The moderating role of supervisor support in the effect of perceived overqualification on workplace loneliness

The purpose of this research was to determine whether supervisor support (SS) plays a moderating role in the perception of overqualification (POQ) on workplace loneliness (WL) among music teachers in Turkey. The research was based on the assumption that loneliness, which is a negative emotional state and expresses employees’ inability to establish a healthy relationship within the organisation, may arise from music teachers’ perceived overqualification and can be shaped with the support of this relationship. With this study we discovered that perceived overqualification has a definite and significant impact on WL. It was also discovered that supervisor support moderated the relationship between perceived overqualification and WL. Consequently, although music teachers felt overqualified, they did not feel lonely at work if their supervisor provided strong support.


Introduction
Music, which is a part of human life, is one of the most essential components of art.Music, as a social and cultural feature, is entrenched in the context of social relations (Sahin, 2008:277).Under these conditions, music contributes to social integration and builds social identity by giving meaning to the social structure (Feld &Fox, 1994:35;Sahin, 2008:277).In particular, music education has significant effects on communication and social abilities (Demirtaş & Üstün, 2023:185;Sakarkaya, Egilmez & Engur, 2019:36).At the same time, music education provides people with the ability to understand, use, and evaluate their emotions (Erdoğdu, 2008).It is known that musical expressions, emotions, and social relationships develop through music education (Piji Kucuk, 2011).Research has shown that individuals who receive professional music education are able to think multidimensionally and have better communication skills than individuals who do not receive this education.The fact that people with music training frequently go on stage and perform has a positive effect on their emotional intelligence (Turkoglu & Gursen Otacioglu, 2020:402-403).This multi-faceted development that takes place during training can also have a positive impact on their self-confidence and self-efficacy in relation to their teaching ability (Akıncı, Sazak Pınar & Alpagut, 2018:1234).
In addition to the qualifications of general educators, music teachers should also possess the requirements, personal characteristics and competencies related to music and music education.In this sense, the qualifications of music teachers do not only involve the role of teaching, but also the personal characteristics of the teacher (Akbulut, 2006;Biasutti & Concina, 2018:264).With their pedagogical and musical skills, imagination, emotional intelligence, and communication skills, music teachers are usually aware of these abilities (Akbulut, 2006;Akıncı et al., 2018;Turkoglu & Gursen Otacioglu, 2020).Because of this, music teachers may develop the belief that they have more knowledge, skills, and experience than their job requires.Ozkanan (2018:665) states that this developed belief leads the individual to have more expectations over time, and individuals whose expectations are not met may develop some negative attitudes and behaviour towards the job, the company, and colleagues.It is hypothesised that music teachers' perceptions of being overqualified may disrupt their social relationships within the organisation, over time leading to workplace loneliness (WL).
The workplace is a social and psychological ecosystem to which employees devote a substantial amount of time.The social and emotional well-being of teachers, who are the basic building blocks of educational institutions, are of great importance.The social and emotional satisfaction of these important individuals, whose job it is to educate people, is an important key to the proper utilisation of their potential.At this point, it is presumed that music teachers' perception of being overqualified will, over time, turn into loneliness in the workplace.The study was based on the hypothesis that the relationship between these two variables can be influenced by the presence of a supportive supervisor.It is well established that SS is important in satisfying the psychological and sociological needs of individuals.Research specifically demonstrates that SS decreases WL (Eroglu, 2018;Oguz & Kalkan, 2014;Wright, 2005;Yilmaz, 2011).In this sense, it is hypothesised that WL, which is likely to be associated with perceptions of overqualification, may be reduced in individuals with high SS in educational settings.Through the literature search we did not locate any research on music teachers' perceptions of overqualification, WL, and SS.With this in mind, it is anticipated that this study will make an important contribution to the literature and be useful to practitioners.

Theoretical Framework Supervisor support
Studies conducted over the years have attributed the most important role in creating, sustaining, and succeeding in organisations to supervisors and their behaviour (Schein, 2010).Today, supervisors in modern organisations are more employee-centred than ever before (Kozlowski, Chao, Smith & Hedlund, 1993).Supervisor support, which is defined as the support that employees receive from their primary supervisor (Yoon & Lim, 1999), refers to the task of appropriately shaping the work environment to create positive work attitudes among employees (Ackfeldt & Coote, 2005).According to Netemeyer, Boles, McKee and McMurrian (1997), this concept expresses the importance that the supervisor places on the employee.

Workplace loneliness
The concept of loneliness, which is becoming more prevalent in society (Killeen, 1998), was discussed by Robert Weiss (1973) in the context of emotional and social isolation (Wright, 2005).Loneliness has a significant impact on the success and performance of organisations and negatively affects employees cognitively, emotionally, and behaviourally, and is one of the topics that have long received attention in the organisational behaviour literature (Demirbaş & Hasit, 2016;Erdirencelebi & Erturk, 2018;Keser & Karaduman, 2014;Nartgün & Demirer, 2012;Oguz & Kalkan, 2014;Peng, Chen, Xia & Ran, 2017).According to Wright (2005:124-125), loneliness consists of a combination of subjective and emotional responses that come together as a unified structure.Most theorists agree that loneliness is a psychological state resulting from qualitative or quantitative deficits in a person's social relationships.
Employees who play a crucial role in achieving an organisation's strategic objectives spend a large portion of their daily lives at work.Therefore, it is critical that employees fulfil their emotional and social needs through the relationships they form with others.As Rhoades and Eisenberger (2002:711) note, employees also need the acceptance and support of their colleagues who spend time together at work and share similar feelings.Loneliness in the workplace is the emotional state that occurs when this need is not completely met, and it has negative effects and consequences for the organisation and the individual.
Studies show that loneliness is an important stressor in work life (Ukwuoma, 1999).Studies also indicate that employees who are exposed to loneliness at work show lower organisational citizenship behaviour (Keser & Karaduman, 2014).In addition, these employees show more defensive behaviour in accepting silence (Nartgun & Demirer, 2012).The decrease in these employees' well-being increases staff turnover intention (Demirbaş & Hasit, 2016;Erdirencelebi & Erturk, 2018), and this situation has financial implications for the organisation (Sparks, Faragher & Cooper, 2001).

Perceived overqualification
Overqualification, the perception that the worker's education, skills, experience, and abilities exceed the requirements of the job, can emerge as a form of underemployment in organisations (Erdogan & Bauer, 2009:557;Johnson & Johnson, 2000:538;McKee-Ryan & Harvey, 2011:963).Overqualification is divided into two categories, namely, objective and subjective overqualification.Objective overqualification is the situation in which the employee's actual qualifications, consisting of educational level, experience, knowledge, skills, and abilities, exceed the requirements of the job, whereas subjectively perceived overqualification is the perception of the employee that his or her qualifications exceed the requirements of the job (Ozkanan, 2018:664).
This perception, which often leads to destructive and negative outcomes, is particularly important to employee job satisfaction (Erdogan & Bauer, 2009;Lobene & Meade, 2013) and performance (Akbiyik, 2016) and leads to a reduction in organisational citizenship behaviour (Pelenk & Acaray, 2019).This perception, which leads to the long-term failure of the employee in the workplace, also causes health problems in employees and brings a loss of morale and motivation (Johnson & Johnson, 1997:1907).Studies on the fit between people and the workplace show that as the harmony between the employee and his or her environment increases, work outcomes also improve (Lobene & Meade, 2010:1).On the other hand, Kanbur and Sen (2020) discovered a negative correlation between perceived overqualification, innovative behaviour and support from the supervisor.

Research Hypotheses and Model
In an organisation, good relationships between employees, based on equality and friendship, are of great importance for them to continue with their work with enthusiasm and remain motivated (Stoica, Brate, Bucuta, Dura & Morar, 2014).Unlike social loneliness, WL only occurs in the place where individuals work.In other words, even if individuals have healthy and reliable relationships in their home environments, they may find it difficult to establish healthy relationships in the work environment (Dogan, Cetin & Sungur, 2009).Although it is recognized that personality, introversion, and social skills play an important role in the development of loneliness (Wittenberg & Reis, 1986), managers have recently recognised that this is an organisational problem (Ozcelik & Barsade, 2011).
The concept of WL, which has attracted attention in the literature because of its implications and outcomes for organisations, still has unresolved elements.Studies have shown that WL supports undesirable and negative attitudes and behaviour in organisations.It is known that individuals who experience WL, which can lead to lower job and life satisfaction (Cindiloglu, Polatci, Ozcalik & Gultekin, 2017;Erdirencelebi & Erturk, 2018), exhibit fewer organisational citizenship behaviour (Keser & Karaduman, 2014).Again, it can be said that these employees have a higher turnover intention (Demirbaş & Hasit, 2016;Erdirencelebi & Erturk, 2018;Wright, 2005).Studies have shown that feelings of loneliness at work affect employees' affective and normative commitments and that leader-member interaction decreases among these employees (Cindiloglu et al., 2017).In the literature review conducted specifically with teachers in mind, a negative relationship was found between teachers' WL and job satisfaction (Cifci & Dikmenli, 2015;Gafa, 2019;Neto, 2015).Similarly, research has shown that teachers' increasing level of loneliness at work decreases their organisational citizenship behaviour (Keser & Karaduman, 2014); on the other hand, it shows that increasing organisational citizenship behaviour increases organisational socialisation and organisational identification (Cavusoglu, 2022).In a study conducted by Demirbas and Hasit (2016) it was found that there is a significant and clear relationship between academics' level of loneliness at work and their turnover intention.
Overqualification refers to individuals' perception that they have more than the needed job requirements in terms of education, work experience, skills, and abilities (Erdogan & Bauer, 2009:557).Studies have shown that employees who perceive themselves as overqualified have a higher turnover intention (Erdogan & Bauer, 2009;Lobene & Meade, 2013;Maynard, Joseph & Maynard, 2006) and are more likely to exhibit unproductive work behaviour (Liu, Luksyte, Zhou, Shi & Wang, 2015;Luksyte, Spitzmueller & Maynard, 2011).These employees also show low levels of performance (Bolino & Feldman, 2000;Feldman, Leana & Bolino, 2002) and job satisfaction (Johnson & Johnson, 1997;Maynard et al., 2006).From the literature review it is clear that perceived overqualification reveals controversial results for educational institutions.For example, Toker (2021) found that it has a positive impact on both task performance and contextual performance.Dar and Rahman (2020), in their study of 193 teachers, found that overqualified employees who perceive their current job as a career activity are more likely to exhibit innovative and creative behaviour for the organisation.On the other hand, in a study of teachers Lobene and Meade (2013) found that perceived overqualification was negatively correlated with job satisfaction and affective commitment.
If employees consider themself superior to other people as a personality trait and always think that they are the "best", i.e. if they consider themself overqualified compared to other employees, they may harm both themself and their environment (Karaman & Cetinkaya, 2019:14).It can be said that these individuals who cannot establish an effective and productive relationship with their environment will find themselves lonely in the workplace over time.At this point, the support an employee receives from his or her supervisor and the level of importance given to him or her can be considered a compensatory factor.Studies conducted specifically in educational institutions confirm this expectation.In a study that Uzun (2018) conducted on teachers, it was found that the perception of support from the supervisor was positively related to identification with the organisation.Gok, Karatuna and Karaca (2015) report on the job satisfaction of SS perceived by teachers; and Nayir (2012) report on trust in the organisation.Oguz and Kalkan (2014) and Yilmaz (2011) found in their research on teachers that social support reduces loneliness at work.Considering that these positive and constructive behaviour mentioned above will shape the relationship between perceived overqualification and WL, the following hypotheses and research models were established and are presented in Figure 1: H1: Perceived overqualification has a definite and significant effect on emotional loneliness.H2: Supervisor support plays a moderating role between perceived overqualification and workplace loneliness.

Research Sample
The population of the study consisted of music teachers working in private and public educational institutions in Turkey.The relational survey model, which is one of the quantitative research designs, was used for the study.With the relational survey model it is attempted to reveal the relationships between two or more variables (Karasar, 2010:79).The data used to determine these relationships were obtained through the survey technique.The snowball method, which is one of the non-random sampling methods, was used in this study.The snowball method made it possible to reach new subjects by using social network connections when the potential subjects were not well known (Gurbuz & Sahin, 2006:135;Noy, 2008).In this method, some participants suitable for the research criteria were identified in accordance with the research objectives.Then, the prepared questionnaire was sent to these participants online, and they were asked to reach out to other participants through their social networks.The population of this study consisted of teachers working in the 2019-2020 school year.According to the National Education Statistics and Formal Education 2019/20 data (T.C.Millî Eğitim Bakanlığı, n.d.), the number of teachers during this time period was 1,117,686, which is equal to the total number of teachers in Turkey.It is known that music teachers make up a small part of this number.However, there is no reliable information about the number of music teachers.Therefore, to calculate the sample, we used the formula of Naing, Winn and Rusli (2006), which is applied when the size of the population is known.According to this formula, the sample of the study was calculated to be 272 people, with a margin of error of 5% at a confidence level of 90%.The research data were collected between February and June 2020.The variance of the population was not estimated as part of the research, but 369 music teachers agreed to participate in the research.

Data Collection Tools
The Perceived Overqualification Scale developed by Maynard et al. (2006) and adapted into Turkish by Yıldız, Ozdemir, Habib and Çaki (2017) was used to measure perceived overqualification, which is the mediator variable of the study.It can be seen that the original unidimensional scale has been divided into two dimensions in its Turkish adaptation, namely work disdain and self-esteem.The Leader Support Scale developed by Netemeyer et al. (1997) was used to measure the degree to which leaders and managers support employees.Finally, the WL Scale developed by Wright, Burt and Strongman (2006) and adapted into Turkish by Dogan et al. (2009) was used to measure WL, which is one of the dependent variables of the study.The scale is a 5-point Likert scale, with responses ranging from 1 = strongly disagree to 5 = strongly agree.

Validity and Reliability Analysis of the Scales used in the Research
An exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were conducted to examine the construct validity of the scale.While the EFA is used to determine which factors generally capture the scale items (Conway & Huffcutt, 2003), the CFA is conducted to determine whether the original structure of the scale has been verified in accordance with the data (Brown, 2006).
In the EFA conducted for the SS scale, the independent variable of the study, it was found that a single dimension was consistent with the original scale.It can be seen that the single factor with an eigenvalue above 1 according to the EFA for the SS scale explains 83.004% or 72.784% of the total variance.As a result of the EFA for the perceived overqualification scale, which is the mediator variable of the study, the original structure of the scale with five items and four items was confirmed.It can be seen that two factors explain 72.784% of the total variance.As a result of the reliability analysis for the scale WL, which is the dependent variable of the study, one expression had to be removed from the analysis.The subsequent EFA revealed that the distribution of two statements on two dimensions required their exclusion from the analysis.All of these analyses confirmed the singlefactor structure of WL.The specified factor explained 67.191% of the total variance.The values of the fit index for the variables of the survey after the CFA of the first stage, which was performed to determine the compatibility of the scale expressions with the factor structures, are included in Table 1.Marsh, Balla and McDonald (1988); Tabachnick and Fidell (2007).

SS WL POQ
The CFA performed for the perceived overqualification scale failed to achieve the expected values for goodness of fit.The 2dimensional structure of the scale was confirmed by removing one item from both dimensions of the scale.Table 1 shows that the CFA results for the SS, WL, and perceived overqualification scales had the expected goodness-of-fit values.Again, the alpha reliability coefficient calculated for the scales was 0.717 for POQ, 0.948 for SS, and 0.799 for WL.
In the study, 56.6% of the music teachers were female, 57.1% were married, 48.4% were between 26 and 36 years old, 60.2% had work experience of 1 to 5 years, and 52.2% worked in a large town.The relationships between perceived overqualification, SS, and WL, which are the variables of the study, and the standard deviation and mean values of these relationships are shown in Table 2. Table 2 shows significant negative and weak correlations between POQ and SS (r = -0.258and p < 0.01) and significant and positive mediumstrong correlations between POQ and WL (r = 0.388 and p < 0.01).A negative medium-strong and statistically significant correlation was again found between SS and the correlation level WL.

Findings Hypotheses
A hierarchical regression analysis was used to examine whether SS plays a moderating role in the relationship between perceived overqualification and WL.The research model included the concept of overqualification (independent variable), SS (moderating variable), and WL (dependent variable).The results of the analysis are shown in Table 3.According to the results of the hierarchical regression analysis in Table 3, the moderating effect of SS for the effect of POQ on WL, the independent variable (perceived overqualification) and the dependent variable (WL) are significant.In the second step, it is between POQ and the moderating variable, SS.In the last step, the interactive term POQ and SS is included in the model.Examination of Table 3 shows that all stages of the hierarchical regression model are statistically significant.It was found that POQ (F (1,367) = 65.174;p = 0.000) in the first group and POQ and SS in the second group (F (2,366) = 48.788;p = 0.000) contribute significantly to the regression model.The corrected R 2 values of the model are 0.151, 0.210, and 0.217 for SS, respectively.This result shows that 15% of the change in WL is explained by POQ and 21% by POQ and SS together.Examination of the coefficients based on models revealed that perceived overqualification in the first group (β = -0.388,p < 0.01), perceived overqualification (β = 0.323, p < 0.01) and support from supervisor (β = -0.253,p < 0.01) included in the model in the second group; and POQ (β = 0.200, p < 0.01), SS (β = -0.421,p < 0.01), and interaction term (β = 0.198, p < 0.01) included in the model in the third group had a significant effect on WL.
To determine the way and direction of the interaction between POQ and SS, Figure 2 shows the effect of a low and high level POQ (where it is the value SS with the mean +1 and -1 standard deviation) on WL.

Figure 2 Moderating role of supervisor support
Figure 2 shows whether music teachers have low or high SS, and the relationship between low and high WL and POQ.The slope test was used to analyse whether the variables were significantly different from zero (Aiken, West & Reno, 1991).Figure 2 also shows that the relationship between POQ and WL of music teachers depends on SS.
From the graph, it can be seen that when SS is low, the POQ is higher, and the WL is higher.In other words, when a person perceives himself/herself as overqualified and also receives low SS, the WL increases.Again, the graph shows that the WL of people who perceive themselves as overqualified but receive high SS, decreases.Based on this result, the relationship between WL and POQ is characterised by WL.

Conclusion
Organisations are social structures in which workers spend a large part of the day and which are of great importance for the satisfaction of spiritual and material needs.In this social structure, the individual's personality traits, self-perception, and level of relationships within the organisation have a long-term impact on the individual's mental health.Based on the expectation that music teachers who perceive themselves as overqualified may experience emotional and social loneliness, with this study we identified the role of SS, which is assumed to shape this relationship.The model was constructed in accordance with the conceptual framework, and the literature was tested within the specified sample.Prior to analysing the data, EFA and CFA were used to determine the appropriateness of the measurement instruments.As a result of the analyses, it was determined that the scales used were reliable and valid.

Discussion and Recommendations
According to the results of correlation analysis, there is a negative correlation between POQ and SS and between SS and WL.It can be seen that there are positive and significant relationships between POQ and WL.Accordingly, WL increases in teachers when SS decreases and WL, when POQ increases.This research finding corresponds with Wright's (2005) study.Wright (2005) found that loneliness at work was related to SS, colleagues' support, and out-of-work support.
As a result of the simple linear regression analysis performed to test the first hypothesis of the study, POQ was found to have a definite and significant effect on WL.Consequently, the WL of music teachers increases with increasing POQ.Based on these results, hypothesis H1, perceived overqualification has a definite and significant effect on emotional loneliness, is supported by findings.Since there is no study in the literature that compares the two concepts, comparing the results from our study is impossible.
In the hierarchical regression analysis conducted to test the second hypothesis of the study, SS was included as a moderator in the model.The analysis revealed that SS moderates the relationship between perceived overqualification and WL.Consequently, it can be stated that SS moderates the relationship between perceived overqualification and WL.According to the research results, WL increases even more for music teachers who have high levels of perceived overqualification (POQ) but low levels of SS.On the other hand, WL decreases for teachers who have high support from their supervisor even if they have high levels of POQ.For this reason, even if music teachers feel highly overqualified, they do not feel alone at work if they have strong support from their supervisor.In this sense, H2, supervisor support plays a moderating role between perceived overqualification and workplace loneliness, is accepted.Similarly, Stoica et al. (2014)

Work
Low Supervisor… supervisors had lower WL scores than employees who did not receive support from their supervisors.Similarly, Oguz and Kalkan (2014), Wright (2005), and Yilmaz (2011) found in their research with teachers that social support reduces loneliness in the workplace.In this sense, we found that social support plays an important role in meeting the needs of workers who feel lonely at work.Music teachers in educational institutions have the task of educating people who are receptive to the arts and have varied knowledge and various skills.The main reason for teachers' interest in WL, which is of great importance to individuals and society, is to improve social and psychological well-being.Naidoo (2019) states that supervisors are very important to school development.Supervisors should, therefore, create a common culture in educational institutions and attend to the social and emotional needs of teachers.In this sense, supervisors in educational institutions need to support teachers and help them connect emotionally and socially with the organisation.To achieve this, it is important to develop fair, transparent, and trusting relationships within the institution and to create an organisational climate that supports this.
Teachers with different personality traits experience many barriers to align on a shared awareness of common goals.Perceived overqualification is just one of these barriers.Teachers who feel that they are capable of more than the requirements of their position can be helped to use this perception positively (SS).Supervisors can support the transformation of these teachers' energies into performance by creating platforms where they can apply their knowledge, skills, and abilities on a large scale.Again, teachers' perceived loneliness is thought to decrease through effective and skilled interaction between supervisors and members.
In addition, supervisors can strengthen intraorganisational relationships by matching teachers who believe that they are overqualified with their colleagues on tasks that require joint efforts.Among other things, we suggest changing perceptions of teachers who believe that they are overqualified by implementing work and business practices that emphasise respect, love, and collaboration within the institution.

Limitations
Like most social science research, this study had several limitations.The first was that the data were only collected from music teachers and, therefore, it is not appropriate to generalise the findings to all teachers.Secondly, any findings obtained should be evaluated in the context of the sample, the preferred survey method, and the limitations of crosssectional research.Finally, a cross-sectional design cannot fully examine causal relationships because intentions or perceptions may change over time.
Therefore, the use of a longitudinal design is suggested for future studies.
in a study with medical professionals and Eroglu (2018) in a study with teachers found that employees who received support from their

Table 1
Confirmatory factor analysis of variable scales/indexes The goodness of fit values: Byrne (2001);

Table 2
Mean, standard deviation and correlation values Note.*Shows that the relationship is significant at the 1% level (p < 0.01).

Table 3
The moderating role of supervisor support in the impact of perceived overqualification on WL