Development and Validation of a Scale Measuring Parental Perceptions of the Special Education Process

The purpose of this study was to develop and assess the psychometric properties of a scale measuring the perceptions parents had of the special education process . Before being distributed to the study sample, the Parent's Perceptions of Special Education Scale (PPSES) was reviewed by parent leaders from the Rhode Island Parent Information Network and students enrolled in a school psychology graduate program to judge the scale items for clarity and relevance. Surveys containing a personal background sheet, the PPSES, the Family-Partnership Scale, and the Marlowe-Crowne Social Desirability Scale were distributed to parents of children in special education at five sites in Rhode Island; two sites were advocacy agencies and three were nontraditional or alternative school settings. The data underwent a principal components analysis which resulted in a 45-item scale with a four factor structure. The four factors were labeled "Interactions at Meetings," "Time Issues ," "Emotional Perspective ," and "Acceptance of Differences. " The first three subscales listed had satisfactory coefficient alpha reliability estimates of .98, .75, and.79 respectively , with the last subscale having an unacceptable estimate of .50. In order to assess convergent validity , the subscales of the PPSES were correlated with the Family-Partnership Scale, developed by Summers et al. (2005), which assessed the satisfaction levels parent s had with their interactions with an individual who provided special education services to their child. The subscale labeled "Interactions at Meetings" had the highest positive correlation with the Family-Partnership Scale (r = .851, df = 71, p = 0.01), while the last three subscales of the PPSES had positive , yet more modest correlations with the Family-Partnership Scale (r = .560, .470, and .453, df = 71, p = 0.01). Based on the correlations, it appears the Family-Partnership Scale and the PPSES are measuring similar constructs . Social desirability did not appear to influence the participants ' responses on the PPSES as measured by correlations between the four subscales of the PPS ES and the Marlow-Crowne Social Desirability Scale; correlations were found to be r = .300, .225, .141, and -.041, df = 65, p = 0.05. Research obstacles and limitations for the present study as well as future directions for research are discussed. Acknowledgements I would like to thank those individuals who helped make this research project possible and who supported me through the duration ofthis process. First, I would like to thank my committee members , Jerry Cohen and Diane Horm for their time and support throughout this process. I would like to give a special thank you to my major professor Margaret Rogers, who took the reigns of this project, for the many contributions she has made to it and for her guidance. I would also like to thank Danel Koonce for the time and support he gave and for being a great mentor for the past 3 and a half years. This project would not have been possible without the help from those individuals I met at each of the sites that participated in the study. I want to thank all of them for their contribution to the project. Next , I would like to thank my parents for their support: My dad for reminding me to keep my eye on the goal and to my ma for reminding me to take it one day at a time. To Stacey, Dawnie, Brian, and all of my friends who kept me laughing during the stressful times. I could not have completed this without all of them. And finally to Shannon, for the strength she gives me everyday to keep moving forward.


Introduction
The importance of parental involvement in the education of children with disabilities has been a topic of considerable concern and importance to policy makers and educators for the past three decades . Prior to the Education for All Children Handicapped Act of 1975, court cases and parent advocacy groups significantly influenced and shaped the way schools interact with families who were seeking a free and appropriate education for their children. One of the most well known cases, Brown v.
Board of Education (1954 ), brought to the forefront the reality that certain groups of students were being denied an equal opportunity to an education. It was at this time during the 1950s that many families of children with special needs began to see progress in the attainment of an equal right to a public education for their children. However an uphill battle still awaited them to achieve the goals they were seeking.
Between the late 1950s and the late 1960s, courts were continuing to rule in favor of the exclusion of students with special needs from schools (i.e., Department of Public Welfare v. Haas, 1958) when children with special needs were thought to not benefit from public education or when the children were judged to be a disruption to others (Yell , Rogers , & Rodgers , 1998). Although compulsory education laws were implemented by 1918, North Carolina passed a statute in 1969 that made it a crime for parents to persistently attempt to enroll their child with special needs into a school after the child had been previously excluded from a public school (Yell et al., 1998). During this era of civil rights, parents began forming national organization s whose mission was to advocate for the right to a public education for children with disabilities (Turnbull & Turnbull , 2001). As a consequence of the tireless effort of these parent-supported organizations , legislation for special education (i.e ., EACHA) and its subsequent amendments (i.e., P.L. 99-457 , P.L. 101-476) were passed to guarantee children with disabilities equal access to an education. A list of common terms used in the special education process (i.e., special education, Individualized Education Program (IEP) , etc.) and their definitions are provided in Appendix A.
A secondary benefit of the efforts of these advocacy groups was that according to these laws parents were considered equal partners with school personnel mandating their involvement in the design and evaluation of special education services (Smith , Polloway, Patton , & Dowdy , 1998;Spann , Kohler , & Soenksen , 2003). Moreover , the law recognized the critical role parents play in their child ' s development , as well as the necessity of having parents oversee the programs the school was providing for their children. The premise of P.L. 94-142 and its amendments was that parents would enter into a partnership with school personnel and develop educational program s appropriate for their children with special needs. However , there is anecdotal evidence that suggests that school and family collaborations are not always productive and conflict free (Lipsky & Gartner , 1997;Rao, 2000 ;Valle & Aponte , 2002). Some parents have reported that they are not in a partnership but are placed in a hostile position whene ver they disagree with school personnel involved in the special education process (Lipsky & Gartner, 1997). Other studie s have found that parents feel intimidated and are not provided with a sufficient or appropriate amount of information to ask questions thereby further impeding their participation (Dinnebeil & Rule , 1994).
Despite the need for and importance of parental involvement in the special education proce ss, the literatur e has shown that it is difficult to actualize the role parents should play as intended by P.L. 94-142 (Vaughn, Bos, Harrell, & Lasky, 2001) . As a result , the field has identified several ways that schools can increase parental involvement throughout the special education process by scheduling meetings with the consideration of parents ' availability, willingly considering new ideas and perspectives presented by parents , obtaining parents' input on decisions about their child's education, using language free of jargon , and encouraging parent s to share information by asking openended questions during meetings (Dunst , Trivette, & LaPointe, 1992;Kroeger , Leibold, & Ryan, 1999;Rock , 2000;Turnbull & Turnbull, 2001 ). Equally important is consideration for the growing diversity among students today. Among professionals , there is a call for a greater degree of understanding and sensitivity to the need s of culturally and linguistically different families who may or may not be active participants in the special education proces s (Harry, 2002).
In spite of these recommendations, parent participation in the special education process remains low (Turnbull & Turnbull, 2001). Researchers have solicited parents' perceptions and experiences of the special education process in an attempt to gather information about their participation (Lynch & Stein, 1982Spann, Kohler , & Soenksen, 2003;Yanok & Deroberti s, 1989). Most of the research exploring the perceptions of parents has used qualitative method s, such as interviews and observational methods , which yields one perspective of the process (Jones & Swain, 2001;Rao , 2000;Soodak & Irwin , 1995Valle, & Aponte, 2002). There has been a lack of systematic quantitative studies, especially those incorporating the use of surveys, to explore this area of research. Therefore, there is a critical need for a measurement tool that explores the relationship between schools and families in the special education process. The present research study is designed to address this need by reporting on the development of a measure that assesses parental perceptions of the special education process.

5
The persistent determination of parents of children with special needs has helped the field of special education evolve into what it is today. Prior to government legislation and regulations, parents and advocacy groups were instrumental in bringing about change through the court systems with cases involving children with special needs (Turnbull & Turnbull, 2001 ). Parents and advocacy groups argued for children with special needs and their inherent right to an education in the United States public school systems.
To fully understand the evolution of the field of special education, the literature surrounding legislative and litigious action must be examined and considered. In 1954 schools were required to racially desegregate in the Supreme Court decision in Brown v.
Board of Education , citing that separate but presumably equal schools did not offer African American students equal protection of the laws under the Fourteenth Amendment (Huefner , 2000). Moreover , the Court asserted in their decision that for any child to reasonably succeed in life they could not be denied an education (Osbourne & Russo , 2003). Although full compliance with this decision would take almost a decade , the decision was ultimately applied to students with disabilities by generalizing the premise of desegregation to this population.
Two pertinent cases that laid the groundwork to improve the education of students with disabilities were Pennsylvania Association for Retarded Children (PARC) v.
Commonwealth of Pennsylvania (1971) andMills v. Board of Education (1972). In PARC v. Commonwealth of Pennsylvania, the settlement approved by the court recognized the right for children with special needs to receive a free appropriate public education (F APE) and established the law of least restrictive environment (Osbourne & Russo, 2003). The PARC decision also established due process , thereby providing parents of children in special education with a formal process for conflict resolution. In the second landmark case, Mills v. Board of Education, a class action suit was filed on behalf of the students with disabilities who had been categorically expelled from the school because of their disabilities. The school district claimed that they could not provide services because they lacked sufficient resources to teach these students.
However, the Court ruled that the financial burden of education should not fall on students with disabilities and that the school district had to expend its funds in such a matter that all students receive an education appropriate for their needs and abilities (Osbourne & Russo , 2003). Because of these court cases, by the early 1970's most states had passed laws that required school personnel to provide an education for students with disabilities. Despite these mandates , states ' educational endeavors were not comparable in the implementation of programs and service s. Limited financial resources also hindered states ' efforts (Gearheart, Weishahn, & Gearheart, 1996;Yell, Roger s, & Rodgers , 1998).
Federal legislation changed the course of history for educating students with disabilities in 1975 when P.L. 94-142 , the Education for all Handicapped Children Act 1975 (EAHCA) was passed by Congress. EAHCA merged different components of state and federal law regarding the education of students with disabilities into a single comprehensive law (Gearheart et al., 1996;Yell et al., 1998). EAHCA also mandated that students with disabilities had certain rights including nondiscriminatory testing , specified evaluation and placement procedures , and education in the least restrictive environment (LRE) .

7
Throughout the years , amendments were added to the EAHCA to further protect the rights of students with disabilities and their families. In 1990, the law was revised and renamed the Individuals with Disabilities Education Act (IDEA). The law stated that parents were to be considered full partners in the educational decisions for their children (Lynch & Stein, 1987). When IDEA was reauthorized in 1997, this practice was strengthened. As a consequence of this legislation, the school's role in educating students with special needs changed, as well as the parent's role in advocating on behalf of their children regarding educational services.
In order to comply with the reauthorization of IDEA in 1997, states had to follow six principles of educating students with special needs. Two of these principles were due process and parent participation. Due process, as discussed earlier, gives professionals and parents the opportunity to formally challenge one another's decisions regarding any part of the special education process (Turnbull & Turnbull, 2001). The parent participation aspect of the reauthorization gave parents the right to access school records concerning their child , to control access to those records by others , to be eligible to participate on state or local special education advisory committees , and to be involved in every aspect (i.e., referral through placement) of the special education process. The reauthorization legalized and legitimized the role of parents as decision makers of their child ' s educational program (Turnbull & Turnbull , 2001). This attention to parental participation stressed the importance of having parents contribute to the meetings held throughout the special education process because their experienc e provides a wealth of information and knowledge about the child that school personnel would not have access to otherwise (Engel , 1993).

8
Historically, the literature suggests that concerns have been raised regarding the quality of parental involvement in their children ' s special education process (Brantlinger , 2001;Lynch & Stein , 1982Rodger, 1995). Although the legislation related to special education mandates participation, parents have generally reported a passive rather than active role in their child(ren)'s special education process (Vaughn , Bos, Harrell , & Lasky, 1988). This lack of collaboration between parents and school personnel has been viewed as contrary to what the federal legislation intended as a joint decision making process (Turnbull & Turnbull , 2001). Despite research studies that report parental satisfaction with their children ' s special education program (Lynch & Stein, 1987), many parents continue to express concerns about the non-productivity of school meetings (Jones & Swain, 2001 ). Specifically, parents of children with special needs find meetings with school personnel lack communication and meaningful dialogue , as well as the exchange of meaningful information (Leyser , 1988) . The combination of these concerns leads to inhibiting parents from actively participating and being dissatisfied with the role that they play in the planning for their child ' s educational need s (Abramson , Wilson , Yoshida, & Hagerty , 1983).
The intent of IDEA was that parents would enter into a partnership with school personnel and develop educational programs that would be appropriate for their children with special needs . However , the atmosphere of collaboration does not consistently emerge when the two sides come together. Many parents report that they are not in a partnership, but rather are placed in a hostile position whenever they disagree with the professionals involved in the special education process (Lipsky & Gartner, 1997). Lipsky and Gartner (1997) found that many parents had to fight for placements for their children and complained that they were left out of the assessment process , suggesting that continued parental advocacy was needed . The findings also indicated that parents felt intimidated and were not provided with the appropriate information required to ask questions, thereby further impeding their participation.

Parental Involvement and the Special Education System
All parents should have the opportunity to participate in their child's education.
Parental involvement in special education differs from parental involvement in general education because of the legal mandates found in IDEA (1990;reauthorization in 1997).
The procedures outlined in IDEA affect both the processes parents experience and the type of knowledge needed to effectively participate in the special education process.
Many studies addressing parents' role in the special education process focus on the legal requirements that safeguard parental involvement requirements in the IEP process (Harry , 1992;Harry, Allen , & McLaughlin, 1995;Katisyannis & Ward , 1992).
Yanok and Derobertis (1989) conducted a telephone survey of 1,702 randomly selected parents to determine if significantly different opinions existed between parents ofregular education students and parents of special education students concerning: (a) schoo l involvement; (b) quality of instruction ; and ( c) equality of educational opportunity. The results indicated that there were no differences between African American parents of children and special education in their perception of opportunities for involvement. Both groups were satisfied with the quality of education and the educational opportunities available to their children. The only difference that emerged was that a greater percentage of parents of special education children reported being contacted more frequently by their child ' s teacher during the past year. Lynch and Stein (1982) conducted a study comparing parents' roles in the IEP process to other studies conducted after the initiation of P.L. 94-142. They surveyed 400 parents measuring several variables including: (1) verbal interactions; (2) parents' perception and knowledge of placement/ IBP meeting; and (3) contacts with the school.
Although 71 % of the participants reported active involvement in the meeting only 14% provided specific recommendations or opinions. Furthermore, the results indicated four major themes of involvement. First , parents asked significantly less questions when compared to the other types of verbalizations they made during the IEP meetings ( e.g. , opinions, suggestions). Later research also found this to occur during meetings in which significant changes in the child's educational plan and the identification of a label or diagnosis was made (Vaughn, Bos, Harrell, & Lasky, 2001). Second, parents reported consistently high levels of satisfaction with the conference. Third, parents indicated a limited understanding of the child's handicapping condition. Fourth, parents reported passive involvement in the IEP process. In comparison to previous studies (Gilliam & Coleman, 1981;Goldstein, Strickland, Turnbull, & Curry, 1980), the findings suggest that despite major changes in legislation about parental participation the perceived roles of parents had not changed.
In a later study, Lynch and Stein (1987) investigated differences in the perceptions of parents' involvement in their child's special education program. They conducted interviews with 63 Hispanic parents and compared the results with data from two earlier studies that included African American and Caucasian parents. Results of the initial investigation revealed that only 45% of Hispanic parents reported they were part of the assessment process. Approximately half of the sample felt they were not active participants during the development of the IEP. Moreover , only a third of the parents actually offered suggestions during the meeting with less than half feeling that they could collaborate with the teacher to work on goals and objectives for their child.
In the same study, Lynch and Stein (1987) found that the difference between Hispanic and African American parents on the concerns above were negligible.
However , the Hispanic parents ' responses were significantly different than the Caucasian parents on the topics of involvement and participation at the school meeting , with the Caucasian parents indicating higher levels of involvement and participation (Lynch & Stein, 1987). These findings indicate some variability among minority and Caucasian parents in their views towards involvement.
In a similar study, Spann, Kohler , and Soenksen (2003) conducted telephone surveys with 45 families of children with autism , examining the families' involvement in and perceptions of children's specia l education services. The majority of parents reported (78%) moderate levels of involvement during the development of the IEP, with only 11 % reporting low involvement. Parents with children in the two younger age groups (i.e., 4 to 5 years and 6 to 9 years-old) reported greater involvement in the IEP process compared to the parents whose children were in the two older age groups (i.e ., 10 to 14 and 15 to 18 years) who reported less input. In terms of the parents' satisfaction with the school's ability to meet the child 's most pressing needs, 44% of the families reported that schools were exerting little or no effort to address these needs. On the other hand, 56% of the parents perceived that schools were expending moderate to high levels of effort in order to meet their child ' s needs . One of the conclusions that can be drawn from this study is that the level of involvement, satisfaction with home-school communication, involvement in the IEP, and the school ' s ability to address a child's needs is influenced by the child's age. In this case , parents of older children reported lower levels of satisfaction across all of these areas .

Roles of Parents in the Special Education Process
Due to changing federal legislation (i.e. , 1997 Amendments to IDEA) parents are considered equal partners with school personnel entitling parents to participate in the design and evaluation of special education services. Turnbull and Turnbull (2001) have suggested that the opportunity for parents to use their expertise to influence their child's education will have a positive effect on the parent's level of involvement. In order to provide a collaborative atmosphere, it is important that all members of the process share their own expertise and resources to benefit the child. However , research has shown that this collaborative emphasis is not being adopted and that parents are not equal participants in the decision making process. Parents are often seen as not being able to contribute useful information regarding the assessment process , and therefore may be perceived as having less expertise and lower status (Lott & Rogers , 2005 ;Rodger , 1995;Valle, & Aponte , 2002 ;Ware, 1994).
According to Rodger (1995), there is a hierarchy of influence that manifests during IEP meetings . The hierarchy lends itself to the assumption that school personnel have roles of perceived expertise that match with the purpose of the IEP meeting. This suggests that the role a parent plays in the special education proces s is not clearly defined and/or not viewed as important by the other members on the multi-disciplinary team.
Equally confusing is the fact that roles of the multi-disciplinary team members in this process are not clearly defined during IEP meetings (Y sseldyke , Algozzine, & Mitchell, 1982).
Some studies that have evaluated the roles a parent plays in the special education process have suggested that these roles vary from having no involvement to having infrequent decisional control (Lusthaus , Lusthaus, & Gibbs, 1981;Scanlon, Arick , & Phelps, 1981;Ware , 1994). Goldstein , Strickland , Turnbull, and Curry (1980) conducted an observational analysis of 14 IEP conferences and found that parents were the recipients of information from the professionals involved in the meeting. In a similar analysis of IEP conferences, Gilliam and Coleman (1981) surveyed the participants (i.e., parents , teachers, social workers , administrators) perceived roles (i.e., influence and contribution) prior to and following the IEP meeting. The participants were asked to rank order the roles in terms of importance and status according to the contributions made regarding diagnosis, planning, placement , and implementation. Results indicated that the parent's role (along with others) was given higher status prior to the meeting but was not considered to be influential after the meeting . Knoff (1983) investigated the influence and status of multidisciplinary team members using four independent samples (20 school psychology graduate trainees , 20 special education trainees, 20 school psychology practitioners , and 20 special education practitioners) based on the intent reflected in P.L. 94-142. The participants rated 11 multi-disciplinary team professions on three separate Likert scales: one rating each profession ' s influence on special education placement decisions, one rating each profession's influence in the participant's actual experience, and one rating the desirability of each profession to chair the multi-disciplinary team meeting (Knoff, 1983).
The result indicated that the parent, medical personnel and classroom teacher were rated as less influential despite the intent of P.L. 94-142. Additionally , parents were rated as the least desirable chairperson of the meeting , with the school psychologist and administrator receiving the highest rating. However , in the participants actual practice the special educator teacher and school psychologist were rated as the most desirable chairperson of the meeting. The pattern of these results differed from the findings of previous research (i.e., Gilliam & Coleman , 1981), suggesting that each multidisciplinary team should be considered a unique entity and the interaction and influences of the team members will vary based on the philosophy and cohesiveness of the group.
Finder s and Lewis (1994) detail barriers to parental involvement from the parents' point of view . Diverse school experiences of the parent themselves influence how they subsequently perceive school relationships (Finders & Lewis, 1994 ). If a parent had a negative school experience, there is a good possibility of it tainting future school relationships; especially for those parents who were not considered a part of the majority culture (e.g., African American , low-income) (Lott, 2001;Lott & Rogers , 2005). Diverse economic and time constraints also affect levels of involvement. Finders and Lewis (1994) relate a parent's story of having three children enrolled in three different schools, catching different buses at different times, and not having private transportation. The illustration of the scheduling issues alone demonstrates the strains that a family can experience. Economic pressure can also affect the ability of a parent to attend school functions. Taking time off work is sometimes difficult and results in economic stress.
Some employers impose pay cuts for missed work, and many families are not able to tolerate lost wages.
Relationships between parents and professionals can greatly serve to enhance the education of children with and without disabilities. Fielder (2000) acknowledges several underlying dispositions before a positive relationship can occur, including 1) the professional must work with the child and with the family ; 2) the professional must value a collaborative relationship as a benefit to themselves professionally and to the child; 3) professionals must lessen family dependency on professional advice and services and seek to empower the family; and 4) professionals must learn not to be irritated or annoyed by parents who actively pursue educational rights and services. Lack of parental involvement cannot always be attributed to the parent; sometimes professionals do not promote an environment in which parents feel comfortable participating (Fielder , 2000 ;Harry, Allen, & McLaughlin , 1995;Swick , 1988) . Fielder (2000) and Harry et al. (1995) acknowledge that IEP meetings often focus on legitimizing the identification of the disability as opposed to designing individualized instruction. These meetings often base programming decisions on "administrative convenience --wheth er there is an existing program in which to readily place a given child" (Fielder, 2000 , p. 8). Parents new to the system without knowledge about the system can be manipulated to seeing these types of meetings as appropriate and beneficial rather than a procedural formality. This can be a major barrier to effective parent collaboration. This type of IEP meeting consists of influencing someone to agree with an idea instead of all parties working with one another in a collaborative effort .
Language and cultural barriers are becoming more frequent as our nation becomes more diverse (Harry , 1992;Kalyanpur & Harry, 1999;Lynch & Stein, 1987;Rueda & Martine z, 1992). Kalyanpur and Harry (1999) write of the "challenge that special education professionals face in collaborating with families and individuals whose implicit and explicit value base may be radicall y different from their own" (p. 8). Rueda and Martinez (1992) relate a story about a Hispanic mother dealing with the school's emphasi s on transitioning her young adult son with mental retardat ion out of the home and into the community . Her response to this proposed action was horror ; in her culture it would be a great failure on her part if her son were to move out of the home . In the Hispanic culture , as she explained , living alone is not important ; however , attending to your children 's needs at home is important. It is presumpti ve for educational professionals to impose their value system on a family . The great range of cultural diversity in the schools today necessitates the need for more communication and collaboration between schools and families.
In the special education arena, parents often defer to educational professionals (Inger , 1992;Lian & Fontanez-Phelan , 2002 ;Lynch & Stein, 1987). This deference seems to be common in some minority cultures. Kalyanpur , Harry, and Skrtic (2000) pointed out that in certain cultures ( e.g. , Chinese and Puerto Rican) questioning professionals is disrespectable, making it difficult for these parents to challenge the recommendations of professionals within the context of a special education meeting. In these instances, parents will often accept the professionals' recommendations without question or suggestion, and unfortunately this deference is often interpreted as disinterest by professionals (Inger , 1992;Lian & Fontanez-Phelan, 2002). Another factor that affects participation is the "bureaucracy , rules , procedures and policies of the school environment " (Seefeldt, Denton, Galper, & Younoszai, 1999, p. 101). The educational setting can be perceived as cold and uninviting , which may negatively affect participation. Harry , Allen , and McLaughlin (1995) found that African-American parents of children with disabilities efforts at advocacy were "fraught with difficulties " (p. 3 70).
Specific deterrents noted by Harry et al. (1995) and Fiedler (2000) included late notices , inflexible scheduling, limited time for conferences, an emphasis on documents rather than participation , the use of jargon , and the structure of power. The untapped potential of parents as partners in decision-making can create an obstacle to involving families as supportive influences in their child's special education.
Gathering information on the perceptions parents hold regarding the special education process may begin to provide researchers and ultimately the schools with insi ght into why parental participation remains at low levels. It may also help to generate strategies to increase active parent participation throughout the special education proce ss.
Researchers past attempts to measure parental perceptions of the special education process have relied mainly on interviews and other qualitative methods (e.g. , Goldstein , Strickland , Turnbull, & Curry, 1980;Lynch & Stein, 1982). While these methods have yielded pertinent information regarding perceptions of parents , the information itself may not be generalizable to a broad population . A review of the literature indicates that there is not a comprehensive tool that assesses parents' perceptions special education meetings.
The aim of this study was to develop and test the psychometric properties of a scale to measure parents ' perceptions of their interactions with the school , their emotional reactions to the special education process , and their views of cultural issues as well as how they view their participation in the special education process .

Statement of the Problem
There is a lack of research incorporating a quantitative measure of parental perceptions of the special education process that focuses on interactions with school personnel , emotional tone of meetings, cultural sensitivity of school personnel, and overall participation. Several studies have used qualitative interviews or case studies to measure parents' perception of the special education process (Jones & Swain , 2001;Rao , 2000;Soodak & Erwin, 1995Valle , & Aponte , 2002); however, the results of these investigations have been antecdotal , subjective , and difficult to generalize to a broader population.

Purpose of the Study
The purpose of the study was to develop and validate a survey that assessed parent perceptions of the special education proc ess . Specifically , the focus of the survey assessed interactions with school personnel , emotional tone of meetin gs, cultural sensitivity of school personnel, and overall participation . Because knowledge of the special education process has been found to influence the level of participation by the parents of children with disabilities ( e.g., Lipsky & Gartner , 1997;Turnbull & Turnbull, 2001;Valle & Aponte, 2002) , a separate measure assessing parental knowledge and rights about the special education process was developed.

Research Questions
The following research questions were addressed to determine the psychometric properties and the factor structure of the Parent ' s Perceptions of Special Education Scale (PPSES).

Research Question I: Did the items on the PPSES detect varying levels of the constructs?
In order to determine if the items on the PPSES could identify varying levels of the constructs , an analysis of the item means , item variances , and item-scale correlations were completed. Desired means were those in the center range of possible scores , 3.5 for this study, and relatively high variances were desirable (DeVellis , 2003). An item-scale correlation of .30 or higher was needed for an item to be retained (Pallant , 2001 ).

Research Question 2: What is the underlying/actor structure of the PPSES ?
A principal components analysis (PCA) was conducted to determine the factor structure of the scale and reduce the large item pool into several factors in the hopes of describing the relationships among the items . Four distinct factors are expected to result from the PCA reflecting the following themes: emotional tone of the meetings , interactions with the school, cultural sensitivity of school personnel, and overall participation in the special education process.

20
Reliability estimates for the overall scale and subscales were calculated using coefficient alpha . The descriptions given by DeVellis (2003) for acceptable alpha levels were used with the minimal acceptable level being between .65 and .70, and it was hoped that the overall and subscale alpha levels had an acceptable value.

Research Question 4: Does the P PSES correlate with variables that it should correlate with demonstrating construct validity ?
Construct validity was assessed through convergent validity which examined the correlation between the PPSES and the Family-Partnership Scale designed by Summers et al. (2005). This analysis was conducted to determine if the PPSES was measuring constructs similar to those on the Family-Partnership Scale. It is expected that the PPSES and the Family-Partnership Scale will have a strong, positive correlation.

Research Question 5: Are responses on the PPSES influenced by social desirability?
Because the PPSES uses a self-report method to gather information on the participant 's perceptions , the Marlowe-Crowne Social Desirability Scale (MCSD) (Crowne & Marlowe, 1960) was used to investigate the participant 's need for approval in responding to the instrument. A correlation between the PPSES and the MCSD is hoped to be weak and non-significant , thereby indicating the participant's responses on the PPSES were not influenced by social desirability .

21
The participants in this study included 74 parents in Rhode Island including 68 mothers , 4 fathers, one foster parent and one stepmother. A majority of the sample was White , non-Hispanic (n = 69) , with one participant self-identifying as White, Hispanic ; Puerto Rican; African American, non-Hispanic; Native American ; and Other , Ecuadorian. Overall the sample of parents for this study was White , non-Hispanic mothers. Seventy percent of the sample was married , with the income and education levels varying within the sample. Table 1 provides the detailed demographic information for those parent s who completed the survey .  Arnold , Michael, Hosley, and Miller (1994). Items from the PAK.AS used in the PPSES reflected issues of cultural sensitivity of school personnel.
Lian and Fontanez-Phelan (2001)  The survey items used from Arnold , Michael, Hosley , and Miller's (1994) study were related to a parent's satisfaction with home -school communication. Survey items generated by Arnold et al. (1994) were reviewed by parents and profe ssionals in special education to establish content validity and the survey was found to have a good reliability estimate (a = .85). The items from these two scales contributed 11 items to the PPSES.
Once the preliminary items were created and pooled together with the items from Lian and Fontanez -Phelan (2001) and Arnold et al. (1994) , leaders from the Rhode Island Parent Information Network (RIPIN) analyzed the 58 items for clarity , readability , and conceptual desirability. Additionally, eight graduate students enrolled in a school psychology program reviewed the items for clarity in wording and phrasing. Based on the feedback received , items were reworded when judged to be confusing and 12 items were removed because they were judged to be redundant or irrelevant, resulting in a 46item scale . The PPS ES was developed to address parental involvement during IEP meetings with regards to emotional tone of meetings , interactions with the school, cultural sensit ivity of school personnel , and overall participation. The items were designed to be rated on a 6-point Likert type scale from strongly disagree (I) to strongly agree (6). A 6-point Likert format designed without the inclusion of a neutral midpoint was chosen by the researcher in order to maximize the information gathered on the perceptions parents had regarding the special education process. The higher the total score on the PPSES the more positive the parental perceptions are with regards to the special education process.

26
A separate scale assessing a parent's knowledge of various aspects of the special education process is also a part of the PPSES and is presented after the 46-item Likert scale. This scale consists of nine items and participants respond by answering yes (Y) or no (N). Examples of these items are "I have to give consent before the school can evaluate my child," and "I am allowed to bring an advocate with me to meetings." This scale is intended to provide information on an individual item level in order to gather information regarding how many parents are knowledgeable about certain aspects of the special education process. No total scores are calculated. For a full list of the items on the PPSES, see Appendix C.
Percentages were generated for the 9-item knowledge scale and it was found that there were three areas parents were less knowledgeable in than the others (for a full list of the results , see Appendix D). Eighty-six percent indicated they were aware that they are allowed to bring an advocate to the meeting with them, 82% were informed on how to work with the school to help their child, and lastly 78% of parents were aware of the actions in place if they do not agree with the school's decision , also known as due proces s. It was also noted that only half (51.4%) of the parents surveyed were given materials about their child's progress prior to the meeting. Receiving materials prior to an IEP meeting may help parents prepare for the discus sion or decisions that need to be made to their child 's IEP.

Family-Professional Partnership Scale. The Family-Professional Partnership
Scale, an 18-item scale developed by Summers et al. (2005) (Summers et al., 2005) . The overall twofactor model had the fit statistics of: i (134) = 270, p<.001, CFI= .90, RMSEA = .08 (Summers et al., 2005). These statistics indicate that the two factor model had an acceptable to excellent fit and can be used in the interpretation of participant's scores on the scale. The Cronbach alpha for the overall 18-item Partnership scale was .96. The Child-and Family-Focused Relationships subscales had alphas of .94 and .92 respectively. The range of scores possible for the Family-Partnership Scale is 18 to 90, with a high score indicating the parent is satisfied with the relationship they have with their child's service provider. For a copy of the items on the Family -Partnership Scale please refer to Summers et al. (2005) .

Marlowe-Crowne Social Desirability Scale . The Marlowe-Crowne Social
Desirabilty Scale (MCSD) (Crowne & Marlowe , 1960) was also used to assess the participant's social desirability , or need for approval. This 33-item scale asks participants to respond "True" or "F alse" to items noting desirable but uncommon behaviors or undesirable but common behaviors. The range of possible scores on the MCSD is between 0 and 33 with high scores representing a higher need for approval. Research on the 33-item version of the MCSD scale has reported means of 13.3 (SD= 4.3) to 16.4 (SD = 6.5) with various populations and environments (Robinson, Shaver , & Wrightsman , 1991). With regards to reliability , the overall alpha level has been reported as ranging from .73 to .88 indicating adequate reliability (Robinson et al., 1991). With regards to convergent validity, high scorers on the MCSD were found to respond more to social reinforcement , inhibit aggression, and were more susceptible to social influence when compared to low scorers (Robinson et al., 1991 ). For a copy of the items on the MCSD , please refer to Crowne & Marlowe (1960) or Robinson et al. (1991 ).

29
Potential sites for subject recruitment were contacted over the Summer and Fall of 2005. Local school systems and agencies working with parents of children with special needs were contacted via phone to inquire about their willingness to participate in the research study. After the initial telephone contact , a meeting was set up to discuss the project in more depth with the special education director or director of the agency.
Overall, 12 sites including schools and agencies were approached to participate in the study and 5 elected to participate . The participating sites included two agencies and three alternative or non-traditional school settings from Rhode Island. One agency was a nonprofit agency that worked statewide providing information and support for parents seeking help for their children, while the other agency worked with individuals and families with low-incomes. One of the schools that participated in the study was a public charter school with a student population of 61 in grades K-2. A public high school located in an urban area of the state also participated and it had a student body of 438 students. This high school is considered a non-traditio nal school setting because it differs in philosophy , structure, and curriculum from most traditional high schools. The third school to participate was located in an urban area and offered kindergarten through 12 th grade. This school ' s student body was composed of children with autism and other emotional and behavioral difficultie s. The number of parents with children in the special education system varied from site to site, ranging from 25 to 130.  Collazo (2005) . However, the Collazo (2005) study was published after the surveys were sent to be translated and it was unknown to the researcher that a translated version was available. Therefore , the translated version of the MCSD was not used in this study.
There is no known Spanish translation of the Family-Partnership Scale at this time.
The completed surveys were returned to the researcher via mail in the envelope provided by the researcher. Only one set of the questionnaires was sent home to the parents, with no follow-up postcard to serve as a reminder to return the survey they had received . Deadlines for data collection were extended twice to accommodate the various sites' time constraints and schedules, with a time frame of initial contact with site to distribution of surveys being between 4 and 6 weeks. In order to help with the extra demands the study may have placed on site staff, the researcher was available to come in and complete the work needed to ensure a successful distribution of materials. Follow-up calls were also made to each site after distribution of the research packets in order to address any questions or concerns that had arisen. Handwritten thank you notes were mailed to each site within two weeks from the date the research packets were distributed to each site. Over the course of the study , a total of 4 to 7 contacts were made with each participating site to discuss the study, plan, and organize the distribution of surveys to the parents within their system.

Personal Background Data
Chapter IV

32
Overall, 74 of the 400 distributed surveys were returned yielding a return rate of 18%. The demographic characteristics for the sample are listed in Table 1. A majority of the parents were married (n = 52), had an education level of some college or higher (n = 61), and had an income of $40,000 or higher (n = 53). The major form of transportation for 97.3% of the parents was automobi les, with 80% of the participants having 2 or more automobiles to use. The parents' children in special education comprised of 58 males and 29 females, who were in grades ranging from pre-school to 12 th grade. A majority of the parents were satisfied with their experiences with special education meetings (63.5%), with 18.9% of the parents indicating they were neither satisfied or dissatisfied with their experiences. A majority of parents (75.7%) attended a special education meeting within the past 3 months of receiving the survey, with 79.7% of parents reporting having attended between 1-3 special education meetings last year. Forty-three percent of the parents (n = 32) reported that they volunteer at their child' s school.

Initial Screening of Data
Before any of the statistical analyses took place , the data was screened for any missing data or potential outliers. Upon screening the data for missing values, it was noted that one parent completed the persona l background information , the PPSES and the MCSD , leaving the Family-Partnership Scale blank. Therefore , the analysis looking at convergent validity , which consisted of finding the correlation between the Family-Partnership Scale and the PPSES , only utilized 73 cases. The principal components analysis performed on the PPSES utilized all 74 cases in order to maximize sample size.
Missing values on the PPSES were replaced with the mean for that individual item. Mean substitution provides a conservative procedure for handling missing data, however the variance of a variable is reduced as well as the correlation the variable has with the other variables (Tabachnick & Fidell , 2001). Overall , 36% of the participants had one or more missing values on the PPSES . In examining the missing values on the PPSES , it appeared that the missing data was concentrated on items 21, 28, 36, and 40 with missing value totals of 5, 7, 13, and 5 respectively. These items included two questions dealing with culture, race, and ethnicity, as well as an item about interacting with the school principal and learning of the news that a child had special needs . These items were left in the analysis and underwent mean substitution. However, because these missing values are concentrated on certain items and not random, the generalizability of the results was affected (Tabachnick & Fidell , 2001). Missing values on the Family-Partnership Scale were also replaced with the mean values Summers et al. (2005) published in their study validating the scale . Overall, 5% of the participants had one missing value on the Family-Partnership Scale.
Next , outliers were explored to ensure no extreme values were distorting the data.
This was done by looking at the data obtained from the frequency analysis . Minimum and maximum values, along with means , and standard deviations were inspected to make certain the values were plausible . Also, univariate outliers were assessed among the continuous variables for the PPSES , Family-Partnership Scale, and MCSD. This was done by identifying standardized scores greater than 3.29 which would indicate potential outliers (Tabachnick & Fidell, 2001). Some cases had scores greater than 3.29, and as a result , the computerized data was reviewed with the hard copies of the data to verify all information had been entered correctly. In each instance, the data had been entered correctly and nothing appeared to be abnormal with the data set.
The multivariate assumption of normality was assessed by examining the distribution of the items. Normality was assessed through skewness and kurtosis (Tabachnick & Fidell, 2001). When the items were tested , it was found that most of the items were negatively skewed, with 6 items showing non-significance when compared to the expected value of zero. When looking into the kurtosis and comparing the values to the expected value of zero, 17 items showed non-significance , while the remaining 29 indicated a positive kurtosis. Although some items approached values that indicated a normal distribution, overall the assumption of normality was not met. Tabachnick and Fidell (2001) recommend the possibility of performing a transformation of the data in this instance. However , no transformations were completed because the sample was not adequate in size and make up. Therefore , due to the composition of the data, it would not have been assumed to be normally distributed.
The findings of the study are organized as follows to address each of the five research questions stated earlier.  Tabachnick and Fidell (2001) for this purpose and were used as methods of extraction in this study. The Kaiser rule retains factors with an eigenvalue greater than 1.0, and when applied to this data set retained 10 factors. For the scree test, the relative values for the eigenvalues are plotted, and those eigenvalues above the elbow are the factors that should be retained. After inspection of the scree plot , it was found that possibly only one factor should be retained. Because the original hypothesis was exploring a four factor solution model, continuing analysis utilized a four factor solution.
In order to help with interpretation of the factors, both varimax and oblique rotations were conducted on the data. After reviewing the results of both rotations, similar findings were found regarding how the variables loaded onto the different factors.
Due to the similarities in factor structure between the two types of rotations , varimax rotation was chosen and used due to its ease of interpretation. The PCA with varimax rotation was conducted using all 46 items of the PPSES. For a variable to be included on a factor, a cut score of .40, as determined by the researcher , was applied. When examined, item 5 on the PPSES failed to load on any of the four factors, and items 1 and 4 failed to load on factor one with loadings of .389 and .367 respectively. Also, item 3 failed to load on factor 2 with a loading of .305. The other 42 items had a factor loading of at least .40 on one or more of the four factors. The first factor contained 28 items, accounting for 34% of the total variance. The second factor had 5 items and accounted for 10% of the variance . Six items loaded on factor 3, accounting for 10% of the total variance , and finally , factor 4 had three items and accounted for 6% of the total variance.
Overall, the four factor structure model accounted for 60% of the variance.
It should be noted that two and three factor models were also investigated due to the instability of the factors due to a small sample size. It was found that the fust factor was strong throughout all structure models, with the remaining factors remaining fairly consistent with regards to which items grouped together. The two and three factor structure models accounted for 52% and 56% of the variance respectively. For the remaining analyses, the four factor model was used . Table 2 include s the factor loadings for the items on the PPSES. Overall , it was found that 34 out of the 46 items loaded on one of the four factors.
Eight of the items had loadings of .40 or higher on two factors and were therefore considered complex variables, and should be considered for elimination from the finalized scale. Three additional items should be considered for elimination because they did not have factor loadings greater than .40 on any of the factors. The researcher is hesitant to eliminate these items in the current study due to the small sample size and instability of the results found using PCA. Item 5 failed to generate a loading on any factor and also had a poor item-scale correlation , therefore it was decided that this item would be dropped from further analy sis resulting in a 45-item PPSES.
After examining the items that loaded, the four factors were labeled: " Interaction s at Meetings ," "Time Issues ," "Emotional Perspective, " and "Acceptance of Differences." Factor One , "Interactions at Meetings ," included 28 items with a sample item being " I felt the other members of the team respected my opinion." Factor Two , "Time Issues," was comprised of 5 items. A sample item for this factor is "I was notified about the meeting in a timely manner in order to attend. " Factor Three, "Emotional Perspective, " was defined by 6 items with a sample item stating "I felt intimidated at the meeting. " Finally , Factor Four , "Acceptance of Differences ," included three items with a sample item being "Conferences or trainings that would enhance the school staffs understanding of cultural diversity are needed." The assumption of normality was asses sed on the four subscales of the PP SES, and was tested by analyzing the skewness and kurtosis of the distribution of the total scores for each subscale (Tabachnick & Fidell, 2001). The analysis showed that the four subscale s were negatively skewed , with three of the subscales , "Interactions at Meetings ," "Time Issues ," and "Emotional Perspective, " having a positive kurtosis . The kurtosis value for the fourth subscale , "Acceptance of Differences ," when compared to the expected value of zero , showed non-significance indicating the distribution of scores for this subscale approach a normal distribution.
Inter-correlations were also calculated between the four subscales of the PPS ES.
The strength of the relationship between the four subscales ranged from strong to moderate. The fourth sub scale, "Acceptance of Differences, " had moderate correlations with the other three subscales on the PPSES , indicating this subscale was measuring a relatively different construct than the others with some overlap. Table 3 lists the correlations between the four subscales.  (Anastasi & Urbina , 1997). One way to explore construct validity is by way of convergent validity which is a validation analysis that needs to show that a scale or test correlates highly with other variables it should theoretically correlate with (Campbell, 1960). To assess the construct validity for the current scale, an analysis of convergent validity was conducted by computing the correlation between the scores for each of the four subscales of the PPS ES and the Family-Partnership Scale. The total score on the Family-Partnership scale was used for the correlation as opposed to using the scores for the two subscales. This was done because it was decided the total score would offer a better examination of the relationship to the subscales on the PPSES, whereas the constructs captured on the two subscales for the Family-Partnership scale might not have had good overlap with those on the PPSES. The scores for the present sample on the Family-Partnership scale ranged from 23 to 90 with a mean score of 79 (SD= 15.6). In comparing the average item means on the Family -Partnership Scale found by Summers et al. (2005) in their two validation studies with the average item mean in this current study, it was found that participants had responded similarly to the Family-Partnership Sclae across the 3 studies. The information on the average item means for the 3 studies are provided in Table 4. The correlation between the PPSES and Family-Partnership Scale was computed using an n = 73. The correlation between the first subscale of the PPSES , "Interactions at Meetings," and Family-Partnership Scale was found to be positive (r = .851, p = 0.01 level). The relationships between the Family-Partnership Scale and the next three subscales , "T ime Issues," "Emotional Perspective ," and "Acceptance of Differences ," on the PPSES were positive , yet more moderate with correlations of .560, .470, and .453 (p = 0.01) respectively .
Overall, this indicated that parents who reported higher levels of satisfaction regarding their relationship with service providers also reported more positive perceptions with regards to : their interactions with school per sonnel , issue s dealing with time or scheduling throughout the process, the emotionality of the meetings , and the acceptance of individual differences, such as culture by school personnel. It is not surprising that the scores on the first subscale of the PPSES and the Family-Partnership Scale had the strongest relationship because both scales assess aspects of the interactions parents of children in special education have with school professionals or service providers.

Research Question 5: Are responses on the PPSES influenced by social desirability?
Social Desirability . The scores for the present sample on the MCSD ranged from 11 to 23 with a mean score of 17.2 (SD = 3.03). A correlation was performed between the four subscale scores on the PPSES and the total score on the MCSD to determine if participants were responding in a way that was viewed as socially desirable. An n = 67 was used due to missing data. The analyses for the first subscale , " Interactions at Meetings," resulted in a correlation of .300 (p = .05). This correlation indicates a positive, yet weak relationship between the score on this subscale with the total score on the MCSD. The next three subscales , "Time Issues, " "Emotional Perspective," and "Acceptance of Differences," had correlations of .225, .141 , and -.041 , respectively.
These three correlations were not significant, thereby indicating no relationship between these three subscales and the MCSD. Therefore , it can be suggested that the participants ' responses on the PPSES were not influenced by a high need for approval on the participants ' part.

Discussion
The current study was conducted to develop and validate a scale hypothesized to assess parent perceptions of the special education process with regards to their participation , interaction with school personnel , emotional tone of the IEP meeting , and the cultural sensitivity of the school personnel. Overall , this project had an 18% return rate, with 74 surveys returned out of 400. The majority of the sample was White , non-Hispanic mothers with varying education and income levels.
Results from the PCA resulted in one item being dropped from the scale becau se it failed to load on any of the factors and had a low item-scale correlation , resulting in a 45-item scale. The findings from the PCA also indicated that eleven items be considered for elimination from the PPSES because they failed to have high enough loadings on a factor (greater than .40) or were considered complex (had loadings of .40 on more than one factor). Results also showed that the PPSES was found to have four distinct factors.
This four factor solution accounted for 60% of the variance. The four factors were labeled: Factor 1-"Interactions at Meetings, " Factor 2-"Time Issues ," Factor 3-"Emotional Perspective, " and Factor 4-"Acceptance of Differences. " The first factor, accounting for 34% of the variance, demonstrated good internal consistency , while factors two and three showed moderate internal consistency. The fourth factor demonstrated poor internal consistency , indicating a potentially unreliable and unstable factor. The resulting four subscales were strongly to moderately correlated with one another , indicatin g a sizable amount of overlap between the subscales.
A four factor solution is being accepted at this time even though the fourth factor was found to have an unacceptable reliability estimate. This was decided because of the limitations involved with running a PCA using a small sample. Due to the small sample for thi s study, the factor structure should be characterized as unstable. Therefore , a similar analysis run with an adequate sample may produce a cleaner, more stable factor structure , possibly affecting the reliability estimates of the factors. An adequate sample is needed to determine the appropriate factor structure for the PPSES , in which the results are considered stable and the researcher could be confident that the factor structure accepted and the reliability estimates generated are appropriate and accurate.
To some degree, the resulting four factor solution corresponded to the hypothesized constructs. Factors 1 and 3 reflected the original hypothesized constructs of emotional tone of the meeting and interactions with school personnel. Factor 4, "Acceptance of Differences, " included two items reflecting cultural sensitivity , which was another of the hypothesized constructs , but also included an item not originally thought to have a relationship with the items related to culture. Factor 2, "Time Issues," was not original ly thought to be a construct that the scale was measuring. The fourth hypothesized construct dealt with parental participation at the meeting. It appeared that many of the items originally thought to measure participation combined with the original items thought to reflect interactions with school personnel to create Factor 1-"Interactions at Meetings." While explaining and defining the constructs to be measured and extracted from the analysis was considered during item creation , the items may be more representative of the overlapping constructs of participation and interaction with school personnel, thereby combining to produce one factor accounting for 34% of the variance.
In an analysis of convergent validity , it was found that the subscales of the PPSES had positive and significant correlations with the Family-Partnership Scale. Both scales assess various aspects of the interactions parents of children in special education have with school professionals or service providers. Therefore , it can be concluded that the PPSES is measuring constructs similar to those measured by the Family-Partnership Scale, which was expected. Lastly , it was found that the scores for the subscales on the PPSES and the total score on the MCSD had positive and weak correlations indicating responses on the PPSES were not influenced by social desirability or need for approval.
The results of this study should be viewed and interpreted with caution due to various research obstacles that were encountered as well as other limitations .

Research Obstacles I Limitations
One of the more challenging aspects of this study was recruiting schools to participate and distribute the surveys to parents of children in special education .
According to Harrell, Bradley , Dennis , Frauman , and Criswell (2000) , schools can provide a large pool of potential subjects, but gaining access to the school and additionally the population of students or parents within that school system can be quite challenging. The recruitment of subjects in many clinical trials has also not been without obstacles and hardship (Beasley, 2004;Hochauser, 2003;Sullivan, 2004). Research conducted in both the schools and for clinical trials have noted difficulty in recruiting the desired number of participants for their studies, and have also expressed how time-consuming the process can be, often delaying research projects for varying amounts of time (Harrell et al., 2000;Sullivan , 2004).

52
In reviewing both the psychological and medical literature , some of the suggestions and recommendations for subject recruitment were followed in the present study. For example, Harrell and colleagues (2000) suggest that in order to achieve credibility with the site or arrange an initial contact with the school, it may benefit the researcher to have an association with individuals working within the school system or to have provided services to the school earlier. The principal investigator and first major professor had contacts at each of the sites. The professional contacts within the schools varied from school psychologist , principal, and special education director. In some instances the individuals appeared enthusiastic about being involved which helped in getting approval from the site' s program or special education director. However in most cases at the schools, having a personal contact within the system did not help in increasing the likelihood that the school would agree to participate.
However , other suggestions presented in the literature were not undertaken in this project during subject recruitment. For example , incentives have been reported as having a positive effect on subject recruitment (Dixon , 1978;Harrell et al., 2000). Monetary incentives or service incentives may grant access to potential research sites such as schools. Guyll, Spoth, and Redmond (2003) demonstrated that monetary incentives were effective for increasing participation rates in community prevention-intervention studies , as it was found that such an incentive positively influenced the participants in their sample. The monetary incentive also had a more positive influence on the decisions to participate made by those with less formal education , thereby potentially reducing sampling bias in studies where those who choose to participate may tend to be more highly educated (Guyll et al., 2003). For this project, there was no incentive for parents as the researcher wanted to collect the information anonymously and allow parents to respond without the worry of having the site find out what was said and potentially affect the services their child(ren) received . The schools and agencies themselves were offered a summary of the results found for their site to serve as feedback about how the special education process is conducted . Incentives were also not undertaken due to the limited funds of the researcher. It is difficult to determine whether participation on the schools or parents part would have increased if given an incentive, however the recommendation of an incentive will most likely be an integral part of the discussion for the next survey project the researcher conducts. Young and Dombrowski (1989) explored the difficulty of working with other agencies in order to complete research projects , and noted the importance of having the researcher keep in mind that the staff have their regular demands placed on them as well as the work they are now doing for the study. At the sites used in this study, attention was paid to the school or agency staff and personnel that were helping with the distribution of the materials. The researcher provided all materials for the research packet and was available to come to the site in order to help with any of the paper work or other duties that needed to be completed in order to distribute the surveys successfully .
It was important to gain and maintain the staff's support throughout the process , and a follow-up call was made near the end of data collection to inquire if there were any questions , concerns, or issues that may have arisen with parents after receiving the survey. The feedback obtained from the contacts was positive, and at the end of the process each site was sent a thank you note for their help in the project.

54
This study had significant limitations. First, the estimated sample size needed for the study was between 200-300 participants as recommended by Comrey and Lee (1992) in order to achieve stable results using PCA. However, the actual sample size was 74 participants, therefore the results should be interpreted with caution given the potential instability of the factor structure. The small sample size also resulted in the study having low power. The concept of power refers to the probability of rejecting the null hypothesis when an alternative hypothesis is true (Keppel , 1991 ). Cook and Campbell (1979) view the concept oflow statistical power as a threat to internal validity, in which the probability of a Type II error occurring (e.g., retaining the null hypothesis when it is false) increases. What this means for the current study is the probability ofrejecting the null hypothesis (e.g., there is no four factor structure present) and retaining the proposed hypothesis is low. At the same time, the probability of retaining the null hypothesis when it is false is increased. Therefore, it is erroneous at this time to come to a conclusion as to whether to retain or reject the null hypothesis as there is an insufficient sample to lend support to the proposed hypothesis , to increase the power of the study, and to decrease the odds of making a Type II error.
A second limitation of the study is the absence of a representative sample resulting in findings that are unable to be generalizable to a broader or more diverse population. The obtained sample can be characterized as being homogeneous with approximately 93% of the sample being White, non-Hispanic females from one geographic location. After reviewing the samples of previous studies that surveyed parents on the special education process, it was discovered that the previous studies had a similar limitation. Previous studies reported a majority of women participants (Lian & Fontanez-Phelan, 2001;Summers et al., 2005) typically drawn from one geographic location (Arnold, Michael , Hosley, & Miller , 1994;Lian & Fontanez-Phelan , 2001).
The resultant sample for the current study is a subset of the initial targeted sample and is not representative of an ethnically and economically diverse group. Because the results of this study were based on a homogeneous sample, the results can only be generalized to samples that are similar to those surveyed in the current study.
Additionally, the participants in this study were identified through local agencies working with parents or from non-traditional school settings (i.e., charter schools), therefore the generalizability of the results is limited to parents involved with parent advocacy agencies or non-traditional school settings. The results may be different when using a sample of parents whose children are enrolled in more traditional school settings.
The third limitation pertains to the grade levels the parents ' children were in at the time of the study. In the initial design of the study, the parents were going to be recruited from elementary school systems and the Family-Partnership Scale was chosen for that reason. Summers et al. (2005) designed their measure to be used with families with children in early childhood or families within specific services (i.e., clinical pediatric programs). However , in order to get a respectable sample size for this study, families who had children in special education up through grade 12 were used. It should be noted that the Family-Partnership Scale was not validated using a sample of older children.
Therefore the findings from the analysis of construct validity should be viewed with caution. Also , when looking at construct validity, is important to note that the PPSES and the Family-Partnership Scale used the same method , pencil and paper , to present the constructs and record the participant's responses. Cook and Campbell (1979) refer to this as mono -method bias. Given this bias, the results of the study would not be generalizable to instances in which the survey was presented or respondent's answers were recorded using other means.
The fourth limitation involved the translation of the research packet materials.
One school system that chose to participate informed the researcher that a significant portion of the parents who would be receiving the survey spoke little to no English , and asked if the research packet could be translated in order to accommodate all parents
This study attempted to create a tool that used quantitative methods in order to collect information about the perceptions parents had with regards to the special education process. What is still needed is a study with a larger and more heterogeneous sample using a quantitative measure that may demonstrate a more powerful statistical analysis and result in a cleaner principal components analysis.
The samples obtained in this study came largely from agencies working with parents or from non-traditional school settings such as charter schools. Future research may want to address the perceptions parents have within more traditional school systems as larger samples of parents of children in special education are present in those settings .
It may also be that the attitudes of professionals and parents vary from agencies and nontraditional public school settings when compared to those parents who place their child in traditional public school settings. Previous research has found that parents who had children with special needs enrolled in charter schools reported higher levels of satisfaction with the services their children were receiving at the charter school than what the children had been receiving in the previous , traditional school setting (Lange & Lehr, 2000). Also , parents with children enrolled in charter schools held the perception that the charter school provided a greater availability of special education services (Lange & Lehr , 2000) . Through qualitative data, parents have also suggested a higher level of satisfaction with regards to the quality of teaching, curriculum, and administrators at charter school s (Lange & Lehr , 2000). Further investigation into the perceptions parents have on the IEP process in traditional school settings and interactions with the traditional school personnel is warranted.
Future research will need to address the role of diversity and how this impacts parental perceptions of the special education process, especially with regards to their perceived level of involvement. Culturally and linguistically diverse parents are reported to have low participation rates in school based planning (Geenan, Powers , & Lopez-Vasquez , 2001) and may indicate a desire for less involvement (Shriver & Kramer , 1993). The ideas behind parent participation tend to be based on ideals that hold high value in the dominant , majority culture (Kalyanpur, Harry , & Skrtic , 2000). The belief that parents are entitled to the same knowledge about their children that professionals have and the expectation that parents will advocate for their children may go against the cultural values of some parents in which deference towards the experts (in this case the school personnel) is the norm (Kalyanpur , Harry , & Skrtic , 2000 ;Lynch & Stein , 1987).
It is important to note that the lack of involvement on the part of culturally and linguistically diverse parents should not be interpreted as a lack of interest in their child ' s IEP (Lian & Fontanez-Phelan , 2001). On the other hand , culturally and linguistically diverse parents who wish to participate have reported in qualitative studies that they often feel stereotyped and disrespected after meetings with school profe ssionals in which they describe themselve s as taking an active role in the process (Rao , 2000;Salas , 2004 ;Zionts , Zionts , Harrison , & Bellinger , 2003) . Under standing the cultural differences in the perceptions of the special education process and specific issues pertainin g to those parents from diverse culture s is an area of research that needs to be explored further.  Please take a moment to answer the following questions: 1. How many months ago was the last special education meeting you attended?
Appendix C

Parent's Perceptions of Special Education Scale (PPSES)
Please take a moment to fill out the survey below. When answering , think about the experiences you have had at IEP (individualized education program) meetings. Check the box that indicates how much you agree or disagree with the statement.      You have been asked to take part in the research project described below. If you have any questions, please feel free to call Kimberly Hill, co-investigator, responsible for this study.
The purpose of this study is to collect data on the perceptions parents have regarding the special education process, particularly the Individual Education Program (IEP) meetings. You will be asked a series of questions asking you to rate how you felt during special education meetings as well as rating your experiences with school personnel. The goal of the research project is to create a tool that will provide valuable feedback from parents who have gone through the special education process in hopes of improving the process.
If you decide to take part in this study, your participation will involve filling out a 7 page survey on your experiences throughout the special education process , and will only take about 20-30 minutes to complete. The possible risks or discomforts of the study 46. I was notified about the meeting in a timely manner 5.14 1.05 in order to attend.