How do relational and personal attributes affect trust in adolescent friendship: An exploratory model

Trust is an essential aspect in a social life and interdependent society (Igarashi et al., 2008; Talwar & Renaud, 2010). We believe in people who grow and process the foods and medicines that we consume; we depend on the person who made the house we live in; we believe in the person who hold our money and our investment; and we believe in the government that keeps us from aggressors. In general, in every phase of our lives, we will rely on others to behave according to our expectations and hope that our expectation will be fulfilled. High levels of trust in society facilitates a fast and wide dissemination of information, (Lindström, 2005; Yip et al., 2007), which functions to reduce uncertainty (Tschannen-Moran & Hoy, 2000), develop and maintenance happiness and well-functioning relationships (Crosnoe et al., 2003; Simpson, 2007). Without trust, the world would be more painful and empty (Shultz, 2007; Simmel, 2011). Even thought trust is important in life, it has received relatively limited attention because of its complex multidimensional construct. Apart from the concept of trust being based on various disciplines, the variety of stages that trust develops within different types of relationships, makes trust a challenging concept to study (Simpson, 2007). There are ART I C LE I NFO ABST R ACT


Introduction
Trust is an essential aspect in a social life and interdependent society (Igarashi et al., 2008;Talwar & Renaud, 2010). We believe in people who grow and process the foods and medicines that we consume; we depend on the person who made the house we live in; we believe in the person who hold our money and our investment; and we believe in the government that keeps us from aggressors. In general, in every phase of our lives, we will rely on others to behave according to our expectations and hope that our expectation will be fulfilled. High levels of trust in society facilitates a fast and wide dissemination of information, (Lindström, 2005;Yip et al., 2007), which functions to reduce uncertainty (Tschannen-Moran & Hoy, 2000), develop and maintenance happiness and well-functioning relationships (Crosnoe et al., 2003;Simpson, 2007). Without trust, the world would be more painful and empty (Shultz, 2007;Simmel, 2011).
Even thought trust is important in life, it has received relatively limited attention because of its complex multidimensional construct. Apart from the concept of trust being based on various disciplines, the variety of stages that trust develops within different types of relationships, makes trust a challenging concept to study (Simpson, 2007). There are Trust is an important aspect in human relationship. According to previous research trust is influenced by 2 main attributes: personal (benevolence, integrity, and competence) and relational (support, closeness, and reciprocity) attribute. The study aimed to find a preliminary model to explain the association of personal and relational attributes in adolescent friendship. Two hundred and twenty participants (male=44.09% and female=55.91%) participated in the study. A survey method was used to collect the data. The design of this study was correlational design. Results from enter method in regression analysis show that six factors of personal and relational attributes influence trust at 44.5%. However, closeness had no significant influence on trust, so closeness was not included in the next analysis, the sequential analysis. The result shows that the relational attribute has higher influence on trust than personal attribute. Support and reciprocity factors which belong to relational attribute, has a higher contribution to trust in adolescent friendship than benevolence, integrity and competencethe latter three factors belonging to the personal attribute. Therefore building trust in friendship requires an emphasis on relational attributes of trust, especially that of support and reciprocity factors. 2 ISSN 2598-6368 (online) / 1693-7236 (print) Rahmanawati et.al relationships where trust persists for long periods of time, while in others, trust is prone to conflict. Therefore, trust is unstable and can change rapidly in both directions (Tschannen-Moran & Hoy, 2000).
Trust is a state where a person is willing to be vulnerable on the basis of positive expectation towards another person (Mayer et al., 1995). Khodyakov (2007) noted that trust is created from a process of maintaining quality in social relation. In line with the definition, Rotenberg (2010) defines trust as a protection against doubt to create a good relationship between each other. These definitions of trust focuses on trust at the interpersonal level, which can be classified in two dimensions, cognitive and affective (Ferris et al., 2010;McAllister, 1995). The cognitive forms of trust consist of perceiving others as reliable, honest, fair, and having integrity. Meanwhile the affective dimension includes a specific relationship with another, which can cause the other party to show concern about a person's welfare.
The trust model from Mayer et al (1995) is one of the models that is quite prominent, which explains that personal factors of a trustee and trustor's propensity leads to trust. Propensity is a condition where a person places general trust towards others, which influences how much trust one exerts towards the trustee. There are three personal factors which influence trust: ability/competence, benevolence, and integrity. This model is similar to the framework on trust in the context of leadership (Burke et al., 2007), trust in online shopping (Connolly & Bannister, 2008), and trust in the organization context (Bews & Martins, 2002;Chong et al., 2003).
Benevolence is when a trustee does something good to the trustor without expecting any profit (Mayer et al., 1995;Tschannen-Moran & Hoy, 2000). From the trustor's perspective, benevolence is the confidence that one's wellbeing will be protected and not hurt by the trusted party. Meanwhile, competence/ability is a set of skills, abilities, and characteristics that enables a person to exert influence over specific aspects. Competence can be interpreted as the personal abilities in individuals as part of the acquisitions and used as knowledge. In a social context, the competence of participants is the knowledge of individuals to place awareness in different situations, as well as individual knowledge in people around (Rotenberg, 2010). Benevolence is not enough, sometimes a good person who does not have enough skills, will not be trusted (Tschannen-Moran & Hoy, 2000). So, competence is also a factor that influences trust. Finally, integrity is defined as how far a trustor perceives the trustee's act in accordance with the set of values the trustor holds (Mayer et al., 1995).
Various factors contribute to the development of trust. The quality of a trusted party, consisting of one's reliability, competence, and integrity (personal attribute), is often used as criteria in developing trustworthiness (Mayer et al., 1995). However, research conducted by Faturochman & Minza (2014) found that trust is influenced by two major attributes, the personal and relational attribute. Relational attributes mentioned in the study consists of support, closeness, and reciprocity (Faturochman & Minza, 2014). Support can be defined as information which leads people to believe that he or she is being cared, loved, and appreciated (Kumalasari & Ahyani, 2012). Closeness can be described as the extent to which the thoughts, affect, and behaviors of trustor and trustee are interrelated (Laursen et al., 2000). Close relationship between two persons determine the quality of trust. Thus trust will increase when the trustor and trustee have a close relationship (Faturochman & Minza, 2014). Reciprocity can be defined as the extent to which mutuality and cooperation figure the relationship (Laursen et al., 2000).
Yet, trust is influential in the fabrication of social life especially in interpersonal relation. In interpersonal relations, trust becomes a key in developing and maintaining happiness and well-functioning relationships (Crosnoe et al., 2003;Simpson, 2007). When a relationship is based a trust, the parties involved will have a strong desire to continue the relationship (Simpson, 2007;Warris & Rafique, 2009).
In this research, we will mainly discuss trust at the interpersonal level, especially in the context of friendship, which is a basic human need (Carroll, 2002). In general, a friend is someone who can be trusted, who can be invited to enjoy togetherness, and to share life experiences (Turner, 2002). However, friendship itself is a social phenomenon that is part of socialization in human life with the aim of improving their social circle for several reasons, such as sharing, enjoying togetherness, and getting some emotional comfort (Warris & Rafique, 2009), an internal need for humans and a tendency to depend on a person (Carroll, 2002). Thus friendship is a human's need in interacting with others to share and support (Hurlock, 2004).
Friendship and trust are strongly related concepts because there is no relationship that can survive without trust (Warris & Rafique, 2009). Trust is a cornerstone on the way we interact with others (Rotenberg et al., 2005), and it causes high-quality interactions and confidence in friendship (Bove & Johnson, 2001). This makes trust in friendship important because it will lead to greater social support and better quality of life (Demerouti et al., 2001;Lunsky & Benson, 2001). Individuals in reciprocated friendships have more positive quality of friendship than non-reciprocated friendships. Reciprocated friendship correlate with less conflict in friendship (Linden-Andersen et al., 2009). Closeness is most common root terms in the study of interpersonal relationship (Johnson et al., 2009), as it is related with the commitment to the continuation of relationship (Simpson, 2007;Warris & Rafique, 2009). Support is related with good quality friendship, and enhances psychosocial well-being (Ciairano et al., 2007).
This study aimed at finding a preliminary model which can explain how trust is influenced by personal attributes (benevolence, integrity, and competence) and relational attributes (support, closeness, and reciprocity) in adolescent friendship. The hypothesis of this study is relational attributes play a major factor on building trust, supported by personal attributes of trustee.

Method
This study was conducted in Yogyakarta. In this study, survey method was used in collecting the data.

Participants
Participants in this study were 220 students from the Faculty of Psychology and Faculty of Engineering at Universitas Gadjah Mada (Indonesia). A breakdown by gender yielded 123 women (55.9%) and 97 men (41.36%) representing the balanced distribution of gender in this student population. The mean age of the participants is 19 years 2 months with a range from 17 to 21 years old.

Procedure
The participants completed a likert scale in Bahasa Indonesia, the national language of participants. The likert scale was developed from the results research conducted by Faturochman & Minza (2014), measuring cognitive, affective, and behavioral dimensions of each variable in this study. The intrument consists of three scales measuring the following variables: personal trustworthiness (benevolence, competence, integrity), relational trustworthiness (support, closeness, reciprocity), and trust. Benevolence, competence, and integrity are considered personal attributes influencing trust, while support, closeness, and reciprocity are considered relational attributes influencing trust. The participants as a trustor (a trusting party) were asked to imagine a specific person, either a best friend or a close friend, to help them decide their answer to the questions that assess the personal characteristics of trustee (a party to be trusted) which includes benevolence, competence, and integrity. The participants were also asked about the relational attributes of trustee such as support, closeness, and reciprocity.

Instrument
Personal trustworthiness scale consists of three subscales as follows: Benevolence Subscale. The benevolence subscale consists of 5 items. It was used to assess the personal characteristic of the trustee, especially in relation to the extent to the trustor believes that the trustee will treat the trustor well. Participants were asked to rate their agreement on statements relating to their friend as a trustee, such as "he/she is friendly" and "he/she cheers the other who is sad." Competence Subscale. The competence subscale was used to measure the competence skill of a trustee. It consists of 6-items.The sample items are "he/she is a broad minded person" and "he/she is a well experienced in his/her field." Integrity Subscale. The integrity subscale consists of 5-item. The items measure how far a trustor sees that the trustee's behavior is in accordance with principles that the trustor finds acceptable. The question in this scale such as "he/she keeps a promise" and "he/she keep the secret." Relational trustworthiness scale consists of three subscales as follows: Support Subscale. The support was used to assess how the trustee support the trustor. This subscale consists of 6-item. Participants asked to rate their agreement on statements related with their friend as a trustee such as "he/she motivates me" and "he/she gives a suggestion to me." Closeness Subscale. The closeness subscale consists of 5 items. The items which assesses the interconnectedness between the trustee and trustor. Participants rated to what extent they agreed or disagreed with the statements related to his/her friend, for instance "he/she is close to me." Reciprocity Subscale. The reciprocity subscale was used to assess the extent of mutuality and cooperation between the trustee and trustor. This subscale consists of 5 items. Participants rated how far their agreed on statements concerning their friend, for instance, "he/she and I understand each other" and "he/she and I believe each other." The trust scale is a 6-item scale which assesses the extent of the trustor's tendency to trust a trustee. Sample of statements is "I can fully trust my friend." The response format was a 7-point likert-type rating scale (ranging from 1= strongly disagree to 7= strongly agree) reflecting the agreement or disagreement of subject in a statement of items. The attributes were tested and has a reliability value of more than .7, which means the measuring instrument can explain the attributes; while the item-total correction range was .3 -.6. Table 1 shows the psychometric properties of the instruments.

Data Analysis
The correlation and regression analysis were used to analyze data with SPSS. First, the correlation analysis with Pearson correlation was done between the personal attribute (benevolence, competence, and integrity) and relational attribute (support, closeness, and reciprocity) as an independent variable with trust. Then enter method in regression analysis was used in the next analysis to find out the contribution of all variables together on trust.
After that, sequential analysis was applied to explore the effect of personal and relational attributes on trust.

Correlation Analysis
The correlation analysis with Pearson Correlation in Table 2 show that all of the factors were positively correlated with trust. All of the correlations were significant at .01 alpha level (2tailed). The means that the six factors can explain trust. The relational attributes which comprise of support (r = .62), closeness (r = .51), and reciprocity (r = .56) have higher correlation coefficient than the personal attributes which comprises of benevolence (r = .46), competence (r = .25), and integrity (r = .46). Support (r = .62) has the highest correlation coefficient with trust among the others factors. Meanwhile competence (r = .25) has the lowest correlation coefficient among the others.

Regression Analysis
The results of regression analysis with enter method in Table 3 show that all factors together influence trust at 44.5%. In this analysis, closeness has no significant influence on trust (β = -.01; p = .981). Therefore closeness was not used in the next analysis. The enter method was also used to explain how the other five attributes influence trust. Table 4 and Table 5 show that relational attributes have higher influence (41.5%) than personal attributes (30.1%).

Sequential Analysis
Sequential analysis was used to explore the model of personal and relational attributes on trust. As shown in Table 6, support and reciprocity has a high influence on trust (R 2 = 41.5%). The enter method which involve all of the factors, shows that contribution of five factors (support, reciprocity, benevolence, integrity, and competence) is 44.5% (see Table 3). Using sequential analysis, the R 2 change between the previous and latest analysis was only 3.0%. As the main variable, relational attribute explains trust at 41.5%.

Discussion
The main purpose of this study was to explore the role of relational and personal attribute toward trust in adolescent friendship. The findings indicate that trust in adolescent friendship is influenced by relational and personal attributes. This result is consistent with Glanville & Paxton (2007), which found that people will trust others based on both personal character and relational aspects. Specifically, the findings explain that relational attributes including support and reciprocity play the main role in forming trust in adolescent friendship, meanwhile personal attributes including benevolence, integrity, and competence play supporting roles. In other words, a good relationship has more contribution than personal 7 attribute in forming trust. This finding strengthens the prior study from Faturochman & Minza (2014) that a good quality personal characteristic of the trustee is not sufficient in forming trust, because there must be a good relationship between actors. Rus & Iglič (2005) also found that quality of trust arises when trustor and trustee have a close relationship.
One of the possible reasons of why relational attributes becomes a main factor in adolescent friendship's trust in this research is because of Indonesian culture, which is predominantly collectivist. Hofstede (2011) stated that less developed and Eastern countries tend to be a collectivistic country. As a collectivist society, Indonesian people are integrated with their group and have a strong attachment (Hofstede, 2011), and it is necessary to first build a relationship and form trust between people (de Mooij & Hofstede, 2010). According to its characteristic as collectivist society, Indonesia tends to place relational attributes first in comparison to personal attribute in forming trust.

The role of relational attributes
Among three factors in relational attributes, support and reciprocity have significant contribution towards trust in adolescent friendship; meanwhile closeness has no significant contribution. In this model, support has higher contribution toward trust in adolescent friendship in comparison with reciprocity. High quality friendship is appraised as a source of support for adolescents (Damon & Lerner, 2008) and support from the environment has a positive correlation with trust (Salazar, 2015). It is possible that social support has an important impact on trust because it can provide a shield from stress. A study from Ying (2008) shows that social support from a friend protects adolescents from emotional exhaustion or stress. Emotional support, which is included as a relational attribute, encourages the formation of trust in friendship (Weber et al., 2004).
The role of reciprocity toward trust in adolescent friendship is strengthened by finding from Das & Teng (2004) and Tanis & Postmes (2005), showing that reciprocity is a fundamental factor in maintaining trust. Reciprocity is returning an act i.e. good deeds are rewarded and bad deeds are punished (Fehr & Gächter, 2011), and a manifestation of the equality matching model of relation. Reciprocity in friendship makes people feel comfortable with the mutual connection they built. Reciprocity is also an important aspect toward trust in adolescent friendship because friendship is an interdependent relation. In interdependence theory, people develop trust depends not only based on personal traits but also on the historical relation with people (Van Lange & Balliet, 2015). If reciprocity can be achieved, where one party can accomplish and be responsive to the other party's need, then trust will follow. In ongoing relationships, partner's actions representing reciprocity are more important than the person's trait. Interdependent theory stated that trust lies in both actors' hands.
As shown previously, the relational attributes of support and reciprocity have an influence on trust; while closeness does not. Nielson (1998) found that closeness is only needed in the beginning of a relationship, while support and reciprocity are needed to maintain the relationships.

The role of personal attributes
Despite the importance of relational attributes in building trust, the quality of a trustee comprising of benevolence, integrity, and competence remains a significant factor in building trust. This finding is consistent with Xie & Peng (2009) who found that benevolence, integrity, and competence has a major role in repairing trust in customers.
The results show benevolence as factor which contributes in forming trust in adolescent friendship. A prior study in different subject (preschooler children) confirms the above result (Johnston et al., 2015;Landrum et al., 2013). The study found children weigh benevolence when choosing who can be trusted. In another context, a client is more concerned with benevolence than the competency of consultants, especially for long-term collaboration (Ko, 8 ISSN 2598-6368 (online) / 1693 Rahmanawati et.al 2010). Wu et al. (2014) also found the same result that benevolence has a higher impact on trust in continuing online social networks.
Apart from benevolence, integrity as a personal attribute of a trustee also leads to trust. Butler (1991) revealed that integrity is one of the basis of trust, which includes honesty and moral character (i.e truthfulness). A person who has integrity will tend to encourage others to be close and trust them. In friendship, a person who has integrity is believed to be a capable person who can be trusted. Palanski et al. (2011) also found that integrity is defined as the consistency of one's behavior, is related with, and will lead to trust at the individual level and at the group level.
However, this study also found that competence as a personal characteristic of the trustee has negative influence on trust. This can be explained by findings from Butler (1991) which found that competence is important and salient condition only in contexts such as in hierarchical relationships (as a superior's downward trust towards his or her staff). Trust in adolescent friendship is a relation which on the most part is equivalent. Parallel to the finding, Ko (2010) mentions that trust towards others no longer focuses on competence.
The implication of this research is that building trust in friendship requires an emphasis on relational attributes of trust, especially support and reciprocity. Along the emphasis on relational attributes, personal attributes of trust (benevolence, integrity, and competence) are also needed to support the foundation of trust in friendship. The limitations of this study lies in the number of subjects. Only 220 subjects participated in this research, and all of which are university students. Therefore, this research cannot be generalized in larger population of adolescents outside the university student population. To obtain a more comprehensive concept of trust in Indonesia, further research needs to include participants which represent a larger variation of adolescents.

Conclusion
This study explored how personal and relational attributes influenced trust in adolescent friendship. The results show relational attribute is the main factor in building trust, meanwhile personal attributes play a supporting role. The findings indicate that in forming trust in friendship, building a good relation through support and reciprocity is more important than constructing good personal characteristics such as benevolence, integrity, and competence.