Abstract
Federal and state policies require all students with significant cognitive disabilities to participate in state assessments and be included in measures of adequate yearly progress. Although these alternate assessments of grade-level content based on alternate achievement standards have been in place for several years, little is known about the knowledge and skills of students with significant cognitive disabilities or about the influence of these assessments on student learning. This study uses an empirically evaluated classification scheme of data from one cohort of students’ participation in a statewide alternate assessment, Connecticut’s Skills Checklist, to validate claims from teacher focus group interviews. Data from both components of the study suggest that nearly half of the teachers found the content on the Skills Checklist inaccessible for their students. Implications of this finding are discussed.
Federal policy requires that all students with significant cognitive disabilities participate in state assessments and be included in measures of adequate yearly progress (U.S. Department of Education, 2003). To meet this requirement, states developed alternate assessments for this population based on alternate academic achievement standards (AA-AAAS) that are aligned to grade-level academic content standards. Though students with significant cognitive disabilities represent only approximately 1% of the general population (U.S. Department of Education, 2005), the diversity of these students’ communication abilities and learning challenges is quite complex. Developing standardized assessments for this diverse group of students requires a balance between the standardization required for large-scale assessments and the flexibility needed for the classroom teacher to successfully work with this population. Although a body of literature on technical issues related to the development of AA-AAAS is slowly emerging (Browder et al., 2004; Gong & Marion, 2006; Perie, 2007), few studies exist to describe what is known about students with significant cognitive disabilities and their participation in large-scale assessment programs.
The struggle to balance flexibility for the students and psychometric rigor is a national issue. The federal government set high standards for the technical quality of these instruments, which states have struggled to meet. In 2006, 38 states did not meet peer review standards for their AA-AAAS. By 2008, 15 states were still working on modifying their assessments to meet federal standards (U.S. Department of Education, 2009). In a national study of alternate assessment practices, respondents from every state reported that the student’s special education teacher administered or assembled the alternate assessments (Cameto et al., 2009). Cameto et al. (2009) also reported that the student’s classroom teacher was allowed to score the alternate assessment in 27 states. These scoring procedures are heavily dependent on disposition of the teacher (Quenemoen, 2008). Task design and scoring place special education teachers at the center of accountability decisions for their students and have a significant influence on the psychometric quality of the assessment results (Elliott & Roach, 2007).
As the measurement field grapples with improving the technical quality of alternate assessments, an effort is also under way to better understand the nature and needs of the student population targeted by these assessments. Results from one study of Learner Characteristics Inventory (LCI; Kearns, Kleinert, Kleinert, & Towles-Reeves, 2006) data from more than 12,000 students from seven states differentiated students with significant cognitive disabilities according to language use (Kearns, Towles-Reeves, Kleinert, Kleinert, & Thomas, 2009). Their study indicated that a majority of the students in this population communicated expressively and receptively at the symbolic level (i.e., they communicate with a wide variety of intents and follow directions independently). A second, much smaller group was at the emergent symbolic level of expressive and receptive communication, meaning that these students use pictures, objects, signs, and gestures in addition to oral speech and require additional cues to follow directions. The final group was at the presymbolic level of communication. These students’ communication efforts require interpretation from the listener or observer, and definition of their receptive language skills is less clear. An earlier study of LCI data from three states indicated that most students in the study had functional reading and mathematics skills but that those skills were strongly correlated with expressive and receptive communication skills (Towles-Reeves, Kearns, Kleinert, & Kleinert, 2009). Almond and Bechard (2005) identified similar patterns of heterogeneity in their study of 165 students participating in a pilot alternate assessment program. With regard to communication, almost 40% of the students were able to communicate using 200 or more words, but approximately 10% did not use words to communicate. Research evidence also exists to demonstrate that students with significant cognitive disabilities are capable of learning some components of reading (Browder, Wakeman, Spooner, Ahlgrim-Delzell, & Algozzine, 2006) and mathematics skills (Browder, Spooner, Ahlgrim-Delzell, Harris, & Wakeman, 2008).
Although these studies provide an important step in understanding the ways in which this population may access academic content, there currently exists neither research evidence nor a theoretical consensus regarding the appropriate expectations for academic content mastery by this student population. Quenemoen (2008) offered,
We do not know as yet what will work the best in teaching and in assessing students with significant cognitive disabilities in the academic content. We are seeing evidence of remarkable achievement, but this group is so varied in characteristics and the field of severe disabilities is still divided on what appropriate outcomes we can and should expect. (p. 23)
Still, it is a federal requirement that all states must assess these students’ academic content knowledge annually. It is not surprising that this situation gives rise to questions regarding the extent of students’ exposure to academic content and the availability of appropriate communication tools. Browder, Wakeman, and Flowers (2009) suggested, “What is not yet known is the extent to which all students have access to the curriculum, appropriate assistive technology, and other means to demonstrate learning” (p. 333). Little is known about these students and their learning environments and the impact of these issues on accountability decisions.
This study uses teacher focus group interviews and an empirically evaluated classification scheme of data from a statewide alternate assessment to examine the accessibility of the grade-level academic content for all students with significant cognitive disabilities. The study was conducted in Connecticut with focus groups composed of special education teachers responsible for implementing the state’s alternate assessment. Unlike most states in which alternate assessments have been administered only in the past few years, Connecticut has been employing a system of alternate assessment since 2000. At that time the statewide assessments, the Connecticut Mastery Test (CMT) in Grades 3 to 8 and the Connecticut Academic Performance Test (CAPT) in Grade 10, were expanded to include the CMT/CAPT Skills Checklist (hereafter “the Skills Checklist”). The current version of the Skills Checklist, implemented in 2006, employs teacher ratings of student knowledge both for use in improving instruction and for statewide accountability purposes. We used existing data from the Connecticut AA-AAAS program to examine teacher perceptions of student academic content knowledge based on teacher ratings. The following sections of the article provide an overview of the CMT/CAPT Skills Checklist, a description of the research methodology, and the results of the qualitative and quantitative analyses.
Instrumentation
The Second Generation CMT/CAPT Skills Checklist was developed in response to 2003 federal guidelines requiring (a) grade-level-specific alternate assessments that reflected grade-level content standards, (b) the establishment of alternate achievement standards in a valid, documented standard setting procedure, and (c) the adoption of these alternate achievement standards by the state board of education (Behuniak & Amenta, 2009). When the Skills Checklist was developed in 2005, it was envisioned that it would transform the educational process for students with significant cognitive disabilities. Specifically, it is stated in the technical manual that teachers will use the instrument to “inform instruction, monitor student growth and progress, and document achievement” and that use of the Skills Checklist will result in “(a) a change in the nature of instruction for students with significant cognitive disabilities, (b) an improvement in the quality of instruction for these students, and (c) the greater inclusion of these students in general education settings” (Connecticut State Department of Education [CSDE], 2006, p. 9).
The Skills Checklist is a nonsecure, stand-alone working document that teachers can use throughout the school year to inform instruction and assess academic skills. All students are assessed in reading and mathematics in Grades 3 through 8 and 10; science is assessed in Grades 5, 8, and 10. The child’s primary special education teacher rates the student’s performance on a 3-point scale (0 = does not demonstrate skill, 1 = developing/support, and 2 = mastered/independent). When rating an item, any mode of communication or responding that is typically utilized by the child is acceptable. Each test item consists of three downward extensions that address the essence of the content standard and expected performance statements but in a simplified form that makes them more accessible for students with significant cognitive disabilities. The CMT/CAPT Skills Checklist is available online at http://www.csde.state.ct.us/public/cedar/assessment/checklist/index.htm.
The Skills Checklist also includes a section called Access Skills, which was designed to assess preacademic skills that students without disabilities typically develop prior to school entry. The Access Skills section covers receptive communication, expressive communication, social-interactive communication, basic literacy, and spatial relationships. Scores from the Access Skills portion of the Skills Checklist are not used for accountability purposes.
Method
The purpose of this study was to examine teacher perceptions of academic content knowledge for students with significant cognitive disabilities in the context of the Skills Checklist. The following research question was of interest: Do teacher ratings of student academic content knowledge from the Skills Checklist data validate claims from teacher focus group interviews that the grade-level academic content included in the Skills Checklist may not be accessible for all students with significant cognitive disabilities?
Several methodologies were used to investigate this question. Focus groups interviews were designed based on the results of a survey of special education teachers who had administered the Skills Checklist in 2009. Quantitative analyses of 2009 checklist data were conducted to validate themes from the focus group data related to assessment administration and the manner in which teachers evaluate student knowledge. These methodologies are described below.
Qualitative Method
Patton (2001) suggested that qualitative inquiry is most appropriate for investigations in which the processes, impacts, or both are largely unspecified or difficult to measure. This is particularly true of teachers’ implementation of the CMT/CAPT Skills Checklist. Development of the focus group interview protocol was based on prior survey research on teacher use of the Skills Checklist to inform instruction and monitor student progress. The final protocol is included in the appendix. Note that the protocol addresses teachers’ use of the Skills Checklist throughout the year for multiple purposes. Only issues related to the accessibility of the academic content for all students with significant cognitive disabilities are addressed in the current article.
Participants were recruited by the consultant responsible for managing the Skills Checklist program at CSDE. A subset of special education teachers who submitted a checklist for a student in March 2009 was randomly sampled and then further limited to a pool of teachers who worked at schools near the focus group location. Those teachers received an e-mail from the CSDE stating that the state was partnering with the University of Connecticut to learn more about how the Skills Checklist was used at schools and its impact on the education of students with significant cognitive disabilities. Efforts were made to recruit an equal number of teachers at each level (elementary, middle, and high school). However, a greater number of teachers at the elementary level opted to participate in the research project.
Three focus groups were conducted in June 2009 separately with elementary (n = 10), middle (n = 5), and high school (n = 5) special education teachers who had administered the Skills Checklist in the 2008–2009 academic year. During the focus group sessions, efforts were made to encourage open and honest responses by assuring complete confidentiality. Focus groups were video recorded with the permission of the participants, and each session lasted approximately 90 minutes. Participants received a gift card for a nominal amount to a local bookstore as compensation for their time. The recorded data from the focus groups were transcribed verbatim, and transcripts were coded using grounded theory (Corbin & Strauss, 2007). Transcripts were analyzed on a line-by-line basis by the research team, and open coding was used to identify key themes in the data. Several themes emerged from the focus group data; in this article, we focus exclusively on teachers’ feedback regarding the range of abilities of students with significant cognitive disabilities and the appropriateness of the Skills Checklist for all of the students for whom the instrument was designed. Specifically, teachers at all levels consistently commented that the content on the Skills Checklist may not be accessible for the range of students they serve.
Quantitative Method
To validate the focus group data and learn more about the administration of the assessment, teacher ratings of students’ academic content knowledge from the 2009 CMT/CAPT Skills Checklist were analyzed. Specifically, we attempted to identify those students who had mastered certain areas of content as well as those who were unable to demonstrate academic content mastery, using data for all fifth grade students (n = 519). The Skills Checklist was first administered in 2006, when these students were enrolled in second grade; thus, the Skills Checklist was in place for a majority of their time in elementary school. In addition, internal analyses at the state level indicated that younger students tended to demonstrate mastery on a greater portion of the Skills Checklist content.
Although checklist ratings submitted to the state were teachers’ summative evaluations of students’ skills, analyses of these data presented an opportunity to study teacher perceptions of student academic content knowledge. Moreover, these ratings were assigned by teachers, and this potentially introduces considerable bias in the data. These limitations notwithstanding, the notion that certain students were unable to engage with the academic content explicated on the Skills Checklist warranted further investigation.
Latent class analysis (LCA; Lazarsfeld & Henry, 1968; Magidson & Vermunt, 2004) was used to identify patterns in the teacher ratings. LCA is a statistical technique used to explore latent variables, which are characteristics of people or places that cannot be directly observed or measured. Examples of such latent variables in the social sciences include phenomena such as economic development, racial prejudice, and religious commitment. LCA is used to explore classification structures in these latent variables through the analysis of variation on observed measures or dependent variables (McCutcheon, 1987). LCA is a useful analytic technique in that it offers detailed information about class membership including the latent class proportions (the proportion of cases in each latent class) and conditional probabilities (the probabilities that a case belongs to a specific class given the pattern of responses to the observed measures).
In this study, the latent variable was student academic content knowledge and the dependent variables were the teacher ratings for each content area in mathematics and reading. Separate analyses were conducted for the mathematics and reading checklist data for the fifth grade students. For both subjects, the mean scores for each content area were used as the observed dependent variables; correlations among the mean scores were indicated in the model. The language arts domains included reading and responding, exploring and responding to literature, communicating with others, and English language conventions/writing; the mathematics domains included algebraic reasoning, geometry and measurement, numerical and proportional reasoning, and probability and statistics. Note that all checklist items were rated on a 3-point scale in which 0 = does not demonstrate skill, 1 = developing/support, and 2 = mastered/independent. After the number of latent classes was established, available additional student data were used to describe the groups of students. Descriptive statistics included demographic data, student disability category data, mean scores for each domain of the Access Skill section of the Skills Checklist, and several items related to communication skills from the LCI (Kearns et al., 2006).
Results
The results of the qualitative research based on the focus group interviews and quantitative analyses based on the LCA are presented below.
Qualitative Analyses
The focus group interview protocol was designed to investigate ongoing use of the checklist for both instruction and accountability. In both contexts, the participating teachers took great care to point out that the Skills Checklist did not necessarily accommodate all students with significant cognitive disabilities. One teacher said,
It’s interesting for a test that’s designed for a population that is supposedly proportionally so small compared to the CMT and CAPT population, within that population there’s such a broad range of ability. It’s hard to use the same tool to assess all the students who we utilize it for.
This theme is particularly salient because it was addressed at the elementary, middle, and high school levels and references to the issue tended to elicit emotional responses from the teachers.
The teachers took time to establish a picture of their classroom in communicating their use of the Skills Checklist. One teacher said,
My students are just learning how to pull their pants down. Some of them are still learning how to feed themselves, doing simple exchanges with pictures and not even understanding that, not even knowing how to play appropriately with any toy.
Another elementary school teacher described working with literature with a particular student in her class:
The one that I love the best was the difference between real and make believe. I have a child who will pull his pants down. We have this discussion with the parent that he doesn’t know . . . that his mom wants him to be able to only pull his pants down when he’s inside rather than outside. He doesn’t have that realization anyways. So we went through that and I said, “How am I ever going to teach real and make believe?”
Yet another teacher, who was particularly emotional in describing her students, said,
My class is the medically fragile class and it’s called that for a reason. We’re running over because Danny’s. . . he’s not breathing and you have to go take care of that before we worry about if he’s looking at the picture.
When asked to describe their use of the instrument, the teachers spoke of a divide between those students who work on Access Skills and those who can engage more deeply with academic content. Access Skills were designed to assess communication and quantitative and preacademic skills that students without disabilities typically develop prior to school entry. Access Skills are included in the Skills Checklist at each grade, but data from this section are not incorporated into performance level scores for accountability. Many teachers described classrooms similar to this teacher:
I would say, of the seven students I teach, I use the Skills Checklist to drive my instruction for three of them. The other students it’s way too high for. I have a student who’s currently functioning on the 18-month level. I feel horrible filling out the checklist because cognitively none of those things are reasonable for her. The only thing that is reasonable for her is the Access Skills.
One teacher said, “I happened to bring a copy of [the Skills Checklist for] my highest functioning child. I chopped his name off of it, but he’s got a few things in the Access Skills and we don’t have anything else.” Yet another middle school teacher described a subset of her students as having limited decoding ability and another subset with decoding skills but limited comprehension skills. When asked how she uses the Skills Checklist, this teacher responded, “I look at the Access Skills, but the other part of the test really does not apply to my students. They’re so far removed from that level of instruction.” Another teacher elaborated, “I have to really extrapolate down, down, down to their level to really picture that in my mind, to see how it would affect my students.”
The teachers were also asked to discuss the extent to which they felt that special education teachers were prepared to teach academic content to students with significant cognitive disabilities. Several teachers mentioned that they were included in all of the academic professional development activities at their schools. Those teachers were more positive, sharing comments such as, “We’re held accountable. We’re involved in all of the regular ed. instruction” and “You’re teaching me how to teach this fifth grade math and I’m never ever going to get anywhere near it,’ but we’re exposed to everything.” Most, however, were concerned: “How well are we prepared to teach students with significant cognitive delays? I don’t think we’re very prepared at all.” The teachers explained that a community of practitioners working with a similar population does not exist for them. One teacher said, “We’re so isolated and there are so few teachers. I just cross my fingers sometimes, hope that I’m on the right track.” One teacher explained, “Those of us who are old are probably not very well prepared. We had one course in mental retardation in college.” She later added,
It just sort of floored me that the stuff that you kind of learn through working with these kids that people don’t grasp. You know, you do first, second, third. You realize, for example, that they don’t know first is one, and second is two. And the nuances of lowness. You teach them all kinds of separate skills and then you have to teach them to generalize.
Another said, “I’m a specialist in this area and I’m still making up my own materials.” In spite of these comments, another teacher added, “I don’t think you can teach somebody in a college course what this looks like.”
Within this context, many teachers suggested that two levels of testing might be more appropriate for these students: “One for students who just have Access Skills and then [one for] the ones where time in the general education environment is appropriate and beneficial.” This teacher added that some students have a diagnosis that leads to a situation in which the student is “either regressing medically or they’re not going to progress past a certain point.” A different assessment would be appropriate for these students because “it’s important for parents to be able to, and teachers to be able to focus on the progress, as opposed to all those zeroes.”
The notion that the Skills Checklist may not be appropriate for all of the students for whom it was designed was an unanticipated and alarming outcome of the focus group interviews. However, the small sample size of this portion of the study left the pervasiveness of this problem unclear. Data from the 2009 administration of the Skills Checklist were analyzed to investigate whether such patterns exist in teacher ratings of student academic content knowledge at the state level.
Quantitative Analyses
For both subjects, two-, three-, and four-class models were fit to the data. Only results from analyses of the mathematics data are presented here because of space limitations (see Note 1). Measures of model fit and the proportion of students in each class for each analysis are included in Table 1.
Model Fit Statistics and Proportion of Students by Class for the Mathematics Data
For mathematics, the Akaike information criterion (AIC) and sample-size-adjusted Bayesian information criterion (BIC) measures of model fit decreased as the number of classes increased; this suggested better fit for a higher number of classes. However, the three-class model had a higher entropy value, providing evidence of its superiority over the other models. In addition, the Vuong–Lo–Rubin likelihood ratio test for the four-class model was not significant (p = .19), which indicates that the three-class model is sufficient to represent the data. Moreover, the three-class model represents a more even distribution of students across classes as compared to the four-class model, which facilitates interpretation of the data given the purpose of the analysis. Thus, the three-class model was selected for the mathematics data, and students were classified into three groups.
After examination of the model fit statistics for the reading data, a three-class model was selected as well. The AIC, the sample-size-adjusted BIC, and entropy statistics suggested a four-class model. However, the Vuong–Lo–Rubin likelihood ratio test was not significant in either the three-class (p = .18) or the four-class (p = .21) model, suggesting that a two-class model was sufficient to represent these data. In the two-class model, however, nearly all of the students fell into the first class; further analyses of these classifications would limit the utility of comparisons across classes. Moreover, the proportion of students in Class 2 of the two-class model and Class 3 of the three-class model was similar, as were the means on the content domains. This suggested that the three-class model incorporated a split of the students in larger class of the two-class model. Use of the three-class model for the reading data allowed for a parallel structure to the mathematics data. Thus, the three-class model was selected for the reading data as well.
The three classes represent variable levels of teacher ratings of content mastery and are subsequently referred to as the students with low ratings, moderate ratings, and high ratings. For mathematics, the data show that 49% of students had mean scores across the downward extensions that are very close to 0 (M = 0.20, SD = 0.22, n = 256). A rating of 0 indicates that the student could not demonstrate the skill explicated in a specific downward extension, even with teacher support. Stated differently, teachers’ ratings indicated that half of the students did not engage with the mathematics content on the Skills Checklist. For reading, 45% of the students were in the group with low ratings (M = 0.18, SD = 0.17, n = 231). Ratings for the other two groups of students indicated that teachers viewed students’ skills as at least developing or supported, but to varying degrees. Students with moderate ratings (mathematics: M = 0.94, SD = 0.26, n = 144; reading: M = 0.69, SD = 0.26, n = 205) and high ratings (mathematics: M = 1.48, SD = 0.26, n = 119; reading: M = 1.25, SD = 0.26, n = 83) engaged with the academic content on the Skills Checklist, at least minimally. Moreover, more students had higher scores in mathematics than in reading. Replication studies were conducted with eighth grade data and produced similar findings: Approximately half of the students had mean scores that were close to zero (62% for mathematics and 45% for reading). In their entirety, the quantitative analyses confirmed the findings of the focus groups: Approximately half of the population did not engage with the academic content on the Skills Checklist.
Available data were used to describe the three groups of students. Again, only results from analyses of the mathematics data are presented here. Review of these data indicated that the distribution of the demographic variables (gender, ethnicity, free or reduced-price lunch status, and English language learner status) is approximately equivalent across the groups within each subject (see Table 2).
Demographic Data by Mathematics Classification
This provides evidence to support the fairness and validity of the instrument in terms of demographic subgroup performance. Examination of the distribution of students within each disability category across each of the three groups, however, may highlight some disparities. In reviewing these data, we consider only the category in which n is greater than or equal to 20. The mathematics data (see Table 3) indicate that most of the students with autism and with multiple disabilities fell in the group with low ratings (55% and 69%, respectively) and most of the students with speech-language impairment (46%) were in the group with high ratings.
Distribution of Mathematics Classification for Each Student Disability Category
Students with intellectual disabilities and other health impairments were approximately evenly distributed across the groups. Similar trends are evident in the reading data. A large portion of the students with autism (44%) and a majority of students with multiple disabilities (66%) were in the group with low ratings. Alternatively, there were few students with speech-language impairments in the group with low ratings (12%).
The additional tables provide a more complete picture of teachers’ views of these students’ skills. Mean scores for each domain in the Access Skills section were lower and more variable for the group with low ratings in both subjects (see Table 4).
Mean Scores and Standard Deviations for Access Skill Domains by Mathematics Classification
Individual items from the LCI provide greater detail on the typical modes of communication for each cluster (Table 5). With regard to communication skills, students in the group with low ratings were more likely to use cries, facial expressions, or changes in muscle tone to communicate whereas students in the group with high ratings were more likely to use symbolic language to communicate. Similarly, students in the group with high ratings were more likely to follow directions without additional cues. In addition, approximately one third of the students in the group with low ratings use augmentative communication during instruction, whereas most of the students in the other two groups do not.
Responses for Communication Items From the Learner Characteristics Inventory by Mathematics Classification
Discussion
The teachers in the focus group interviews referred to two types of students with significant cognitive disabilities: those for whom grade-level academic content is accessible and those for whom it is not. These findings were confirmed by analyses of checklist data in mathematics and reading. Our research indicated that nearly half of the teachers found the content inaccessible for their students. Moreover, the data showed these students had poor communication skills, used augmentative communication as part of their instruction, and were more likely to have disability classifications of autism or multiple disabilities. Ratings for the remaining students suggested teachers felt their students could engage with the academic content with some support and with varying degrees of success. Students who could engage with the academic content had higher ratings of their communication skills and were less likely to use augmentative or assistive communication devices.
These data highlight the challenges of bringing an academic curriculum to a population of students who do not share the same cognitive and communication abilities as the general student population. Students with significant cognitive disabilities have specific information processing challenges related to short-term memory and the need for increased frequency of skill repetition with instructional feedback (Kleinert, Browder, & Towles-Reeves, 2009). Teachers of students with significant disabilities encounter many more obstacles in working to cover the curriculum with their students, even a curriculum judged against alternate achievement standards. The finding that certain disability categories were more prominent in the group with low ratings is also of interest. These data are only suggestive of a relationship between disability category and teacher ratings, and further investigation into this pattern and its implications for accountability practice is needed. The present study also brings focus to teachers’ inadequate preparation to instruct their students on content of this nature. Several teachers in the focus groups were quick to note deficits in their own knowledge and skills. Inclusive accountability practices have questionable meaning and impact in the absence of effective instruction. This finding suggests the need for resources devoted to helping teachers develop instruction to meet their students’ learning challenges through preservice teacher training programs as well as professional development programs for practicing educators. Certainly, more research is needed to explore the interplay of these learning challenges and teachers’ efforts to expose students to an academic curriculum in combination with functional skills. The current study provides a preliminary examination of variability in academic content knowledge for one cohort of students in one state.
The results of this study suggest that stratification of the population of students with significant disabilities may be necessary to improve both the instruction and the assessment of these students using the Skills Checklist. This is not a new concept. Browder, Flowers, and Wakeman (2008) conducted an empirical evaluation of a classification scheme for the communication abilities of students with significant cognitive disabilities, based on teacher ratings of 189 students. Their study provided support for the notion that students with significant cognitive disabilities can be classified by symbolic level (awareness or presymbolic, early symbolic, and symbolic), and they offer this classification scheme as a tool for academic planning. The authors further suggest that it may be appropriate to establish alternate expectations for achievement on alternate assessments based on these levels of communication, as most alternate assessments are biased toward students who communicate at the symbolic level. Using the LCI, Kearns et al. (2009) found that students at the presymbolic level of receptive and expressive communication were also identified as having other issues that increased the complexity of their learning needs. These issues included low levels of social engagement, limited motor skills, sensory impairment, and health-related issues that affected school attendance. Their data suggested these students, students with the “most significant cognitive disabilities,” represented approximately 10% of the population. The present study, which incorporates data from both the LCI and Connecticut’s Skills Checklist, suggests this number may be much higher. Although the results of the current study can be considered only preliminary, the finding that the grade-level academic content of one state’s alternate assessment may not be accessible for a large proportion of the students for whom it was targeted must be viewed as a cause of concern.
Implications and Directions for Further Research
The Skills Checklist is more than an assessment of academic progress for students with significant cognitive disabilities. It is also expected to fill a role as a comprehensive guide for teachers to instruct students on AA-AAAS. The present study suggests that these standards may not be accessible for the entire population of students it was designed to serve. Specifically, the data indicate that additional classification of this small population of students may be necessary for both effective instruction and effective assessment. A distinction within this population would have important implications for educators and researchers alike.
This study raises a number of questions regarding reasonable expectations for students at presymbolic levels of communication. Are alternate assessments based on grade-specific content an appropriate accountability measure for these students in their current form? What is the relationship between information processing challenges and the extent to which the academic curriculum can be covered? Do these students have sufficient opportunity to learn the target academic content? Would it be beneficial to focus instruction, and subsequently assessment, primarily on “preacademic” skills such as communication and other adaptive activities for these students? Clearly, more research is necessary to address these questions.
One alternative to the study of student academic content knowledge is to focus research on the classroom teacher. What knowledge and skills must a special education teacher have to effectively deliver academic content to students with limited communication abilities? Can we assume that all special education teachers have such knowledge? Would alternate assessments based on grade-specific content be an appropriate accountability measure for this population if teachers felt comfortable providing grade-specific academic content instruction for these students? Furthermore, do educators have sufficient access to assistive technology or augmentative communication devices to allow for efficient communication with these students? Do educators with access to these technologies understand how to use them effectively? Although these questions cannot be answered at this time, the results of this study provide evidence of the importance of addressing these issues in the future.
Conclusion
The development and validation of alternate assessments based on alternate achievement standards for accountability constitute a nascent field. Connecticut’s Skills Checklist was created to change the nature of education for students with significant cognitive disabilities. The quantitative analyses indicated that approximately half of the target population can engage with the academic content on the Skills Checklist. However, for the other half of the target student population, the effects of participation in the alternate assessment are unclear. In the focus groups, teachers explained that these students may focus more on communication and other preacademic skills while in school. Our study raises both practical and conceptual issues that relate to preservice education, ongoing professional development, availability of resources within schools, and the broad issue regarding the impact of a large-scale assessment program on the educational environment of students with significant cognitive disabilities.
Footnotes
Appendix
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interests with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This manuscript was supported, in part, by the U.S. Department of Education Office of Special Education Programs (Grant No. H373X070003). However, the opinions expressed do not necessarily reflect the position or policy of the U.S. Office of Special Education Programs and no official endorsement should be inferred.