). While behaviorist practices may be "effective", we must
  question if the means, or even the ends, are appropriate in all cases.

If data showing that effectiveness (regardless of how it is defined) is achieved with given methods, yet those methods are proscribed as culturally inappropriate, then we have no way to judge the effectiveness of interventions. Data showing better outcomes, even if desired by an individual, could be judged convincing only if the methods used to obtain them are judged culturally appropriate. Perhaps a judgment of cultural appropriateness should not trump data showing effectiveness for individuals, such that data showing better outcomes are subject to censure if the methods do not pass a test of cultural sensitivity. Another way of putting this is to ask: Could methods resulting in better outcomes or effectiveness for individuals be inherently culturally insensitive, culturally unresponsive, or culturally inappropriate, such as the situation described by McIntyre (2007)?

Some writers have advocated the use of scientifically groundless practices for teaching children from certain groups on the basis of the children's cultural identity. For example, in attempting to demonstrate sensitivity to Native American students' culture Holiday, Bitseedy, and Russell (1995) stated that "whole language, literature, and cooperative learning should be used to improve academic skills in the child's native language as well as in English" (p. 52), although the type of instruction known as whole language finds no support in empirical research (Kauffman & Sasso, 2006a).

We do not question the fact that the performance of schools in educating culturally and linguistically diverse (CLD) children has been dismal, nor do we deny the mistreatment of minorities, especially African Americans, that was a catalyst for the development of multicultural education. The goal of multicultural education is to transform educational systems so that all children (e.g., male and female, exceptional children, and CLD children) have an equal opportunity for academic success (Banks, 1997). Thus, we are challenged to consider carefully how we educate children most effectively in a diverse, democratic society. Carefully designed multicultural education should recognize the uniqueness of majority and minority children, better preparing all children who receive it to live in a world with increasing diversity. Indeed, if educators are to use the cultural diversity in schools as an advantage, then careful analysis and planning of how culture is incorporated in instruction is critical (Hallahan & Kauffman, 2006). However, in spite of the meaningful contributions of multicultural education (e.g., curriculum materials that reflect America's diverse population), CLD children continue to have high rates of academic failure, which has sparked calls for "culturally appropriate" or "culturally responsive" pedagogy (e.g., Shealey & Callins, 2007; Sorrells, Webb-Johnson, & Townsend, 2004).

One of the immediate problems of describing culturally sensitive, culturally competent, or culturally responsive interventions is defining culture. Although others have offered various definitions of culture (see Hallahan & Kauffman, 2006), there are many dimensions of cultural identity, including gender, sexual orientation, skin color, socioeconomic status, nationality, religion, exceptionality, and so on (Banks, 1994). All of these dimensions raise difficult issues about how cultural identity should be defined and what the implications of any such identity should be for educators.

For example, culture may defined by the place in which one's ancestors lived, one's skin tone or other physiognomic feature, or some other feature of individuals. The designation of groups as African-American, European-American, Asian-American, and so on is usually thought to be important in designating culture and making cultural sensitivity possible. However, the continents of Africa, Europe, and Asia are comprised of very heterogeneous groups. We only need to think of the religious conflicts in Northern Ireland, the genocides in Rwanda and Darfur, the Japanese-Chinese atrocities, the Sunni-Shiite conflicts, the ethnic cleansing in the former Yugoslavia, the Palestinian-Israeli fighting, the Israel-Lebanon wars, slavery in the Western hemisphere (including the United States), mistreatment of Indians in the Americas, maltreatment of the aboriginals in Australia, and so on to understand that no continent or people is free of ethnic and religious heterogeneity, discrimination, and brutality. Furthermore, some people who consider themselves Africans do not have black skin, and some people who consider themselves European or Asian do. Claiming an ancestral home may depend on how many generations of one's ancestors have lived in a particular place. We do not know when people become culturally English or Dutch or American or any other nationality, regardless of their ancestors. Whether someone can adopt a cultural identity that he or she does not have at birth remains an issue. The number of generations of one's ancestors that must have practiced a particular religion, how devout they must have been in their religion, and how devout an individual must be before he or she can claim to be of a particular religion for cultural purposes is unknown. Mixed ethnic origins or mixed religions also present difficult issues for defining culture. Gender and gender preferences are also cultural issues. The basic question for those who want to apply behavior principles to education is this: Under what circumstances is group membership more important than how an individual behaves or what an individual believes or wants?

As proponents of behavioral interventions, we need to be able to say what cultural competence or cultural sensitivity or cultural appropriateness means, and we should be able to answer questions such as these: Can a person be sensitive to an individual but insensitive to that person's culture? Can a person be sensitive to a culture but insensitive to an individual? Which is more important, sensitivity to a culture or sensitivity to an individual? And when, if ever, is the use of behavior principles itself sensitive or insensitive to an individual or to a group? Most of all, we are concerned about what the evidence suggests regarding cultural sensitivity in education.

We note that the Carter Center's work with poor people in various nations in Africa and Asia helped people produce more food and avoid disease, not by accommodating people's cultural traditions, which in fact had perpetuated or exacerbated their problems of starvation and illness, but by teaching them scientifically-based techniques that were contrary to their cultural traditions (see Carter, 2005, pp. 180-184). We wonder whether we could be selling children and adults short by deferring to what some describe as cultural--by allowing what is said to be cultural to prevail over scientific evidence. Writing from the perspective of the third world, Nanda (1998) suggests that the rejection of science in favor of cultural tradition is a mistake.

Although many dimensions of culture could be explored, we concentrated attention on three that we think are most relevant to problems of current practice in schools: ethnicity, gender, and religion. For each of these dimensions, we sought behavioral research that would help us understand the ways in which behavior principles apply in particular ways to specific cultural groups and what culturally sensitive practice entails for specific cultural groups.

Ethnicity

Free and effective public schooling is often considered a cornerstone of democracy, and the primary purpose of schools is frequently interpreted to mean teaching children skills that allow them to become successful and productive members of society. Indeed, teaching may be seen as the only profession upon which all other professions depend. If a person is to become an engineer, then she must first learn basic math facts; if a person is to become a poet, he must first learn to read and write. Some see education as the key to the American dream for the current generation of immigrant and impoverished children. Unfortunately, the education system is failing to educate appropriately and effectively far too many children. About 30% of students drop out before completing high school, and innumerable graduates are poorly prepared for life beyond high school (Bainbridge, 2006).

One undeniable fact is that CLD children, especially those from poor families, are not faring well in American public schools. Data provide a clear pattern of outcomes related to ethnicity and socioeconomic status (SES):

1. Impoverished minority children in urban schools experience reading failure at levels of 60-70% (Moats, 2005).

2. African American, Hispanic, and Native Americans are at higher risk than other groups for receiving a special education disability label and less likely to be placed in gifted education (Artiles, Trent, & Palmer, 2004; Donovan & Cross, 2002; Hosp & Reschly, 2004; Losen & Orfield, 2002).

3. African American, Hispanic, and Native American learners are at heightened risk for dropping out of school and for being suspended from school (Losen & Orfield, 2002).

Observations of these patterns have rightly created great concern for equity in schooling and for instruction that is sensitive to the cultures of children--specifically, to ethnicity and SES, often linked with ethnic identity. The data suggest that CLD students are often poorly taught, but the data do not suggest why they are poorly taught or whether interventions specific to ethnicity are important (e.g., Osher et al., 2004).

Recommendations for ethnicity-specific interventions and remedies have been presented without supporting data, and in some cases in the face of contradictory research findings (e.g., Holiday et al., 1995; McIntyre, 2007). These recommendations are troubling for another reason. Race, language, country of origin, religion, gender, or any single attribute of a person can lead to simplistic answers that do not provide clear information on the educational needs of individuals in the designated category.

Race, language, country of origin, and religious differences per se provide some useful information about individuals. However, these categories typically do not provide classroom teachers with the level of information required to provide meaningful academic instruction. A child whose ethnicity is Latino may be from any of many different cultures in South America, Mexico, or the Caribbean. The label gives some probability information about the child, such as the likelihood of his or her exposure to Spanish. However, the Latino label alone does not guarantee that the child is fluent in Spanish or even can speak Spanish at all! More valuable information about the child might be what language is spoken at home, the child's previous schooling, the presence of a disability, and what skills the child currently possesses.

Knowing the ethnicity of a student may be useful in some ways, but ethnic identity alone provides insufficient knowledge, just as knowing that a child has been given a special education label (e.g., specific learning disability--SLD) is not sufficient. The SLD label tells a teacher that the child has a higher probability of having reading difficulties than a child without the label. However, additional information is needed because a child's label of SLD does not guarantee that the child has reading problems. Those who advocate culturally specific interventions may be making the assumption that there is an instructional "silver bullet" for children based on their ethnicity. Although the problem is pedagogical, the solution to the problem is not necessarily ethnic-specific instructional strategies but empirically validated instructional strategies that are effective for the individual student.

The best contribution a teacher can make to the success of students is to employ empirically validated pedagogy and then monitor its effects on each student's academic progress (Bateman, 2004; Heward, 2003). Denying students effective instruction based on their ethnicity is, in our judgment, unethical and will only expand the achievement gap between CLD children and their peers.

We do not assert that behaviorism has all of the answers to education's ills, but we do believe that using behavior principles can play a critical role in correcting problems in education. Specifically, behavior principles are particularly well suited to addressing the challenges found in meeting the needs of a diverse student population, These principles include the following:

1. The student is always right (i.e., the student's behavior guides instruction).

2. Assessment includes an evaluation of reinforcement history (including behaviors and reinforcers influenced by culture, including ethnicity).

3. Teaching is seen as a science that involves the task analysis of skills, assessment of students' behavior, the careful development of an intervention, ongoing monitoring of the effects of instruction on the terminal skill, and systematic planning for generalization and maintenance of skills. Indeed only through science will we be able to address the complex problems facing educators.

4. Instructional techniques are completely identified and described (and, therefore, are easier for other educators to replicate).

5. The methodology requires robust, socially relevant results (thereby systematically eliminating strategies with negative, nonexistent, or weak results).

6. Single-subject research methodology involves the analysis of each participant's behavior so that the independent variable can be analyzed across individuals.

7. Single-subject designs have high internal validity.

Behavioral researchers have identified critical variables for promoting academic achievement, such as active student responding (Heward, 1994), corrective feedback (Barbetta, Heward, & Brady, 1993), schedules of reinforcement (Ferster, 2002), stimulus control (Tiger & Hanley, 2005), three-term contingency (Cooper, Heron, & Heward, 2007), establishing operations (Michael, 1988), functional analysis (Wallace, Doney, Mintz-Resudek, & Tarbox, 2004), maintenance and generalization (Baer, Wolf, & Risley, 1968) and so forth. Although educators who are behaviorists have not been the only professionals examining these variables, they have been prominent among those doing and interpreting research with direct implications for teachers (see Bateman, 2004; Engelmann, 1997; Heward, 2003; Kauffman, Mostert, Trent, & Pullen, 2006; Lloyd, Salzman, & Kauffman, 1981; Walker, Ramsey, & Gresham, 2004).

Furthermore, numerous studies have demonstrated that children from all cultural backgrounds benefit from behavioral interventions such as Precision Teaching (Therrien, Gormley, & Kubina 2006), peer tutoring (Al-Hassan, 2003; DuPaul, Ervin, Hook, & McGoey, 1998; Greenwood, Dinwiddie, Terry, Wade, Stanley, Thibadeau, & Delquadri, 1984), Direct Instruction (Engelmann, Meyer, Johnson, & Carnine, 1999), response cards (Gardner, Heward, & Grossi, 1994), guided notes (Heward, 1994), self-monitoring (Maag, Reid, & DiGangi, 1993), interteaching (Saville, Zinn, Neef, Van Norman, & Ferreri, 2006), and so forth. The robust effects of behavioral interventions would seem to demand that these strategies should be widely used by educators. However, education is still in the early stages of converting from a profession based on the art of teaching to a profession based on the science of teaching (Heward, 2003; Reyna, 2004). What this means is that, unfortunately, not all of our colleagues in education accept scientific data as the universal standard. Reyna (2004) stated:

Research is the only defensible foundation for educational practice.

If not scientific evidence then what?

Tradition -- is the way we always did things but that is not enough.

Philosophy -- offers appealing rhetoric but is often lacking evidence.

Superstition -- gives the illusion of correlation.

Anecdote -- is not representative.

Intuition -- is unreliable without additional study. (p. 48)

Much the same conclusions were reached by Kauffman and Sasso (2006a, 2006b) and Heward and Silvestri (2005). Science may not be perfect or provide all answers, but it is our best tool for helping people.

An issue of great importance is the social validity of the target behaviors and interventions employed with students differing in ethnicity. Educators must be aware of the conventions of language, dress, food, recreation, social interaction, and learning of the families and communities from which their students come and understand that these may be sources of proscription, prescription, or preference for certain behavioral targets and interventions. Some of these may be part of gender roles, religious teachings, or cultural traditions different from those of the teacher. However, educators must not just assume that a student's ethnic identity will determine these but must actively engage individual students and families to determine these matters. Some individuals may be members of ethnic groups in which certain language, dress, behavior, and so on are generally practiced but may also prefer alternatives to their group's traditions or norms that more closely resemble those of the larger or dominant society.

In fact, behavioral patterns that are normative in a given school may be misleading when analyzed at the individual level. For example, data from some schools suggest that a high proportion of students in these schools exhibit behavior that is seriously, persistently maladaptive and impairing. The stressors or risk factors students in these schools experience "may combine to create a school culture in which noncompliance is the norm and where peer reinforcement leads to students' acceptance and expectation of disruptive behavior in their peers" (Warren et al., 2003, p. 82). Nevertheless, students themselves, and particularly their parents, may prefer behavior that is non-normative in such a school. The only way to confirm the social validity of target behaviors and interventions for individuals is to have discussions with the individuals involved and take their thinking into consideration.

We offer cautions here about two important matters related to social validity. First, educators must guard against ethnic stereotypes in choosing target behaviors and interventions. Knowledge that a student is from a particular ethnic group must not be allowed by itself to determine the prescription or proscription of behavioral targets or interventions. Nothing can substitute for sensitivity to the ideas of the individual student and family. Second, judgments of social validity also involve the knowledge and willingness of the educator to make appropriate adaptations. For example, corporal punishment may be a cultural tradition or a preference of parents but may not be acceptable to the educator, and in such a case the educator may seek an alternative, noncorporal form of punishment. Likewise, tangible reward may not be a socially valid intervention in an individual case, and the educator may be well advised to use a more subtle form of social approval for desired behavior.

Conclusions Regarding Ethnicity

America's educators are faced with the challenge of educating an increasingly diverse school-age population (Gollnick & Chinn, 2002). Behavioral intervention strategies are effective for CLD students, and they appear to work effectively across all ethnicities. Thus, behavior principles appear to be culturally neutral. Unfortunately, educational decisions are often not made based on empirical data. Behavior analysts expect educators still clinging to the belief that education is an art and that all knowledge is constructed to be swayed solely by data, but too often they are not. Some time ago, one leader in the application of behaviorism to education said, "Effective educational methods are available. They have been available for a long time. They are mostly behavioral, structured, fast paced, and require a high proportion of regular daily practice" (Lindsley, 1992, p. 21).

Unfortunately, constructivist thinking tends to dominate the literature on culturally responsive instruction. Kauffman and Sasso (2006a) note that constructivism is not supported by empirical evidence, whereas direct instruction is (see also Bateman, 2004; Engelmann, 1997; Gardner, Sainato et al., 1994; Heward, 2003). Certainly, instruction should be important and useful to students, and the teacher should know students' families and understand their cultures. Nevertheless, attention to individual and cultural relevance alone results in ineffective instruction, further reducing the ability of CLD students to acquire the skills they need to think critically and solve problems creatively. The value of direct instruction is denigrated by constructivist views, leading many educators to see direct instruction as disadvantageous to CLD students if it is dominant, when in reality it is superior instruction that is advantageous for CLD learners (Gardner et al., 1994). The following quotation illustrates the point about constructivist views of education:

  In embracing constructivist views of learning, we do not mean to
  suggest that there is no place in schools for direct instruction,
  memorization, and basic skills instruction. When such transmission-
  oriented strategies predominate, however, their pedagogical value
  diminishes, much to the students' disadvantage. Such an approach to
  teaching does not give students opportunities to actively engage in
  learning and integrate new ideas and frameworks into their own ways of
  thinking. Therefore, students are less likely to learn to think
  critically, become creative problem solvers, and develop skills for
  working collaboratively--all qualities that are essential for success
  in life and work. (Villegas & Lucas, 2007, p. 30)

Educators must use behavior principles to improve outcomes for CLD learners. They can clearly address the issue of diversity by making CLD populations an explicit and major part of their agenda. Too often we hear the mantra that CLD status does not matter--it is just a matter of good instruction. Essentially we agree that the data support that assertion, but then we must ask this: "Why is good instruction not getting to CLD populations more reliably?" Educators can do more to infuse the existing technology that we know works into the school programs for struggling diverse learners (Heward, 2003). Currently, the majority of these at-risk learners are the responsibility of educators who persist in using failed practices--assumed to be culturally sensitive but having little or no support in reliable empirical evidence. We believe that educators sometimes use ineffective practices because these methods are the ones (a) they know, (b) with which they feel most comfortable, (c) they have been taught by mistaken teacher educators are best, and/or (d) they believe are most culturally sensitive in spite of evidence to the contrary.

All things considered, we believe that we can be passionately committed to education as a science while maintaining sensitivity to the cultural differences of individual children and their families. However, first and foremost we must recognize that the most culturally responsive practices are empirically validated instructional strategies (Heward, 2003). At this point the data seem to suggest that this applies to all children, regardless of their ethnicity.

Gender

Similar to issues related to ethnic diversity, the sensitivity of behavioral practices to differences in gender is an empirical question: "a small matter of proof," as Baer (2001) titled a book chapter. At least in human societies, gender can be viewed as a setting factor that may increase or decrease the likelihood of a particular behavioral interaction. For instance, consider the following question: Does being a female or male increase or decrease the likelihood that a child will be asked to play tackle football on the playground or be asked to play dress-up with dolls in the housekeeping area? Although these illustrations are stereotypical, we could argue that in American society the reverse-gender expectations or interactions are far less likely than the more traditional ones. Moreover, we recognize that in our diverse society there are different perceptions of appropriate gender roles, although subjugation and abuse of women are simply unacceptable.

Nevertheless, the answer to the question of how gender is related to behavior principles is not necessarily straightforward or simple when we evaluate individual differences. In many or most social situations, an individual's gender may increase or decrease the likelihood of a particular request or comment. However, until the data are gathered and examined for an individual student in a particular social interaction, the influence of a marker variable such as gender is unknown for that individual. On the surface, the role of gender as a setting factor that affects the occurrence of various behavioral interaction patterns appears to be a logical assumption. Nevertheless, we must ask whether data support this assumption.

Relatively few behavioral studies have examined the influence of participants' gender on their responses to behavioral interventions, and the few studies that have been conducted have produced mixed findings. For example, Schroeder, Schmittner, Esptein, and Preston (2004) found gender differences in treatment responses to heroin and cocaine dependency. Carpenter and Gatchel (1994) found gender differences in response to a videotape intervention designed to reduce dental anxiety. However, White, Sloan, and Spirito (2001) found no gender differences in headache characteristics and sleep adjustment of pediatric male and female participants following a behavioral treatment program. We found no studies at all examining the influence of gender on students in educational settings, and our conclusions are therefore only extrapolations from the examination of gender in other settings.

As in the case of ethnicity, the social validity of targets and interventions is an important goal, and one must guard against forming stereotypes. Gender roles, proscriptions, prescriptions, and preferences may be related to ethnic or religious values. The only way we see to avoid stereotyping and making the kind of errors associated with cultural insensitivity is to engage the individual student and family in decisions about targets and interventions.

Conclusions Regarding Gender

Given the lack of research focused on examining the role of gender in educational settings and the outcomes of behavioral interventions in education, it appears that gender has not been given much experimental attention in the behavioral literature. Adequately addressing the role of gender may be difficult or impossible with single-subject design methodology. On the one hand, documenting the influence of gender (or its absence) is an empirical question, and without data to support or reject the role of gender there is little we can say as behaviorists. On the other hand, determining the exact nature of gender's role in the behavioral operant might not serve a useful function for designing effective interventions for students. One of the strengths of a behavior analytic approach to science and educational interventions is the examination of individual differences--finding out what works and what does not work for the individual. It is better to analyze behavior within the individual, examining the direct evidence, rather than marker variables such as gender, age, religion, or ethnicity. Obviously, individual reinforcement histories affect how a student responds to a particular treatment. Perhaps, in a particular societal context, females are likely to receive reinforcement for specific behaviors and males are likely to receive reinforcement for different behaviors; thus, individuals may respond to different types of stimuli based on this reinforcement history and, indirectly, their gender.

Replication of research and comparisons of research findings across studies may yield important data. We may want to consider examining the role of gender on the effectiveness of education by letting the data speak for themselves. To date, however, we were able to find no data indicating that behavior principles work differently for males than for females or are inherently insensitive in educational settings to individuals differing in gender.

Religion

The issue of religion and behavioral intervention can be examined at four levels: (a) philosophy or world view, (b) principles, (c) practices, and (d) behavioral targets. At each level we attempted to assess the influence of differences in religious heritage and belief on the application of behavior principles.

At the level of philosophy or worldview, we might ask whether behavioral theory is compatible with any religious belief. Contrast B. F. Skinner's assumption that events and behaviors in the world are entirely the product of the interaction between individuals' genetic, biological, constitutional makeup and their environment (e.g., Skinner, 1938, 1971) with, on the other hand, the assumption of many of the world's religions that some events and behaviors are the result of the influence of nonphysical, spiritual, or supernatural factors.

Virtually all of the world's religions are morally prescriptive in that they promote a specific view of what is good or ideal. Behavioral theory itself is essentially amoral, that is, "neutral with respect to what would constitute a personal ideal or ideal person" (Woolfolk & Richardson, 1984, pp. 780-781, in Jones, 1988). This issue is partially resolved by acknowledging that behaviorism is not anti-moral; its theory simply does not address the issue of moral judgments. Furthermore, we note that values and moral judgment have been evaluated by behaviorists (e.g., Skinner, 1971) and without appeal to religious dogma (e.g., Dawkins, 2006). Nonetheless, fundamentally different views of morality may affect how persons of differing religious perspectives respond to behavioral intervention. Furthermore, persons without belief in any god may hold views of moral behavior that conform to or surpass (in prescriptions of prosocial behavior and proscriptions of cruelty) those of persons with religious beliefs (see Dawkins, 2006). Thus, a basic issue in applying behavior principles to individuals differing in religion is the social validity of the ideas--the acceptability of the behavioral targets and interventions chosen by the educator for use with the student in the light of the student's and family's religious beliefs.

A fundamental principle of behavioral theory is that a person's behavior can be altered by introducing a change to setting events, discriminative stimuli, or consequences associated with a specific behavior. We have not found instances in which the tenets of a religious faith would require an alteration of basic behavioral principles or render them invalid. Indeed, some mental health interventions based on religious or spiritual principles have been "translated" into language consistent with behavior principles, particularly in the treatment of addiction. Twelve-step programs generally include reference to a "higher power" from which one seeks forgiveness and solicits help. Neff and MacMaster (2005) reinterpreted the effects of such programs in terms of behavior principles. They emphasized that social learning processes such as peer modeling of desired behaviors, enhancing self-efficacy, reinforcing skills, and providing social support are easily identified in many twelve-step programs. Examples such as this support the position that fundamental behavioral principles apply to interventions that on the surface appear to be specific to a given religious perspective.

In some circumstances, the effectiveness of a specific practice may depend to some extent on religious heritage. For example, we may have evidence to support the position that extinction is the preferred intervention to reduce a child's cursing. Theoretically, if the practice of extinction is implemented with integrity, the effect should be the same whether the context is a conservative Christian home or a liberal atheist home. Although a parent may agree that extinction is the best procedure for dealing with cursing and he or she may agree to go along with completely ignoring the child's cursing, the parent may be incapable of withholding all reaction because of religious beliefs, and the intervention will therefore fail. The literature on the social acceptability of interventions has provided a useful framework for understanding this issue (see Elliott, Witt, Kratochwill, & Stoiber, 2002, especially pages 259-268).

In the example involving extinction of cursing, a family's religious heritage may not affect the result that an intervention would have if it were implemented perfectly, but rather the ability of the parent to implement the intervention with integrity. On that basis, we may choose a different intervention. This is not an example of a differentially effective practice, but in practical terms the result is the same. We may choose not to recommend an effective intervention for some deeply religious families because we understand that they will never be able to implement it successfully. Lau (2005) makes this point when she notes that given religious communities may respond poorly to certain therapy components because of "factors that constrain engagement in the intervention" (p. 4). In the end, however, there is no evidence to suggest that an extinction procedure for cursing, implemented by a classroom teacher, will be any more or less effective depending on the religious heritage of the child.

Another way in which religious heritage may affect the practice of behavioral intervention relates to the selection of reinforcers. Families differ, in part as a function of religious belief, with respect to what they are willing to offer their children as rewards (e.g., comic books, attending a school dance, watching the Simpsons or South Park, listening to hip-hop music). Interventions may be derailed because parents will not allow their children access to certain reinforcers based on religious values or beliefs, or because children have internalized their parents' values and beliefs to the extent that the available rewards do not serve as reinforcers. Such differences clearly may affect the practice of behavioral interventions in school settings, but religious heritage is only one of many possible determinants of whether a given event is a reinforcer. Behavioral theory accommodates these multiple determinants, and the principles regarding the effects of reinforcing events are unaltered.

In sum, we found no evidence that response to behavioral intervention is a direct function of religious heritage, although in practice the efficacy of interventions may vary across religious groups because of differential willingness or ability to implement an intervention with integrity. Variations across religions may occur because of one or more of the following: (a) components of the practice are more acceptable or less acceptable to various religious groups, (b) the effect of the intervention is based on the individual identifying with an idea or a metaphorical construct that may be differentially meaningful depending on religious values and beliefs, (c) different religious heritages value differently the specific factors that mediate an intervention's effect.

It has long been recognized that the selection of targets for behavioral interventions extends beyond the science of behavior change into the realm of ethics and philosophy. The selection of target behaviors is an expression of values, and values differ substantially across subgroups defined by religious heritage as well as other dimensions of culture. Although some targets are likely to be nearly universally endorsed across religious groups (e.g., being able to read or reducing students' aggression in the classroom), others may be more controversial. Consider behaviorally-based public health interventions to increase child immunizations and the effect of those interventions among Christian Scientists or the Amish. Regardless of the salience of the prompts or the expected power of the rewards, some targets will not be equally affected across faith groups because of beliefs that find those targets unacceptable.

Contemporary educators generally accept the premise that teaching critical thinking skills is a valued enterprise, and that increasing critical thinking is an appropriate target for intervention. However, there are religious subgroups whose practice, if not their explicit policy, is to favor acceptance of authority over critical thinking. If we have an intervention that can be empirically demonstrated to increase critical thinking skills, these groups might reject it because it works, i.e., precisely because it effectively fosters a class of behaviors that the group has defined as undesirable. For example, an intervention that successfully fosters assertiveness may be rejected if it is offered to women in a traditional Islamic society or a conservative Mennonite community. An intervention that reliably produces an outcome that the community defines as undesirable or harmful would generally not be viewed as effective treatment; indeed, the civil powers in a given society might label such interventions "crimes."

Conclusions Regarding Religion

Research suggesting that a student's response to behavioral operations and treatments is a function of religious identity is relatively underdeveloped. There are conceptual and empirical strands in the literature that bear on the topic and that will lead to productive investigation of behavioral theory and practice. Such investigation might focus on areas such as a re-examination of the philosophical roots of behavioral theory, further exploration of how the social acceptability of interventions is affected by religious belief, or an investigation of how the process for selection of target behaviors affects the diffusion and the effects of behavioral practices in social subgroups, including those defined by religious belief. We do not find, however, any evidence that school-based behavioral instruction and intervention will be differentially effective depending on the religious heritage of the student.

General Conclusions

Our purpose was to review literature related to the culturally sensitive application of behavior principles to education. Specifically, we wanted to know whether behavioral interventions would have different outcomes depending on the cultural identity of the target individuals.

We found no research evidence that behavioral interventions work differently with students depending on their ethnicity, gender, or religion. Of course, there may be differences in behavior and likely reinforcers in groups defined by these variables. Moreover, socialization may play a role in shaping behavior, selecting reinforcers, and judging the acceptability of interventions. The basic principles of behavior, however, appear to operate without regard for the dimensions of culture we examined. We found no empirical reason to believe that a science of behavior is anything other than applicable to or transportable across all cultures.

We found no evidence suggesting that interventions based on the cultural markers we examined (ethnicity, gender, and religion) are more likely than others to be successful. In fact, instruction or other intervention based on a single variable or combination of variables unrelated to instructional or behavioral needs of the individual student (but by which students and others are commonly grouped into cultures; e.g., females, Europeans, Muslims, Native Americans, urban black males) finds no support in the behavioral literature. Furthermore, we fear that proposing or basing interventions on such group classifications may be counterproductive, even in sincere attempts to achieve fairness.

The fact that we found no evidence of cultural differences in responding to behavior principles does not mean that these differences do not exist. We note that behavioral psychologists and educators need to study such possible differences with much greater focus and detail. Many studies reported in leading behavioral journals (e.g., Journal of Applied Behavior Analysis, Behavior Therapy, Behavior Research and Therapy, Behavior Modification, Journal of Behavioral Education, and so on) do not report sufficient detail about the cultural identities of participants to allow analysis of cultural differences and do not analyze cultural differences specifically. We also note that ABA research designs are uniquely suited to the study of cultural differences. Whereas group designs violate principles of cultural sensitivity by their very nature (because results are assumed to apply to all members of a sample or are generalized to a population without regard for individuality), ABA designs evaluate individual responses to interventions.

We conclude that cultural sensitivity in the application of behavior principles to education means measuring the effects of scientific interventions in achieving socially valid behavioral objectives through procedures that are acceptable. We understand that definitional confusion remains at the edges of many useful concepts (e.g., Sidman, 2006 in discussing positive and negative reinforcement). Nevertheless, we attempt to clarify some of the more obvious points of possible confusion about our definition. First, behavioral intervention depends on a science of behavior, and this requires measurement of outcomes. Interventions not supported by scientific investigation (e.g., whole language reading instruction, facilitated communication, exorcism) have no place in our definition. This does not mean that behaviorally-oriented educators must eschew all extrascientific practices (e.g., prayer) in their personal lives, but it does mean that they must avoid nonscientific practices in helping students learn. Second, socially valid behavioral objectives are those targets of intervention judged to be desirable after discussion with the parents and students themselves (if the students are able to understand the issue). That is, parents and students must see the behavioral objective as meaningful or worthwhile. Third, parents and students must deem the intervention procedures acceptable. Procedures may be expressed in terminology that parents and students find acceptable, but this does not mean abandoning the science of behavior. Describing scientific principles in terms that are acceptable to a particular person is defensible; misrepresenting the principles involved is not.

In short, we believe that cultural sensitivity in education is based on scientifically grounded regard for the individual student. In our opinion, it is impossible for educators to be sensitive to the needs of a student while being insensitive to that student's culture. Ultimately, ineffective procedures are insensitive to both the needs of the student and to his or her culture in spite of any attempts to cast them as "culturally responsive." In all cases, cultural sensitivity requires using the most effective acceptable means to help students achieve socially validated objectives.

Educators have a responsibility to explain to those whose cognitive ability allows informed choice the benefits of applying the science of behavior. A long cultural tradition may perpetuate a condition for which treatment is desired (cf. Carter, 2005), and it is incumbent upon those with knowledge of science to explain the likely benefits and costs of applying scientific principles to the solution of a problem. In this regard, Nanda (1998) stated, "We prefer the cold, objective facts of science to the comfortable, situated knowledge of our ancestors for the simple reason that we refuse to subordinate what is good to what is ours" (p. 299). Of course, sensitivity to cultural tradition, which we prefer to call social validity, necessitates ultimately deferring to the informed choice of the responsible party. However, if that party insists on an intervention that is nonscientific, then the educator should decline to offer the intervention. Finally, students or parents could request help in achieving objectives inimical to the educator's cultural values, in which case the educator should rightfully decline to help, explain why, and refer the parent and student to others.

References

Al-Hassan, S. (2003). Reciprocal peer tutoring effect on high frequency sight word learning, retention, and generalization of first-grade and second grade urban elementary school students. Unpublished dissertation, The Ohio State University.

Artiles, A. J., Trent, S. C., & Palmer, J. (2004). Culturally diverse students in special education: Legacies and prospects. In J. A. Banks & C. M. Banks (Eds.), Handbook of research on multicultural education (2nd ed.) (pp. 716-735). San Francisco: Josey-Bass.

Baer, D. M. (2001). A small matter of proof. In W. T. O'Donohue, D. A. Henderson, S. C. Hayes, J. E. fisher, & L. J. Hayes (Eds), A history of the behavioral therapies: Founders' personal histories (pp. 253-265). Reno, NV: Context Press.

Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied behavior analysis. Journal of Applied Behavior Analysis, 1, 91-97.

Bainbridge, W. (2006, April 8). High-school dropouts cost the nation billions. Columbus Dispatch, A14.

Banks, J. A. (1994). Multiethnic education: Theory and practice (3rd ed.). Boston: Allyn & Bacon.

Banks, J. A. (1997). Multicultural education: Characteristics and goals. In J. A. Banks & C. A. Banks (Eds.), Multicultural issues and perspectives (3rd ed.). Boston: Allyn & Bacon.

Barbetta, P. M., Heward, W. L., & Bradley, D. M. C. (1993). Relative effects of whole-word and phonetic-prompt error correction on the acquisition and maintenance of sight words by students with developmental disabilities. Journal of Applied Behavior Analysis, 26, 99-110.

Bateman, B. C. (2004). Elements of successful teaching: General and special education students. Verona, WI: IEP Resources.

Carpenter, D. J., & Gatchel, R. J. (1994). Effectiveness of a videotaped behavioral intervention for dental anxiety: The role of gender. Behavioral Medicine, 20, 123-133.

Carter, J. E. (2005). Our endangered values: America's moral crisis. New York: Simon & Schuster.

Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis (2nd ed.). Upper Saddle River, NJ: Merrill/Prentice Hall.

Dawkins, R. (2006). The god delusion. New York: Houghton Mifflin.

Donovan, S., & Cross, C. (Eds.) (2002). Minority students in special and gifted education. Washington, DC: National Academy Press.

DuPaul, G. J., Ervin, R. A., Hook, C. L., & McGoey, K. E. (1998). Peer tutoring for children with attention deficit hyperactivity disorder: Effects on classroom behavior and academic performances. Journal of Applied Behavior Analysis, 31, 579-592.

Elliott, S. N., Witt, J. C., Kratochwill, T. R., & Stoiber, K. C. (2002). Selecting and evaluating classroom interventions. In M. R. Shinn, H. M. Walker, & G. Stoner (Eds.), Interventions for academic and behavior problems II: Preventive and remedial approaches (pp. 243-294). Bethesda, MD: National Association of School Psychologists.

Engelmann, S. (1997). Theory of mastery and acceleration. In J. W. Lloyd, E. J. Kameenui, & D. Chart (Eds.), Issues in educating students with disabilities (pp. 177-195). Mahwah, NJ: Lawrence Erlbaum Associates.

Engelmann, S., Meyer, L., Johnson, G., & Carnine, L. (1999). Corrective reading. Columbus, Ohio: SRA/McGraw Hill.

Ferster, C. B. (2002). Schedules of reinforcement with Skinner. Journal of the Experimental Analysis of Behavior, 77, 303-311.

Gardner, R., Heward, W. L., & Grossi, T. A. (1994). Effects of response cards on student participation and academic achievement: A systematic replication with inner-city students during whole class science instruction. Journal of Applied Behavior Analysis. 27, 63-71.

Gardner, R., Sainato, D., Cooper, J., Heron, T., Heward, W., Eshleman, J., & Grossi, T. (Eds.). (1994). Behavior analysis in education: Focus on measurably superior instruction. Belmont, CA: Brooks/Cole.

Gollnick, D. M., & Chinn, P. G. (2002). Multicultural education in a pluralistic society (6th ed.). Upper Saddle River, NJ: Merrill/Prentice Hall.

Greenwood, C. R., Dinwwiddie, G., Terry, B., Wade, L., Stanley, S. O., Thibadeau, S., & Delquadri, J. C. (1984). Teacher-versus peer-mediated instruction: An ecobehavioral analysis of achievement outcomes. Journal of Applied Behavior Analysis, 17, 521-538.

Hallahan, D. P., & Kauffman, J. M. (2006). Exceptional learners: Introduction to special education (10th ed.). Boston: Allyn & Bacon.

Heward, W. L. (1994). Three "low-tech" strategies for increasing the frequency of active student response during group instruction. In R. Gardner, D. Sainato, J. Cooper, T. Heron, W. Heward, J. Eshleman, & T. Grossi (Eds.), Behavior analysis in education: Focus on measurably superior instruction (pp. 283-320). Belmont, CA: Brooks/Cole.

Heward, W. L. (2003). Ten faulty notions about teaching and learning that hinder the effectiveness of special education. The Journal of Special Education, 36, 186-205.

Heward, W. L., & Silvestri, S. M. (2005). The neutralization of special education. In J. W. Jacobson, R. M. Foxx, & J. A. Mulick (Eds.), Controversial therapies for Developmental Disabilities (pp. 193-214). Mahwah, NJ: Erlbaum.

Holiday, L., Bitseedy, W., & Russell, G. W. (1995). Through Navajo eyes: Curriculum guidelines from a teacher's perspective. Multiple Voices, 1(1), 50-54.

Hosp, J. L. & Reschly, D. J. (2004). Disproportionate representation of minority students in special education: Academic, demographic, and economic predictors. Exceptional Children, 70, 185-199.

Jones, S. L. (1988). A religious critique of behavior therapy. In W. R. Miller & J. E. Martin (Eds.), Behavior therapy and religion: Integrating spiritual and behavioral approaches to change (pp. 139-170). Newbury Park, CA: Sage.

Kauffman, J. M., Mostert, M. P., Trent, S. C., & Pullen, P. L. (2006). Managing classroom behavior: A reflective case-based approach (4th ed.). Boston: Allyn & Bacon.

Kauffman, J. M., & Sasso, G. M. (2006a). Toward ending cultural and cognitive relativism in special education. Exceptionality, 14, 65-90.

Kauffman, J. M., & Sasso, G. M. (2006b). Certainty, doubt, and the reduction of uncertainty: A rejoinder. Exceptionality, 14, 109-120.

Lau, A. (2005). The whys, whens, whats, and hows of cultural adaptation of evidence-based treatments. In Balance, 20, 4-5.

Lee, C. D. (2003). Literacy and culture. In J. W. Guthrie (Ed.), Encyclopedia of education (2nd ed.) (pp. 1509-1513). New York: Macmillan Reference.

Lindsley, O. R. (1992). Why aren't effective teaching tools widely adopted? Journal of Applied Behavior Analysis, 25, 21-26.

Lloyd, J. W., Salzman, N. J., & Kauffman, J. M. (1981). Predictable generalization in academic learning by preskills strategy training. Learning Disability Quarterly, 4, 203- 216.

Losen, D. J., & Orfield, G. (2002). Racial inequality in special education. Cambridge, MA: Harvard Education Press.

Maag, J. W., Reid, R., Digangi, S. A. (1993). Differential effects of self-monitoring attention, accuracy, and productivity. Journal of Applied Behavior Analysis, 26, 329-344.

McIntyre, T. (2007). Are behaviorist interventions inappropriate for culturally different youngsters with learning and behavior disorders? http://maxweber.hunter.cuny.edu/pub/eres/ED-SPC715_MCINTYRE/CBehModR.html. Retrieved April 14, 2007.

Michael, J. (1988). Establishing operations and the mand. The Analysis of Verbal Behavior,6, 3-9.

Moats, L. C. (2005). Speech to print: Language essentials for teachers. Baltimore: Paul H. Brookes.

Nanda, M. (1998). The epistemic charity of the social constructivist critics of science and why the third world should refuse the offer. In N. Koertge (Ed.), A house built on sand: Exposing postmodernist myths about science (pp. 286-311). New York: Oxford University Press.

Neff, J. A., & MacMaster, S. A. (2005). Applying behavior change models to understand spiritual mechanisms underlying change in substance abuse treatment. American Journal of Drug and Alcohol Abuse, 31, 669-684.

Osher, D., Cartledge, G., Oswald, D., Sutherland, K. S., Artiles, A. J., & Coutinho, M. (2004). Cultural and linguistic competency and disproportionate representation. In R. B. Rutherford, M. M. Quinn, & S. R. Mathur (Eds.), Handbook of research in emotional and behavioral disorders (pp. 54-77). New York: Guilford.

Reyna, V. F. (2004). Why scientific research? The importance of evidence in changing educational practice. In P. McCardle, & V. Chhabra (Eds.), The voice of evidence in reading research (pp. 47-58). Baltimore: Paul H. Brookes.

Saville, B. K., Zinn, T. E., Neef, N. A., Van Norman, R., & Ferreri, S. J. (2006). A comparison of interteaching and lecture in the college classroom. Journal of Applied Behavior Analysis, 39, 49-61.

Schroeder, J. R., Schmittner, J. P., Esptein, D. H., & Preston, K. L. (2004). Adverse events among patients in a behavioral treatment trial for heroin and cocaine dependence: Effects of age, race, and gender. Drug and Alcohol Dependence, 20, 45- 51.

Shealey, M. W., & Callins, T. (2007). Creating culturally responsive literacy programs in inclusive classrooms. Intervention in School and Clinic, 42, 195-197.

Sidman, M. (2006). The distinction between positive and negative reinforcement: Some additional comments. The Behavior Analyst, 29, 135-139.

Skinner, B. F. (1938). The behavior of organisms. New York: Appleton-Century-Crofts.

Skinner, B. F. (1971). Beyond freedom and dignity. New York: Knopf.

Sleeter, C. E. (1999). Making choices for multicultural education: Five approaches to race, class, and gender (3rd ed.). New York: Wiley.

Sorrells, A. M., Webb-Johnson, G., & Townsend, B. L. (2004). Multicultural perspectives in special education: A call for responsibility in research, practice, and teacher preparation. In A. M. Sorrells, H. J. Reith, & P. T. Sindelar (Eds.), Critical issues in special education: Access, diversity, and accountability (pp. 73-91). Boston: Allyn & Bacon.

Therrien, W. J., Gormley, S., & Kubina, R. M. (2006). Reread-adapt and answer-comprehend: Using repeated reading and question generation to improve reading achievement. Teaching Exceptional Children, 38(3), 22-26.

Tiger, J. H., Hanley, G. P. (2005). An example of discovery research involving the transfer of stimulus control. Journal of Applied Behavior Analysis, 38, 499-509.

Villegas, A. M., & Lucas, T. (2007). The culturally responsive teacher. Educational Leadership, 64(6), 28-33.

Walker, H. M., Ramsey, E., & Gresham, F. M. (2004). Antisocial behavior in school: Strategies and best practices (2nd ed.). Pacific Grove, CA: Brooks/Cole.

Wallace, M. D., Doney, J. K., Mintz-Reudek, C. M., & Tarbox, R. S. F. (2004). Training educators to implement functional analyses. Journal of Applied Behavior Analysis, 37, 89-92.

Warren, J. S., Edmonson, H. M., Griggs, P., Lassen, S. R., McCart, A., Turnbull, A., & Sailor, W. (2003). Urban applications of school-wide positive behavior support: Critical issues and lessons learned. Journal of Positive Behavior Interventions, 5, 80-91.

White, K.S., Sloan, A.C., & Spirito, A. (2001). Characteristics of children presenting to a behavioral treatment program for pediatric headache. Journal of Clinical Psychology in Medical Settings, 8, 109-117.

James M. Kauffman

University of Virginia

Maureen Conroy

Virginia Commonwealth University 

Ralph Gardner, III

Ohio State University

Donald Oswald

Medical College of Virginia 

This paper is based on presentations made by the authors at the 32nd Annual Convention of the Association for Behavior Analysis in Atlanta, GA on May 28, 2006.

Correspondence to James M. Kauffman, Department of Curriculum, Instruction and Special Education, Curry School of Education, 405 Emmet Street South, University of Virginia, Charlottesville, VA 22904; e-mail: [e-mail].