1. Introduction
Relationships are an important source of life satisfaction and emotional well-being. Likewise, and almost without exception, theories of psychological well-being include positive relationships with others as a core element of mental health and well-being.
Most theories view satisfying relationships as a determinant of well-being, while some of them go so far as to posit “positive relations with others” as an intrinsic component of psychological well-being and not just as a cause of it (Keyes, 2002; Ryff, 2014). In the theoretical model of well-being called PERMA, Martin Seligman (2011) focuses on positive relationships as an essential connective piece of the puzzle of well-being, which also is facilitated by experiencing positive emotions, engagement, sense of meaning and accomplishment. In her book, Positive Relationships, Roffey (2012) describes relationships as the heart of a person’s life and explains that we rely on others for care, love, laughter, and a shoulder. Seligman (2011) describes how relationships can turn from positive to negative and how important it is to identify, engage in and nurture the true positive relationships that can enhance well-being. Roffey (2012) pinpoints the need to understand the nature of a positive relationship in order to build them so as to feel a sense of value for ourselves and others.
In recent years, there is also an increase in empirical studies that examine the role of deep and meaningful relationships to every aspect of human life, from mental health and well-being (e.g., Chandoevwit & Thampanishvong, 2016; Chu et al., 2010; Keyes, 2005, 2010; Krause & Hayward, 2015; Lyubomirsky et al., 2005; Ryff & Singer, 2008; Seligman, 2002; Tan et al., 2017; Vaillant, 2012) to morbidity and mortality (e.g., Holt-Lunstad & Smith, 2012; Kawachi & Berkman, 2001; Lakey & Cronin, 2008; Uchino, 2009; Vaillant, 2012). Most of them show that people, who have more supportive and rewarding relationships have better mental and physical health and higher levels of subjective well-being.
In fact, the integration of individuals into support networks provides them resources needed to successfully cope with depression, anxiety, loneliness, obesity, alcohol abuse, smoking and many other physical and mental health difficulties. To add more, having a positive network of people has another significant impact on well-being. A person’s chance of being happy increases by 15%, if he is related to a happy person, and by 10% if he is related to a person, who is related to a happy person. Therefore, studies around social networks argue that happiness is transmitted through positive relationships (Fowler & Christakis, 2008).
The positive impact of relationships has been examined in all stages of the life span. Positive friendships contribute to better adaptation, development and well-being during childhood (Kornieko & Santos, 2013; Laursen et al., 2007), while one positive relationship is capable to diminish the negative consequences of a negative relationship and provoke a sense of balance among teenagers (Wang & Eccles, 2012). Teenagers, who lack positive relationships, are less likely to receive emotional support during stress periods, which makes them more vulnerable to the negative consequences of conflicts. Emotional support, in general, provides teenagers with the necessary resources to confront and overcome stress and diminish the probability of dangerous behaviours (Telzer et al., 2015). The quality of relationships at the age seems to increase even if they are chatting (Best et al., 2014). On-line friends have also a positive impact on well-being of teenagers, which becomes bigger as the number of on-line friends increases (Wang et al., 2014). The existence and number of on-line friends influence the perceptions of teenagers about social support, their status and sense of being a member of a group, which consequently impact their wellbeing.
During adulthood, a positive relationship depends on the balance between autonomy and intimacy (Merz & Huxhold, 2010). Support may have a different impact on adults regarding the person, who provides it. There has been empirical evidence that high levels of practical support may have a negative impact on people’s wellbeing as it is related to a loss of autonomy, an increase of needs and a sense of dependence, while emotional support is positively related to well-being (Reinhardt et al., 2006). Those, who perceive their relationships as positive and of high-quality tend to feel more intimacy and closeness and succeed to keep a balance among their autonomy and their interdependence.
Adult friendship is one of the most valuable interpersonal relationships (Demir, 2015), which contributes in various ways to individual’s well-being (Demir & Weitekamp, 2007; Diener & Biswas-Diener, 2008; Pezirkianidis, 2020) and significantly fulfils the fundamental human need for social interaction (Lyubomirsky, 2008). The friendship components that have been found to correlate with higher levels of well-being are stimulating companionship, support, reliable alliance, and self-validation (Demir & Weitekamp, 2007; Pezirkianidis, 2020). Also, marriage, as the “ultimate” positive relationship, seems to significantly impact well-being through the life span. It seems that the ability of spouses to communicate effectively and to show empathy and emotional support to each other works beneficially for their well-being (Verhofstadt et al., 2008). In a longitudinal survey, where 24,000 subjects were assessed for a period of 15 years, married were significantly happier during the whole period of their marriage, whether they were at their first or second marriage, in comparison with singles, divorced or widowers (Lucas et al., 2003).
At last, among elderly people, positive friendships are positively related to life satisfaction and negatively related to depressive symptoms, especially among women (Antonucci, 2012; Park & Roh, 2013). In particular, research shows that older people with positive relationships report higher levels of well-being, since their ability to regulate their negative emotions regarding the loss of internal and external resources is enhanced (Urry & Gross, 2010). In addition, positive relationships are positively associated with maintaining the physical health of the elderly (Lachman & Agrigoroaei, 2010). For example, they are negatively associated with the development of senile dementia and Alzheimer’s disease (Fratiglioni et al., 2004), as well as high levels of cognitive impairment (Fu et al., 2018). In other words, positive relationships are a determinant of the well-being of the elderly (Zielińska-Więczkowska et al., 2012).
Most of the above studies are focused on the importance of supportive relationships in the context of stress and adversity giving emphasis on how relationships act protectively for individuals and foster psychological resilience (Pezirkianidis & Stalikas, 2020). Nevertheless, there is also evidence that relationships may have an impact in well-being even in the absence of specific stressors (Lakey & Orehek, 2011). Furthermore, Feeney and Collins (2015) support that relationships can protect and enhance well-being by enabling individuals to fully participate in life’s opportunities for growth and development in the absence of adversity and enhance their efforts to learn, grow, explore, achieve goals, cultivate new talents, and find purpose and meaning.
Despite the frequent use of the term “positive relationships”, little is known about the exact meaning of it. The term “positive” is usually used as a synonymous to close, meaningful, supportive or important. At the same time, it is used for relationships with others that contribute to individual’s well-being. Social scientists often describe close relationships as those with whom one discusses important issues or spend his free time with. Half of them are friends while the others are spouses, lovers, parents, brothers and sisters but also colleagues, neighbours, coaches and counsellors. However, the qualities and content of a relationship characterized as positive have not yet been fully examined and described (Mertika et al., 2020).
A qualitative study has found through thematic analysis that positive relationships are characterized by three components: self-growth, emotional support, and readiness to support (Mertika et al., 2018; Mertika et al., under preparation). The researchers found that positive relationships enhance individual’s self-growth by assuming other’s responsibility, learning through modelling, having shared values, confronting fears, and discovering personal talents and strengths. Secondly, they found that positive relationships provide high levels of emotional support through stimulating companionship, experiencing interpersonal positive emotions and gratitude, and putting faith and motivating to the other. Thirdly, it was found that positive relationships are characterized by reciprocal readiness to support the other in the future. This means that both members of the relationships feel safe and secure and they perceive that the other will be available to provide support, whenever they need it. Also, Mertika and colleagues (2018, under preparation) have identified two processes within positive relationships. The first concerns a mutual and equal communication process that enhances acceptance of the other and the second concerns a change over time in proximity and autonomy (positive relationships combine high levels of both).
The aim of this article is to develop and pilot validate the Positive Relationships Questionnaire (PRQ), a novel questionnaire that has been specifically constructed to measure the qualities and content of positive relationships. The construction of this questionnaire will be a first step in the understanding of the term “positive relationship”. Also, as mentioned above, several studies and wellbeing theories talk about positive relationships, without defining them, describing their qualities and functions or suggest a way to measure them. This study will make the first attempt to address these gaps.
2. Method
2.1. Sample
The sample of the present study consisted of 331 adults, 228 women (68.8%) and 103 men (31.3%), aging from 17 to 67 years old. The mean age of the total sample was 37.15 years (SD = 12.38). Most of the participants were unmarried (48%), without children (59.8%) and employed (66.46%). Regarding their educational level, 19.63% were high school graduates, 10.27% university students, 38.67% university graduates and the others had different educational status.
2.2. Measures
The Positive Relationships Questionnaire (PRQ)
The development of the PRQ was based on the results of a qualitative study (Mertika et al., 2018). The aim of that study was to explore in-depth the specific characteristics of positive relationships and suggest an operational definition with exploratory interviews.
The PRQ includes 35 self-report items. The participants are asked to bring to their minds the important relationships they have in their lives, regardless of the context in which they were formed e.g. friendships, work relationships and/or family relationships. The first 12 items measure the characteristics of the relationship. The participants report, if they feel specific emotions and support in their lives, using a 10-point Likert-type scale (1-not at al/never to 10-absolutely/always). The following 20 items measure the characteristics of the persons involved and the participants report, if they are related with persons that make them feel positive emotions, are supportive and contribute to the relationship using the same 10-point Likert-type scale. The last three items ask participants to report 1) the number of the relationships they have in their lives with either one or some of the above characteristics at a high level, 2) if they have a relationship that contributes to their happiness and 3) if so, how many relationships of this quality (i.e., contributing to their happiness) they currently have.
PERMA Profiler (Butler & Kern, 2016; Greek version: Pezirkianidis et al., 2019).
The PERMA Profiler is a 23-item multidimensional questionnaire, which measures the five components of wellbeing based on Seligman’s (2011) theory of flourishing: positive emotions, engagement, positive relationships, meaning in life, and accomplishments. Each component is measured by three items and eight items are fillers assessing overall wellbeing (single item), negative emotions (three items), loneliness (single item) and physical health (three items). An overall wellbeing score can be also computed by summing the items of the five components of wellbeing and the overall wellbeing single item. Each item is scored on an 11-point Likert-type scale anchored by “0 = never” to “10 = always”, “0 = not at all” to “10 = completely”, or “0 = terrible” to “10 = excellent” based on the item content. The Greek version of PERMA Profiler, based on a sample of 3.696 participants demonstrated good concurrent, convergent and discriminant validity, acceptable internal consistency and test-retest reliability for the overall wellbeing items and for almost all the domains. It is a reliable and valid measurement of wellbeing as a total and of its main components.
2.3. Procedure
All the measures were administrated in an electronic Greek version of PRQ using google forms. Research data was acquired 1) with the assistance of students of the Panteion University of Social and Political Sciences, 2) from social media and 3) a webpage that was created specifically for the study of positive relationships. The university students got trained to recruit adults of their social milieu without providing them any external incentives or compensation. The participants took part in the study following a piece of brief information about study aims and anonymity of their responses and they provided informed consent.
3. Results
3.1. Item Analysis
We estimated means and standard deviations for the first 32 items of the scale in order to examine item quality and locate possibly dysfunctional items. The item analysis did not reveal any item that if deleted it would contribute significantly to the reliability of the scale. The mean of each item ranged from 7 to 8.6 (SD ranges from 1.6 to 2.3), which is high for an anchored 10-point scale but quite acceptable and expected as the scale measures qualities and characteristics of positive relationships, which are supposed to be beyond the mean so as the relationship to be a positive one (see Table 1). We also carried out a correlation analysis between all scale items. Item analysis revealed Pearson r correlations between the 32 items that ranged from .32 to .84 (see Table 2).
3.2. Factor Analysis
In order to examine the factorial structure of the scale we proceeded to Principal Components Analysis. Four factors with an eigenvalue over 1 were extracted using Principal Components Analysis with Varimax Raw Rotation. According to the analysis and the Kaiser criterion (eigenvalues greater than 1) four factors were extracted with loadings higher than .50. These four factors explained the 75.6% of the total variance and were named as follows: Self Improvement (19.7%), Practical Support (11.7%), Emotional Support (28.1%) and Shared Enjoyment (16.1%). Table 3 shows the results of Principal Components Analysis.
The first factor consists of eight self-improvement items, the second of three practical support items, the third of fifteen emotional support items and the fourth of six shared enjoyment items. Five items had double loadings. The item 16, “I have relationships with people that make me happy”loaded both in Self Improvement and Emotional Support, the item 18, “I have relationships with people,who make me laugh” to Self-Improvement and Shared Enjoyment and items 23, 27 and 29, “I have relationships with people to whom I offer practical support”, “I have relationships with people I understand very well”, and, “I have relationships with people that when we argue I know we will find a solution” all loaded to Emotional Support and Shared Enjoyment. We put these items where they loaded most taking also into consideration the meaning of the item (see Table 3).
![]()
Table 1. Item analysis Means and SD per item of PRQ.
![]()
Table 2. PRQ inter-item correlations (n = 331).
Note. Every correlation is significant at p-value < .001. n = sample size,p-value = value of statistical significance.
![]()
Table 3. PRQ item loadings per factor.
3.3. Reliability
We estimated the scale reliability using Cronbach alpha index. According to the analysis, the Cronbach Alpha for the scale was a = .93, for the factor of Self-Improvement was a = .94, for the factor of Practical Support was a = .90, for the factor of Emotional Support was a = .96, and for the factor of Shared Enjoyment the Cronbach Alpha was a = .93. Further item analysis exploring the possibility to strengthen the scale reliability if any of the items was deleted gave negative results.
3.4. Concurrent, Convergent and Discriminant Validity Analysis
In order to further examine the validity of the scale we used as criteria the PERMA Profiler factors. We hypothesized that the presence of positive relationships in life would correlate positively with the seven subscales of the PERMA profiler: positive emotions, engagement, relationships, meaning, accomplishment, physical health and happiness. We also hypothesized that they would correlate negatively with the subscales of negative emotions and loneliness. As hypothesized, PRQ showed 1) high positive correlation with Positive Emotions (r = .60, p < .001, n = 331) and Relationships (r = .60, p < .001, n = 331) indicating adequate convergent and concurrent validity, 2) moderate positive correlations with Engagement (r = .51, p < .001, n = 331), Meaning (r = .48, p < .001, n = 331), Accomplishment (r = .41, p < .001, n = 331), Physical Health (r = .36, p < .001, n = 331) and Happiness (r = .55, p < .001, n = 331) indicating adequate convergent validity, and 3) low negative correlation with Negative Emotions (r = −.11, p = .032, n = 331) and Loneliness (r = −.22, p < .001, n = 331) indicating adequate discriminant validity.
We also examined the correlation of each factor with the subscales of the PERMA Profiler (see Table 4). Self-Improvement factor showed high positive correlation with Positive Emotions (r = .61, p < .001, n = 331), Engagement (r = .54, p < .001, n = 331), Relationships (r = .76, p < .001, n = 331) and Happiness (r = .55, p < .001, n = 331) and moderate positive correlation with Meaning (r = .49, p < .001, n = 331), Accomplishment (r = .43, p < .001, n = 331), and Physical Health (r = .31, p < .001, n = 331).
Practical Support showed high positive correlation with Positive Emotions (r = .41, p < .001, n = 331) and Relationships (r = .55, p < .001, n = 331), moderate positive correlation with Engagement (r = .32, p < .001, n = 331), Meaning (r = .33, p < .001, n = 331) and Happiness (r = .39, p < .001, n = 331) and low positive correlation with Accomplishment (r = .25, p < .001, n = 331) and Physical Health (r = .28, p < .001, n = 331).
Emotional Support showed high positive correlation with Positive Emotions (r = .61, p < .001, n = 331), Relationships (r = .80, p < .001, n = 331) and Happiness (r = .53, p < .001, n = 331), moderate positive correlation with Engagement (r = .46, p < .001, n = 331), Meaning (r = .47, p < .001, n = 331), Accomplishment (r = .40, p < .001, n = 331) and Physical Health (r = .36, p < .001, n = 331).
![]()
Table 4. Correlations between PRQ factors and PERMA factors (n = 331).
Note. *p < .05, **p < .001.
Finally, Shared Enjoyment showed high positive correlation with Positive Emotions (r = .54, p < .001, n = 331) and Relationships (r = .70, p < .001, n = 331), moderate positive correlation with Engagement (r = .47, p < .001, n = 331), Happiness (r = .49, p < .001, n = 331), Meaning (r = .39, p < .001, n = 331), Accomplishment (r = .36, p < .001, n = 331) and Physical Health (r = .33, p < .001, n = 331).
All the four factors showed low negative correlation with negative emotions and loneliness.
3.5. Analysis of the Last Three Items
The last three items of the PRQ were analyzed separately due to their different form, they are rather open-ended questions and are not scored on a Likert scale. The participants reported that they have a mean of four relationships in their lives with either one of some of the characteristics described in the items 1 to 32 of PRQ (M = 4.23, SD = 3.35). Most of the participants (97.1%) reported that they have a relationship that contributes to their happiness and similarly to the previous question they reported a mean of four relationships that contribute overall to their happiness (M = 4.23, SD = 3.35).
Finally, in order to examine the number of relationships required in one’s life to have a high score in the PRQ, which indicates good quality and richness of positive relationships, we performed ROC analysis. Using the individual scores on the four factor scales and the total score, participants were divided into two groups, those who scored higher than 60 (i.e. MEAN + 1 SD) and those who scored lower than that. It was assumed that scores higher than MEAN + 1 SD indicate a higher state of “Emotional Support”, “Practical Support”, “Self Improvement”, “Shared Enjoyment” and overall higher “Total” positive relationships. Using this classification as a “golden standard” we performed ROC analysis in order to assess the number of relationships required to reach such a state and be happy. Our results indicated that at least three relationships were required for emotional support (AUC = .657, z = 3.370), four relationships were required for practical support (AUC = .629, z = 3.167), and five relationships were required for shared enjoyment (AUC = .668, z = 3.370) and self improvement (AUC = .599, z = 2.273). A total of at least two positive relationships (AUC = .634, z = 2.934) were required in order for the participants in our sample to have high ratings of strong positive relationships. However, although all the above results were statistically significant the AUC values ranged between .6 - .7, which is considered moderate.
4. Discussion
The purpose of the present study was the development of a novel questionnaire specifically constructed to measure the qualities and content of positive relationships. The initial findings are promising since the new instrument has robust psychometric properties. In terms of its reliability, overall and factorial reliability are satisfactory and item analysis and item inter-correlations indicate that all items contribute significantly to the reliability of the scale. Admittedly, the scale yielded consistently high ratings resulting in positively skewed results. It is important though to note that high ratings are both acceptable and expected because the scale measures qualities and characteristics that ideally exist in surplus.
The factorial structure of the PRQ echoes the results of previous studies on social relationships and their usefulness. The PRQ factors reflect two basic functions of positive relationships that have been described in the existing literature (Feeney & Collins, 2015). More specifically, our questionnaire includes two factors (Emotional and Practical support) that describe social support usually needed during times of adversity and two factors (Self-Improvement and Shared Enjoyment) that foster self-growth and are probably most useful during more quiet and serene periods of a person’s life. According to previous findings, social relationships provide four types of social support, which depict PRQ factors. First of all, social relationships provide emotional support to individuals. In other words, they provide sympathy, concern, love, trust, acceptance, intimacy, and affection. Emotional support is also called “support of self-esteem” (Fehr & Harasymchuk, 2018; Li et al., 2014). The second type of support is called practical support and concerns the provision of financial assistance, material goods or services to help the other, when needed (Nguyen et al., 2016). Thirdly, social relationships provide informational support, which is about giving advice, guidance, suggestions or useful information to a person in order to solve a problem (Fehr & Harasymchuk, 2018; Wood et al., 2015). Finally, relationships with others provide companion support. This form of support refers to the presence and companionship in joint social activities or in individual activities, where the other person is present, and enhances the sense of belonging (Yang et al., 2014).
Overall, the PRQ’s stronger factor is the one that addresses emotional support. The “Emotional Support” factor has 16-items, more than any of the other three factors. The observed imbalance is though fully understood and explained considering the predominant view that relationships have stress-alleviating qualities. Merz and Huxhold (2010) suggest that relationships promote well-being by providing intimacy. The term intimacy describes a state in which the individuals, who are in a social relationship enjoy and share feelings of safety, trust, and acceptance; oftentimes at the expense of personal independence (Merz & Huxhold, 2010). The same finding is supported by the results of a previous qualitative study, where authors highlighted a multitude of emotional states and feelings that are experienced in a relationship that is considered positive and meaningful by the participants (Mertika et al., 2018, under preparation).
In terms of PRQ’s convergent validity, the results further indicate that the PRQ measured the quality of positive relationships and this measurement was indeed highly correlated with current ratings of well-being components measured by the PERMA Profiler. Previous literature support this finding, since perceived social support from positive relationships has been found to predict wellbeing levels more than any other variable (Chandoevwit & Thampanishvong, 2016; Chu et al., 2010; Demir, 2010; Krause & Hayward, 2015; Tan et al., 2017).
More specifically, the findings of the present study indicate that the quality of positive relationships positively correlates with the experiencing of positive emotions, a sense of meaning in life, engagement in everyday activities, accomplishments, and physical health. The aforementioned findings are in accordance with current literature. Feeney and Collins (2015) support that relationships can protect and enhance well-being by enabling individuals to fully participate in life’s opportunities for growth and development in the absence of adversity and enhance their efforts to learn, grow, explore, achieve goals, cultivate new talents, and find purpose and meaning. Also, the findings of previous studies support that daily experiences of positive emotions increase social support levels, the sense of connection with others, trust, and relationship quality (Dunn & Schweitzer, 2005; Fredrickson et al., 2008; Gable et al., 2006; Waugh & Fredrickson, 2006). Moreover, previous findings support that 1) the sense of belonging in relationships with others leads to a life with a strong sense of meaning (Sørensen et al., 2019; Zhang et al., 2019), 2) social support correlates with high levels of engagement to everyday activities in society and academic settings (Heck & Fowler, 2007; Lascano, 2016), 3) positive relationships with others lead to achieving goals in school, work and personal life (Bronkema & Bowman, 2017; Koestner et al., 2012; Liu et al., 2013; Silvy et al, 2009), and 4) positive relationships predict longevity (Holt-Lunstad & Smith, 2012).
Furthermore, PRQ demonstrated adequate levels of concurrent validity, based on its high positive correlation with other existing instruments of measuring positive relationships (namely PERMA Profiler). In terms of discriminant validity, the PRQ exhibited negative correlations with loneliness and negative emotions as it was expected, since social relationships characterized by support have been found by the existing literature to diminish the experiencing of negative emotions, undo their unpleasant effects and reduce loneliness effects on depression and wellbeing (Fredrickson et al., 2008; Gariepy et al., 2016; Liu et al., 2016; Pierce & Quiroz, 2019). It is important, here, to reiterate that existing literature underscores the fact that social relationships act as a buffer to stress and anxiety and are seen as a powerful mediator that leads to resilience and improved well-being even in periods of hardship and suffering (Pezirkianidis & Stalikas, 2020; Rueger et al., 2016; Mohd et al., 2019).
In conclusion, the findings of the present study are a first step in understanding the term “positive relationships”, their qualities and characteristics. However, there are some limitations in the present study. Mainly the method of sampling, since a part of the sample was self-selected and probably more eager to provide information about their relationships. Also, a random sampling method did not take place. Yet our findings indicate that PRQ is a reliable and valid questionnaire and future studies should focus on further examining its psychometric properties, confirming its factorial structure and explore the effects of positive relationships on well-being.
As positive psychology blooms, more and more people and practitioners focus on the cultivation of positive relationships as an integral component of wellbeing; this instrument will provide a unique and useful tool for evaluating and understanding the quality and content of positive relationships and will facilitate interventions aiming at the enhancement of psychological flourishing through building positive relationships. In the era of COVID-19 pandemic, the studies that focus on understanding the positive relationship characteristics in different samples and their impact on well-being components are of high social and clinical importance. Also, a better understanding of positive relationships will facilitate the design of novel interventions that can be useful tools for mental health promotion in different settings, like schools, individual and group counselling, and psychotherapy.