1 Introduction

The Federal Government has started since 2007 the construction and implementation ofthe policy of Attention to the Health and Safety of the Federal Public Servant – PASS, within the principles of universality and integrality of actions, facing the obligation to respond to actions in the health and safety at work of the federal public servant, linked to the need to improve quality of life, promote health of the servants, standardize proceedings, and optimize the resources. The PASS is sustained through the inter-relationship between the axles ofsurveillance and promotion, medical expertise and assistance to health funded by the SIASS – -Integrated Subsystem of Attention to the Health of the Federal Public Servant. From its elaboration on, various federal sectors included in their context the PASS actions. Studies are being carried out in order to verify its efficiency as highlighted by Zanin et al. (2015). In this context, in 2011, the Federal Institute of Education, Science and Technology of Bahia- IFBA, an institution of education specialized in the offer of professional and technological education in all the levels and modalities of education, started the implementation of the PASS afterreaching an agreement of technical cooperation with the Federal University of Bahia.

However, the implementation of a public policy is influenced by an array of factors that interfere in its operability, like resources, discretion, organizational culture, political influence, contextual variables, etc., and consequently, in its performance (LOTTA, 2010; LIMA; D'ASCENZI, 2013; LIPSKY, 2010; WU et al.,2014). Besides these factors, another important concept in the comprehension of implementation is the one about state capacities. These represent an important element in the comprehension of the success or failure of a public policy, (GOMIDE; PIRES, 2014; WU; RAMESH, 2015; MENDES; FERREIRA, 2021; GRIN, 2021).Thus, this study has as a goal to contribute with a discussion about the process of implementation of public policies as suggested by Barret (2004) and Mendes and Aguiar (2017, p.1105), according to whom “the analysis of the process of implementation of public policies is a field of study under construction, an object of few academic analyses in the area of public administration and of scarce discussion in all the government spheres”. It also supports the concept by DeGroff and Cargo (2009), Proctor et al. (2013), Tummers and Bekkers (2014), Mendes and Aguiar (2017) and Souza (2017), who describe methods of implementing public policies through testing new variables by means of a quantitative approach as an addition to the qualitative approaches of the implementation of public policy processes.

That stated, the goal of this research is to analyze, in the public servants ́ perception, the factors that influence the efficiency of the implementation of the policy Attention to the Health and Safety at work of the federal public servant at IFBA.

The research, therefore, justifies itself to make up for the nonexistence of data about the implementation of this policy in the scope of the Institute. Thus, it is important to point out that from the time passed between the beginning of this policy at IFBA and the period when the research was carried out, one should expect a larger amount of actions and information about this policy which in the research, showed to be the opposite. It can also be able to subsidize the elaboration and planning of the programs, projects, and actions guided by the information fromthe field research, direct the implanting servants in a way to optimize the process of implementation and furthermore, assist the Institute in spreading the policy among the servants and make them aware of the importance of their active participation in the construction of the policy.

As to the academic sphere, this research is justified by the insufficiency of public studies that explore the PASS in the context of the Federal Institutions of Education, above all with regard to the implementation phase of the policy. Using the quantitative approach, this research offers an additional contribution to the academic studies, because an important part of the research about the implementation of public policies uses a qualitative approach: a search in the portal of journals CAPES in the last five years under the phrase “implementation of public policies” shows 121 articles. However, only 2 of them mention quantitative research. Thus, this article is structured from this introduction. Followed by the discussion on public policies and the stage of implementation. Then, i nformation is presented about the PASS and the IFBS. Hereafter comes the method and then the results, and data discussion are shown. At last, come the final considerations.

2 Public Policies

According to Dias and Matos (2012) public policy is a set of parameters and guidelines that enable the state ́s management to solve national problems aiming to achieve unanimous conditions of life quality for all the citizens. The authors Secchi (2012) and Castro and Oliveira (2014) also see public policies through the bias of society. This one considers he public problem as the theoretical foundation of the public policies composed by two basic elements: the public intentionality and the answer to the political problem that construe the treatment or resolution of the problem. These suggest that it is the state ́s responsibility to provide them (or even through delegation) to attend to the challenges and problems of society ́s aspiration.

Public policy is divided into stages that together are usually known as “The Cycle of Public Policies”. However, the stage of implementation is the theoretical basis of this research.Secchi (2012) holds that in this stage, political intentions are transformed into actions, i.e. the public policy is implemented. It can be understood as the completion, the performance, and the delivery of products and services to society. (PRESSMAN; WILDAVSKY, 1984; MENY; THOENIG, 1992; FERREIRA; MEDEIROS, 2016). Furthermore, in the view of Proctor et al. (2013), implementation consists in a set of techniques and methods to enhance the sustainability of a public policy. As to Pressman and Wildavsky (1984), the importance of implementation lies in the fact that it shapes or influences in a significant way the success of the very policy. In thissense, implementation represents an important stage in the context of public policies, mainly because an array of factors can affectits performance. Aspects such as political fights, bargaining, negotiations, and disputes for power can be part of this process and decisively have an influence on the success or failure, not only of its performance but also of the public policy itself. (BARRET, 2004; BEUNEN et al., 2013; BEUNEN; DUINEVELD, 2010; DEGROFF; CARGO, 2009; TUMMERS; BEKKERS, 2014; ELIAS, 2017; JANUÁRIO et al., 2017). It can even be influenced by factors such as the performance of the bureaucrat, its discretion, organizational culture, financial, human, and material resources, the capacity of the net, power disputes among others (BARRET, 2004; PROCTOR et al., 2013; TUMMERS; BEKKERS, 2014; WILSON et al., 2018).

That is why this stage of public policy must embody a set of elements such as planning, assessment of performance indicators of the process of implementation, and mapping of factors that can interfere in it. (BARRET, 2004; CERNE, 2013; BEUNEN et al., 2013; ELIAS, 2017; WILSON et al., 2018; TUMMERS; BEKKERS, 2014), as a way of minimizing the possible difficulties encountered in the process of implementation.

In this article, state capacities and factors that influence the process of a public policy implementation are discussed, because both are relevant in this process and affect the success or nonsuccess of a specific public policy. State capacities are understood as deriving from the institutional arrangements of a state (GOMIDE; PIRES, 2014), and are defined as “the set of bureaucratic elements that involve the administrative and political abilities and competence in the goal-setting, in the allocation of resources, in the efforts and implementation of policies needed to achieve planned results “(MENDES; FERREIRA, 2021, p. 133). Its importance lies in the fact of being composed of abilities or competence needed to accomplish public functions (WU; RAMESH; HOWLETT, 2015). It is this set of characteristics that will define the possibility or impossibility of a specific public policy to be implemented or not.

It is important to highlight that in spite of the various definitions suggested by several authors (WU; RAMESH; HOWLETT, 2015), there exist similarities among them.Grin (2021, p. 7), from the literature review, considers that the concept of state capacities encompasses fiscal management, financial resources, human resources, organizational infrastructure, technical abilities, and control of its own operations. In contrast, Gomide and Pires (2014) understand that the concept of state capacities encloses two dimensions: the technical-administrative one and the political one. In the technical-administrative dimension, stand out. the competence of public agents in materializing public policies as planned to reach the expected results. In contrast, in the political dimension one includes the competence of the public agents to interact with the agents of society, including the capacity to solve conflicts of interest around public policies.

On the other hand, Wu, Ramesh and Howlett (2015) uphold that state capacities can be set up from two dimensions (level of resources, capacities, abilities and competence). In the dimension level of resources and capacities, there are the individual, organizational and systematic capacities, whereas in the dimension abilities and competence there are analytical, operational, and political capacities making up a total of nine types of state capacities: individual analytical capacity, organizational analytical capacity, systemic analytical capacity, individual operational capacity, organizational operational capacity, systemic operational capacity, individual political capacity, organizational political capacity, and systemic political capacity.

The importance of these capacities can be understood in the expression described by Grin (2021, p. 7), when he states that “to identify state capacities means to identify organizational structures, whose absence or presence are critical for the governmental action”. This same perception is understood by Gomide and Pires (2014), when they defend that it is the existing state capacities that produce desired results”.

For the execution of a public policy, the mobilization of political actors is essential since they are catalyzing elements of this process and do not in their majority belong to the top of the political pyramid (RUA, 2013).

With regard to the performance of the actors, the bureaucrats decide about the services offered by their organization, as well as about the quality and amount of improvements and sanctions/restrictions for the users (LOTTA, 2010; OLIVEIRA, 2012; SANTOS, 2017; TUMMERS; BEKKERS, 2014). Literature has given emphasis to the role of the bureaucrats of high, mid or street-level (PIRES et al., 2018), since “the bureaucrats have become key actors to understand the course of decisions and their consequence regarding the results of public policies” (LOTTA; SANTIAGO; 2018), in spite of the limited studies about mid-level bureaucrats (NOVATO; NAJBERG; LOTTA, 2020). High ranking bureaucrats are understood as those who occupy the highest positions in the organizational structure; the street-level bureaucrats are those who interact directly with the beneficiary of the implemented public policy, while the mid-level bureaucrats hold intermediate positions in the bureaucracies and are in charge of the articulation between the bureaucrats of the high and street-level. (NOVATO; NAJBERG; LOTTA, 2020).

In this study, due to the fact that it is an organization of public education, we have the figure of the dean at the high level, the vice deans and directors at mid-level, and the coordinators of the offices who represent the responsible bureaucrats for the implementation and interaction with the beneficiaries of the policy being studied. The implementing agents, confirms Oliveira (2012), hold the authority to benefit or punish a citizen with regard to the materialization of the policy. In this perspective Howlett, Ramesh and Perl (2013, p. ix) assure that “the policies implemented by the street-level bureaucrats can distance themselves considerably from what has been viewed in the stage of formulation of the policy. It is in this aspect that the bureaucrat ́s discretion and autonomy are set up, which in the view of Lotta and Santiago (2018), could configure into an advantage as they could adapt the actions of the policy implementat ion to local circumstances, thus this process becoming more efficient. Although it seems somewhat common, discretion can happen due to various factors. However, what characterizes the existence of discretion is the agent ́s freedom to make decisions regarding the implementation of a public policy. As already mentioned, this decision making can be an advantage in the sense of a decision being more favorable for the beneficiaries of the public policy or disadvantageous to them if the intention of the bureaucrat is to take profit on the implementation process acting with opportunism at the expense of the most important, which is society.

Whatever the result of discretion, it exists, according to Lotta and Santiago (2017) and Lipsky (2010) among other factors due to the scope, ambiguity, overlapping and conflicts of rules, divergence between the street-level bureaucrats and their managers or the bureaucrat ́s preference. The performance of the bureaucrats involves furthermore their capacity to coordinate the net of organizations (WU et al., 2014).

Organizational culture, in turn, can be understood as a set of shared values, customs and beliefs, comprehension of the group members, ideologies, historic traditions, and customs, rules of behavior, rules of social interaction practices, collective aspiration, activities of communication, and sanctions (SILVA, 2011). It represents the behavior considered right by the members of each group (STRESE et al., 2016). In this aspect, culture guides the behavior and can facilitate or hamper the process of a policy implantation. Barret (2004), DeGroff and Cargo (2009) and Elias (2017) point out that the implementation of a public policy can suffer restrictions when it affects symbolic structures, and the way things are built in the organization. This portraits organizational culture.

Administrative discontinuity is also a characteristic of the culture in public management.It can occur after a change in management, and the results become obvious in the bad use of the public machine, in the discouragement and apathy of the teams involved, in the waste of public resources (NOGUEIRA, 2006) and inserts the need for innovation in the negotiations (LIMA; D'ASCENZI, 2013).

Lipsky (2010) highlights the resources that are seen as human resources, material, input, and financial resources as an additional factor that can affect the process of public policy implementation. The amount of these resources will guarantee that the implementing bureaucrat is able to execute such a policy (LIMA; D'ASCENZI, 2013; WU et al., 2014).

The goals of public actions that indicate precision and distinctness to avoid ambiguous and conflicting situations are also factors that affect implementation. Vague or multiple missions and bad concepts are also included (WU et al., 2014). Lima and D'Ascenzi (2013) highlight that the organization of the administrative apparatus and contextual variables also affect the implementation of a policy.

The organization of the administrative apparatus refers to organizational structures, availability, size, vitality, and competence of human resources and hierarchy degrees (VIANA, 1996; LIMA; D’ASCENZI, 2013) and bad management structure (WU et al., 2014).

Besides the factors inside the organization, it is important to consider that those being outside the inner environment and its control can also affect the process of a public polity implementation. This aspect is relevant, because external elements in which the organization lies, can facilitate or hamper the execution of a public policy. External factors can even hinderthe adequate assessment of the internal component effects like financial and human resources in the process of implementation.

Among the contextual variables that affect implementation one can list in the first place the social context related to the public support and opinion that can be sometimes unstable. In 5308this dimension, the support of public opinion can be relevant since it can or contribute to the process of implementation if the public perceives in the public policy to be implemented, benefits for them, or it can disseminate a discourse contrary to the policy, thereby impede the process of implementation (BEUNEN et al., 2013).

The economic context reflects the influence in the transfer of financial resources and the political context that refers to government changes, support of the elites, and position of oppositional party groups (VIANA, 1996; LOPES; AMARAL; CALDAS, 2008; LIMA, D’ASCENZI, 2013). According to Proctor et al. (2013), Barret (2004) and DeGroff and Cargo (2009), the political context reveals itself as the most important among the three contexts mentioned since political instabilities tend to decisively affect implementation processes especially in conjunction with administrative discontinuity.

Political influences or political barriers such as slow approval, weak political support, bureaucratic opposition, lack of incentives for the implementer, and power disputes affect the implementation of a public policy because they are directly related to the amount and degree of the cooperation that exists between the organizations involved (WU et al., 2014; LOPES; AMARAL; CALDAS, 2008).

2.1 The Policy of the Attention to The Health and Safety at Work for the Federal Public Servant – PAS

The policy of attention to the health and safety at work for the federal public servant is upported by an integrated system of information in the health of the servant (BRASIL, 2010a).Concerning the consolidation of this policy, the government proposes several strategic actions among them, the Integrated Sub-system of Attention to the Health of the Servant – SIASS, which, according to Zanin et al. (2015), is the structuring basis of the PASS that actually has a major governmental action and thus is the most advanced one, The SIASS is a structuring system of people management which has as goal coordinate, implement and integrate actions and programs of the PASS, as well as articulate the existing resources (BRASIL, 2012a), The modal aims humanization in expert assistance, standardization and improvement of the services of attention to the health of the public servant (PEREIRA, 2012).

SIASS is operated by a multi-professional team formed exclusively of federal servants which integrates a policy sustained by 3 (three) interconnected axles: a) assistance to health:actions aiming the servant ́s health regarding prevention, detection, repair, and treatment of disease; b) official expertise: dental or medical action that consists in the technical evaluation of questions related to the health condition and the work capacity of the servant; c) health promotion, prevention and follow up: has the goal to offer actions of intervention in the processes and workplaces (BRASIL 2010b).

In order to fulfill the Pass Bahia State, the first SIASS unit was installed in 2010 at the Federal University of Bahia – UFBA. In this context UFBA as an agency participating and aligning with the policy organized the Department of Life Quality – DEQUAV, support and intermediation of the IFBA with the SIASS making it responsible in organizing programs, projects and action based on the three PASS axles with the goal to promote health and quality of life at work of their servants.

3 Methods

It is a study of descriptive nature, cross-cut end, and quantitative approach. The participants are the active and statutory faculty members and administrative technicians of IFBA. The institute has 2.518 servants, 1.107 being administrative technicians, and 1.411 faculty members. The choice for these groups was because they represent the permanent servants of the organization. The others are outsourced and have a temporary permanence at IFBA. For that reason, they did not participate in the research. Data collection was carried out between August and September 2018 by means of a printed questionnaire provided to the servants in the departments and further sectors of the rectory and Campi. 346 questionnaires were answered and completed and handed in for the analysis of the results.

The questionnaire contained closed questions and was segmented in two parts. The first one contained questions 1 to 7 and in it were data of characterization of the respondents in the research such as socio-demographic and occupational data. The second part with 41 affirmatives (variables), was built using a Likert scale with five levels of intensity (disagree in full, disagree in parts, neither disagree nor agree, agree in part, agree fully). The level of agreeing was evaluated considering the sum of the percentage of the answers for options 4 (agree in parts) and 5 (agree fully) of each affirmative present in the questionnaire.

The data obtained were moved to software R and analyzed through factorial analysis and multiple regression in order to measure the conditional relationships between the implementation of the PASS, measure by the perception of efficiency, and the factors that affect the implementation process of a public policy evaluated by the factors; comprehension of the policy, resources, goals of the public policy, organization of the administrative apparatus, performance of the actors, political influence, organizational culture and, contextual variables.

The regression analysis is a statistical procedure that allows analyzing the relationshipbetween the independent variable (explainable) and the dependent variable (explained): if the independent variable and the dependent variable move in the same direction in their averages, than there is a positive association between them. If on the other hand, the variables move in opposite directions, then there is a negative relationship. The formula used to explain the results has a general form described in equation 1:

Y =β0+β1X1 +β2X2+β3X3+... +βkXk +e(1)

Where Y is the dependent variable (percetion of the efficiency of the policy), X1, X2....,Xk... are the independent variables (factors that influence the implementation of the policy) β0 the parameter of the intercept, β1, β2...Βk, the coefficients of each independent variable and the control variable and the error associated to the model ((MALHOTRA, 2012). The control variables used in this work are gender, age group, occupation, time of service, campus, education, hierarchy level.

To use the model of regression described above since it is the questionnaire composed of 41 affirmations (variables) the factorial analysis was carried out as a goal of reduction of the number of variables in a way to make them manageable in the model of regression (FIGUEIREDO FILHO; SILVA JÚNIOR, 2010). In the questionnaire, each affirmative represents a variable or an item. Each factor is built of a specific amount of items in each variable, like comprehension of the policy (5), resources (4), goals of the public policy (4) organization of the administrative apparatus (4) performance of the actors (4) political influence (4) organizational culture (4) and contextual variables (4), The variables included in the model of regression are described according to the following chart 1:

4 Results and Analyses

346 valid answers were collected. They represent 13.74% of the faculty members and administrative technicians. 42.2% of the people who answered represent faculty members, whereas 57.8% were administrative technicians in education. Furthermore, if we consider the answers for the total of servants from each group, we found out that 200 are administrative technicians, which represents 18.07% of this career, while 146 are professors representing 10.35% of the total faculty members. Also, 52.3% are male, 24.9% act in the rectory, 31.2% at the Salvador campus, 8.1% in the metropolitan areas of Salvador, and 35.8% are in the interior of the state. Only 7.2% are less than 30 years old. Furthermore, 70.8% of the servants that participated in the research are up to 10 years in the institution. Finally, 7.8% hold management positions, 26.3% are in the remunerated functions while 65.9% do not have a trust function. As to the level of education of the respondents, only 5,9% hold a superior level.

The average of the servants ́ agreement regarding the presence or absence of definite factors influencing the PASS implementation, can be observed in Figure 1. The factors resources (11.2%), policy goals (11.8 %), administrative apparatus (12.9%), and performance of the actors (16.6%) are conditioning factors that, if positively present in the organization, according to Viana (1996) and Lima and D’Ascenzi (2013), strongly represent the capacity of implementation of a public policy.

These results can be the reason that the PASS did not receive the insertion that, in theory, it would need to reach the expected success when it was conceived. This insertion, which is not very significant, can be inferred by the averages of agreement of the variables: level of policy comprehension (27.3%), perception of police efficiency (21.9%) and level of servant participation in the actions associated to the PASS (35.9%) The factors political influence (40.4%), organizational culture (47.7%) and contextual variables (52.7%) obtained agreement percentages that were higher compared to the other factors mentioned above.However, their presence in the organization has an opposite effect upon the actions of the policy implementation as they represent elements of a negative feature upon the organizational actions. Especially the contextual variables (economic, social, political) according to table 1 and table 4 as seen by Wu et al. (2014) and Lima and D'Ascenzi (2013), have relevance in the implementation success of a public policy since they can materialize important support so that the organization is able to get the necessary resources for the execution of the actions of the policy institutionalization; Nevertheless, in the view of the respondents, the result shown in these factors reveals more the possibility of inhibiting than contributing to the actual PASS implementation.

Analyzing these results based on the perspective of state capacities that according to Grin (2021) is to identify structures whose absence or presence are critical to the state action,some observations can be made. The first one is about the consideration of Mendes andFerreira (2021) that states that all state competencies refer to the set of political and administrative abilities and competence in the formulation, allocation, implementation and achievement of results. The results put into evidence the low state capacity of this organization considering the numbers shown, especially in the variables resources, goals of public actions, organization of the administrative apparatus, the performance of the actors, and participation in the actions.

The variables goals in the political actions, organization of the administrative apparatus, the performance of the actors that can be associated with the technical-administrative dimension of Gomide and Pires (2014) and the analytical and operational capacities by Wu, Ramesh and Howlett (2015), are the lowest in the perception of the respondents. This may have been the cause of the low level of the PASS implementation in the Institute since the state capacities affect significantly the implementation and results of the policies (MENDES; FERREIRA, 2021), which can intuitively be seen from the low percentages of the variables of the PASS comprehension and efficiency.

Still from the point of view of state capacities, the variable “contextual variables” that evaluates the external influences upon the implementation project of the PASS, can also be the object of thought. So one could expect that the political capacities (GOMIDE; PIRES, 2017), political abilities, and competence and systematic category (WU, RAMESH; HOWLETT, 2015), when they exist, are capable of articulating external actors, especially resources and legitimacy in the PASS implementation. However, the results show a negative perception of this variable in the actions of the Institute, which can represent the low political and systematic capacity of the managers looking for external political support. This might also have been reflected in the perception of a lack of resources for the policy because “resources” showed a lower degree of agreement by the respondents.

If the policy had as a goal to improve the health of the public servant, it can be said that its goal has not been reached since the data show a climbing tendency of illnesses in the organization, confirmed by the increase of leaves of absence from work that occurred during the time of the analyses, especially in the years from 2016 to 2018 (table 1).

lthough the 2018 number of days of leave of absence from work was less than in 2016 and 2017, it is noted that these numbers were computed until August 2018. Considering the increase from 2016 to 2017, the tendency is that the numbers of 2018 are between 2016 and 2017 if they are not surpassing them. This finding can also be done in the answers given in the questionnaire for the item “my health improved after the implementation of the PASS” where only 4.9% signaled an improvement in the actions related to the PASS. However, when detecting that only 27.3% of the servants indicate knowing the PASS while the level of general participation was 35.9%, there seems to exist an important limiting factor in the policy of implementation.

With regard to regression, in the first place, the model used explains 52.04% (adjusted R- squared: 0.5204) of the dependent variable (perfection of the policy effectiveness). Secondly, of the variables used, only the factors contextual variations, cultural organization, organization of the administrative apparatus and resources are significant and explain the perception of effectiveness levels that would express the improvement of the servants ́ health. However, the variables comprehension of the policy, performance of the actors, policy goals, participation of the servants, and political influence were not significant and indicate that these variables do not affect the perception of the servants ́ effectiveness. Future studies can show with more accuracy the causes of these variables not being significant, since with the stated theoretical discussion it was formerly expected that they would be significant. All the variables presented the awaited indications of the estimators including the variable political influence with a negative sign since its existence and increase would provoke less effectiveness in the policy results as it refers to aspects of opposition and limitation in the actions of the public policy implementation being studied. Furthermore, of the control variables used, only gender, time of service in the Institute and level of hierarchy (management position) showed to be significant according to table 2.

Regarding the variable gender, the results indicate that men, more than women, perceive an improvement in the health conditions at the workplace, Also the time of service has a stimulator with a negative sign, this means that the more time the servant has dedicated to the Institution, the worse the perception about health conditions becomes. Moreover, those people who hold management positions are able to see improvements in the health conditions since the start of the PASS implementation more than those individuals in positions of a lower hierarchical level. This seems natural because members of management positions are more prone to assessing positively the actions within the organization. Adopting, as an example, the variable level of policy comprehension, it can be observed that the degree of agreement by the managers is almost twice the average of the agreement of other servants as can be seen in table 3.

What calls the attention in the table above is the big difference between the perception of the servants in general and the managers in their answers to the item “I am fully aware of the PASS content concerning promotion, prevention and follow-up of health”. This reinforces the tendency of the managers to be more benevolent in their analyses towards the organization than its other members.

In the result of regression, the variable organization of the administrative apparatus was significant and positive, indicating that the greater the existence of elements that compose this variable, and especially the performance of the managers, the greater the perception of policy effectiveness. These results point to the importance of this variable by presenting, among all those that were significant, the highest coefficient (0.39475), indicating that it is the onevariable that most influences its implementation. Thus, an institutional action to accelerate the implementation of the PASS should focus efforts on the administrative apparatus due to the fact that, according to Viana (1996) and Lima and D ́Ascenzi (2013), it, among other elements, contemplates availability, size, vitality, and competence of human resources, levelsof hierarchy and performance of the bureaucrats, can put into jeopardy the success of the implementation of any public policy (LIPSKY, 2010; WU et al.,2014)

From the point of view of state capacities, it can be said that the result is adequate. The organization of the administrative apparatus would be associated with the technical-administrative dimension of Gomide and Pires (2014) and the analytical and operational skills of Wu, Ramesh and Howlett (2015) regarding the capacities of managers to materialize actions within organizations that make viable the implementation and effectiveness of public policy (MENDES; FERREIRA, 2021; GRIN, 2021).

Another discussion that deserves importance in this study and that may have affected the process of implementing the PASS at IFBA is the cultural aspect. In percentage terms, 47.7% of the respondents considered that the Institute's culture affects the implementation of the policy. In the item organizational culture, elements such as administrative discontinuity and characteristics of implementing agencies stand out as influencers of actions to implement a public policy, since culture shapes the behavior of subjects in an organization (SILVA, 2011; STRESE et al., 2016). The respondents consider that culture is an important impediment to changing behaviors towards more preventive health since they do not get involved with the actions promoted by the organization and which aim to improve health conditions, such as lectures, courses, etc. And, in fact, the results reveal that there is a low participation of public servants, as only 24.9% declare that they participate in the actions promoted by the Institute. Also, in this variable, the reduced participation of individuals can as well signal low state capacity in the technical-administrative dimension (GOMIDE; PIRES, 2014), since managers would lack the ability to articulate the various resources and people around the policy.

In the regression, the results indicate a significant and negative relationship between the perception of the policy's effectiveness and the cultural aspects that involve the Institute. The negative coefficient (-0.12364) indicates that the more those cultural aspects measured in the questionnaire are present in the organization, the less improvement is seen in health conditions within it.

An important reflection was the perception of the lack of actions that took into account the organizational culture during the implementation of the PASS. Considering that organizational culture can affect this process (BARRET, 2004; DEGROFF; CARGO, 2009; ELIAS,2017), it was observed that some aspects of the Institute's culture, such as administrative discontinuity and the culture of cure rather than prevention, are incompatible with the essence of the policy itself. This incompatibility may have hampered the implementation of the PASS, as highlighted by Nogueira (2006) and Lima and D’Ascenzi (2013).

The context factor or contextual variables, in which the organization is inserted, also proved to be relevant as an intervening factor in the implementation of the PASS. Following the conception of Wu et al (2014), Viana (1996), Lopes, Amaral and Caldas (2008) and Lima and D'Ascenzi (2013), factors that are external to the organization and that are not always under its control affect its actions in a significant way. The regression shows that the contextual variable has a positive statistical significance over the dependent variable, meaning that whenthose factors indicated as belonging to the context external to the organization increase, the perception of policy effectiveness varies positively.

From the perspective of state capabilities, the variable “contextual variables” also refers to the organization's ability to seek political support and legitimacy for policy implementation actions. In this sense, from the point of view of Gomide and Pires (2014), it could be said that the organization lacks the political dimension of state capacities, which signals the bureaucrat's capacity to interact with other actors in society and other organizations. This same political capacity, in the view of Wu, Ramesh and Howlett (2015), would be responsible for the search for trust and legitimacy on the part of external actors and society, who could obtain the necessary support for the effective implementation and search for policy results (GRIN, 2021; MENDES; FERREIRA, 2021).

Among the items in this factor (Table 4), the perception of 73.1% of the respondents that changes in the political environment affect the Institute's processes stands out. This result converges with the view of Wu et al (2014) when they consider that the degree of political stability and the degree of stability of the political and economic environment reveal themselves as potential difficulties in the implementation of a public policy. In this sense, it also highlights to what Barret (2004), DeGroff and Cargo (2009) and Proctor et al. (2013) draw their attention, considering that the political dimension of the external context reveals itself as the most impacting in the process of implementing a public policy.

These findings reinforce what Viana (1996), Lopes, Amaral and Caldas (2008) and Lima and D'Ascenzi (2013) consider as contextual factors, involving economicaspects, which refer to transfers of financial resources,social, which refers to public opinion, which in this case, 43.9% of respondents believe is important and influences the functioning of the Institute's activities, and political opinion, which refers to changes in government and the positions of support or opposition party groups.

The factor resources, which encompasses financial, human, material and technological resources, was also statistically significant and positive, indicating that insofar asindividuals note that there are more resources allocated to the activities of the Institute in conducting the PASS, one can see a greater positive effect on them. This factor had the lowest level of agreement among all those analyzed in this study. Only 11.2% of the respondents consider that there exist sufficient and adequate resources for the IFBA to carry out the activities associated with the PASS. From the point of view of the possibility of more effective implementation, the study suggests that there is a difficulty in the implementation process of the policy, once the servants understand that there are not enough resources for such a process.

Among the elements that make up the state capacity, the lack of resources may reflect the shortage of other state capacities, such as political capacity (GOMIDE; PIRES, 2014) or systemic and political capacity (WU; RAMESH; HOWLETT, 2015), since the resources of public organizations derive from the public budget that can be increased by political articulations through amendments or other resources arising from partnerships with other organizations. As the organization presented low political and technical-administrative capacities (GOMIDE; PIRES, 2014) and low capacities presented by Wu, Ramesh and Howlett (2015), the results found both in the regression and in the percentages in the respondents' perception may reflect the perceptions of policy effectiveness.

These results may mean greater difficulty in conducting the policy, considering that, according to Lipsky (2010), resources, together with the objectives of actions and the performance of the actors, are among the main factors of success or failure in the implementation of a public policy. Finally, in the regression result, the coefficient of the variable resources is 0.21886, being the second-highest among the significant variables. It comes only behind the variable organization of the administrative apparatus (0.39475). This means that it is the most impacting factor for the success of the policy followed by the variable resources that includes budget allocation, people, material and technologies.

5 Final Considerations

The goal of this research was to analyze, in the public servants' perception, the factors that influence the effectiveness of the implementation of the IFBA policy of Attention to the Health and Safety at Work of the Federal Public Servants. An initial finding of this research is the identification of the low institutionalization and implementation of the policy at the Institute. Servants know little about PASS and its goals. Also, in general lines, the organization presentedlow state capacity, measured by the variables included in the research, which may also have reflected in the percentages of each variable as to the low perception of the policy's effectiveness.

An important observation of this study is what the quantitative approach revealed. As the literature discusses, an array of factors should affect the process of implementing a public policy. However, the results of this research, through regression, indicated that only the factors Resources, Organizational Culture, Organization of the administrative apparatus, and Contextual variables affected the perception of policy effectiveness at IFBA. Compared to what the theoretical framework presents with regard to the intervening elements that determine the progress of public policies, the dimensions of policy comprehension, the performance of the actors, policy goals, and political influence did not affect the perception of the effectiveness of the policy. In this perspective, the results indicate the need for greater reflection from the public managers on the factors that proved to be significant in influencing the behavior and performance of the PASS implementation at the Institute.

As a contribution, this study can support the elaboration and planning of programs, projects and actions guided by the information collected in the field research, guide the implementing servants in order to optimize the implementation process and also help the Institute to promulgate the policy among the servants, as well as alert them about the importance of their active participation in its construction. As far as the academic sphere is concerned, this research expands the published studies that explore the PASS in the context of the Federal Institutions of Education, especially with regard to the phase of the implementation of the policy.

Another relevant contribution of this work is the use of the quantitative approach, with the use of multiple regression, to test the factors that influence the process of implementingsince most of the works that discuss implementation processes have a qualitative approach. Therefore, this study can be important in the sense that public managers, in their processes of implementing a public policy, focus their actions on factors that were presented as significant and that can optimize the possibilities of effectiveness in the execution of a public policy.

An important limitation of this study was the fact that it used efficiency based on the perception of the respondents. This limits the conclusions and analyses carried out. Furthermore, future studies may expand quantitative research in different areas and also in other policies to verify the influence of factors in the implementation process with the intent to map new impacting elements in the implementation process. In addition, further research can expand the analysis of state capacities of institutions of education, as they can represent an important source of knowledge in this subject.

IR