The Importance of Positive Affect: the Role of Affective Personality in Predicting Organizational Citizenship Behavior

Previous research demonstrates inconsistent results in predicting how affect influences organizational citizenship behavior (OCB). This study aims to solve the inconsistency by taking the position that positive affect and negative affect are orthogonal, and their interaction produces four types of affective personality. They are ‘Self-fulfilling' (high positive affect and low negative affect), ‘High affective' (high positive affect and high negative affect), ‘Low affective' (low positive affect and low negative affect) and ‘Self-destructive' (low positive affect and high negative affect). The study hypothesizes that the self-fulfilling group displays the highest mean of OCB while the self-destructive displays the lowest. The high affective and low affective groups lie somewhere in between the two groups. The participants of this study were 227 employees, consisting of 151 males and 76 females with ages ranging from 20 to 60 years old (mean=38). They were measured using the Organizational Citizenship Behavior Scale (OCBS) and Positive and Negative Affect Schedule (PANAS). Based on the scores of their positive and negative affect dimensions, they were classified into four groups of affective personality types. One-way ANOVA analysis supported the hypothesis. The self-fulfilling group revealed the highest mean of Organizational Citizenship Behavior while the Self-destructive group revealed the lowest. The High affective and Low affective groups were located in between the first two groups. This paper discusses this contribution and highlights how it is potential to explain organizational behavior.


Introduction
Organizational citizenship behavior (OCB) is defined as behavior that is not directly or explicitly recognized by formal rewards, which aggregately promotes the effective functioning of the organization (Dennis W Organ, 1988;p. 4). OCB implies that individuals contribute beyond the required demands of their workplace. It comprises forms of behavior such as helping others, volunteering for extra tasks and complying with the rules and procedures of the workplace. This behavior constitutes employees' added value, which includes various forms of pro-social behavior including positive, constructive and meaningful social behavior (Kreitner & Kinicki, 2009). Organ, Podsakoff, and MacKenzie (2006) propose five primary dimensions in OCB: (1) Altruism, voluntarily helping other employees with tasks in an organization's operational activities; (2) Civic virtue, voluntarily participating in and supporting organizational functions in both professional and social contexts; (3) Conscientiousness, performing beyond minimum required standards; (4) Courtesy, solving problems related to others' tasks; and, (5) Sportsmanship, being reluctant to engage in issues that may damage the company's or team's work despite feeling personally irritated. Yen and Niehoff (2004) explain how OCB relates to organizational effectiveness in many ways. Among those are as follows. (1) An employee with prosocial behavior tends to help new employees speed up their training process so that they can adapt with the work faster. (2) Employees who help each other need less managerial supervision, freeing up the manager's time for other urgent matters. (3) Employees with positive attitudes towards one another are more cooperative and they avoid destructive conflicts with other workers. (4) Organizational communication improves because the employees voluntarily meet to informally discuss their work matters. (5) Support is available to make a working environment more positive. And finally, (6) employees are more willing to take on new responsibilities or learn about new technology or work systems. A study that examined this argument was conducted by Johnson (2008). He found that the positive affect correlates positively with OCB while the negative one correlates negatively with OCB.
OCB can be directed towards individuals (OCBI) and groups in a working environment or towards the organization (OCBO) to improve its functionality. Positive affect can be described as feelings of enthusiasm, activeness and alertness, while negative affect refers to feelings of distress and lack of pleasure.
Positive affect is correlated to both OCBI and OCBO, while negative affect is not (Dávila & Finkelstein, 2013). These two researchers explained that the absence of correlation between negative affect and OCB is due to the different nature of negative affect compared to positive affect, as argued by Lee and Allen (2002). Specific negative emotions correlate with OCB more than general negative affect (Lee & Allen, 2002). This is the reason why negative affect is less likely to correlate with OCB, as the measurement of negative affect encompasses the general negative affect rather than only specific negative emotions. If the negative affect is measured by only specific items such as hostility, the correlation with counterproductive behavior becomes higher and more significant compared to the correlation between anxiety and counterproductive behavior.
There have been numerous studies on Organizational Citizenship Behavior concerning personality, cognition and affective aspects, organizational commitment and job satisfaction (Organ et al., 2006). In Indonesia, Oetomo (2007) investigated the impacts of job satisfaction and leadership on OCB, and he showed that transformational leadership style and job satisfaction have significant impacts on OCB. A plausible explanation for these results is that people will experience positive affect if they work with transformational leaders who are inspirational and motivating, ideally influential, individualized attentive and intellectually encouraging. All these forms of behavior along with job satisfaction create positive feelings at one's workplace. These positive affects make individuals want to do things beyond the call of duty as manifested in OCB.
Several factors support the practice of OCB such as fair treatment and mood in the organization, and they all influence OCB in both individual and group settings. Desmet (2015) suggests that a person's intention to help others is also influenced by his/her mood. Personality is a set of characteristics that is relatively permanent, while mood refers to the characteristics that change from time to time. Positive moods increase the odds of a person to help others. Although mood is influenced (partially) by personality, it is also influenced by situations like working groups and factors originating from within the organization. Therefore, in cases where an organization appreciates its workers and treats them fairly, and where positive group working environments exist, the workers are more likely to develop positive moods. As a result, they tend to volunteer to help others.
Affect, which comprises phenomena such as moods and emotions, has been proven to be associated with OCB (Wright & Sablynski, 2008). The findings above suggest, however, that the degree to which affect predicts and influences OCB differs. Lee and Allen (2002) studied the roles of affect and cognition on OCB. They found that organizational behavior tends to be controlled and planned naturally, in contrast to the expressive characteristics represented by affect. Lee and Allen eventually concluded that cognitive impacts play a larger role compared to affect. This study, which included a sample from numerous industries, organizations, and positions in Singapore, found that the positive affect experienced by workers influenced their intentions to display organizational citizenship behavior. Different from the aforementioned studies, Wright and Sablynski applied an experimental approach to study two antecedents that may influence organizational citizenship behavior, including procedural justice and mood. The findings succeeded in establishing an influence for procedural justice towards organizational citizenship behavior, however mood was not found to be influential.
It is necessary to find the causes of the inconsistent findings in the influence of affect on OCB. One possible reason for this inconsistency is that the measurement of affect in previous studies mostly used questionnaires or experimental approaches that induced temporary positive and negative affect, rather than a more stable affect. Instead of observing the influence of temporary affect, the present study looks at more stable affects as a characteristic of personality traits. An attempt to categorize personality types based on affect was initiated by Galen, who divided human personality into four orthogonal types i.e. melancholic, choleric, sanguine and phlegmatic (cited from Eysenck, 1967). Melancholic and Choleric represent negative affect, while Phlegmatic and Sanguine represent positive affect. Eysenck (1967) categorized personality into two types, which are extraversion and introversion. The extraversion is dominated more by positive affect, while the introversion by negative affect. The more recent theory of the Big Five Personality, formulated by McCrae & John (1992), also categorizes personality based on positive and negative affect, in which the types of Extraversion, Agreeableness, Conscientiousness and Openness are characterized as positive affect, while Neuroticism as negative affect. Tellegen (1985) used the terms of Positive Emotionality and Negative Emotionality to describe positive affect and negative affect. Based on this concept, Watson, Clark, and Tellegen (1988) developed the PANAS (Positive Affect Negative Affect Schedule) scale, an instrument that aims to measure both positive and negative affect. Positive Affect and Negative Affect are two orthogonal traits that are not correlated with each other. PANAS has been vastly used in several psychological contexts. A number of studies found that positive affect correlates with greater life satisfaction, high commitment, self-control, self-confidence and feeling secure. In contrast, negative affect has been found to be correlated with negative characteristics such as anger, shame, fear and depression (see Norlander, Bood, and Archer, 2002 for a review). Since positive affect is associated with positive behavior, it is expected to bring positive influence on OCB, while negative affect is supposed to correlate negatively with OCB.
Using the PANAS concept, Norlander et al. (2002) developed a method that allows the establishment of four affective profile types emerging from the PANAS scale. These authors viewed that scoring combinations of positive affect and negative affect may increase explanatory power, and therefore propose four personality types; Self-fulfilling (high positive affect and low negative affect), High affective (high positive affect and high negative affect), Low affective (low positive affect and low negative affect), and Selfdestructive (low positive affect and high negative affect). Numerous studies have indicated that different affective personalities also differ in the level of stress experienced (i.e. Adrianson, Ancok, Ramdhani, & Archer, 2013;Archer, Adrianson, Plancak, & Karlsson, 2007;Arnten, Jansson, & Archer, 2008;, coping with stress (Arnten et al., 2008), anxiety levels Norlander, Johansson, & Bood, 2005), depression levels Norlander et al., 2005), quality of sleep Norlander et al., 2005) and abilities of thriving following traumatic events (Norlander et al., 2005). The advantage of using the four types of affective profile is that they are superior compared to the twodimensional model, because affective profiles have also been found to be associated with personality characteristics (Bood, Archer, & Norlander, 2004;Garcia, 2012;. Since OCB has been found to be correlated with the Big Five Personality traits (Organ & Ryan, 1995), it is expected that affective profile types will also influence OCB. To this day, the authors have not yet discovered any studies relating the four types of affective profile with organizational citizenship behavior.
The present study intends to investigate the extent to which affective profile influences organizational behavior. Based on the elaborations above, we hypothesize that affective profile influences organizational citizenship behavior; the self-fulfilling group will display the highest level of OCB, while the self-destructive group will display the lowest. High affective and low affective groups will be intermediaries between those two groups. The reasoning for the hypotheses is that the subjects of the self-fulfilling group have the highest positive affect and the lowest negative affect; consequently, they tend to be the most willing to help others and volunteer for extra tasks. In contrast, subjects in the self-destructive group have the highest negative affect and the lowest positive affect so that they tend to be most reluctant to help others or volunteer for extra tasks.
Meanwhile, subjects of the other two groups do not have this kind of combination. If the findings of this study indicate that there is a significant influence of types of affective profile on OCB, this could contribute to human resources management practices in organization. Affective profile type is one aspect of personality that needs to be considered as a part of selection criteria to improve the accuracy of selection and placement process of employees in a certain job position. In addition to the affective profile, this study also intends to reveal the influence of demographic factors, such as the length of service in the company and employees' sexes, on OCB.

Methods
Participants and Procedures. The participants of this study were employees of a state-owned electrical company operating in the areas of Kudus, Pati and Blora in the Province of Central Java, Indonesia. This company was chosen as the field of study because of the nature of the company's working system, in which the employees work mostly for a long times in order to ensure no electricity power shortage. Compared to other areas in the same province, in this area the power plants and transmission systems are relatively old and thus need more time and effort to maintain the system. In such working conditions, OCB plays an important role for obtaining better performance. The participants were asked for their consent to participate in this study.
A purposive sampling technique was applied by taking the employees as the research participants. This was applied with the consideration that the dependent variable strongly associates with work environment in industrial organizations. There were 227 employees involved in the study, consisting of 151 male and 76 female with ages ranging from 20-60 years old (mean = 38). Their length of service in the company also varied, ranging from 1 to 36 years (mean = 14). All 227 questionnaires indicated valid responses and were subsequently included for further analysis.
Materials. This study used two types of scales; the Positive Affect and Negative Affect Schedule (PANAS) and Organizational Citizenship Behavior (OCB) Scales. The PANAS that was developed by Watson, et al. (1988) consists of two sub-scales that represent two primary dimensions, namely Positive Affect (PA) and Negative Affect (NA). Each of these subscales consist of 10 items, so there are 20 items in total. Each item contains a word representing the subjects' feelings. Examples of the items comprise the words proud, enmity, inspired, assertive, etc. Subject responses are based on the 5-point Likert scale; from 1 for strongly disagree to 5 for strongly agree. Reliability of the scales was satisfactory, with Cronbach's alphas value of 0.74 for positive affect and 0.77 for negative affect. After performing correlation analyses towards these two dimensions, no correlations were evident. (r = -0.056). Therefore, it confirmed that the positive and negative dimensions are statically independent. Both questionnaires were self-reported. Classification of the four personality types (Self-fulfilling, Self-destructive, High affect, and Low affect) applied the scoring model developed by Norlander et al. (2002). The first step was to calculate the sum of both positive and negative affect, and was the continued by defining the median for both positive and negative affect. The code "1" was given for scores below the median, while the code "2" was used for scores equal to and above the median. Both of these codes classify scores between those with high or low affect, with the code 2 representing high affect and code 1 representing low affect. The results of these two scales were combined according to a procedure that assigned each one of the four affective personality groups, as follows: individuals showing high PA and low NA ('Self-fulfilling'), High PA and High NA ('High affective'), low PA and low NA ('Low Affective'), low PA and High NA ('Self destructive'). The validity of this procedure in classifying personality types is supported by a recent study. MacDonald and Kormi-Nouri (2013) used k-means cluster analysis to test the affective personality profiles model and found a similar affective personality profile as formulated by Norlander et al. (2002).
The OCB scale used in this study was a modification from the Organizational Citizenship Behavior Checklist (OCB-C) developed by Podsakoff that adapted for Asian context (Kumar & Shah, 2015). This version was employed 10 items out of Podsakoff's 24 items, were selected to avoid participants' boredom of answering too many questions. These 10 items translated into Indonesian with some modification such as reversing the items from favorable item and vice versa. OCB scale were arranged with the five OCB-shaping aspects, namely (1) altruism, voluntarily helping other employees related to tasks concerning organizational operational activities, (2) civic virtue, voluntarily participating in and supporting organizational functions, in both professional and social contexts, (3) conscientiousness, performance beyond minimum required standards, (4) courtesy, solving problems related to others' work, and (5) sportsmanship, employees' reluctance to engage in issues that may damage the company or team work although the particular employee is irritated. One exemplary item used for revealing the altruism aspect was "I am actively involved in giving creative and innovative suggestions to my co-workers'. An example item showing civic virtue was "I try my best to join extra trainings offered by the company in order to improve my job performance". Meanwhile, an example of the item of conscientiousness aspect was "I try to finish my job as quick and well as possible without wasting much time". A structured questionnaire was directed to measure OCB with alternative responses following the 5-point Likert scale model between Strongly Disagree = 1 to Strongly Agree = 5 for favorable responses and reversed for unfavorable responses. Reliability analysis revealed that only 10 items were valid and two items were excluded. The results of the reliability analysis indicated that the OCB scale had good validity (Cronbach's alpha = 0.70). Procedures and all measurements that were used in this research have been reviewed by the Research Ethic Committee of Faculty of Psychology, Gadjah Mada University.

Results
One-way ANOVA was used to test the hypothesis that affective personality influences organizational citizenship behavior. Homogeneity tests revealed that no difference was evident between groups (Levene's test = 1.93, p = 0.13). The results of the analysis as presented in Table 1 confirmed the hypothesis; the self-fulfilling group (M = 38.722, SD = 2.750) displayed the highest OCB score and the self-destructive group (M = 37.140, SD = 2.675) displayed the lowest, while both low affective (M = 37.633, SD = 3.146) and high affective (M = 37.895, SD = 3.014) individuals were intermediaries between those two groups (F (3,223) = 2.848, p = 0.038). After discovering the influence, analysis on differences of affective personality and organizational citizenship behavior between groups was conducted. To test for differences between groups, post hoc analysis was performed using Tukey HSD. The results of the analysis revealed the difference between mean OCB for the Selffulfilling group (M = 38.722, SD = 2.750) and selfdestructive group (M = 37.140, SD = 2.675), with p = 0.022. The results confirmed the hypothesis that the Self-fulfilling Group has the highest OCB, while the Self-destructive group has the lowest mean, and the High Affective Group and the Low Affective Groups lie in between the self-fulfilling and self-destructive groups.
In order to see the impact of the length of service on OCB, a simple regression analysis was used to test if the length of service in the company predicted the participants' OCB. It was found that length of service predicted OCB (R 2 = 0.034, F (1, 210) = 7.286; p = 0.008). Furthermore, analysis on the impact of the subjects' gender on OCB was conducted using a t-test. There was a significant difference between the genders on OCB. Male subjects showed higher OCB (M=30.967, SD=2.569) compared to female subjects (M=30.275, SD=1.909); t (1) = 1.993, p = 0.047.

Discussion
This study discovered that organizational citizenship behavior is significantly influenced by affective personality. The finding that self-fulfilling groups have the highest organizational citizenship behavior is in line with previous research concerning the relationship between organizational behavior with affect. The selffulfilling type itself is represented by high positive affectivity and low negative affectivity (Dalal, Lam, Weiss, Welch, & Huiin, 2009;Norlander et al., 2002), and it was found that positive affect is positively correlated with OCB and negative affect showed a positive correlation with CWB (Counter-productive Work Behavior). These findings are in line with the findings in the present study, since CWB is the opposite of OCB. In a similar direction, Janssen, Lam, and Huang (2010) found that positive affect correlated positively with OCB-I and OCB-O, meanwhile negative affect showed a negative correlation with OCB-I and OCB-O.
Based on the above explanation, lower positive affectivity leads to lower OCB. This is confirmed by the findings of the present study. Two personality types with low positive affectivity, namely self-destructive and low affective respectively, indicate the lowest levels of OCB. The low level of OCB for the self-destructive group is also explained by the high levels of its negative affectivity. These findings convincingly show that high positive affect combined with low negative affect lead to high OCB, while low positive affect in combination with high negative affect produce low OCB. The findings are consistent with Arnten et al.'s study (2008) which demonstrated that self-destructive groups and low affective groups had the highest negative affectivity compared to other personality types. In the context of organizations and working environment, negative affect is associated with deviant behavior (Appelbaum, Iaconi, & Matousek, 2007) such as arriving late to work, leaving early without any notice, claiming sick leave despite being healthy, insulting co-workers in front of other people, etc. This kind of behavior obviously contradicts OCB dimensions like loyalty, helpfulness and compliance. However, what came as a surprise was that the high affective group indicated not only high levels of negative affectivity but also high OCB (second after self-fulfilling). In relation to this, Fredickson (2001) explains that positive emotions may undo the aversive impacts of negative emotions, as documented in the Broad and Build Theory. As a consequence, the adverse impacts of negative affect towards OCB may be undone by the high positive affectivity from high affective individuals.

Table 1. Organizational Citizenship Behavior in the Four Types of Affective Profile, Mean and Standard Deviation
The findings in the study have confirmed the importance of affectivity in the organizational context and certainly the implications on organizational life. Taking into consideration the influences of positive affect and negative affect, organizations may have to make adjustments to increase workers' experience of satisfaction and positive affect, and to reduce situations that may lead to stress. This may relate to job designs, work environments, transformational leadership style and interaction with others, which can all be tailored to workers' needs. The findings of this study contribute to the validity of the notion that affective personality types are influential towards organizational behavior such as OCB. This has implications for the practice of personnel selection, where the Self-Fulfilling type of personality shows the best OCB.
The study also found that length of service was correlated with OCB. The longer the employees work in the company, the higher their OCB is. Presumably, the longer the employees work at a company, the more meaningful their lives are since their loyalty towards the company is higher. People may feel their lives have been supported by the company, that they have more friends in the company, and that they have more knowledge in mastering their work. Having a meaningful life, being a member of a big family, and getting more experience from work may lead to stronger commitment to their organization. These findings are in line with the concept of Three M (Meaning, Membership and Mastery) as factors that motivate people to work (Kanter, 2001). Another finding indicated that the employees' gender influenced OCB. Male employees showed higher OCB. One possible reason for this difference is the nature of work at power plants that deal with electricity supply, that in Indonesian context is deemed to be more appropriate for male employees than female employees.

Conclusion
The findings of this study confirm that positive affect plays an important role in creating high organizational citizenship behavior. The self-fulfilling personality characterized by high positive affectivity and low negative affectivity produces the highest organizational citizenship behavior. Meanwhile, two other personality types with low positive affectivity, namely selfdestructive and low affectivity, both produce the lowest level of organizational citizenship behavior. The low level of organizational citizenship behavior for the selfdestructive group is caused by the high level of negative affectivity.
Limitations. One limitation of this study is that the measurement of Negative Affect was not specific enough to make it more powerful in influencing the impact of negative affect on OCB. Since the four categories of Affective Personality were divided based on several combinations of Positive Affect and Negative Affect, using a more specific item of negative affect may have led to different results. In addition, there was some concern over the generalizability of this study. The study has limitations in terms of its external validity because it analyzed one company only, in which the number of female subjects was under-represented. There were 76 female subjects out of 226 subjects. In addition, the design of the study is cross sectional, not a longitudinal design. In order to get more convincing results it is necessary to conduct more studies covering different types of organization with more respondents, using both cross-sectional and longitudinal designs.
Future Research. The findings of this study suggest that more research needs to be done because no significant differences between the high affective group and the low affective group were found. Apparently, the reason for this insignificant difference is that the personalities of the participants in these two categories were not emotionally stable. Further research must include personality characteristics such as emotional stability or other variables in the Big Five personality traits as mediating variables. In addition, the measurement of negative affectivity must use more specific items rather than general items to make it more powerful in predicting the impact of affective personality types on the OCB. In addition, further studies are necessary to assess the impacts of affective personality on other organizational behavior such as work engagement (Schaufeli, 2017;van Wijhe, Peeters, Schaufeli, & van den Hout, 2011), and organizational commitment as proposed by Allen & Meyer (1990, 1993. In addition, further studies may need to be done using an experimental approach by inducing positive and negative affects, to see whether affective personality will correlate with real affect experienced by subjects and how it impacts OCB.