This study considers one way to make more productive use of information in a recognised survey instrument, the Parental Attitudes to Inclusion (PATI) scale and, thus, to enhance inclusive classroom practice for students with special needs. The instrument, designed to elicit views about inclusion, was initially administered to a large sample of Californian parents of inclusion-eligible children with severe cognitive disabilities. The present study gathered responses from a convenience sample of 10 Queensland parents of children diagnosed with autistic spectrum disorder and enrolled in regular Queensland classrooms. This study demonstrates that it is possible to identify specific parental attitudes that differ significantly from those of the group, and that this information can be used to target points for discussion between teacher and parent. This methodology raises general issues about the context in which an attitude is expressed. For example, attitudes may refer to the inclusion of a specific child (parent view) or a group of children (teacher view). This study raises the issue of the moral weight of a group-based view of inclusion. This study also discusses the need to take into account the effect of context on the formulation of views about inclusion, and proposes refinements to the PATI scale that would allow it to be used to identify the locus of such contextual differences.
Disability advocacy and consumer rights, enacted in antidiscriminatory legislation, have prepared the way for students with special needs to acquire a right to regular education (Fuchs & Fuchs, 1994). Studies of stakeholder attitudes to inclusion have affirmed the principled, philosophical support for inclusion promoted by some parent groups and by sectional interests within the educational community. Most research on inclusion has sought to clarify the consensus about inclusion and to document the boundaries of values-based beliefs that support inclusive practice. This work has involved surveys of views about the enrolment of children with special needs in the regular classrooms.
Characteristically, attitudinal research related to inclusion has not provided guidance to improve inclusive practice. Salend (1999) argued that the ease and speed of surveys make them particularly suitable for teachers to evaluate inclusive practices. Salend (1999, 2002) speculated that the Parent Attitude to Inclusion (PATI) scale designed by Palmer, Borthwick-Duffy, and Widaman (1998a) was one possible tool for change, but did not indicate how this survey might enable changes to be made. The present study examines Salend's proposal for teacher use of parent data from the PATI scale, and considers how a survey used mostly to map the sharing of current values might be used as a tool for changing practice. Given that the PATI scale was used to survey group-based patterns of shared belief about inclusion, the critical issue identified in this study is how to use survey data to communicate better with individual parents.
In practice, the right of children with special needs to be enrolled in regular schools and to be included in regular classes has created difficulties. Parental aspiration, teacher expertise, school resources, and system policy have influenced how this right is exercised. Parents and educators often hold conflicting views, with the issues confronting parents and schools typically seen as multidimensional (e.g., appropriateness of educational services, reciprocal benefits, social context for learning). Attitudinal research on inclusion has confirmed multiple, sometimes discrepant, and ultimately unresolved perspectives on people's attitudes (Elkins et al., 2004). Therefore, the process of finding consensus and identifying areas of agreement traverses a range of educational issues based on what is affirmed and what is experienced.
Perhaps it is because achieving consensus about inclusive acceptance among stakeholders is such a dynamic process, that studies of general attitudes towards inclusion and its benefits (i.e., what stakeholders want and why they want it) have continued to attract the most research effort. The need to articulate who to include, why to include, and when to include has maintained a research focus on perspectives, values, and beliefs. The present study acknowledges this diversity of views, in opting to target the specific points at which parent responses affirm or depart from a general consensus with other parents of children with special needs.
Surveys of principals and educational administrators in Australia and the USA have indicated positive opinions about inclusion, but reported caution about training and resourcing (Bailey & du Plessis, 1998; Downing, Eichinger, & Williams, 1997). Bailey and du Plessis (1998) identified three factors in their study of Queensland principals. First, previous special education training was associated with more positive attitudes to managing inclusion. Second, aggressive behaviour was seen as a problem. Third, primary principals were more negative than secondary principals.
Recent studies of teachers in general education have been characterised by cautiously positive opinions (Cook, Tankersley, Cook, & Landrum, 2000; Coots et al., 1998; Wright & Sigafoos, 1998b). Despite their positive views, teachers have expressed frustrations about in class planning and instruction. For instance, one sample of 15 Queensland teachers (Wright & Sigafoos, 1998a) generated supportive comments about regular classroom placement, but expressed caution about implementation issues.
The attitudes of non-disabled peers have been comparable with those of other school-based stakeholders. In inclusive classrooms, non-disabled peers have reported a mixture of positive attitudes and ambivalence about benefits. They have also reported some reluctance in socialising (Fisher, Pumpian, & Sax, 1998). Upper primary Queensland students reported mixed views of personal disadvantage and benefits to all students (Wright & Sigafoos, 1998a), over and above their broad awareness of included students in the school community.
Parental views of what happens to their children when they are included in regular classrooms have been surveyed (Ryndak, Downing, Morrison, & Williams, 1996; Wright & Sigafoos, 1998b; Palmer, Borthwick-Duffy, Widaman, & Best, 1998; Palmer, Fuller, Arora, & Nelson, 2001). In line with their placement decision, parents have been positive about inclusive settings and cautiously positive about instructional and social benefits.
In summary, reviews of attitudinal studies and interviews serve to confirm the common agreement on a general right to inclusion and the need for extra support and training. Over and above this general expression of support for inclusion, the attitudes, questions, answers, and evaluation of attitudes of stakeholder groups become dissonant. That is, despite in principle agreement about inclusion, opinions remain mixed about practices related to inclusion.
Parents and teachers are most closely concerned with inclusion of the student with special needs in a regular classroom. Existing data have indicated negative perceptions, for example, teacher attitudes about working in inclusion programs; parental demands for inclusive placements irrespective of the disabling condition; teacher and parent fears about lack of academic, social, and behavioural benefits. Further examining these negatives might help teachers and parents, either to reach agreement on effective inclusive practice, or to specify areas of disagreement, as precursors to changes in expectations, practice, or both. Palmer et al. (1998a) have argued that parents' views of what is working for their child constitute an important but neglected teaching resource. Salend (1999) argued that, within the classroom, teachers and parents need tools that help them to examine their reflections on existing and desirable practice and to share perspectives. Yet he did not explain how to achieve outcomes that are useful in practice. The present study set out to bridge that gap.
The PATI's construction deliberately incorporated the "multidimensional nature of parent perceptions regarding inclusive practices for children with significant cognitive disabilities" (Palmer et al., 1998b, p. 273). In particular, they specified three dimensions: quality of educational services, mutual benefits for the included child and non-disabled peers, and socio-emotional issues of peer acceptance and self-feelings. Two of the factors, labelled Quality of educational services (Items 5, 6, 9 and 10) and Child acceptance and treatment (Items 2 & 3), focused on the benefits of inclusion for these students. The third factor, labelled Mutual benefits of inclusion (Items 1, 4, 7, 8 and 11), focused on relations with other students.
|01||The more time my child spends in a regular classroom, the more likely it is that the quality of his/her education will improve.|
|02||The more time my child spends in a regular classroom, the more likely it is that he/she will be mistreated by other nondisabled students in that room.|
|03||The more time my child spends in a regular classroom, the more likely it is that he/she would end up feeling lonely or left out around the regular education students.|
|04||When a student with severe disabilities is enrolled in a regular education classroom, the positive benefits to the regular education students outweigh any possible problems that this practice may present.|
|05||It is impossible to modify most lessons and materials in a regular classroom to truly meet the needs of my child.|
|06||If my child were to spend a lot of time in a regular classroom, he/she would end up not getting the extra help he/she needs.|
|07||If my child were to spend much of his/her day in a regular classroom, he/she would end up becoming friends with nondisabled students in that room.|
|08||The quality of a regular education student's education is enriched when a student with severe disabilities participates in his/her class.|
|09||If my child were to spend much of the day in a regular classroom, he/she would end up not getting all the necessary special services that would be provided in a special education classroom.|
|10||A regular education classroom provides more meaningful opportunities for my child to learn than does a special education classroom.|
|11||The more time my child spends in a regular classroom, the more likely it is that he/she will be treated kindly by the nondisabled students in that room.|
The 11 items of the PATI scale (Table 1) each listed one probable effect of placing a child with significant cognitive disabilities in a regular classroom. Parents were asked to respond on a 6-point Likert rating scale that ranges from 'strongly agree', scored as 1, to 'strongly disagree', scored as 6. Reverse scoring was used in the present study to ensure that the lowest score invariably reflects the most positive perception.
In comparison with many previous surveys of attitudes to inclusion, PATI was tested in the USA with a relatively large sample of 460 parents (Palmer et al., 1998a). Findings confirmed the effect of differential influences on parental perceptions. Clear influences included the child's individual characteristics and perceived needs, the parent's history and socialisation, and the child's placement history. When behaviour problems were paired with more severe intellectual impairment, parents favoured self-contained classrooms and expressed more negative perceptions about inclusive practices. These parents viewed their child as 'fragile' or in need of special handling or intense services. Thus, PATI was shown to be sensitive to school, family, and child issues.
In summary, the PATI i s a short survey that draws together familiar themes in the inclusion literature of values and beliefs. Moreover, the survey's administration in the USA to a comparatively large sample, relative to other studies of attitudes to inclusion, enhanced the reliability of reported mean scores, deviation scores, and factor structure. The reliability provided by this data set was the main reason for selecting the PATI scale to consider how data on attitudes to inclusion might be used to improve inclusive practice. In short, the USA sample provided a standard against which the current sample could be measured.
If it could be confirmed that the USA statistical values reported for PATI items were replicated in a local setting in Queensland, Australia, then the major purpose was to explore how data from that survey might be used to construct a tool of some benefit to teacher-parent communication about the child's inclusion. The question was how to bring a more productive focus to parent-teacher discussions about inclusion concerns by constructing a profile of parental inclusion beliefs.
An important characteristic of the 10 children selected for the Queensland study is that, like those in the Californian study, they were all eligible for entry to specialised educational settings. They differed from children in the US study in being diagnosed as mildly rather than severely intellectually impaired and in having ASD. These children attended regular Education Queensland primary schools within the capital city. They were withdrawn from the regular classroom for special educational support throughout the school week. Data were not collected, however, on patterns of withdrawal, because the format of the field trial did not provide opportunities to obtain such data.
An important characteristic of this sample of parents is that inclusion was neither promoted nor discouraged in the support group. This neutral approach was nurtured by Advisory Visiting Teachers (AVTs), employees of Education Queensland, the government provider of regular and special education in the state. Thus, these AVTs did not set out to direct the views of individual parents about inclusion either one way or the other.
First, the variability of individual responses to particular items in the Queensland study was compared with that of the American sample. Second, the variability of group responses to particular items was compared with that of the American sample. Third, a simple but novel test administration key was used as a tool to identify subsets of items that differentiated between subgroups of parents in terms of their adherence to or departure from consensus.
A starting point for the three analyses was to convert the raw scores of this sample into z-scores. Subtracting the mean score from a raw score and then dividing the deviation score by the standard deviation provides a z-score. By definition, raw scores that equate to the mean score have a z-score value of zero; scores that are one standard deviation from the mean score have a z-score value of one (plus or minus). The value of this system is that it becomes easy to identify extreme scores. By convention and by assuming a relatively normal distribution, z-scores with a value of 1.96 (plus or minus) or greater have no more than 5% probability of being part of the comparison distribution.
Hence, conversion of raw or mean scores for the Queensland sample into the z-score format allows ready identification of departures from the norms reported for the American sample (see Table 2). An important aspect of this comparison is that the mean z-score for each item is assumed to be 0.00 in the Palmer et al. (1998a) study. For purposes of comparison, this study uses the obtained z-score to calculate the mean score per item for each of the 11 items (see Table 3).
Table 2 presents the z-scores for individual Queensland parental responses to each item within the context of the American study. In this first analysis, the scores of the 10 parents on the 11 items were transformed into z-scores relative to the mean and standard deviation scores reported for each respective item in the American sample. The resulting z-score represents the distance of each score from the mean for that item in the American study, with an American mean score of 0.00 and an American standard deviation score of 1.00. Inspection of Table 2 shows that most of these z-scores as well within 1.96 of American mean scores.
Table 3 presents the Queensland sample's mean scores for the 11 items, computed from the 10 z-scores per item listed in Table 2. The usefulness of these scores is that they provide an additional indication of the high degree of similarity between Australian and American ratings. For instance, the mean score of 0.65 for Item 1 in the Queensland sample was not significantly different from the mean z-score of 0.00 for this item in the Palmer et al. (1998a) study. In Table 4, the curious inversion of values allocated to response categories by Palmer et al. means that the relative 'positivity' in Australian sample scores actually reflects an increased negativity. That is, with the exception of item 8 (indicating agreement with the benefits of regular classroom attendance), this group tended to be more cautious about the benefits of inclusion than were the American sample.
Table 4 provides a simple but novel scoring key to recode parental responses. This test administration key highlights the pattern of scores in a way that avoids the need for teachers using PATI prior to or at the start of meetings with parents to calculate the z-scores considered in Table 2. This key highlights the items where the American sample were least likely to strongly agree or least likely to strongly disagree (after reversing item scores so that strong agreement reflects the most positive perception). To facilitate judgements, one response category has been excluded from the range of normal scores for each of the items. Response categories were excluded on the basis of having the lowest likelihood of being selected by the American participants, with the probability of selection approximating a z-score of ±1.96, calculated on the basis of American norms, a z-score equivalent to that obtained by chance in about five out of 100 testing occasions. This simplified key provides a tool that allows the test administrator, typically a teacher, to discriminate between (a) normal responses that fall within the range of shared perceptions and (b) other responses that fall outside the normal range of shared parental judgements about inclusion.
Parental responses to PATI could be identified as 'normal' in relation to the American sample by the use of z-score transformation procedures for (see Results). Teachers working with parents of children with disabilities such as II or ASD can facilitate planning sessions that accommodate the common core of parental beliefs about inclusion in terms of parental responses to the 11 items (and 3 factors). For instance, a teacher who meets with parent B, none of whose responses differ significantly from that of the USA sample (see Table 2), could use B's responses to these items (and factors) to initiate discussions of B's typical attitudes to inclusion (e.g., reassurance about normal kind of concerns, awareness of attitudes and beliefs that are widely shared and characteristic). In short, one might expect the tool profile obtained from PATI responses to set part of the agenda for teacher interviews with the parents of children with ASD or II about instructional progress specifically and student progress generally.
Parental responses that fall outside of the 'normal' response range could also be addressed. Some parents of children with either II or ASD will give responses that fall outside the normal range as defined in terms of z-scores based on the Palmer et al. study. Such atypical responses from a parent of a child with special needs might serve, therefore, as a likely indicator of, for example, home-school friction, and confusion about inclusive practices used in the school. In the present Queensland sample, certain parents gave extreme responses (z = ±1.96) to certain items (see Table 2). For instance, parents E and F gave strongly positive responses to Items 6 (equivalent to strong disagreement in a reverse-scored item) and 10 (equivalent to strong agreement), both related to the factor Quality of educational services. One might engage these parents in terms of these extreme responses to these items and to this factor. For example, are these parents being unrealistic and over-optimistic to expect that their child would get extra help or would learn better in a regular classroom? Alternatively, are they expressing a firm commitment to their child's inclusion? What are these parents' expectations about, for example, the teacher's instructional strategies and access to resources? Such extreme 'outlier' responses by parents warrant dialogue about the prospects for this specific student with either II or ASD in an inclusive setting. For instance, the post-survey meeting with parent C might explore the strong disagreement with Items 8 and 11, both related to the factor Mutual benefits of inclusion.
In that respect, this study has demonstrated that it is possible to make the kind of improvements recommended by Salend (1999), if teachers use Likert-response data from questionnaires in conjunction with follow-up interviews. When there is norming data available for a survey such as PATI, then the z-score transformation procedure adopted in the administration key indicates that the profile of consensual and discrepant attitudes might supply a useful focus for discussing, reviewing, and reflecting on inclusive practice. Moreover, this 'tooling up' of an attitudinal survey recognises that the range of views expressed by parents may have important implications that address these issues directly. Further investigation is needed to establish what actual value that parents and teachers obtain from the use of this administration key. The tool might be useful in initial interviews when a student enters a new classroom and might help teachers to engage parents in meaningful collaborations in ongoing practice implementation (Salend, 2002). Whether the PATI scale requires further adaptation for item clarity, unambiguity, and lay meaning (Salend, 2000) is yet another matter.
The pressure on individuals to work to a consensus may not be either a comfortable or an independent reflection of actual concerns (Salend, 1999). When a teacher conducts a follow-up interview of an individual parent participating in an educational survey, the use of a z-scored key is also a prompt to the teacher to be cautious in judgement. Two major issues become evident with this step from teacher awareness of the general (group-based) attitudes of parents of an included child into teacher discussion with an individual parent (one parent within that group). One is the moral relativism entailed in making these judgements about what parents as a group might agree on in the way of inclusion. The other is the relevance of context to group consensus on inclusion. The former requires recognition that the agreement of parents (consensus) does not necessarily equate with good judgment. The latter requires recognition that the consensus of parents may differ from that of other stakeholders, such as teachers. Our willingness to grapple with these issues is essential if the current research climate of attitude measurement is successfully to be translated into good practice.
It is of particular interest that the deliberate shift in context encouraged by the use of impersonal and plural contexts (children, students, classrooms) can produce instances of teachers and parents disagreeing about individual items (Stanley, Grimbeek, Bryer, & Beamish, 2003). Stanley et al. reported the outcome of administering a modified version of PATI, the Teacher Attitudes to Inclusion (TATI) scale, to 17 teacher delegates at a special education state conference in Queensland, Australia, in which the language of PATI was altered so that 'child' became 'children.' The responses of these teachers as a group did not vary significantly from those of the 10 parents reported in the present study except for two items. This shift in context was important in relation to the two items affirming the likelihood of regular students benefiting or becoming friends with students with disabilities. Teachers were significantly more likely to disagree with this position than parents in their responses to these two items (4 and 8).
As stated by Stanley et al. (2003), one way of interpreting such parent-teacher differences was that parents and teachers disagree on these issues. Another was that the change in context afforded by the shift from personal to impersonal contexts produced the measurable differences. If the latter, then one might conclude that regular students are more likely to benefit from the company of 'my child' than from the generic company of 'students with disabilities.' Likewise, regular students are more likely to become friends with 'my child' than with the more impersonal 'children with disabilities.' It follows that such shifts in context might well have influenced the responses of parents as readily as teachers.
An associated difficulty with the Stanley et al. study was that the coincidence of shifts in the choice of stakeholders coupled with shifts in the use of context in that study, on one level, constituted an obstacle to arriving at a straightforward interpretation of the outcome. Certainly, it would be have been better to ask either teachers or parents to answer both versions of these items as a way to partition the effect of specific context from effect of belonging to specific classes of stakeholders.
At another level, despite these shortcomings in the research design, such outcomes highlight the potential importance of context in teasing out stakeholder-related incongruities in attitudes to inclusive education. Certainly, the deliberate use of contextual variations could benefit Salend's (1999) agenda of developing tools that help teachers and parents to reflect on existing and desirable practice and to share perspectives with each other. It seems very likely that the perceptions of stakeholders might be explored effectively if the contexts for eliciting these reflections were varied deliberately . Future research could investigate further the influence of context on the attitudes toward inclusive education of larger samples of teachers and parents of students with ASD.
With respect to the PATI scale, it can be suggested that, there is a need for refined item analysis, reworking of reverse-scored items that interfere with ease of item interpretation and comparison of item data, and conversion of the rating scheme to the more generally preferred method of scoring strong disagreement as 1 and strong agreement as 6 (i.e., the higher score). The evident relevance of PATI norms to an Australian setting and to another group of children with disabilities (ASD) does not preclude a larger investigation to improve scale validity and to renorm and 'retool' the attitudinal field of items. In one sense, this study is a minor exercise based on an existing scale, in order to demonstrate a possible application from belief to practice. In another sense, the search for ways to use data from attitudinal surveys to improve inclusive practice raises some larger conceptual and methodological issues about measurement, meaning, and interpretation of stakeholder views.
Cook, B. G., Tankersley, M., Cook, L., & Landrum, T. J. (2000). Teachers' attitudes toward their included students with disabilities. Exceptional Children, 67, 115-135.
Coots, J. J., Bishop, K. D., & Grenot-Scheyer, M. (1998). Supporting elementary age students with significant disabilities in general education classrooms: Personal perspectives on inclusion. Education and Training in Mental Retardation and Developmental Disabilities, 33, 317-330.
Dorries, B., & Haller, B. (2001). The news of inclusive education: A narrative analysis. Disability & Society, 16, 871-891.
Downing, J. E., Eichinger, J., & Williams, L. J. (1997). Inclusive education for students with severe disabilities: Comparative views of principals and educators at different levels of implementation. Remedial and Special Education, 18, 133-142.
Elkins, J., Kraayenourd, C. E., & Jobling, A. (2003). Parents' attitudes to inclusion of their children with special needs. Journal of Research in Special Educational Needs, 3, 122-129.
Fisher, D., Pumpian, I., & Sax, C. (1998). High school students' attitudes about and recommendations for their peers with significant disabilities. Journal of the Association for Persons with Severe Handicaps, 23, 272-282.
Fuchs, D., & Fuchs, L. (1994). Inclusive schools movement and the radicalization of special education reform. Exceptional Children, 60, 294-309.
Palmer, D. S., Borthwick-Duffy, S. A., & Widaman, K. (1998). Parent perceptions of inclusive practices for their children with significant cognitive disabilities. Exceptional Children, 64, 271-282.
Palmer, D. S., Borthwick-Duffy, S. A., Widaman, K., & Best, S. J. (1998). Influences on parent perceptions of inclusive practices for their children with mental retardation. American Journal on Mental Retardation, 103, 272-287.
Palmer, D. S., Fuller, K., Arora, T., & Nelson, M. (2001). Taking sides: Parents' views on inclusion for their children with severe disabilities. Exceptional Children, 67, 467-484.
Pereira Dos Santos, M. (2001). Inclusion and/or integration: The debate is still in Brazil . Disability & Society, 16, 893-897.
Ryndak, D. L., Downing, J. E., Morrison, A. P., & Williams, L. J. (1996). Parents' perceptions of educational settings and services for children with moderate or severe disabilities. Remedial and Special Education, 17, 106-118.
Salend, S. J. (1999). So what's with our inclusion program? Evaluating educators' experiences and perceptions. Teaching Exceptional Children, 32(2), 46-54.
Salend, S. J. (2000). Strategies and resources to evaluate the impact of inclusion programs on students. Intervention in School and Clinic, 35(5), 264-270.
Salend, S. J. (2002). What do families have to say about inclusion. Teaching Exceptional Children, 35(1), 62-66.
Stanley, A., Grimbeek, P., Bryer, F., & Beamish. W. (2003). Comparing parents' versus teachers' attitudes to inclusion: When PATI meets TATI. In B. Bartlett, F. Bryer, & D. Roebuck (Eds.), Re-imagining practice - Researching change (Vol. 3, pp. 62-69). Brisbane, Australia: School of Cognition, Language, and Special Education, Griffith University.
Wright, S. L., & Sigafoos, J. (1998a). Teachers and students without disabilities comment on the placement of students with special needs in regular classrooms at an Australian primary school. Australasian Journal of Special Education, 21, 67-80.
Wright, S. L., & Sigafoos, J. (1998b). Parent views on the placement of students with special needs in regular classrooms at an Australian primary school. Australasian Journal of Special Education, 22, 30-37.
|Authors: Dr Fiona Bryer|
Applied developmental psychologist, interested in research-to-practice in education (e.g., school-wide practice in inclusion and positive behavioural support, middle school, teachers' emotional competencies, teacher practices in relation to socioeomotional development) E-mail: F.Bryer@griffith.edu.au
Dr Peter Grimbeek
Griffith University methodologist, statistical advisor to Faculty of Education, lecturer in research method, psychological counsellor.
Special educator and researcher in best practice and program improvement in early childhood special education, educational services for students with intellectual impairment and autistic spectrum disorder, and inclusion practices and positive behavioural support.
Teacher, university tutor, and doctoral student in education researching teacher-parent communication issues supervised by Wendi Beamish and Fiona Bryer.
Please cite as: Bryer, F., Grimbeek, P., Beamish, W. and Stanley, A. (2004). How to use the Parental Attitudes to Inclusion scale as a teacher tool to improve parent-teacher communication. Issues In Educational Research, 14(2), 105-120. http://www.iier.org.au/iier14/bryer.html