1.1 The role of the Civil Society Organizations in legitimizing the human rights of people with disabilities

Civil Society Organizations⁶ are made up of voluntary partnerships with the public administration, on a non-profit basis, under a mutual cooperation regime, involving or not transfers of financial resources. Their objectives are to stimulate democratic management through the formulation, execution, monitoring and evaluation of policies and actions of public and reciprocal interest for the promotion of human rights. Fundamentally, they dedicate themselves to issues that involve political and social minorities in the areas of health, education, culture, science, technology, agrarian development, social assistance and housing (^{Institute for Applied Economic Research [IPEA], 2020}).

With regard to people with disabilities, the current 781,921 Civil Society Organizations⁷ represent an important reinforcement to the more than 4,000 specialized Brazilian institutions⁸ in the process of breaking stereotyped and negatively discriminatory views, reformulating policies and reframing cultures from different spheres of society. In this sense, through the involvement of people with disabilities and their various professionals, Civil Society Organizations have invested in training human resources, empowering families and public awareness of the rights and services aimed at this population segment.

However, in the context of these experiences, some questions have become more evident: What conceptions regarding disability do professionals from different areas of public services present? Are there differences in conceptions when considering the contexts of the actors, their training and area of expertise? Can training or continuing education courses contribute to changing conceptions?

In the face of such concerns, this study aimed to identify and analyze the conceptions of disability of managers and technicians from different public areas, rights councils and professionals from other similar organizations that work in the service networks of four municipalities in the State of São Paulo. Specifically, the aim was:

The study is justified by the need to foster, together with the scientific, academic, institutional, governmental and social communities, possible paths for the identification and analysis of conceptions in relation to people with disabilities. It is also noteworthy to highlight its direct relationship with cooperative, affirmative and accessibility actions involving the Civil Society Organizations, public, private and non-governmental initiatives, aimed at recognizing the identities and differences of people with disabilities, as political and social minorities, and the protection against violence and its prevention.

2.1 Research pa rticipa nts

The study was carried out in the scope of four São Paulo municipalities. It was identifed¹¹ and selected those: a) that participated in some action of the CSO training program and/or a state program on the theme; b) with a population above 100 thousand inhabitants; c) with more than 100 inhabitants with disabilities, according to the last census carried out by the Brazilian Institute of Geography and Statistic (^{IBGE, 2012})¹²; d) with geographic distribution within the state (close to the state capital; on the coast; and in the hinterlands of the state).

In these contexts, 97 professionals (46%)¹³ that worked in social assistance, health, education, public security, urban mobility, human rights issues, tutelary council, municipal rights councils, specialized institutions and/or other similar organizations participated. The average age of the participants was 44.65 with a standard deviation of 9.30 and a range between 22 and 70 years old.

2.2 Data collection procedures and instruments

Using the ethical procedures of the research, the referred professionals who agreed to participate in this research took the training course¹⁴ offered by the CSO. On the occasion, the research subjects performed several activities within the scope of the four training modules and, among them, responded to the Scale for Conception of Disability (SCD)¹⁵ before the first module (test) and at the end of the last module (retest).

2.3 Data processing and analysis

The data collected through the SCD were tabulated in the Excel software, respecting their origins in relation to the municipalities investigated. For each municipality, scores¹⁶ for the four dimensions of disability conceptions that make up the SCD were inserted and calculated, which, it should be noted, allows only one answer option for each statement (Table 1):

For the quantitative analysis of the data, the software IBM SPSS Statistics Base (Version 21.0) was used. The analyzes were based on the procedures of descriptive and inferential statistics (^{Dancey & Reidy, 2006}). For all data sets, statistical procedures were performed to verify normality¹⁷. In this sense, the application of the Shapiro-Wilk test predominated and, depending on each case, we proceeded with Analysis of Variance (ANOVA), Kruskal-Wallis and Wilcoxon non-parametric tests, Post Hoc Tukey's or Dunn's and Effect Size Calculator (Cohen's D).

Considering the significance level of α = 5%, the quantitative analyzes were discussed with the scientific literature, with a view of understanding the conceptions of disability of the professionals participating in the research and, using tests and retests carried out in the context training courses, identifying their possible relationships with geographical origin (municipalities) and professional categories.

3.1 Professionals and their conceptions of disabilities

The disability conceptions of the 97 research participants showed statistically significant differences among them, with greater agreement prevailing in relation to statements associated with social and biological conceptions (Table 2).

It is possible to notice fluctuations in the values of the means among the conceptions. To verify whether these fluctuations can be considered significant, an Analysis of Variance (ANOVA one Way) was performed and, with a 95% confidence level, the result reported p <0.05. Therefore, it can be said that there is a significant difference among the means of the four conceptions, confirming what the values in Table 2 indicate.

An expected result is considered, given the political context and the understanding of disability in the Brazilian reality. This is so because, even though the Statute of the Person with Disabilities has determined the social perspective, the narratives with a biomedical perspective identified indicate that the classification character and focused on the individual tendentiously predominate.

This result shows that understandings and conceptions can coexist in the same society and that, not necessarily, one is canceled so that the other exists. The understanding about the various social phenomena, as well as the ones of disabilities, is not static and remains in a constant process of changes that, according to ^{Dempsey & Nankervis (2006)}, occur as "society's values in general change" (p. 6).

3.2 Relationship between professional category and dimensions of disability conceptions

The analysis of the relationship between the variables Professional Category (Management; Social Assistance; Special Education; Education; Health; Psychology; and Security)¹⁸ and Dimensions of Disability Conceptions (Social; Historical-Cultural; Biological; and Metaphysics) was carried out under the hypothesis that, for the same conception, there would be different agreements, depending on the participants' duties. In fact, at first, when the data average was specifically considered, it was possible to infer differences among professionals (Table 3).

However, it was necessary to verify the significance of this variation, for which we proceeded with the normality test (Shapiro-Wilk) and the homogeneity of variance. According to the result, we proceeded with the appropriate test, in that case the Analysis of Variance (ANOVA) and the Kruskal-Wallis test. Based on these procedures, the data presented in Table 4 are the comparisons between the Professional Category and intra-dimensional factors.

The data indicate that the distribution of scores measured for each Dimension of Conception of Disability behaves practically homogeneously among the Professional Categories. This means that the Occupation Position factor does not significantly influence the agreement or disagreement of statements of a certain disability conception. Thus, the hypothesis that the professional's performance could influence their conception of disability could not be corroborated, which may indicate that "definitions of disability describe not individuals but society's perceptions and expectations of people associated with that definition or classification" (^{Dempsey & Nankervis, 2006, p. 7}).

3.3 Relationship between the dimensions of conceptions of disabilities and municipa l contexts

From the perspective of the Policy Cycle Approach, the micro-contexts (training of participants, management of the municipality, municipal policies, etc.) of the participating municipalities and the understandings of professionals regarding disability were considered. The hypothesis in which it was conceived that the same dimension of conception (dependent variable) would diverge (agreement/disagreement) significantly among the four participating municipalities (independent variable) was verified.

In this sense and based on the Analysis of Variance (ANOVA) and the Kruskal Wallis test¹⁹, the data presented in Table 5 indicate the descriptive values for each municipality in relation to each dimension of the SCD, as well as their respective p values, which indicate conclusive test results.

Fundamentally, the highlighted results indicate statistically significant differences for the Biological and Metaphysical conceptions, that is, it corroborates our hypothesis that the same conception of disability may not show agreement among the municipalities. Through Post Hoc tests it is possible to accurately interpret where this difference is found.

The Post Hoc Tukey's test specifies that the descriptive results of Municipalities 3 and 4 differ from each other in relation to the biological conception, with the first municipality tending to agree more with this conception according to the descriptive data. The Post Hoc Dunn's test, in turn, reveals that the difference in metaphysical conception is present in Municipalities 2 and 3, and this occurs in greater agreement in the second municipality. The unreported municipalities showed practically similar values.

However, when considering all study participants, without relating them to their respective municipalities, a greater agreement with the social conception of the disability and less agreement with the metaphysical conception was found. This result indicates that the actors who work in the services of the network for the violence prevention and protection against it seem to be in line with the social perspective, advocated by the International Convention on the Rights of Persons with Disabilities (^{United Nations Educational, Scientific and Cultural Organization [UNESCO], 2006}) and the Brazilian Inclusion Law (^{Law no. 13,146, July 6, 2015}).

3.4 Disability conceptions: compa rison between test and retest

The comparison among the conceptions of disability of all study participants considered the data from the first application of the SCD (test), carried out at the beginning of the training course and, at its conclusion, from the second application of the SCD (retest), using descriptive and inferential statistics. This is so because, if we considered in this comparative analysis only the mean and the standard deviation of the participants' disability conceptions, at the time of the test and the retest, it could be inferred that the training course does not influence the change in disability conceptions, especially when we find that the variations between the data do not indicate major changes (Table 6).

When proceeding with inferential statistics, in turn, the Normality test indicated that the data set did not follow normal-type distributions and, therefore, we proceeded with Wilcoxon's non-parametric test, in order to support the comparison between the pairs of measures of test/retest conceptions, considering each dimension in each municipality (Table 7).

When considering, in comparing the four conceptions of disability measured in the test and retest of all municipalities, the 5% index (<0.05) as a parameter of statistical significance by the Wilcoxon test, it is noted that there is a significant difference only for the Biological conception (Municipality 2) and for the Historical-Cultural conception (Municipality 3).

Specifically for these two measures, the effect size was calculated and, considering the parameters indicated by the literature, the differences cannot be considered as having an effect of great magnitude²⁰. Thus, this result reinforces that the training carried out promoted simple changes in the conceptions of some participants, but without great effects on those of all participating professionals.

It should be noted, however, that most professionals had a tendency, since the beginning of the training course, to agree with the social concept of disability (Tables 3, 4 and 5), even though the agreement/disagreement rates in relation to the same concept of disability have not been homogeneous among the municipalities.

These indications converge with those found by ^{Baglieri et al. (2011)}, ^{Mattos (2016)} and ^{Carballo et al. (2019)}, who identified the dominance of social conception in social discourses, due to more recent historical-cultural influences, even with some remnants of agreement with understanding biological disability.

In addition to these results, the study revealed that the professional categories, in themselves, do not influence the conception, even though the tests have indicated that the conceptions of disability differ significantly among the municipalities. These nuances contribute to support the statement that the conceptions of disability may not be an easily observable phenomenon. However, improving strategies to identify them in different social spheres is a fundamental action to encourage the expansion of scientifically based debates and initiatives.

Identifying and deconstructing negatively discriminatory conceptions about the phenomenon of disability is essential in the process of overcoming potentially violent attitudes, in their most diverse faces, present among the most diverse social actors.