Labour integration of young people at risk of social exclusion through the development of key competencies

Abstract This study examines the key competencies that young people at risk of exclusion must develop to achieve integration into the labour market, and assesses their level of integration after their participation in the programme proposed. A programme for developing Social Skills is undertaken for three weeks. A methodology is presented to that end, based on constructing three synthetic indicators which aggregate and measure the different components proposed: Appearance, Confidence, Attitude, and Organisation and Planning. These three models measure the aspects before the programme and the aspects after the programme and the improvements are analysed. The results of the study are based on a sample of 373 young people between 18 and 30 belonging to vulnerable groups. An analysis of the results shows that the competencies that underwent the most significant changes were Attitude and Organisation and Planning. Control variables are applied (gender, age, experience and level of education). The results obtained in the model studying the aftermath of the programme show that the competence of Confidence influences the student's recruitment and incorporation into employment.

leaving teachers free to create their own content, are committed to providing free education, and encourage commitment to their culture and patriotism (Informe PISA, 2019). However, Spain obtains its worst results in science, and remains below the OECD average.
Focusing on Spain, the OECD conducted an educational thematic review in 2005 based on Equity in Education, which highlighted shortcomings in studies on educational equality between groups which had economic difficulties and those which did not. This problem became more acute over the years.
It has become an essential subject for discussion in Spain in recent years (Caparr os et al., 2010;Parrilla, 2008). The last reports of the Programme for International Student Assessment (PISA) compiled by the Organisation for Economic Co-operation and Development (OECD) have identified the critical situation in the Spanish education system. In 2009, the European Commission pointed out that the school drop-out rate should have fallen by 14.5% (in Spain), but not only had it not fallen, but instead it had increased to 31.9% (Commission of the European Communities, 2009).
The concept of 'basic competencies' has been introduced in both areas of Basic Education (Primary and Secondary) for this reason. The concept of competencies began to be discussed in the late 1960s and early 1970s (Spencer, 1992). The concept has gradually evolved, and the current understanding of it is different. The knowledge acquired in school alone is not enough (Boer et al., 2011;Chiner & Cardona, 2013). It is therefore necessary to develop other basic competencies, related to the individual's physical appearance, ability to establish relationships, and their awareness of their abilities and limitations, among others, in order to adapt to complex environments and integrate into them (Bol ıvar, 2008;Mitrani et al., 1992;Morris et al., 2007).
Training programmes combining theoretical and practical components have been created to improve these skills. These programmes not only measure their technical knowledge acquired in final examination, but also observe the progress made and skills acquired by the students throughout the programme, using the tools provided to them. It is at this point when the term key competencies arose. These included teamwork, collaboration, initiative, communication skills, self-confidence and active listening (Edwards-Schachter et al., 2015;N uñez et al., 2018;Ortega-Goodspeed, 2016), in addition to the importance of the improvement and integration in the social and work environment of factors including physical appearance, punctuality, good behaviour in class, etc. (Edwards-Schachter et al., 2009;Morris et al., 2007). All these competenciesboth attitudes and skillswill be decisive factors in any individual's employability. A specific aim is to investigate the profile of vulnerable groups in the proposed programme in greater depth and in more detail.
There are several models which study how to measure the quality of life, and which determine that a group is vulnerable (Ecclestone, 2012;Ecclestone & Hayes, 2009;H€ ausermann et al., 2015). The Schalock (2009) and Verdugo (2009) model includes eight dimensions, which are: emotional well-being, personal development, interpersonal relationships, social inclusion, physical well-being, self-determination, material well-being and rights. Schalock and Verdugo (2006) propose a model that identifies where this group requires the most support. The specific support areas are: human development, education and education, home life, community life, employment, health and safety, behavioural, social, protection and defence.
Numerous debates lead to social concern, not only because they affect the lack of understanding in the education system's programmes, but also because they are affected by the country's politics and legislation (Casal et al., 2006). Spain has undergone numerous political and economic changes in recent years. J. Casal et al. (2006) argue that both the biographical dimension and the political structure are factors which determine how adolescence unfolds, and integration in the social sphere is affected as a result.
The country's lack of stability means that it is impossible to finance programmes and this leads to the current situation in the labour market (Abi etar-L opez et al., 2017). This enhances the difficulties involved in joining the labour market for the most disadvantaged groups and those at risk of social exclusion, which have fewer opportunities to acquire employability skills, i.e. academic training and qualifications. This leads them to continue becoming poorer and less employable (Danziger & Ratner, 2010;Settersten & Ray, 2010;Terzian et al., 2009;Weinberg, 2004). According to the VIII AROPE (At Risk Of Poverty and/or Exclusion) Indicator Annual Monitoring Report there were a total of 12,338,187 vulnerable people in Spain in 2017. In other words, 26.6% of the Spanish population was at risk of poverty and social exclusion. It also notes that there were 6.4 million women compared to 5.9 million men at risk of poverty and social exclusion in 2017 (European Anti-Poverty Network (EAPN), 2017). This study has focused on a sample of young people, since they are one of the most vulnerable groups (Gaviria et al., 1995;Glewwe & Hall, 1998;Luis Vives Foundation, 2007;Subirats, 2004).
All these factors mean that undertaking this study and developing a programme that promotes labour integration among vulnerable groups at risk of exclusion is a subject of interest. This study focuses specifically on disadvantaged population groups at high social risk. This vulnerability or risk plays an important role in the social sphere. According to Dom ınguez Lostal o (1999), there is a failure in group and community containment, due to the inability to guarantee access to essential human rights. These problems may be caused by bullying, illness, lack of attention at school, and family conflicts, among many other reasons. This is supported by Whitson and Keller (2004), Morris et al. (2007), Einarsen et al. (1994) and Balducci et al. (2012), among others, in their previous studies, who examine various factors (physical, behaviour, emotional management, demographic, level of education, economic and social) that make this group vulnerable. These situations lead to students dropping out from school, and failing to continue with their academic training. Two types of socialisation are therefore constructedindividuals who obtain academic qualifications, get a job and gain self-confidence; and on the other hand, those who fail at school, do not obtain an education, have greater difficulties in being employed by a company, and who are therefore affected in terms of their own self-esteem (Rossini, 2003).
After this introduction to the Spanish educational system, the problems for young people in vulnerable groups, and their difficulties in finding work, in the first section, this paper reviews the literature on the training programmes and the skills identified as necessary for development on the programmes for subsequent incorporation into the labour market in the second section. The methodology used, the indicators developed and the statistical techniques implemented are presented in the third section. The main results obtained in the study are discussed in the fourth section. Finally, a section containing the main conclusions of the study and the future research lines is included.

Review of the literature
There are many studies analysing the educational model in countries of origin and how it affects the success or otherwise of students (Dheer, 2017;Kangasharju & Pekkala, 2002;Padilla-Mel endez et al., 2014;Reilly, 2013;Wan, 2017). According to Morales (2006), the programmes are effective when the primary objective is for young people to participate in an education that trains them, both in educational terms and in the attainment of basic skills, and which fosters their processes of transition and integration towards adult, social, educational and productive life (Morales, 2006).
Other studies discuss the lack of appreciation of the Spanish education system, its improvements and restructuring, and the programmes that seek integration for students in Spain (Escudero & Mart ınez, 2012;Gonz alez, 2015;Gonz alez & Moreno, 2013;L opez & Palomares, 2012;Mar ın et al., 2013). In Spain, the areas with the highest levels of economic activity, such as the Community of Madrid, the Basque Country and the Valencian Community, are those that are most severely affected by major economic crises. Training that provides enhanced educational and employment alternatives for young people is therefore an essential factor in ensuring their integration into society and employment (Marhuenda & Navas, 2004).
The study conducted by Miriam Abi etar-L opez et al. (2017) uses an analysis of the progress made by training programmes over the last ten years as a framework of reference. In addition to this study, we aim to include our own programme that develops social skills to enable vulnerable groups to join the labour market. The programmes needed are those that involve trainers/teachers with high levels of psychopedagogical competencies, in terms how to work with the content and adapt them to the group, and transfer them to the labour market, in addition to motivating students to continue studying and improving their self-esteem (Jurado et al., 2015). Furthermore, when the students are selected, it is important that the ideal student is someone who has some capacity for work and knows how to use it in the time spent in the classroom or on external activities (Abi etar-L opez et al., 2017). According to M ıriam Abi etar-L opez et al. (2017): 'it is clear that if they work in school terms, even if it is a little, it is because they have the capacity for effort and are able to focus their future in a different direction than they have been doing so far. The ability to change is closely related to the possibility of demonstrating the maturity necessary to face the outside world'.
The literature review highlighted other training programmes that include providing help and facilities for integrating vulnerable groups into the world of work (Jurado and Olmos, 2010;Jurado et al., 2015;Llinares et al., 2012;Olmos et al., 2011;Weller, 2009). There is also extensive literature on studies related to applying training programmes for the development of social skills among young people at risk of exclusion (Durlak et al., 2011;Jovarini et al., 2018;Leme et al., 2016;Pereira-Guizzo et al., 2018). As regards the key competencies to be developed in programmes for young people from vulnerable groups, and which will subsequently facilitate their incorporation into the world of work, there is a study that uses a self-perception questionnaire of basic and labour competencies (AUTOCOM). Olmos et al. (2011) mentions the competencies that must be emphasised in the proposed training programme on employability for young people belonging to vulnerable groups. This study reports that teachers must pay greater attention to the competencies in the 'health' dimension (physical, emotional, mental and social health) and 'adaptive behaviour' dimension (autonomy and initiative) (Olmos et al., 2011). However, few studies measure the percentage of employability after the social skills programme has been completed. According to D. Wayne Osgood, E. Michael Foster and Mark E. Courtney (2010), there is a need to reinforce all programmes for young people in the transition to adulthood, there needs to be an improvement in the systems of care for adolescents, and more aid must be provided to facilitate access to employment once they are adults. Nevertheless, awareness is increasing among governments as regards improving support for vulnerable young adults (Osgood et al., 2010).
In 2016, a study by Katz, C. M. and Artiga, R. E. examined seven programmes carried out in Central America over the previous five years, in which the main objective was to improve employability among young people, albeit from different perspectives in each programme. The study concluded that the major competencies are: the importance of planning based on the real needs of the labour market, and the involvement and cooperation of potential employers (Katz & Artiga, 2016).
An increasing number of changes are taking place, which require individuals to adapt more quickly and flexibly. An important factor mentioned by Katz and Artiga (2016) is adaptation to the real needs of the labour market. Future workers must be prepared and trained according to the demands of the market (Cuello, 2006). More than ever, the incorporation of new technologies and analytical profiles are today subject to the highest levels of demand (Jarque Puntada, 2018). The breakneck speed of technological development is a key factor influencing the education system (Gallego Arrufat et al., 2010). Despite the country's high rate of unemployment, Spain has problems filling job vacancies, but the country's lack of stability means that it is impossible to finance programmes, and this has created the current situation in the labour market.

Methodology
In order to study the influence of the various factors analysed in the survey on the result pursued by the training programme (the employability of the vulnerable individual), various probit models were used to estimate the influence of the individual's characteristics and the skills and attitudes on the final result. This estimate was made from three points of view. First, how the initial provisions and conditions (those before the training course) influence the final result was analysed. Second, the influence of the levels achieved in the employment training programme is examined. Finally, the improvement experienced in the different skills and competencies by the participants in the programme and its influence on obtaining employment are studied.
The methodology proposed in the present study is unique and original, and it uses the studies carried out by Katz and Artiga (2016), the performance appraisal of the USAID programme (United States Agency for International Development) by Cambronero et al. (2012) and consequently the various studies carried out by USAID (2013aUSAID ( , 2013bUSAID ( , 2013cUSAID ( , 2013d as its benchmarks. This analysis was therefore performed using the data from a survey conducted by the trainers on a group of young people participating in a social skills development programme. These students belong to socially vulnerable groups at risk of exclusion: i.e. in a situation of difficulty, risk and/or social conflicts. A sample of 373 participants from vulnerable groups was used, with a mean age of between 16 and 30 years old. The sample participated in the social integration plan of various foundations, and were subject to follow-up to confirm their employment or otherwise within six (6) months. The selection processes were similar for all students in the programme, and the inclusion criteria were: 1) Participation in a foundation's employment and social integration plan, 2) Aged between 16 and 30 years old, 3) Situation of difficulty, risk and/or social conflict, 4) Employment situation: unemployed and 5) Enthusiasm, motivation and a willingness to learn and work. Meanwhile, the exclusion criteria were: 1) Failure to participate in a foundation's employment and social integration plan, 2) Age not in the range between 16 and 30 years old, 3) No situation of difficulty, risk and/or social conflicts, 4) Employment situation: employed and 5) Lack of enthusiasm, motivation and unwillingness to learn and work. In general, all the participants had similar options, offered partly by the institutions that organised the course, which had job banks and offered employment opportunities to all the members of the group using the same information channels.
The duration of each programme was three (3) weeks, involving five (5) hours each day spent in face-to-face training sessions. Each group was composed of between 15 and 20 participants. The programme was based on paying particular attention to their individual needs in order to achieve those objectives, and subsequently their integration in the labour market. Table 1 shows the variables that were measured for each competence: Physical Appearance was assessed in terms of care of the individual's image rather than the attribute of beauty, and whether the student takes care to maintain the standards of decorum and cleanliness suitable for the cultural environment with which he/she is engaged was analysed. It is also important to interact and be able to maintain good relations with their peers, making good use of his/her communication skills, and maintain the minimum levels of courtesy (good morning, goodbye).
It is also essential to observe their degree of action and effectiveness in the activities carried out, when working under time pressure and dealing with disagreement and adversity. It is important that there is no sign of the effects of tiredness on the loss of control of behaviour, and that no efforts are made to blame other people, but instead they seek to learn from their mistakes.
At the same time, it is vital that the students maintain a positive attitude in any situation, they are tireless when faced with discouragement, prevent and deal with feelings of frustration in the group, and take the initiative to restore the vision. Where he/she copes with ease in situations of stress and does not harm the good climate or the loss of initiative. Engages with the team, plays and is able to brighten up the day for people around him/her, without losing sight of the continuous improvement of the purpose.
Finally, the ability to establish objectives to be achieved and priorities, knowing how to identify what is important in urgent matters, efficiently selecting and distributing the tasks, time and resources required to achieve it, being able to obtain the maximum possible results in the time established and with the resources available. In this study, all the activities in the classroom are covered and the theoretical examinations are passed. The programme implemented is shown in Table 2 below: As noted above, different competencies were worked on throughout the programme and could be grouped into four general dimensions, which were as follows:  Physical Appearance, Self-confidence, Attitude, Organisation and Planning. The aspects were measured with four indicators that could adopt five values on an ordinal scale from 1 to 3, where 1 indicates the lowest attainment of the level or attitude of the item evaluated, and 3 the highest value, with intermediate values possible in fractions of 0.5. An indicator for each aspect was obtained by a linear combination of the scores obtained in these items. Accordingly, if different weights are also added to each aspect, each indicator can adopt a wide range of values. The structure of each indicator composed of the four dimensions considered would be as follows: Where I j i the score for item j in dimension i, where the subscript i ¼ 1, 2, 3, 4 indicates the different aspects or capacities measured (Appearance, Self-confidence, Attitude and Organisation and Planning) as basic dimensions, and where the weights a ij attribute the importance of each item in each indicator composed of each aspect considered. The weighting will be equal to one for each item in this case. The sum of the four scores is therefore used as an indicator of each dimension. Various calculations and estimates of the probit models with other weights were also performed, and similar results were obtained, which prove the robustness of the approach used.
The indicators of each dimension were therefore the basic explanatory variables for analysing their influence on employment. As noted above, three models are estimated: one in which the indicators are obtained for the scores before the training, a second for the indicators obtained after the training process, and a third model that explains employment in terms of the improvements experienced in each dimension after receiving the training.
Meanwhile, in order to complete the explanatory variables, a further set of variables included different characteristics of interest about each of the people interviewed, such as their age, work experience, level of education and the gender. This information was grouped in a vector of control variables and incorporated into the explanatory variables of the probit models.
The dependent variable of the three probit models is the probability of obtaining employment. This is a latent, unobserved variable that must cross a threshold for the dependent variable in order to take the value of one or zero, given that by nature it will depend on the values taken by the specified control variables such as sociodemographic and socioeconomic variables and the various measurement indicators represented by a linear combination.
The model would therefore be expressed as follows: Where: b 0 is the independent term of the model. CV: a vector of control variables that contains three exogenous variables that influence the dependent variable (experience, age, gender and level of education).
These models cannot be estimated by the ordinary least squares method, and as such the maximum likelihood estimate is used, making assumptions about the distribution of errors. In this case we assume that the errors are considered to be normally distributed.
Meanwhile, the interpretation of the coefficients and slopes of the probit models does not match the marginal effect on the dependent variable of a change in one of the explanatory variables; however, these estimates are used to quantify the marginal effects of the explanatory variables on the probability of finding employment. Accordingly, the results of the coefficients and the estimation of the marginal effects on the means of the independent variables will be presented in the results tables, as these effects vary according to the different values of the explanatory variables.
The goodness of fit is analysed using various measures and tests. First, the value of the Neperian logarithm of the likelihood function will be considered, although this value acquires significance when compared to similar models. The results of the Wald tests and the likelihood ratio will also be provided in all cases.
Another measure of goodness of fit is the percentage of correct predictions, which compares the results for the sample with those predicted by the model for each observation. We use the usual construction rule, choosing a threshold of 50% of the estimated probability to consider the prediction correct, i.e. the prediction is considered correct when it is greater than 0.5 and the observed value is one, or when the observed value is 0 and the prediction is less than 0.5.
To measure the goodness of fit, McFadden's pseudo-R 2 values are also provided. This measures the degree of improvement in the fit of the model when comparing the logarithm of the likelihood function of a model without explanatory variables, which contains only the constant and the logarithm of the likelihood function of the estimated model. Instead of variances, this coefficient compares the logarithm of the naive model and the logarithm of the likelihood function of the complete model, taking values between 0 and 1.

Results
After collecting the data from the pre and post questionnaires, the descriptive statistical analysis of the variables that comprise the questionnaires and their comparison was carried out. Table 3 shows that the variable with the lowest average at the beginning of the training was Organisation and Planning and Attitude. However, the variable that had improved the most at the end of the programme was Self-confidence, followed by Attitude and Organisation and Planning. Within the Physical Appearance competence, the variables that had the most positive changes were: v2. 'Takes care to maintain the standards of decorum and cleanliness suitable for the cultural environment with which he/she is engaged' and v3. 'Interacts and maintains good relationships with others by making good use of his/her communication skills (does not swear, does not insult)'. In the Self-confidence competence, the variables that improved the most were: v6. 'Deals with disagreement and adversity' and v7.'Does not show tiredness affecting a loss of control of behaviour'. The variables with the greatest changes in the Attitude competence were: v10. 'Takes the initiative to recover the vision' and v11. 'Copes with ease in situations of stress and does not harm the good climate or the loss of initiative. Finally, for the Organisation and Planning competence, all the variables improved to a large extent and in a similar way, and the following were subject to a slightly larger difference than the other variables: v13. 'Is able to set goals to achieve, as well as priorities'.
The estimates of the coefficients and the estimated values of the marginal effects of the three models suggest first, that the dimensions selected as elements for measurement and training objectives are important when finding employment. Indeed, all the dimensions, except for Self-confidence in the estimated model with values of indicators prior to the programme, have significant effects on the probability of finding employment, and are also positive and have values between 12 and 28% (see Table 4).
In comparative terms, the marginal effects of the four dimensions are greater for the values of the indicators after carrying out the programme, with the sole exception of the Attitude Indicator, which has a greater effect on the probability of finding employment in the indicator values before the course.
If the average for each indicator is taken as the starting point, the marginal effect of an additional point in the score for Physical Appearance entails an increase of the probability of finding employment of 12.5% before the programme and 15.6% after the programme. An additional point in the score for Organisation and Planning represents an increase of 16.2% before the programme and 27.7% after the programme. The marginal effect of Attitude falls from 24.7 to 18.3% after the training received, i.e. with the increase in the importance of the two previous factors, its marginal effect is reduced. Finally, the Confidence Indicator goes from not being significant to having a marginal effect on the probability of finding employment of 24.3%. The model that includes the improvements experienced in each dimension as explanatory variables has very balanced average marginal effects for each dimension. As a result, an additional point in the improvement in the dimension related to Physical Appearance entails an increase of 7.5% in the probability of finding employment. This increase is 6.3% for the Confidence Indicator, 5.5% for the Attitude Indicator and 6.4% for the Organisation and Planning Indicator. Due to their significance and the magnitude of the effect, the dimension of Appearance and the dimension related to Organisation and Planning are somewhat more influential. This means that in addition to the influence of these aspects on the levels obtained before and after the training, the programme had a positive effect on its participants obtaining employment.
It is also interesting to note the influence of the control variables on the probability of obtaining employment. For this group of vulnerable people, having previous experience therefore seems to be a factor that significantly reduces the probability of obtaining employment in all the estimated models, and specifically by 53.2% and 48.2% before and after the training, and by 22.9% in the case of improvements. Having an education is not a variable that is significant, other than for the first model and also in a negative way. Being a man also reduces the probability of finding employment, and age is a factor that also reduces this probability. Unexpected effects therefore occur for both experience, the level of education and being a man which exert a reducing effect on the probability of finding employment. In studies based on samples from large surveys, these variables usually lead to increases in the probability of finding employment. The estimated models have an adequate specification according to the results of the Wald test. Meanwhile, testing the likelihood ratio also indicates a significant improvement in the models estimated using the model with an intercept. The values of McFadden's pseudo-R2 show a good performance of the first and second models and of the estimated model for improvements to a lesser extent. The percentage of correctly classified individuals is very high for the first two models, exceeding 90%, and is acceptable for the third model, which correctly classifies almost 70% of the individuals. As a result, in general terms, the models have an adequate fit, especially the estimated model for the final values of the indicators.

Conclusions
This study analysed the effects of the implementation of a Social Skills programme on young people belonging to vulnerable groups in order to help them find work after completing it. This programme worked on the development of various key competencies, which were Physical Appearance, Self-Confidence, Attitude and Organisation and Planning. These dimensions were defined and subsequently analysed individually, beforehand and afterwards, using three synthetic indicators.
At the beginning of the programme, the dimensions that had a deficit, and thereby involved more work within the programme, were the dimensions of Organisation and Planning and Attitude. In general, the participants' profile involved a lack of motivation at the beginning of the course, as they were lost, did not know what to do with their lives, and showed no interest in studying. After completing the programme, the dimensions where they showed improvements were Self-confidence, Attitude and Organisation and Planning. In other words, they were able to reinforce the dimensions that had the most limitations at the beginning. This conclusion is closely related to the study conducted by Katz and Artiga (2016), mentioned in the literature review section. This author concluded that the main competencies that should be reinforced and worked on in this type of programme were planning and the involvement and collaboration of the participants.
It was the variable that underwent the most positive changes in the Physical Appearance dimension. This applies above all to awareness of the importance of attending job interviews neatly turned out, according to the circumstances and respecting the basic rules of politeness. For the Confidence and Organisation and Planning dimension, there were major improvements in the presentation of their ideas and persevering in order to achieve their objectives.
The three models proposed suggest that the dimensions selected and defined are important for finding employment, as noted by authors including Edwards-Schachter et al. (2015); Ortega-Goodspeed (2016) andN uñez et al. (2018), among others. Selfconfidence at the beginning of the programme was not significant, but it did appear once the programme was completed. In other words, the students learn how to sell themselves to the labour market, feeling more confident about what they want. According to the results, the most important dimensions with the most direct effects when seeking employment are Organisation and Planning and Physical Appearance, for which the chances of finding employment increased by 6.4% (Organisation and planning) and 7.5% (Physical Appearance). These variables are related to preparing for the interview and the performance during it. Attending the interview suitably dressed for the position being applied for, doing research about the work done by the company and preparing for the most frequently asked questions are key points for gaining the interviewer's attention (Bolino et al., 2016;Proost et al., 2010;Tews et al., 2018). The Self-Confidence dimension also increases the probability of finding a job by 6.3%, and Attitude does so by 5.5%.
Unexpected results contradicting many studies were also obtained which for this group were having had previous experience, being a man and being older has negative effects when it comes to finding employment. For this reason, a planned future project involves carrying out a broader analysis of these variables and seeking answers, since as mentioned above, there is a higher unemployment rate among women belonging to vulnerable groups than among their male counterparts (VIII AROPE Indicator Annual Monitoring Report in Spain, 2017).
Finally, the recruitment of the participants after training was monitored, and 60% of the sample obtained their first job within six (6) months. Most of the participants obtained jobs in the logistics sector, as warehouse operators and perishable food handlers. These were followed by sales staff at petrol stations and retail stores, and finally, customer service positions such as cashiers. The duration of the employment was temporary and subject to periodic renewal, in which those who were employed and and those who were not continue to be monitored. The participants benefited from their participation in the programme, as did the recruitment companies, who reported that the selection process and costs related to the recruitment and training of new employees were streamlined.

Disclosure statement
No potential conflict of interest was reported by the authors.