Foundations-4

THE JOURNAL OF POSITIVE PSYCHOLOGY, 2018 VOL. 13, NO. 5, 522539 https://doi.org/10.1080/17439760.2017.1414298

Joy is a distinct positive emotion: Assessment of joy and relationship to gratitude and well-being

Philip C. Watkinsa, Robert A. Emmonsb, Madeline R. Greavesa and Joshua Bella

aDepartment of Psychology, Eastern Washington University, Cheney, USA; bDepartment of Psychology, University of California, Davis, CA, USA

ABSTRACT In three studies we investigated joy and its relationship to subjective well-being (SWB). We developed measures of joy based on recent conceptualizations of joy in the humanities and social sciences. In Studies 1 and 2 we developed reliable measures of state and trait joy. In Study 3 we used a two- month prospective design to investigate the relationship of joy to gratitude and SWB. We found that dispositional gratitude predicted increases in state joy over time. We also found that trait joy predicted increases in state gratitude, providing evidence for an intriguing upward spiral between joy and gratitude. Finally, we found that trait joy was associated with increases in SWB over time. Factor analyses indicated that joy loaded separately from other positive emotions. We conclude that joy is a discrete positive emotion, it can be measured reliably with self-report instruments, and that it may be an important component of well-being.

When baseball’s loveable losers – the Chicago Cubs – won the World Series in 2016, joy was the one word that perhaps best captured the reaction of Cubs’ supporters. The sobbing response of Dina Mansaour was typical of many Cubs fans: ‘I can’t even explain how I feel right now. This is so epic. I’ve waited my whole life for this. This is the most amazing feeling in the world’ (McLaughlin & Palmer, 2016). It had been 108 years since the Cubs had last won the championship, and Jason Nye’s reaction echoed that of Mansaour: ‘I’ve been waiting my whole life for this, my grandfather has been waiting his whole life for this, everyone has been waiting for this’ (McLaughlin & Palmer, 2016). Longsuffering Cubs fans had indeed waited long for this event, and joy seemed to be the dominant response. Although joy appears to be essential to the human con- dition (Vaillant, 2008), outside of a handful of studies, this emotion has rarely been empirically examined as a discrete emotional state. In an attempt to fill this vacuum, our pur- pose in these current studies is to develop valid measures of joy, and to investigate the relationship of joy to gratitude and well-being.

Need for joy research

In his treatise on spiritual emotions, Vaillant stated, ‘We cannot understand human beings unless we understand joy and how joy comes to be’ (Vaillant, 2008, p. 131).

Indeed, the understanding of joy has been identified as important to the goals of positive psychology (Fredrickson, 1998, 2001; Seligman, 1998; Watkins, 2016). Moreover, joy is invariably included on notable lists of basic emotions (e.g. Ekman, 1992; Fredrickson, 1998; Frijda, 1986; Lazarus, 1991; Plutchik, 2003; Tompkins, 1962). Because of the importance of positive emotions to subjective well-being (SWB) (Myers & Diener, 1995), it would seem that under- standing joy is crucial to grasp human flourishing. But despite the apparent importance of joy, it appears to be the least studied of the positive emotions (Vaillant, 2008).

Joy as a discrete positive emotion

Why has psychological science, and more particularly pos- itive psychology, neglected the systematic study of joy? First, it may be because joy is often used as a synonym for happiness, and thus joy is usually treated as a non-specific positive affect (e.g. Ellsworth & Smith, 1988). Because of this, the appraisal most often proposed for joy is interpret- ing any situation as indicating progress toward or comple- tion of a personally relevant goal. For example, Fredrickson (1998, p. 304), states that joy results from events that are ‘construed as accomplishments, or progress towards one’s goals.’ One issue here is that ‘progress towards one’s goals’ may be viewed as the basic appraisal of any positive emo- tion, and thus joy is reduced to any positive emotional

© 2017 Informa UK Limited, trading as Taylor & Francis Group

KEYWORDS Joy; gratitude; subjective well-being; happiness

ARTICLE HISTORY Received 16 October 2017 Accepted 7 November 2017

CONTACT Philip C. Watkins [email protected]

THE JOURNAL OF POSITIVE PSYCHOLOGY 523

done something important for me’), pride (‘I have achieved something of value’), and interest (‘there’s something new about this situation that might be important to me’) seem fairly straight forward compared to joy. We have seen that the object that activates joy is appraised as good, but what makes it good to the individual? People often experience joy when they receive good news, but what makes the news good? Here is where a discussion of the appraisal structure of joy is crucial.

First, there is an implication that the ‘good news’ is news that one has been waiting for. In this sense, the news is good because it in some way fulfills the individual’s hope (Wright, 2015). Many approaches to joy have emphasized human longing: we experience joy when we are longing for something and we finally receive news that it’s coming to be (e.g. Lewis, 1955; Vaillant, 2008). Thus, joy is not the emotional response to all things encountered that one deems good, but rather one is more likely to experience joy in response to a hoped for event. Even though a latte may give us pleasure, we probably won’t experience joy when consuming our daily latte because it is simply part of our daily routine, not something that we have been waiting for. On the other hand, when we’ve been apart from a loved one for some time and we have longed for their presence, joy is the probable response when the reunion occurs.

Second, the object of joy is deemed to be good because it indicates that we are, in some expansive way, blessed (Motmann, 2015). Thus, when an event suggests to us that we are favored, joy is the probable consequence. Volf expands on this idea by suggesting that the object of joy is ‘intentional’ (presumably from an outside source) and is viewed as ‘unowed’ (Volf, 2015). For example, in response to a raise given for excellent performance on the job, one might be likely to experience pride related emotions, but not joy. We included items related to this aspect that allowed us to evaluate whether the objects of joy are typically viewed as ‘unowed.’ Both Frijda (2007) and Mathewes (2015) have also argued that joy is more likely to be experienced when the good object is in excess of what one expected.

Third, theorists have emphasized that the object of joy is deemed good because we appraise the situation as indicating some kind of turning of fortune or a redemp- tive twist, (McAdams, 2006) where something good fol- lows on the heels of something bad. Redemption is ‘a deliverance from suffering to a better world’ (McAdams, p. 7). According to Vaillant (2008), prototypical joy events involve some kind of triumph after all seemed lost. Thus, Easter (resurrection after Christ’s death) and Israel’s Red Sea exodus (when Israel evaded certain death at the hands of the Egyptian army in their escape through the Red Sea), are viewed by Vaillant as prototypical joy events.

response. In this paper, we view joy to be a discrete pos- itive emotion, and seek to investigate it as a specific and distinctive positive emotional response.

Is there evidence that joy may be seen as a discreet positive emotion? Several studies have provided evidence that joy can be distinguished from other positive emo- tions both in appraisals and function (e.g. Bagozzi, 1991; de Rivera, 1977; de Rivera, Possell, Verette, & Weiner, 1989; Egloff, Schmukle, Burns, Kohlmann, & Hock, 2003; Roseman & Evdokas, 2004). Thus, suggestive evidence supports the theory that joy is a specific and distinct positive emotion, and deserves to be investigated in its own right.

What is joy? Recent theoretical proposals on joy

In order to pursue a systematic investigation of joy, we must have a clear understanding of exactly what it is. As Vaillant noted, ‘There is no easy definition of joy’ (2008, p. 134), and this may be one reason systematic research on joy has been absent. Recent scholarship in the theology of joy has resulted in advances in conceptualizing this emo- tion and how it may differ from happiness (Volf & Crisp, 2015). Almost all descriptions of joy in that literature state that joy is a response to a ‘good’ object – usually a positive event or circumstance. Different from mood states, joy as an emotional state is always about something, and usually this is news about something good in one’s life. Joy is a response to some good object. This does not mean that joy is completely beyond one’s control, however. Although the emotion of joy is in response to some good event, one can prepare for joy (see Mathewes, 2015, for a helpful dis- cussion of the ‘middle voice’ of joy). Thus, in responding to a positive event, some individuals may experience joy and some may not. We propose that the disposition of gratitude is one way that individuals can prepare for expe- riences of joy.

Most positive emotions are responses to positive events, so the important question is: how is the event peculiarly good for a response of joy to result? Most emotion theorists believe that the appraisal of an event is crucial to determining specific emotional responses, so here we discuss various theoretical suggestions for the appraisal structure of joy. First, we should be quick to emphasize that the construals of joy are not likely to be captured in a sentence or two. Unlike fear (‘There’s some- thing here now that might harm me’) or disgust (‘There’s something repugnant to me that I must get rid of’), the appraisal structure of joy is not likely to be so simple, and this might be another reason that joy has not been system- atically investigated in emotion science. It is possible that the appraisal structure of any positive emotion is complex, but we submit that joy may be unique in this respect. For example, the appraisal themes of gratitude (‘someone has

524 P. C. WATKINS ET AL.

Those involved with the recent integrative theoretical project on joy have emphasized other subjective aspects of joy. For example, Mathewes (2015) suggests that joy is self-conscious in that when we experience joy we are aware that we are joyful. He goes on to suggest that this self-consciousness is not to the extent of becoming self-consumed. Various theorists have also emphasized the sense of freedom in joy – particularly freedom to play (e.g. Moltmann, 2015). Finally, theorists have also argued that there is a sense of timelessness in joy; when we experience joy, time flies, time does not ‘drag on’ (Mathewes, 2015; Moltmann, 2015). The apparent timelessness of joy high- lights a spiritual aspect of joy: that of a sense of eternity.

Finally, theorists have proposed that the good news that leads to joy is good because it indicates that things are working out right, life is going well, in accordance with one’s perception of the purpose of their life and of the world (e.g. Volf, 2015). Thus, not just any positive event should result in joy, but only those events that seem to be consistent with one’s life plan. This proposition sug- gests several interesting implications. For example, only if one has a clear narrative identity would they be able to appraise a situation as corresponding to their evolving life story. Thus, one prediction from this theory is that joy is more likely to be experienced in individuals who have a clear narrative identity, especially one that contains ele- ments of redemptive suffering.

All of the above suggestions are important theoretical proposals for the appraisal structure of joy. We have found Vaillant’s (2008) approach to joy particularly helpful to developing a theory of the appraisal structure of joy. First, Vaillant proposes that, ‘Joy is connection’ (2008, p. 124). By this he means that joy is relational both in its appraisal and in its function. We experience joy when we appraise a situation as indicating increased connection with some- thing good, or appraising a connection with some good object that we did not previously feel connected to. Here, Vaillant refers to spring as an important metaphor for joy. When spring arrives we are reminded of the end of winter, and the beginning of new life. Thus, out of the deadness of winter, spring brings the good news that we will be connected to life again.

Furthermore, Vaillant theorizes that the construals of ‘reunion’ and ‘Separation and return’ (2008, p. 127) are central to the appraisal structure of joy. Thus, when we appraise a situation as indicating that we are returning to that which is good, we will experience joy. Importantly, Vaillant cites a parent and child game of ‘peek-a-boo’ as a prototypical joy interaction. The mother hides her face from the gaze of her child, only to peek back into the child’s gaze, which invariably brings bursts of joyful laughter in both parent and child. This example highlights Vaillant’s theory that joy can only be elicited in situations of secure

attachment. If the child is insecurely attached, the parent’s hidden face will not be a playful aspect of a joyful game, but only a threat that the parent may abandon her again.

Finally, Vaillant suggests that when we experience joy it is the result of an appraisal that we haven’t had a lot to do with the situation that has caused us joy. This is similar to Volf’s (2015) idea that joy is the result of an ‘intentional object’ that seems to require an appraisal of an external source. We interpret Vaillant as proposing that it is not that an external source appraisal is required for joy to result, rather it is simply that an internal attribution for the joy object is not salient, thus, the individual concludes that she doesn’t have a lot to do with the joy object.

Summarizing the important theoretical proposals of the appraisal structure of joy, we have seen that joy results from construals of: an intentional good object that is ‘unowed’, something that one has been longing for, something that indicates we have been blessed or favored, something that indicates a turning of fortune, and that life is working out well and that life is being well lived. We submit that Vaillant’s approach may most parsimoniously capture the appraisal structure of joy: when we appraise a situation as indicating a return to that which is good, we are most likely to experience joy.

What are the thought/action tendencies of joy? Various authors have proposed that when one experiences joy one wants to celebrate, and usually celebrate with others. Although the joyful celebration may be simply communi- cating one’s success with another, Wright (2015) empha- sized that joy activates a more physical celebration: when we experience joy we want to celebrate with clapping, singing, and dancing. Relatedly, celebratory play is often the result of joy (Moltmann, 2015; Vaillant, 2008). Frijda expertly describes this play as ‘free activation’; the thought/ action tendency of joy ‘is in part aimless, unasked-for read- iness to engage in enjoyments’ (1986, p. 89). Vaillant (2008) argues that one of the main functions of this celebratory play is connection – perhaps with others but with any important good. When we celebrate something with oth- ers this should provide for a strong social connection. The appraisal of separation and return results in joy, and thus for Vaillant, ‘Joy is the motivational system that reinforces return’ (p. 127).

Finally, recent theorizing has emphasized that joy results in spiritual longing: joy activates a desire for transcendence – when we experience joy we long for something beyond the self. Some have argued that joy is a spiritual emotion in that it is not closely associated with physical sensa- tion pleasures (Roberts, 2007), but others argue that the spiritual aspect of joy is in that joy encourages us to long for something transcendent (e.g. Thompson, 2015; Volf, 2015; cf. Lewis, 1955). We include items in our joy measures that relate to both aspects of the spirituality of joy.

THE JOURNAL OF POSITIVE PSYCHOLOGY 525

reason to believe that joy may be important for gratitude. People who are high on the affective trait of joy should be able to rejoice over many things, and should be able to find joy even in difficult situations. They should see the good in their life as coming from sources outside of themselves, and they should be able to see and celebrate being recon- nected with good things in their life, particularly reconnec- tions with important others. All of these may be conducive to experiencing gratitude. When we recognize gifts, rec- ognize the goodness of the gift, recognize the goodness of the giver, and recognize the gratuitousness of the gift, a grateful response is likely (Watkins, 2014). It is easy to see how a disposition for joy may enhance these appraisals, and thus increase the frequency of gratitude. Not only may gratitude be foundational for joy, here we can see how a disposition for joy may be conducive to gratitude. Thus, gratitude and joy may have a reciprocal relationship that results in an upward spiral that enhances well-being.

We have argued that joy may have be a uniquely spiritual emotion in that it promotes our longing for tran- scendence. Thus, in the current studies we investigated the relationship of joy to a number of spiritual qualities. Foremost among these qualities was gratitude to God (Scheibe, Watkins, & Uhder, 2017). For religious people, gratitude to God may be particularly important to their experience of joy. Likewise, a disposition for joy may be important for enhancing one’s gratitude for God.

Summary and overview of the studies

Despite the apparent importance of joy, systematic inquiry of this positive affect is virtually absent. In order to properly understand any phenomenon, we must first develop effec- tive measurements of the construct. Thus, the purpose of these studies was to develop valid measurements of joy as a state and trait, and to investigate the relationship of gratitude to joy. In Study 1 we develop a measure of the joyful state, in Study 2 we sought to develop a measure of trait joy. In Study 3 we utilized a two-month prospective design to investigate whether gratitude predicts increased levels of joy over time. In this study participants completed measures of state and trait joy and gratitude at two points, seven to nine weeks apart. We predicted that Time 1 levels of trait gratitude would predict increased levels of state joy at Time 2, after controlling for Time 1 levels of joy.

Study 1

In Study 1 we sought to develop a reliable self-report meas- ure of state joy, the State Joy Scale (SJS). We initially created 35 items in an attempt to assess the crucial aspects of joy as a discrete positive emotion. Because the good news of joy is not likely to happen frequently, we reasoned that

Another important insight from the recent integrative effort to understand joy (Volf & Crisp, 2015) is that joy can be experienced in the midst of trials and difficult events. There are various reasons proposed for this possibility. For example, Thompson (2015) proposes that joy can be experienced in the midst of trouble when one has the positive outcome of the trial in view. As Vaillant points out, pain is often intimately involved with experiences of joy: ‘Without the pain of farewell, there can be no joy in reunion. Without the pain of disapproval, there can be no joy in forgiveness. Without the pain of captivity, there is no joy in exodus’ (2008, p. 131). In the current studies, we tested these notions by including items in our joy meas- ures related to enduring joy and also experiencing joy in difficult circumstances.

The relationship of gratitude to joy

If joy is indeed a discrete positive affect important to the good life, it is important to explore factors that might be conducive to joy. We suggest that the disposition for grati- tude should be conducive to joy. Grateful people – people high in the disposition of gratitude – see all of life as a gift (Watkins, Woodward, Stone, & Kolts, 2003). This implies that they engage in psychological processes that may be important to the experience of joy. For example, people disposed to gratitude should look for the good in their life. In order to experience joy, one must of course notice the good that might bring them joy. Second, grateful peo- ple should interpret the good in their life in a benevolent fashion. Grateful people see positive things as good gifts; benefits given intentionally for their well-being. Although one may feel grateful for a benefit not intentionally given, research shows that when individuals interpret benefits as given intentionally for their well-being, a grateful response is more likely (see McCullough, Kilpatrick, Emmons, & Larson, 2001 for a review). When people tend to interpret benefits in a positive fashion, this should also enhance their experience of joy. For example, a grateful person should be more likely to notice benefits, and they should also be more likely to appraise them as valuable and as given from altruistic motives. All of these cognitive ten- dencies should also promote experiences of joy. Earlier we argued that people experience joy when their life is well lived: when they live in accordance to their true nature. Grateful people live in gratitude in a way that continually acknowledges that they are contingent beings: dependent on others and the earth’s resources. Because we are indeed contingent beings, gratitude helps us live in accordance to our true nature, and thus helps us live well in attunement with the world, which should result in joy.

One goal of the current studies is to test whether grat- itude is foundational for joy. However, there is also good

526 P. C. WATKINS ET AL.

to evaluate the construct validity of the SJS. In order to assess emotional state, we administered the short form of the PANAS (Watson, Clark, & Tellegen, 1988). Participants completed the PANAS with reference to the previous week. The PANAS is one of the most commonly used and well validated assessments of emotion. However, there are no items related to gratitude in the PANAS, so we included the three adjectives of the GAS (grateful, thankful, and appre- ciative; McCullough, Emmons, & Tsang, 2002). Because the PANAS has been criticized for only including adjectives for high activation emotional states, we also included several items related to positive and negative low activation states (sad, contented). We also added items that should be more directly related to joy, as indicated by our earlier analyses of the PANAS-X (joyful, happy, delighted, lively). Finally, we added two items related to Roberts (2007) idea of the spiritual nature of joy (pleasurable sensations, unpleasant sensations). If Roberts is correct, these items should show small to negligible correlations with the joy items.

For our eudaimonic well-being measure we used a short psychological well-being measure that has good psychometric characteristics: the Flourishing Scale (FS, Diener et al., 2010). Eudiamonic well-being refers to hap- piness as defined by ancient Greek philosophers. In brief, eudaimonic happiness relates to fulfilling one’s true nature – fulfilling one’s authentic human needs and capacities (Watkins, 2016). The FS provides for an assessment of the fulfillment of basic psychological needs, as well as other factors important to eudaimonic well-being. Participants respond to items such as ‘I actively contribute to the hap- piness and well-being of others’ on a 7-point agree/disa- gree scale. We also administered one of the most utilized measures of SWB: the Satisfaction with Life Scale (SWLS; Diener, Emmons, Larsen, & Griffin, 1985). We also included a spiritual well-being measure derived from Kashdan and Nezlek (2012). This measure included two items: ‘This week, the spiritual part of my life was very important to me’, and ‘This week, my relationship with a power greater than myself was important to me.’ Participants responded to these items on a 7-point Likert-type scale ranging from ‘very uncharacteristic of me’ to ‘very characteristic of me.’ This scale had excellent internal reliability (  = .940, r = .887). We also included two items related to gratitude toward God that we have found to reliably assess this con- struct (Scheibe et al., 2017): ‘This week, I felt very grateful toward God’, and ‘This week, I felt very appreciative of God.’ Participants responded to these items with the same scale as the spiritual well-being items, and this scale also showed excellent internal consistency (  = .989, r = .979).

In order to asses trait gratitude, we administered two of the most used and validated measures of dispo- sitional gratitude, the Gratitude Questionnaire (GQ-6; McCullough et al., 2002), and the Gratitude, Resentment,

participants would be unlikely to have just encountered this ‘good news’ immediately prior to taking the measure, thus we chose to use a one-week time frame rather than immediate momentary joy reports. First, we based several of our items on factor analyses of the PANAS-X from previ- ous data sets (McCurrach, 2015; e.g. ‘In the past week, how often have you felt joyful?,’ ‘In the past week, how often have you felt enthusiastic?,’ ‘I felt exuberant this week’). Second, we created several items related to theoretical appraisal aspects of joy (e.g. ‘This week, good things hap- pened to me I did not expect’, ‘This week, my life went well’, ‘This week, life just made sense to me’), and theoretical subjective and thought/action tendencies of joy (e.g. ‘In the past week, how often have you felt like celebrating?,’ ‘This week, I felt playful’, ‘This week, I felt the presence of something much bigger than myself’).

Method

Participants and procedure

We administered the initial joy measure along with other measures at three time points. Time 1 and 2 were admin- istered one week apart, and we administered the meas- ures at Time 3 four weeks after Time 2. Following are the demographics for the participants who completed the measures at all three time points: 33.33% were male, 64% were female, and 2.67% endorsed ‘other’ for gender. The age range for our participants was between 18 and 43 years old, with a mean of 22.38. At Time 1, 185 partici- pants completed the measures, 189 and 190 participants completed the measures at Time 2 and Time 3 respectively. All measures were administered in paper and pencil form in a group format. Participants were all university students in psychology courses, and received partial course credit in exchange for their involvement in the study. This study fol- lowed the ethical guidelines of the American Psychological Association (APA), and was approved by the IRB of Eastern Washington University.

Measures

Our primary measure was the SJS, the purpose of which was to provide a valid assessment of state joy, the frequency of the emotional experience of joy. The original instrument contained 35 items to which participants responded on a 7-point Likert-type scale. The first eight items referred to frequency of experience and thus were anchored ranging from ‘not at all’ to ‘frequently.’ The remaining items were in a agree/disagree format. Examples of these items may be found in the appendix which includes the final SJS.

We also included a number of questionnaires that served to investigate various theoretical proposals and

THE JOURNAL OF POSITIVE PSYCHOLOGY 527

theoretical components of joy tend to cluster together. Exploratory factor analysis using PCA showed that one dominant factor emerged explaining 52.8% of the vari- ance. Although four other factors achieved eigenvalues greater than 1.0, none of these factors explained more than seven percent of the item variance. This suggests however, that a number of items were not loading highly on the primary joy factor. From the original 35 items, we retained items that loaded highly on the primary factor and also appeared to provide a unique contribution to the assessment of state joy. Thus, after eliminating statistically and conceptually redundant items, we retained 11 items (see Table 1). Exploratory factor analysis of these 11 items revealed only one factor with an eigenvalue greater than one. This factor explained 65.17% of the variance and no other factor explained more than 6% of the variance. Thus, we submit that a one-factor solution best explains the 11 item final version of the SJS. Internal consistency of this version of the SJS was again quite good (  = .945). In Table 1 we present the items of the SJS along with factor weights and corrected item-to-total figures. In order to test tem- poral stability we conducted test-retest reliabilities across the three administrations. We found that temporal stabil- ity was what should be expected for a state measure that might be expected to vary somewhat over time (rs = .58, .53, .56). Finally, we also compared our shorter final form to the longer original form, and found that they were highly correlated (rs = .976, .977, .979), so researchers will prob- ably elect to use the shorter form.

Construct validity

Construct validity correlations with the SJS can be found in Table 2. As can be seen from these relationships, in general the pattern of correlations were as predicted. Specifically and most importantly, the strongest correlation was between the SJS and the joy subscale of the PANAS, and this association exceeded that of the general positive affect scale and the attentiveness subscale of the PANAS. The SJS was negatively related to PANAS negative affect. These correlations provide preliminary evidence that joy may be

and Appreciation Test, Short Version (GRAT-S; Watkins et al., 2003). The GQ-6 includes 6 items assessing one’s dis- position to experience gratitude (e.g. ‘I have so much to be thankful for’). Participants responded to these items on a 7-point Likert-type scale ranging from ‘strongly disagree’ to ‘strongly agree.’ The GRAT-S contains 16 items that theoret- ically assess an individual’s appreciation for the giftedness of life (e.g. ‘I think it’s important to appreciate each day that you are alive’). Participants responded to these items on a 9-point Likert-type scale ranging from ‘I strongly disagree’ to ‘I strongly agree with the statement.’

Finally, all participants completed the Cynicism and Lack of Trust Scale (CLOT; Floberg, Sestrap, Bart and Watkins, 2014). We reasoned that if one consistently questions the motives of those who provide them with social benefits, it would be difficult for them to respond with joy to benefits provided by others because rather than seeing favors from others as an indication of increased connection, cynical individuals may even see these benefits as further evi- dence of their disconnection with others. The CLOT was developed to assess local cynicism, i.e. how cynical indi- viduals feel recently about their immediate social circle. This instrument contains 23 items (e.g. ‘Lately, I’ve noticed that when others do something for me they often have ulterior motives’), and participants respond to the items on a 9-point agree/disagree format. The CLOT appears to have good psychometric properties supporting its use (Floberg et al., 2014). At all three assessments, the above questionnaires were administered in the following order: PANAS, FS, spiritual well-being, gratitude to God, SWLS, GRAT-S, GQ-6, CLOT, SJS.

Results and discussion

Development and reliability of the SJS

In the development of the SJS, our primary purpose was to develop a unifactorial measure of state joy. We first ana- lyzed the internal reliability of the 35 items and found that the measure was indeed internally consistent (  = .971). The fact that all items achieved a corrected item-total correlation above .40 provides some evidence that the

Table 1. Final Items and statistics for the State Joy Scale (SJS).

Item Factor loading Corrected item-total 1. In the past week, how often have you felt joyful? .811 .766 2. In the past week, how often have you felt enthusiastic? .858 .820 3. Because of the joy I experienced this week, time just seemed to fly .828 .788 4. This week I found myself enjoying something so much that I lost track of time .741 .690 5. This week I felt free .780 .730 6. This week, I felt ready to enjoy whatever opportunity presented itself .837 .795 7. Something happened this week that made me feel like celebrating .765 .713 8. This week, the reality of my life was the way I feel it should be .803 .755 9. This week, I felt free to play .795 .748 10. This week, my life went well .813 .767 11. This week, life just made sense to me .842 .797

528 P. C. WATKINS ET AL.

for religious individuals. However, the evidence did not support Robert’s (2007) idea of joy as a spiritual emotion. Because joy is spiritual emotion, Roberts reasoned that joy would not be associated with material pleasures. We found strong positive correlations between reports of pleasura- ble physical sensations and joy, contradicting this idea of a spiritual emotion. We believe that the spirituality of joy is likely to be incarnational, and thus will be experienced in a variety of pleasures, including physical pleasures. Whatever the case, these results call for more research on the spirituality of joy. In sum, we conclude that the SJS shows adequate validity for assessing the emotion of joy and researchers may find this to be a useful tool in joy research.

Study 2

The purpose of Study 2 was to develop a valid measure of trait or dispositional joy: the Dispositional Joy Scale (DJS). As clearly explained by Rosenberg (1998), emotion can be described at several different levels of analysis: emo- tional state, mood state, and affective trait. In Study 1 we developed an instrument that appears to be an accurate assessment of the emotional state of joy, and in Study 2 we attempted to develop a measure of trait joy. An affective trait refers to a person’s disposition for a particular emo- tional state. Thus, people high in the trait of joy should have a low threshold for joy; they should experience joy more frequently across a wider variety of circumstances. We are aware of only one measure designed to assess dis- positional joy, the joy subscale of the Dispositional Positive Emotions Scale (DPES; Shiota, Keltner, & John, 2006). The DPES attempts to measure seven different positive emo- tion dispositions, but questions have been raised regard- ing this measure about whether the subscales actually assess distinct emotional dispositions. Thus, it appears that a measure of dispositional joy is needed. Based on theoreti- cal propositions of joy and on preliminary analyses from six studies using the PANAS-X (McCurrach, 2015), we created 31 items we thought would assess the disposition of joy. These items were interspersed with items from the DPES, which was administered with other measures to assess the construct validity of the DJS. In addition to descriptions from the DPES that would obviously tap into trait joy (e.g. ‘I often feel bursts of joy’), we included items related to vari- ous facets of a theory of the joyful disposition (Volf & Crisp, 2015). For example, we included items relating to the the- ory that life is going well (e.g. ‘I consistently feel like my life is going well’), appraising positive events as blessings (e.g. ‘I often feel blessed’), the idea that joy may be experienced even in difficult circumstances (‘Even when things aren’t going well, I can still fell joy’), the notion that the objects of joy are ‘unowed’ (e.g. ‘I consistently receive blessings that

important to well-being. The SJS was strongly correlated with hedonic well-being (as indicated by the SWLS and the CES-D), but was also strongly associated with eudaimonic well-being (as indicated by the relationship with the FS).

Also as predicted, we found that grateful emotion was strongly associated with joy. Although this was a strong relationship, it still appears that gratitude and joy may be viewed as distinct affects. In Table 2, we show correlations of the validity variables with the GAS as well as the SJS. These correlations support the theory that although grat- itude and joy are related, they are distinct positive affects. Whereas joy is more strongly correlated with the joy sub- scale of the PANAS, the GAS was more strongly correlated with both trait gratitude measures, and gratitude to God. Somewhat surprisingly, the SJS was somewhat less strongly correlated with grateful emotion as assessed by the GAS than attentiveness and pride. But we should note that the SJS was more strongly correlated with the GAS at T2 and T3, supporting the idea that gratitude and joy may have a unique relationship. Both of our trait gratitude measures showed strong relationships with the SJS, supporting the idea that dispositional gratitude may provide an important mindset for experiencing joy. Of course, cross-sectional correlations such as this leave open a number of alterna- tive explanations. It could be that the correlation is the result of joy promoting gratitude, a reciprocal relationship between gratitude and joy, or some third variable (such as a positive self-report bias) that explains this relationship.

Correlations with the measures of spirituality pro- vided preliminary evidence that joy may be an impor- tant spiritual emotion. Consistent with our predictions, gratitude to God may be important to experiencing joy

Table 2. Construct validity correlations with the SJS and GAS in Study 1.

Note: All correlations, p<.001. SJS: State Joy Scale. GAS: Gratitude Adjectives Scale.

Scale SJS GAS Affect and well-being scales SJS – .527 PANAS joy .767 .673 PANAS positive affect .667 .657 PANAS attentiveness .551 .526 PANAS ‘Proud’ .543 .520 ‘Pleasant Sensations’ .542 .446 Gratitude adjectives scale (GAS) .527 – Flourishing scale .599 .452 Satisfaction with life scale .561 .479 Spiritual well-being .303 .384 Gratitude to God .284 .385 PANAS negative affect −.443 −.276 PANAS ‘Sad’ −.434 −.168 ‘Unpleasant Sensations’ −.386 −.216 CES-D −.450 −.260 Personality variables GQ-6 .424 .507 GRAT-S .437 .502 Cynicism and Lack of Trust Scale (CLOT) −.576 −.368

THE JOURNAL OF POSITIVE PSYCHOLOGY 529

and are rated on a 5-point disagree/agree scale (Hendin & Cheek, 1997). Because narcissism involves an orienta- tion to the self and joy often involves reconnection with others, we predicted negative relationships between joy and narcissism. We hypothesized that the self-orientation of vulnerable narcissism would be most strongly related to dispositional joy. Relatedly, we predicted that disposi- tional envy would be inversely associated with disposi- tional joy because when one consistently envies others, it should be difficult to experience joy in response to an increased connection with others. We assessed envy with the Dispositional Envy Scale (DES, Smith, Parrott, Diener, Hoyle, & Kim, 1999), which showed good internal consist- ency (   =  .89). To investigate relationships with self-es- teem we used the Rosenberg Self-Esteem Scale (RSE: Rosenberg, 1965), which demonstrated strong internal consistency (  = .91). Although joyful people should not be preoccupied with the self, we predicted that they would have a positive view of themselves. We also included a measure of humility (General Humility Scale, Hill, Laney, & Edwards, 2014) that showed adequate internal consistency (  = .725), and was significantly related to the narcissism measures in the predicted direction (Grandiose Narcissism: r = −.37, p < .001; Vulnerable Narcissism: r = −.32, P = .001). For our measure of cynicism we used the CLOT, as in Study 1.

Results

We followed the approach we took in Study 1 in an attempt to develop a unifactorial measure of trait joy. First, consist- ent with past studies, both exploratory and confirmatory factor analyses testing several different rotations failed to confirm the structural validity of the DPES. This suggests that more development of this measure is needed. As would be expected because of the number of joy items in this questionnaire, the first and strongest factor that emerged was a factor related to joy. After eliminating ineffective items, the long version of the DJS contained 27 items and had excellent internal consistency (  = .973).

In order to develop a shorter more unidimensional measure, we eliminated statistically and conceptually redundant items and only used items that loaded highly on the joy factor. This resulted in 16 items and the internal consistency of the short version of the DJS was excellent (  = .954). Table 3 includes the final DJS items along with item statistics. Factor analysis showed that one factor best explained the variance and this factor accounted for 59.81% of the variance. Items that loaded most strongly on this factor were ‘I consistently feel a subtle but enduring feeling of joy’, ‘I am an intensely cheerful person’, and ‘In a typical day, things often happen that tell me my life is working out the way it should.’ This provides some support

seem undeserved’), and that joy can be experienced as a more subtle but enduring state (‘I consistently feel a subtle but enduring feeling of joy’).

Method

Participants and procedure

In this study, 106 participants completed the measures in exchange for partial course credit. Participants com- pleted measures in two separate sessions, one to two days apart. In session 1, participants were administered questionnaires in the following order: modified PANAS (as in Study 1), a general humility scale, social cynicism, dispo- sitional envy, vulnerable narcissism, trait gratitude (GQ-6,

 =  .66), self-esteem, trait gratitude as measured by the GRAT-S (  = .87), the DPES – which included our additional items in an attempt to develop the DJS, the CLOT (  = .93), and finally a measure of grandiose narcissism. In the sec- ond session, participants first completed a mood induction procedure. Participants were randomly assigned to one of three mood induction conditions: gratitude, pride, or neutral. For the gratitude condition, they were asked to ‘recall an event where someone did something important and valuable for you.’ For the pride condition, participants were asked to ‘recall an event where you accomplished something important and valuable for yourself.’ The neu- tral condition merely requested that participants ‘recall the last day nothing truly notable happened: you completed your daily routine without interruption, were about as pro- ductive as average, and spent your afternoon doing what you most often do.’ Participants were then asked to write about their assigned event for 5 min. Following the induc- tion participants completed the PANAS. We predicted that the DJS scores in Session 1 would predict increases in joy in Session 2. This study was completed according to the ethical standards of the APA and was approved by the IRB of Eastern Washington University.

Measures

For the Narcissism measures, we used the Narcissistic Personality Inventory (NPI), as well as the 10-item Hypersensitive Narcissism Scale (HSNS; Hendin & Cheek, 1997). The 40-item Narcissistic Personality Inventory is the most frequently used measure in narcissism research, used in this case as our grandiose narcissism measure, and demonstrated satisfactory internal consistency, (  = .86). The HSNS was used to measure vulnerable narcissism. It demonstrated high internal consistency (  = .95) and was not redundant with the NPI, showing no correlation with the NPI (r = .02). Items on the scale include statements such as ‘I often interpret the remarks of others in a personal way’,

530 P. C. WATKINS ET AL.

predicted, the DJS was inversely associated with envy (r = −.53), vulnerable narcissism (r = −.51), global cynicism (r = −.39), and local cynicism (r = −.63).1

The disposition of joy implies that people high in this trait should experience more joy in relevant situations than those low in the trait. Thus, the DJS should be able to predict levels of joy after a positive mood induction. Indeed, the DJS scores in Session 1 were reliably corre- lated with the PANAS joy subscale (see Study 1) in ses- sion 2 after our mood induction (r = .44, p < .001). More importantly, we found that the DJS scores in Session 1 predicted significant increases in joy after a positive mood induction in Session 2. We conducted hierarchical regressions predicting Session 2 state joy scores for the positive and neutral mood induction conditions. In step 1 we included Session 1 state joy, and in step 2 we entered Session 1 DJS scores. For the positive mood induction condition, Model 1 was significant, F = 11.913, p =  .001, as was Model 2, F = 10.11, p < .001. Importantly, the DJS significantly predicted Session 2 joy after controlling for Session 1 joy (ΔR2 = .093, Fchange = 7.027, p = .01, DJS stand- ardized β = .013). Hierarchical regression analysis for the neutral mood induction condition showed that the DJS did not significantly predict Session 2 joy after controlling for Session 1 state joy (p = .77). Thus, as expected, the DJS predicted significantly enhanced joy, only after a positive mood induction.

In sum, the short DJS showed good construct validity, and results provided preliminary evidence that disposi- tional joy may be important to emotional well-being. Further validation of both the SJS and the DJS could be provided if they were included in one study. Furthermore, if trait gratitude is an important disposition for experienc- ing joy, this should be investigated in a prospective design. In Study 3 we sought to deal with these issues.

for the idea that joy involves both intense pleasant feelings and more subtle long-term emotions. This also supports the idea that joy results from the perception that one’s life is being ‘well lived.’ Because items such as ‘Even when things aren’t going well, I can still feel joy’ were retained in the final DJS, this supports the proposal that joy can be experienced in aversive situations. The appraisal that the object of joy is ‘excessive’ and beyond expectations was supported, as items such as ‘I often feel blessed’, and ‘I often find myself amazed at all the good things I have received’ were retained. The short DJS related very well to the longer measure (r = .99), and thus researchers will probably elect to use the shorter version.

Some results did not support proposals about joy. In particular, the proposition that joy results from benefits that are perceived to be ‘un-owed’ was not supported (Volf, 2015). In our initial factor analyses of the modified DPES the item ‘I consistently receive blessings that seem undeserved’ loaded on a separate factor that seemed to be related to gratitude. This provides some evidence that joy and gratitude are distinct but related constructs. As in Study 1, we did not find support for Roberts (2007, 2014) theory that joy should not involve pleasant physical sen- sations. Items such as ‘I really enjoy physical pleasure’ and ‘I frequently experience pleasurable sensations’ loaded most strongly on the joy factor in our initial factor analyses (however, these items were not loaded strongly enough on this factor to be included on a unidimensional measure).

The short DJS showed good construct and predictive validity. As expected, the short DJS was related to imme- diate state joy (r = .68) and state joy 2 days later (r = .41). The short DJS was also positively associated with state gratitude (r = .58) and state attention (r = .45), trait grat- itude as measured by the GRAT-S and GQ-6 (rs = .68 and .53), self-esteem (r =  .64), and humility (r =  .24). Also as

Table 3. Final items and statistics for the Dispositional Joy Scale (DJS).

Item Factor loading Corrected item-total 1. I often feel bursts of joy .783 .742 2. I consistently feel like my life is going well .718 .679 3. I can find joy in almost any occasion .828 .719 4. I would say that most of the occasions in my life bring me joy .817 .787 5. My life is always improving .703 .662 6. Many things bring me delight .771 .734 7. I would say that I am an enthusiastic person .756 .726 8. Good things happen to me all the time .733 .696 9. I often feel blessed .754 .711 10. Even when things aren’t going well, I can still feel joy .802 .766 11. I often find myself amazed at all the good things I have received .743 .706 12. I am an intensely cheerful person .834 .800 13. Even in the midst of bad situations, I can usually find something to rejoice about .779 .747 14. In a typical day, things often happen that tell me my life is working out the way it should .834 .802 15. I consistently feel a subtle but enduring feeling of joy .843 .808 16. I often feel overjoyed when something good happens .710 .662

THE JOURNAL OF POSITIVE PSYCHOLOGY 531

occurring in succession. In order to achieve a general per- sonality profile for joy participants then completed the Big Five Inventory (BFI; John, Donahue, & Kentle, 1991; John, Naumann, & Soto, 2008). This measure provides an assess- ment of the Big Five personality characteristics: extrover- sion, agreeableness, conscientiousness, neuroticism, and openness. After completing two scales not relevant to the purpose of this study, participants were administered the Balanced Inventory of Desirable Responding (BIDR, Paulhus, 1991). This questionnaire was used to control for response bias and contains two subscales: impression management (  = .74) and self-deceit (  = .67). In order to continue our interest in the spirituality of joy, participants then completed the Gratitude toward God scale (GTG: Uhder, Webber, & Watkins, 2010; Watkins, Uhder, Webber, Pichinevenskiy, & Sparrow, 2011). This scale adapted items from the GQ-6 and the GRAT by changing the wording so as to be directed toward a divine benefactor, and appears to have good reliability and validity (  = .95; .95). Finally, participants completed a demographics form.

Results and discussion

Factor analysis of the PANAS-X

In order to further investigate the discrete nature of joy, we conducted an exploratory factor analysis on the positive affect items of the PANAS-X, current state form.2 Because the PANAS-X contains a number of additional positive adjectives from the PANAS and because we added the adjectives from the GAS, this allowed us to investigate the distinct nature of joy and whether joy and gratitude are separate affects. For this analysis we used a PCA with a Varimax rotation. Six factors emerged with Eigen val- ues greater than 1.0 explaining 68.72% of the variance. The strongest factor was clearly related to joy, and items loading stronger than .60 included cheerful, joyful, happy, delighted, enthusiastic, and excited (respectively in order of load strength). The second factor related to boldness and strength (e.g. bold, fearless, strong), and the third factor contained the 3 items of the GAS – all loading on this factor above .80. The fourth factor related to surprise/ astonished, and the fifth factor was related to calm states (calm, relaxed, at ease). Finally, the sixth factor was related to attentiveness (attentiveness, concentrating, alert, inter- ested). These results correspond to the positive affect fac- ets as initially identified by Watson and Clark (1994), with the caveat that the addition of the items from the GAS clearly added a gratitude factor. The important conclusion from this analysis is that consistent with our past studies (McCurrach, 2015), joy clearly emerged as a discrete posi- tive emotion and was distinct from other positive affects, including gratitude.

Study 3

In Study 3 we used a prospective design to investigate the relationship of gratitude to joy, and further investigate the validity of our joy measures. Participants completed our joy measures along with other relevant questionnaires seven to nine weeks apart.

Method

Participants and procedure

Participants completed all measures at two points in time, seven to nine weeks apart. Students were awarded partial course credit for completing the study. At T1, 215 students completed the measures, and 176 completed the ques- tionnaires at T2. Because not all students participating at T1 completed the measures at T2, 153 participants com- pleted the measures at both time points. At T1, 27.3% were males, 53.1% were females, 0.8% reported ‘other’, and 4 participants did not report gender. For T2, 20.8% were males, 44.2% were females, and 0.4% reported ‘other.’ Six participants did not report their gender. Ages at T1 ranged between 18 and 51 with mean age of 21.15 (SD = 4.16). At T2, ages ranged between 18 and 52 with a mean age of 21.11 (SD = 4.21). At T1, 55.0% reported ‘white’ ethnicity, 2.7% African American, 1.9% Asian, 9.6% Hispanic, and the remainder reported either other ethnicities, mixed ethnic- ity or did not report. At T2, 46.2% reported ‘white’ ethnicity, 1.5% African American, 2.7% Asian, 9.6% Hispanic, and the remainder reported either other ethnicities, mixed ethnic- ity or did not report. This study was approved by the IRB of Eastern Washington University and was conducted accord- ing to the ethical principals of the APA.

Measures

In order to have a more complete assessment of emotional state, participants completed two forms of the PANAS-X, one for how they were feeling right now, and one for ‘the past few weeks.’ The PANAS-X is the expanded form of the PANAS, containing 60 adjectives describing positive and negative emotional states (Watson & Clark, 1994). As in Studies 1 and 2 we included the three items of the GAS interspersed with the original items. Using the PANAS-X for ‘the past few weeks’, we subtracted scores from the 10-item negative affect scale from the 10-item positive affect scale as a measure of hedonic well-being. Following the two forms of the PANAS-X, participants completed the SJS (  = .95, .96), the DJS (  = .96, .97), the GRAT-S (  = .84, .90), and the GQ-6 (  = .81, .87). Because of the importance of these measures to our hypotheses and the possibility of order effects, we created four different orders of these measures, with the two forms of the PANAS-X always

532 P. C. WATKINS ET AL.

DJS (r = .714, p < .001), which is the expected relationship between a state and trait emotion measure.

We would also like to highlight the strong associa- tion between our joy measures and the measure of SWB. We further tested the relationship of joy to SWB by con- trolling for the Big-5 personality traits. We conducted a hierarchical regression predicting SWB where we entered scores from each Big-5 personality trait in step 1, and then entered DJS scores in step 2. Model 1 was significant, F(4,205) = 35.33, p < .001, R2 = .463. Model 2 was also sig- nificant, F(6,146) = 16.71, p < .001. The Fchange for Model 2 was significant, F(1,204) = 29.42, p < .001, and the ΔR2 was .068. Standardized β for the DJS = .369. Thus, dispositional joy predicted SWB above and beyond the Big-5 personality traits. This is a promising preliminary result supporting the theory that joy is an important facet of the good life.

Correlations of joy with personality traits provided fur- ther support for the validity of the SJS and DJS. The per- sonality measures showed stronger associations with the DJS than the SJS, supporting the validity of the DJS as a measure of an affective trait. Of the Big-5 traits, extraver- sion and neuroticism showed the strongest correlations with our joy measures. This provides a personality por- trait of the joyful person: they tend to be extraverted but emotionally stable. If extraversion is indeed reflective of the Behavioral Activation System that purportedly medi- ates positive affect in the brain (Gray, 1986) and joy is an important emotion to positive affectivity, then the strong correlation of dispositional joy with extraversion would seem to follow. The strong negative association of trait joy with neuroticism suggests that joy may counteract negative affectivity. Alternatively, negative affectivity may inhibit joy, and future research on this relationship should be revealing.

As predicted, we also found strong correlations between joy and trait gratitude. To further examine whether joy has a unique relationship to gratitude, we conducted a hierarchical regression where we investigated if trait joy predicted trait gratitude after controlling for the Big- 5. For simplicity, we created a trait gratitude composite score by combining the z scores for the GRAT-S and the GQ-6 (  = .88).3 Model 1 was significant, F(5,207) = 16.37, p  <  .001, and more importantly, the Fchange for Model 2 was significant, F(1,206) = 75.44, p < .001, ΔR2 = .192, DJS standardized β  =  .621. Thus, after controlling for Big-5 personality traits, dispositional joy strongly predicted trait gratitude. This provides more evidence for the important relationship of joy to gratitude.

Finally, we note the relationship of joy to gratitude to God. As seen in Table 4, although both measures of joy showed significant relationships with the GTG, clearly the relationship is much stronger with dispositional joy. As with our other analyses, we sought to further examine

Reliability and validity of the SJS and DJS

Because of the importance of cross-validation, we again conducted reliability and validity analyses for our joy measures from the data gathered in this study. Again the internal consistency of the SJS was quite good, both for the long (  = .950; .965), and shorter versions (  = .935; .955). Similarly, the DJS showed excellent internal consistency for the long (  = .964; .972) and short (  = .942; .955) ver- sions. Test-retest reliability was also good for the SJS (short: r = .60, long: r = .56) and the DJS (short: r = .64, long: r = .66). As expected, temporal stability was somewhat stronger for the trait than for the state joy measure.

We evaluated structural validity by conducting fac- tor analyses on the short measures of the SJS and DJS. For the SJS, in both administrations only one factor had an Eigenvalue greater than one and this factor explained greater than 61% of the variance (T1 = 61.10%; T2 = 69.69%). All items loaded above .7 on the factor. The short DJS also showed good factorial validity. One factor explained 54.18 and 60.32% of the variance, and no other factor explained more than 7% of the variance. On both administrations of the DJS, only one item fell below the factor loading of .60 (‘I often feel blessed’), and this item was still strongly loaded on the primary factor (.59 and .58).

Construct validity correlations are shown on Table 4. We present only correlations from T1 because T2 results were similar. PANAS-X results are from the form instructing par- ticipants to respond ‘for the last several weeks.’ Observation of this table shows that in general, results support the construct validity of our joy measures. In particular, the highest correlations were found between our joy measures and the joy subscale of the PANAS-X. We should also note that we found a strong relationship between the SJS and

Table 4.  Construct validity Correlations for the SJS and DJS in Study 3.

Notes: *p = .01; All other correlations, p < .001. PANAS subjective well-being = PANAS PA – PANAS NA.

Construct Validity Scale

Joy scale

SJS DJS

Affect and well-being scales

PANAS-X Joy .637 .749 PANAS-X Gratitude .342 .530 PANAS-X Bold/Strength .488 .498 PANAS-X Surprised/Astonished .398 .364 PANAS-X Calm .470 .377 PANAS-X Attentiveness .319 .338 PANAS-X Subjective Well-being .616 .643

Personality scales

Extraversion .388 .532 Agreeableness .291 .369 Conscientiousness .320 .364 Neuroticism −.467 −.547 Openness .286 .297 GRAT-S .538 .606 GQ-6 .489 .603 Gratitude to God (GTG) .184* .331

THE JOURNAL OF POSITIVE PSYCHOLOGY 533

SWB. Model 1 was significant, F(1,151) = 128.81, p < .001, as was model 2, F(2,150)  =  67.86, p  <  .001. Crucially for our prediction, Fchange(1,150)  =  4.18, p  =  .04, ΔR2  =  .015, standardized DJS β = .159. Thus, dispositional joy predicted increased SWB over time, supporting the theory that joy may be important to human flourishing.

In summary, Study 3 provided additional evidence for the construct validity of our state and trait joy measures, and also provided evidence that the disposition of joy may be important to SWB. Furthermore, results supported a model of an interesting upward spiral where gratitude promotes joy and joy in turn enhances gratitude.

General discussion

In three studies, we sought to develop valid measures of state and trait joy, and investigate the relationship of joy to gratitude and well-being. We believe our findings offer four main contributions to the literature. First, results from all three studies support our claim that joy is indeed a discrete positive emotion. Second, joy can be reliably measured by self-report. Third, joy is related to, but distinct from grat- itude. Moreover, evidence from our studies supports the supposition that gratitude is important for experiences of joy. Finally, our studies present promising evidence that joy is important to overall SWB. We elaborate on each of these contributions below, discuss the theoretical implications of our findings, and suggest future directions for joy research.

Both as a state and as a trait, in all of our factor analy- ses joy emerged as a separate and distinct factor from the other positive affects. This is consistent with analyses of six past studies with the PANAS-X (McCurrach, 2015; see also Watson & Clark, 1994). These results suggest that joy is not a diffuse, general positive response, but rather, joy has its own distinct phenomenological and appraisal profile.

Our results provided evidence for the validity of two measures of joy. First, in Study 1 we developed a unidimen- sional measure of the joyful state: the SJS. This measure showed good reliability and construct validity and should be useful to researchers interested in investigating joy (see Appendix 1). Because we felt that joy may not be a frequent emotional experience, this measure was devel- oped by asking respondents to refer to the ‘past week.’ This form is probably most useful to personality and well-be- ing researchers, but may not be as useful to those who are interested in experimental designs and would pre- fer to have an assessment of an individual’s momentary experience of joy. In future research, we are interested in modifying the SJS so as to provide an assessment of momentary joy, but evidence from our analyses of the PANAS provides another option. Across studies, several items from the PANAS-X loaded strongly on the joy factor: joyful, delighted, and enthusiastic. Thus, we suggest that

this relationship by controlling for the Big-5 personality characteristics. Thus, in step 1 we entered the Big-5 per- sonality traits, and this model was significant in predict- ing the GTG, F(5,205) = 3.43, p = .005. More importantly, the F change of Model 2 was significant, F(1,204) = 18.73, p < .001, ΔR2 = .078, standardized β for the DJS = .394. Thus, dispositional joy predicted gratitude to God above and beyond the Big-5 personality traits. This provides some evi- dence for the spiritual dimensions of joy, but clearly, there are many other spiritual variables that call for investigation.

Prospective analyses

Cross-sectional correlations have provided evidence for the importance of gratitude to joy, and that joy might be a significant contributor to SWB. However, a much stronger case could be made for these relationships in prospective analyses. First, we investigated the relationship of trait gratitude to joy using hierarchical regression analyses. In order to provide a strong test of this relationship, in step 1 we entered T1 SJS scores along with the scores from the two response bias scales from the BID-R for predicting T2 SJS scores. We then entered T1 composite trait gratitude scores in step 2. Model 1 was significant, F(3,133) = 20.28, p < .001, but importantly, the Fchange of Model 2 was also significant, F(1,132) = 8.24, p = .005, ΔR2 = .040, trait grat- itude standardized β = .241. Thus, after controlling for T1 state joy and response bias, T1 trait gratitude predicted increased T2 joy, supporting the theory that the disposi- tion of gratitude is important to joy.

It is also possible however, that dispositional joy may prepare the way for increased gratitude. In order to test this idea we conducted a hierarchical regression where we entered T1 response bias and grateful emotion scores (GAS) in step 1, and then entered T1 DJS scores in step 2, for predicting T2 gratitude states. Model 1 was signifi- cant, F(3,148) = 15.18, p < .001, but importantly, the Fchange for model 2 was also significant, F(1,147) = 6.49, p = .01, ΔR2 = .032, DJS standardized β = .230. Thus, dispositional joy predicted increased state gratitude over time. Perhaps this is simply the result of positive emotions tending to activate each other, but when we conducted prospective partial correlations with the other positive affect subscales of the PANAS-X, none were found to predict significant increases in joy or gratitude. Thus, gratitude and joy may have a particularly unique relationship among the pos- itive emotions, where gratitude supports future joy and joy enhances future gratitude; providing evidence for an interesting upward spiral that may be important to SWB.

The design of this study also allowed us to investigate the prospective relationship of joy to well-being. In step one of our regression analysis we entered T1 SWB scores, and in step 2 we entered the DJS scores for predicting T2

534 P. C. WATKINS ET AL.

We also found promising evidence that joy might be important to SWB. In our cross-sectional correlations, joy correlated strongly with SWB and flourishing. More impor- tantly, in Study 3 we found that joy prospectively predicted increases in hedonic well-being. Our understanding of joy and well-being would be advanced by exploring this rela- tionship with experimental designs. For example, future research could use experimental designs to see if joy inter- ventions enhance SWB. What might a joy intervention look like? One way to promote joy would be to use interven- tions similar to those designed to encourage trait gratitude (e.g. Froh et al., 2014). Another approach would be to use an experimental design to investigate a joy adaptation of grateful recounting. For example, one could ask partici- pants to daily recall three events that brought them joy. Careful thought would need to go into the planning of these interventions – particularly to insure they are not simply versions of gratitude treatments – but this research should prove to have both practical and theoretical value.

The results from our studies have several implications for theories of joy. First, as suggested by the recent inte- grative joy project (Volf & Crisp, 2015), joy is clearly a very pleasant hedonic response to a positive object. Various authors in this project (e.g. Volf, 2015) also suggested that joy is the result of the perception that life is going well and life is being lived well. Psychometric analyses of our meas- ures provided some support for this idea in that items such as ‘This week, my life went well’, ‘This week, life just made sense to me’, and ‘In a typical day, things often happen that tell me my life is working out the way it should’ were retained in our final measures. Also, our finding that joy was strongly related to a measure of eudaimonic well-be- ing (the FS), supports this notion.

Theories of joy have also emphasized that although joy may be experienced as an intense short-lived emotion, it can also be experienced as a more subtle enduring state. Somewhat surprisingly, the item that loaded most strongly on the DJS was ‘I consistently feel a subtle but enduring feeling of joy’, despite the fact that items such as ‘Many things bring me delight’ and ‘I am an intensely cheerful person’ were also retained on this measure. Results also supported the idea that joyful people are able to experi- ence joy even in the midst of difficult situations. The items ‘Even when things aren’t going well, I can still feel joy’ and ‘Even in the midst of bad situations, I can usually find some- thing to rejoice about’ were retained in the final DJS. This is an intriguing aspect joy, and more research should be devoted to investigating how joy might be experienced in difficult situations.

Although several theoretical propositions of joy were supported, several were not supported. For example, the proposal that the good object causing joy is perceived as ‘unowed’ (Volf, 2015) was not supported. Perhaps it is not

these three adjectives may provide for a good measure of momentary joy. Indeed, in Study 3, the internal reliability of this short scale was quite good (  = .901; .890). The dis- advantage of this scale is that because it is so short, it may not provide an assessment of the full experience of joy.

We were also able to develop a measure of the joyful disposition (DJS), and this showed good reliability and construct validity. Correlations of the well-being measures with the SJS and DJS suggested that the disposition of joy is more important to flourishing than isolated joyful expe- riences. If the disposition of joy proves to be important to well-being, the DJS may be valuable for investigating aspects of a joyful lifestyle that are important to the good life.

Our studies also provided evidence that joy is related to, but distinct from, gratitude. First, consistent with McCurrach (2015), in factor analyses gratitude and joy always emerged as distinct factors. Indeed, even though the DPES does not have an explicit gratitude factor, one seemed to emerge that was distinct from the joy factor, and the item ‘I consistently receive blessings that seem undeserved’ (designed to tap the ‘unowed’ aspect of joy), loaded strongly on this factor. These findings contra- dict theories maintaining that gratitude is a form of joy. Nevertheless, although our results showed that joy and gratitude are distinct, they are clearly related. Indeed, joy was strongly related to trait gratitude (correlations ranged from .42 to .54), supporting the idea that the disposition of gratitude may prepare one for experiences of joy. Our prospective study (Study 3) provided the strongest results supporting this theory. We found that trait gratitude pre- dicted increases in joy over time. Why would the disposi- tion of gratitude prepare the way for experiences of joy? Grateful people have been found to have several cognitive biases that should also promote joy. Research has found that trait gratitude is associated with positive biases in the encoding and interpretation of events (Scheibe, Watkins, McCurrach, & Mathews, 2016; Watkins & McCurrach, 2016). When one has an inclination to notice and affirm events as good, this should enhance the likelihood of joyful experi- ences. One implication of these findings is that the culti- vation of gratitude should result in enhanced joy, and we explore this possibility below.

Not only does gratitude promote joy, in Study 3 we found that over time dispositional joy predicted enhanced gratitude. This suggests an intriguing upward spiral between gratitude and joy: as one cultivates the disposi- tion for gratitude, this increases the frequency of experi- ences of joy, which in turn should foster the disposition of joy, thus increasing gratitude. More research should inves- tigate this cycle of virtue, but if future research confirms this relationship, it would seem that this upward spiral might be important to SWB.

THE JOURNAL OF POSITIVE PSYCHOLOGY 535

interesting to compare our measures to non-self-report measures, such as life events recall bias and valenced judg- ment of neutral words. One of the most notable problems with self-report measures of positive constructs is with social desirability: people like to present themselves in a positive fashion. Although this is still a clear limitation of these studies, in Study 3 we used the BID-R, which allowed us to control for self-presentation biases. We look forward to research that validates these instruments using inform- ant reports. Finally, the correlational nature of these stud- ies limits the strength of conclusions that can be made, and we encourage the use of experimental designs to investi- gate the contribution of joy to well-being. Although our prospective design in Study 3 provided more conclusive results supporting the contribution of joy to well-being, experimental designs would strengthen our understand- ing of this relationship. Experimental designs using exer- cises to induce joy should provide more definitive results.

In summary, results from these studies provide evi- dence that joy is a discrete positive emotion, it can be measured reliably with self-report instruments, gratitude promotes joy and joy promotes gratitude, and joy pro- spectively predicts increases in SWB. If joy is important to a flourishing life, it behooves us to continue to advance our understanding of joy. In particular, we believe inves- tigations that shed light on the appraisal structure of joy are needed. Indeed, joy research may have lagged behind the other positive emotions because previously, a clear appraisal structure has not been proposed. In this regard we believe that Vaillant’s (2008) approach may be espe- cially insightful. In brief, Vaillant argues that we experience joy when we appraise an event as bringing us back to, or increasing our connection with, something that we deeply value as good. Usually, these good things are reconnec- tions with other people we deem important in our life, but it can be an increased connection with anything that we hold in high value. For Cubs fans, it was something that for 108  years they had been apart from, and when the wait was finally over, when they were finally once again reunited with baseball’s summit, there was nothing but ‘joy in Wrigleyville.’

Author note

Jeffrey Froh served as action editor for this manuscript.

Notes

1. All ps < .001. 2. Results with the PANAS-X ‘last few weeks’ form were

similar. 3. Separate analyses with the GQ-6 and the GRAT-S were

similar to the analysis with the composite measure.

so much that the benefits that produce joy are viewed as undeserved, but rather that the benefits are in some way seen as excessive – they exceed one’s expectations. Items such as ‘I often feel blessed’ and ‘I often find myself amazed at all the good things I have received’ were retained in the final DJS, which seems to support the idea that the objects of joy are perceived as going above and beyond what one could expect.

Our studies also provided evidence as to the spiritual nature of joy. Some authors have argued that joy is a spiritual emotion in that it is an enjoyment that is not asso- ciated with physical sensations of pleasure (Roberts, 2007). We actually found strong positive correlations between items measuring the experience of physical pleasure and joy, which appears to contradict this theory. Perhaps this is the wrong approach to joy as a spiritual emotion. We propose that pleasurable physical sensations often result in joy, but that one’s experience of joy is enhanced in these situations when there is a spiritual dimension to these experiences. For example, one is more likely to experience joy in a sexual encounter when one sees the transcend- ent giftedness of the interaction. When enjoying a glass of wine, one is more likely to experience joy when one sees the wine as a divine blessing. Of course, these ideas are highly speculative at this point, and demand further investigation.

We have seen that joy is not constrained to non-material pleasures. But what is the spiritual dynamic of joy? Much work remains to be done on this front, but in our studies we found small to moderate relationships between joy and spiritual well-being and gratitude to God. Thus, there is clearly a spiritual dimension to joy, in the words of C.S. Lewis, ‘Joy is the serious business of Heaven’ (1963, p. 93). The correlations between gratitude to God and joy support our suggestion that when a good object is experienced as a divine gift, joy is more likely, but clearly our results are far from definitive on this issue. We propose, along with Lewis (1955) and Nietzsche, that the spiritual dimension of joy may be in that it promotes spiritual longing – a longing for the transcendent. In Nietzsche’s words (2006, p. 264):

Yet all joy wants eternity –

– Wants deep, deep, eternity.

In sum, we propose that the experience of joy is enhanced when there is a perceived spiritual purpose to the joy object, and that a spiritual thought/action tendency of joy is that it promotes a desire for the transcendent.

As with most research, these studies have limitations and perhaps raise more questions than they answer. Our use of college students seriously limits the external validity of our conclusions, and we look forward to research that uses more diverse populations with these measures. Also, all of the measures used were self-report, and it would be

536 P. C. WATKINS ET AL.

Hendin, H. M., & Cheek, J. M. (1997). Assessing hypersensitive narcissism: A reexamination of Murray’s narcism scale. Journal of Research in Personality, 31, 588–599. doi:10.1006/ jrpe.1997.2204

Hill, P. C., Laney, E. K., & Edwards, K. E. (2014). Unpublished intellectual humility scale. La Mirada, CA: Rosemead School of Psychology, Biola University. [email protected].

John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory – Versions 4a and 54. Berkeley, CA: University of California, Berkeley, Institute of Personality and Social Research.

John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the integrative big-five trait taxonomy: History, measurement, and conceptual issues. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and research (pp. 114–158). New York, NY: Guilford Press.

Kashdan, T. B., & Nezlek, J. B. (2012). Whether, when, and how is spirituality related to well-being? Moving beyond single occasion questionnaires to understanding daily process. Personality and Social Psychology Bulletin, 38, 1523–1535. doi:10.1177/0146167212454549

Lazarus, R. S. (1991). Emotion and adaptation. New York, NY: Oxford Press.

Lewis, C. S. (1955). Surprised by joy: The shape of my early life. Orlando, FL: Harcourt.

Lewis, C. S. (1963). Letters to Malcolm: Chiefly on prayer. San Diego, CA: Harcourt Brace Jovanovich.

Mathewes, C. (2015). Toward a theology of joy. In M. Volf & J. E. Crisp (Eds.), Joy and human flourishing: Essays on theology, culture, and the good life (pp. 63–96). Minneapolis, MN: Fortress Press.

McAdams, D. P. (2006). The redemptive self: Stories Americans live by. New York, NY: Oxford University Press. doi:10.1093/ acprof:oso/9780195176933.001.0001

McCullough, M. E., Emmons, R. A., & Tsang, J. (2002). The grateful disposition: A conceptual and empirical topography. Journal of Personality and Social Psychology, 82, 112–127. doi:10.1037/0022-3514.82.1.112

McCullough, M. E., Kilpatrick, S. D., Emmons, R. A., & Larson, D. B. (2001). Is gratitude a moral affect? Psychological Bulletin, 127, 249–266. doi:10.1037/0033-2909.127.2.249

McCurrach, D. (2015, May). The phenomenology of joy and gratitude. In P. C. Watkins (Chair), Joy and gratitude: Exploration of a relationship important to well-being. Symposium presented to the Annual Convention of the Western Psychological Association, Las Vegas, NV.

McLaughlin, T., & Palmer, K. (2016, November 3). World series 2016: Tears of joy as Chicago Cubs fans celebrate first triumph in 108  years. The Independent. Retrieved from http://www. independent.co.uk/sport/us-sport/major-league-baseball/ chicago-cubs-fans-celebrate-world-series-title-a7394736. html

Moltmann, J. (2015). Christianity, a religion of joy. In M. Volf  & J. E. Crisp  (Eds.), Joy and human flourishing: Essays on theology, culture, and the good life (pp. 1–16). Minneapolis, MN: Fortress Press.

Myers, D. G., & Diener, E. (1995). Who is happy? Psychological Science, 6, 10–19.

Nietzsche, F. (2006). Thus spoke Zarathustra. New York: Cambridge University Press.

Paulhus, D. L. (1991). Measurement and control of response bias. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.),

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This research was funded by a grant from the John Templeton Foundation [grant number ID60876].

References

Bagozzi, R. P. (1991). Further thoughts on the validity of measures of elation, gladness, and joy. Journal of Personality and Social Psychology, 61, 98–104. doi:10.1037/0022-3514.61.1.98

de Rivera, J. (1977). A structural theory of the emotions. Psychological Issues,10(4, Mono 40), 1–178.

de Rivera, J., Possell, L., Verette, J. A., & Weiner, B. (1989). Distinguishing elation, gladness, and joy. Journal of Personality and Social Psychology, 57, 1015–1023. doi:10.1037/0022- 3514.57.6.1015

Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale. Journal of Personality and Social Psychology, 49, 71–75.

Diener, E.,  Wirtz, D., Tov, W., Kim-Prieto, C., Choi, D., Oishi, S., Biswas-Diener, R.  (2010). New well-being measures: Short scales to assess flourishing and positive and negative feelings. Social Indicators Research,  97, 143–156.

Egloff, B., Schmukle, S. C., Burns, L. R., Kohlmann, C., & Hock, M. (2003). Facets of dynamic positive affect: Differentiating joy, interest, and activation in the positive and negative affect schedule (PANAS). Journal of Personality and Social Psychology, 85, 528–540. doi:10.1037/0022-3514.85.3.528

Ekman, P. (1992). Are there basic emotions? Psychological Review, 99, 550–553.

Ellsworth, P., & Smith, C. (1988). Shades of joy: Patterns of appraisal differentiating pleasant emotions. Cognition and Emotion, 2, 301–331.

Floberg, C., Sestrap, N., Bart, B., & Watkins, P. C. (2014, April). The costs of cynicism: Development of a new measure. Poster presented to the Annual Convention of the Western Psychological Association, Portland, OR.

Fredrickson, B. L. (1998). What good are positive emotions? Review of General Psychology. Special Issue: New Directions in Research on Emotion, 2, 300–319.

Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The broaden-and-build theory of positive emotions. American Psychologist, 56, 218–226.

Frijda, N. H. (1986). The emotions. Cambridge: Cambridge University Press.

Frijda, N. H. (2007). The laws of emotion. Mahwah, NJ: Lawrence Erlbaum.

Froh, J. J., Bono, G., Fan, J., Emmons, R. A., Henderson, K., Harris, C., … Wood, A. (2014). Nice thinking! An educational intervention that teaches children how to think gratefully [Special Issue: Theoretical Frameworks in School Psychology Intervention Research: Interdisciplinary Perspectives and Future Directions]. School Psychology Review, 43, 132–152.

Gray, J. A. (1986). Anxiety, personality and the brain. In A. Gale, J. A. Edwards, A. Gale, & J. A. Edwards (Eds.), Physiological correlates of human behaviour, Vol. 3 Individual differences and psychopathology (pp. 31–43). San Diego, CA: Academic Press.

THE JOURNAL OF POSITIVE PSYCHOLOGY 537

Uhder, J., Webber, A., & Watkins, P. C. (2010, August). Favors from heaven: Sources and benefits of gratitude toward God. Poster presented at the Annual Convention of the American Psychological Association, San Diego, CA.

Vaillant, G. E. (2008). Spiritual evolution: How we are wired for faith, hope, and love. New York, NY: Broadway Books.

Volf, M. (2015). The crown of the good life: A hypothesis. In M. Volf & J. E. Crisp (Eds.), Joy and human flourishing: Essays on theology, culture, and the good life (pp. 127–136). Minneapolis, MN: Fortress Press.

Volf, M., & Crisp, J. E. (2015). Joy and human flourishing: Essays on theology, culture, and the good life. Minneapolis, MN: Fortress Press.

Watkins, P. C. (2014). Gratitude and the good life. Dordrecht: Springer.

Watkins, P. C. (2016). Positive psychology 101. New York, NY: Springer.

Watkins, P. C., & McCurrach, D. (2016). Exploring how gratitude trains cognitive processes important to subjective well-being. In D. Carr (Ed.), Perspectives on gratitude: An interdisciplinary approach (pp. 27–40). Oxford: Taylor and Francis.

Watkins, P. C., Uhder, J., Webber, A., Pichinevenskiy, S., & Sparrow, A. (2011, May). Religious affections: The importance of gratitude toward God to spiritual well-being. Poster presented at the Annual Convention of the Association for Psychological Science, Washington, DC.

Watkins, P. C., Woodward, K., Stone, T., & Kolts, R. (2003). Gratitude and happiness: Development of a measure of gratitude, and relationships with subjective well-being. Social Behavior and Personality: An International Journal, 31, 431–451.

Watson, D., & Clark, L. A. (1994). The PANAS-X: Manual for the positive and negative affect schedule. Expanded form. Retrieved from https://www2.psychology.uiowa.edu/faculty/clark/panas-x.pdf

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. doi:10.1037/0022-3514.54.6.1063

Wright, N. T. (2015). Joy: Some new testament perspectives and questions. In M. Volf & J. E. Crisp (Eds.), Joy and human flourishing: Essays on theology, culture, and the good life (pp. 39–62). Minneapolis, MN: Fortress Press.

Measures of personality and social psychological attitudes (pp. 17–59). New York, NY: Academic Press.

Plutchik, R. (2003). Emotions and life: Perspectives from psychology, biology, and evolution. Washington, DC: American Psychological Association.

Roberts, R. C. (2007). Spiritual emotions: A psychology of Christian virtues. Grand Rapids, MI: Eerdmans.

Roberts, R. C. (2014). Cosmic gratitude. European Journal for Philosophy of Religion, 6, 65–83.

Roseman, I. J., & Evdokas, A. (2004). Appraisals cause experienced emotions: Experimental evidence. Cognition and Emotion, 18, 1–28. doi:10.1080/02699930244000390

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: University Press.

Rosenberg, E. L. (1998). Levels of analysis and the organization of affect. Review of General Psychology, 2, 247–270. doi:10.1037/1089-2680.2.3.247

Scheibe, D., Watkins, P. C., McCurrach, D., & Mathews, A. (2016, May). Cognitive characteristics of gratitude. Poster presented to the Annual Convention of the Association for Psychological Science, Chicago, IL.

Scheibe, D., Watkins, P. C., & Uhder, J. (2017, April). Gratitude toward God: Exploring an important source of spiritual well- being. In P. Watkins (Chair), New research in the science of gratitude. Symposium presented to the Annual Convention of the Western Psychological Association, Sacramento, CA.

Seligman, M. E. P. (1998). Positive social science. APA Monitor, 29(4), 2–5.

Shiota, M. N., Keltner, D., & John, O. P. (2006). Positive emotion dispositions differentially associated with Big Five personality and attachment style. The Journal of Positive Psychology, 1, 61–71.

Smith, R. H., Parrott, W. G., Diener, E. F., Hoyle, R. H., & Kim, S. H. (1999). Dispositional envy. Personality and Social Psychology Bulletin, 25, 1007–1020. doi:10.1177/01461672992511008

Thompson, M. M. (2015). Reflections on joy in the Bible. In M. Volf & J. E. Crisp (Eds.), Joy and human flourishing: Essays on theology, culture, and the good life (pp. 17–38). Minneapolis, MN: Fortress Press.

Tompkins, S. S. (1962). Affect, imagery, and consciousness (Vol. 1). New York, NY: Springer.

538 P. C. WATKINS ET AL.

Appendix 1

SJS

Your Experiences Last Week

For the following items we would like you to indicate your opinions and experience for the last week, as honestly as you can. For each item, please circle the number that is associated with your response. Please note that the scale changes at item 3.

1. In the past week, how often have you felt joyful?

1 2 3 4 5 6 7 Not at all Hardly at all A few times Several times A number of

times Fairly Often Frequently

2. In the past week, how often have you felt enthusiastic?

1 2 3 4 5 6 7 Not at all Hardly at all A few times Several times A number of

times Fairly Often Frequently

3. Because of the joy I experienced this week, time just seemed to fly.

1 2 3 4 5 6 7 Completely

Disagree Neither agree or

disagree Strongly Agree

4. This week I found myself enjoying something so much that I lost track of time.

1 2 3 4 5 6 7 Completely

Disagree Neither agree or

disagree Strongly Agree

5. This week I felt free.

1 2 3 4 5 6 7 Completely

Disagree Neither agree or

disagree Strongly Agree

6. This week, I felt ready to enjoy whatever opportunity presented itself.

1 2 3 4 5 6 7 Completely

Disagree Neither agree or

disagree Strongly Agree

7. Something happened this week that made me feel like celebrating.

1 2 3 4 5 6 7 Completely

Disagree Neither agree or

disagree Strongly Agree

8. This week, the reality of my life was the way I feel it should be.

1 2 3 4 5 6 7 Completely

Disagree Neither agree or

disagree Strongly Agree

9. This week, I felt free to play.

1 2 3 4 5 6 7 Completely

Disagree Neither agree or

disagree Strongly Agree

10. This week, my life went well.

1 2 3 4 5 6 7 Completely

Disagree Neither agree or

disagree Strongly Agree

11. This week, life just made sense to me.

1 2 3 4 5 6 7 Completely

Disagree Neither agree or

disagree Strongly Agree

THE JOURNAL OF POSITIVE PSYCHOLOGY 539

DJS

Please respond to the following items in terms of how much you agree or disagree with each statement. With each item, please use the following scale:

1 2 3 4 5 6 7 Strongly Disagree Strongly Agree

1.___I often feel bursts of joy.

2.___I consistently feel like my life is going well.

3.___I can find joy in almost any occasion.

4.___I would say that most of the occasions in my life bring me joy.

5.___My life is always improving.

6.___Many things bring me delight.

7.___I would say that I am an enthusiastic person.

8.___Good things happen to me all the time.

9.___I often feel blessed.

10.___Even when things aren’t going well, I can still feel joy.

11.___I often find myself amazed at all the good things I have received.

12.___I am an intensely cheerful person.

13.___Even in the midst of bad situations, I can usually find something to rejoice about.

14.___In a typical day, things often happen that tell me my life is working out the way it should.

15.___I consistently feel a subtle but enduring feeling of joy.

16.___I often feel overjoyed when something good happens.