This paper presents the results of a research trying to evaluate of internal validity of placement and effectiveness of type 8 of special education, e.g. reintegration of these pupils in regular classes. Belgium has a specialized educational system, divided into eight types of teaching. Although, an increasingly amount of students join this educational system, mainly type 8 (intended for the children presenting learning disabilities), we have very few data on the performances of our special education system, on its capacity to achieve its own objective and its effectiveness. The special education’s type 8 is organized only at the elementary school level for children with learning disabilities. The requirements and the school organization imply that the child must pass first through regular education before being directed in special settings of type 8. It acts of like a "wait-to-fail model" (Lyon, Fletcher, Shaywitz & al., 1999). In fact, the average age of entry in special education of type 8 was 9.40 years (standard deviation 1,64 years), the most important group being those arriving towards 10-11 years, i.e. suppose to be in 5th grade (21,07%). Moreover, 81.14% of the students arrive with an "official" delay of less then one school year. Placement in special education of type 8 follows an evaluation of a special center (CPMS) leading to the drafting of a protocol with the agreement of the parents. The goal of the special education of type 8 is the reintegration of the pupils in regular classes. This finality caracterizes the specific organization of type 8 (exists only in primary education) and is undermaked by various texts. The decision of rehabilitation relies upon a decision of a staff meeting, assisted by the CPMS of the specialized school. However, it should be noted that a come back in regular education, except project of partial or permanent integration, imply total ending of intervention from the specialized school. The parculiarities of the special education of type 8 and its contextualisation, difficulties both agreed on a definition of learning disabilities and on reliable diagnostic tools, make it very difficult to compare with foreign school systems. Beside, there are very few researches in the French Community on special education in general and type 8 in particular. However, in the case of the Dutch-speaking Belgium, several researchers have focused on this problem. De Fever & Coppens (1995) and Ruelens, Dehanschutter & al. (2001) realized a study on the quality of the global functioning of the type 8. They notice a large lack of valaidity of placement and weak outcomes. With regard to the passage towards the secondary, De Fever & Coppens (1995) observe that 82% of the children are reintegrated into regular education (62 % in 1B and 20 % 1A). There are 18% of the children who are directed towards secondary special education of the 3rd form (15%) or 2nd form (3%) mainly in types 1 or 3. This paper presents the results of researches conducted in 2006-07 about the internal validity and effectiveness of special education of type 8 (Tremblay, 2007). All things considered, the central question of this investigation is: does one achieve the goals laid down for the prescribed population? We analyzed the course of students having forwarded towards a special education intended for children with learning disabilities (type 8), existing only at the elementary education level in Belgium. We followed the course of these pupils in secondary education during a period of three (3) years.

In an attempt to answer our assumptions, we have established a methodology for tracking a cohort of students leaving special education of type 8. This was to take into consideration all students leaving type 8 schools during or at the end of the 2002-2003 school year and to know the reasons that led to their orientation at entrance, as well as their journey up to three years after the end of primary school-type 8. On checking the internal validity, we have five sources of information to characterize the students and their school: the CPMS protocol, CPMS documents, the school administrative documents and the telephone contact with the school and / or with parents. A consultation record presents obvious limitations. It is indeed very difficult to know the students on this basis and evaluate its real level of skills. Another approach would have been to control ourselves the validity of the recrutment process with an individual assessment of each student. However, in addition to the persistent problems regarding the diagnosis of these students, this approach would not permit, unless substantial resources, to create a large sample. The objective of this research is to compare rather than formally data from the prescribed requirements on the matter. This limits in part our investigation on more formal aspects but also allows a better understanding both of the field of investigation as of the scope of our analysis. This will enable us to determine wether the requirements for the diagnosis of these students, and any exclusion are well respected (Communauté française de Belgique, 1992). Three main precautions seemed essential to carry out the implementation of our methodology and the analysis of our results. A first precaution concerns the difficulty of ensuring, as we have just seen, the internal validity of this type of education is if students attending this type of education really all have learning disabilities. Indeed, given our previous results and our first hypothesis, it is likely that this type of education is not directed exclusively to a population of children with learning disabilities. However, that they have common caracteristics, as having learning difficulties such as maintaining regular education, seems more possible. Our research interest is for pupils who have attended special education of type 8 and not for those who strictly have a learning disabilities (whether they go or not special education). We select therefore all children who have passed through this type of education, regardless why they are there. However, those reasons will be taken into account in our analysis. A second caution is regarding the validity of the information gathered in the records of students. We choose to take into account only the documents and information corresponding to a legal prescription necessary for the placement, (e.g. protocol, IQ, medical tests, route, etc.) or whose value is factual (e.g. date of birth, name, nationality, etc.). This will offer us an opportunity to see if these assumptions are the same for all students, according to various characteristics, which may or may not have impact on their orientation. A general confirmation can hide behaviour in some sub-samples, as it have been demonstrated in our previous work (Tremblay, 2003). Finally, other informations, although of interest, are not taken into account because, for example, they do not figure in a sufficiently large number in the records or are subject to many interpretations. Wishing to become acquainted with the learners' achievement three years after the end of their schooling in formal type 8 education, we have contacted their schools or their parents during the year 2006. We were consequently able to know the course for students (grade…), repetition, change of school, etc.) during the three years that followed their orientation in secondary education. The specialized schools in which we have taken into account all the outgoing were chosen along geographic (five provinces), and according to the five networks to which they belong, taking care to select more than 50% of schools in each province. For example, if there were several school networks (state, city/town, province, independent confessional or independent non-confessionnal) represented in the province, we wanted at least one school in each of the school networks. We have also varied the sites. This means that within the same geographical area, when two schools of different networks were present, we select only one. The schools that we have selected (n = 38) accounted for 58.46% of Walloon’s schools organizing type 8 classes that year (2002-2003) and scholarizing 2531 on 3957 placed students (63.96%). In the constitution of our sample, we also represented a situation that characterizes obviously schools. Indeed, we found three types of situations in the organisation of classes. o Group A: Schools where classes have only one type of education - students are grouped by type and level (45.32% of the students out-going) o Group B: Schools where mixing occurs only in 1st grade, the policy then was a clear separation (10.74% of the students out-going) o Group C: Schools where classes have several types of education - students are grouped by level (40.57% of the students out-going)

This research has attempted both to assess the internal validity and effectiveness of a special setting designed for children with learning disabilities. For the first matter, it's necessary to check if this type of education is directed to the specified peoples. To do this, we analyzed the records of the students under the legal provisions. Considering the second matter, we evaluated, with the help of the secondary schools and parents, the aim of reintegration for these students into regular education. The difficulty of defining these specific disabilities, their reality for all children in the sample and the relative accuracy of diagnoses as real or as formal (e.g. protocols), makes it difficult to verify the first hypothesis. In the Walloon Region, a minority of students is diagnosed by an estimate of learning disabilities (less than 5%). However, the presence or absence of this information in the protocol does not necessarily mean that the child has no such disability. We must for that reason look for other indicators. Relying on the various exclusions from the definition and the circular of 1992, one can observe a certain gap between the people in this type 8 and prescribed requirements. Considering the first hypothesis, we can say that this category of students does not include children with solely and exclusively learning disabilities. Although we cannot conclude with conviction that a lack of validity, reliability, comparables and occurrences prevents us to have an opinion. We believe, on the basis of the data to our knowlegde, about 10% to 15% of the students should not be in this type of education. This phenomenon is much more present in Brussels (25% to 33% of students) than in Wallonia (Tremblay, 2003). The latter hypothesis focused on achieving the goal of integrating students into regular education. Using the course of the students in secondary education, we can observe a heavy reliance on secondary special education structures. About half of the outgoing students of the special education of type 8 will attend a high school organized for students with mild mental retardation or behavioral disorders (types 1 and 3). These changes resulted from the inability to propose a reintegration in secondary education. This orientation in secondary special education drastically limits their academical hope and their selection of schools;even if they might have access to professional training. For students in secondary education, they reintegrated regular (+ / -50%), after theses three years studied, we can observe that they are moving in large numbers to a vocational education, with a less demanding curriculum in "classical" subjects (languages, mathematics, sciences…). This trend in vocational education is also very fast. Few students will go to a general or technical tracks, even those with the opportunity to do it. For example, nearly all students who obtain their CEB in 1B are continuing in 2P. After three years, there is still less than 4% of the students who are attending general and technical education. In the end, very few students who have passed through that type of education will have a diploma giving them access to higher education. Overall, we can say, with regard to the verification of this second hypothesis, that a the large number of students going into special secondary education in types 1 and 3 or in regular secondary vocational education reveals a significant failure to achieve the purpose of reintegration both quantitatively and qualitatively. We cannot conclude that this type of education is the most successful regarding student’s achievement. These results are significantly different than those found in Brussels and Flemish areas where the rate of reintegration into regular education ranges from 75 to 82% and the use of vocational is more nuanced. Paradoxically, while Brussels recruitment validity seems to be lower than in Wallonia, their effectiveness looks better. Although, enrolling twice children with a dubious validity, the proportion of students returning in regular schools is far superior in Brussels. Some might think that the type 8 in Brussels receives “different” students than in Wallonia. This is true in part (nationality, language spoken at home, IQ, etc.). But, theses factors disadvantage rather the Brussels region. For example, we clearly observed that in an equal socioeconomic level, the Brussels returning more regular education. Various others factors (heterogeniety, orientation practices, schools offer….) may be involved in these differences observed between regions. Indeed, an orientation in a specific type of special education is losing its relevance for about 40% of students (in group C), i.e. they will receive an other kind of education. These forms of classroom oganization did not seem neutral. The purpose of reintegration may not be prosecuted for all students in mixed classes and have an effect on practices. Of course, the methodology used here does not verify such a case. We don’t know either, the effect of this practice on students in types 1 and 3. It might be possible that the mixed type didn’t have the same effect for all of these students. The latter could possibly take advantage of this heterogeneity (emulation, indisturbed class…). In spite of this, this type of educational organization does not affect diagnostic’s validity, but its educational consequences. The way than special intervention for children with learning diabilities is organized in Belgium doesn’t seem efficient to allow these students to carry out their school acheivement. The idea of giving specialized school services in specialized structures created serious problems in accessing early services, the validity of the recruitment and with the aim of retegrating regular education.