Personality and Individual Differences 119 (2017) 152–159
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
Development and validation of the State Contentment Measure Tamasin A. Taylor b,a,⁎, Oleg N. Medvedev c, Richard G. Owens b, Richard J. Siegert a a b c
Auckland University of Technology, North Campus, 90 Akoranga Drive, Northcote, Auckland 0627, New Zealand University of Auckland, Tamaki Campus, 261 Morrin Road, St Johns, Auckland 1072, New Zealand The University of Auckland, School of Medicine, 85 Park Road, Grafton, Auckland, New Zealand
a r t i c l e
i n f o
Article history: Received 27 April 2017 Received in revised form 4 July 2017 Accepted 11 July 2017 Available online xxxx Keywords: State Contentment Measure Psychometrics Well-being Positive emotions Affect Contentment Satisfaction
a b s t r a c t In this study, the authors developed and evaluated the psychometric properties of a new self-report instrument the State Contentment Measure (SCM1). Study 1 (n = 352) used Exploratory Factor Analysis (EFA) to investigate the preliminary psychometric properties of the SCM. The results indicated two related factors; a cognitive and a physiological factor, with both individually showing satisfactory internal consistency. Confirmatory factor analysis (CFA) in study 2 (n = 206) showed satisfactory fit for the 2-factor structure of SCM and slightly better fit for the single factor solution. Study 3 (n = 83) demonstrated that the measure was significantly related in the expected directions to theoretically-related and unrelated psychological constructs including happiness, mindfulness, depression, anxiety, stress, and life satisfaction, supporting its convergent and divergent validity. Study 4 (n = 70) indicated low agreement between test scores at different time points giving evidence for scale sensitivity to state changes. The SCM is a brief, valid and reliable self-report measure of state contentment that was previously lacking in psychological research. This is an important contribution given the rapid increase of studies using variables that would be likely to produce a state of contentment, such as, mindfulness, music, physical activity and yoga. © 2017 Elsevier Ltd. All rights reserved.
1. Introduction Attention to positive emotions has gained momentum over the past few decades with an increasing number of links being found with improved health and well-being (Weiss, Westerhof, & Bohlmeijer, 2016; Sin & Lyubomirsky, 2009). Correspondingly, many governments have included happiness as an important indicator of a country's well-being (Huppert & So, 2013). The Oxford Dictionary of English (Stevenson, 2010) defines contentment as “a state of happiness and satisfaction: he found contentment in living a simple life in the country.” and the Dictionary of Psychology includes contentment as one of the ‘valued’ psychological phenomena along with satisfaction, hope and optimism, and happiness and well-being (Coleman, 2015). Contentment has been categorised as an affective state (Berenbaum, 2002), that is posited to engender tranquillity and relief arising in situations appraised as safe and as having a high degree of certainty and low degree of effort (Ellsworth & Smith, 1988). Additionally, as a physiological state, contentment is considered to be subtly discernible and at the same time, its specificity has been identified by neurophysiological evidence (Stephens, Christie, & Friedman, 2010), as well as in studies ⁎ Corresponding author at: Auckland University of Technology, Faculty of Health and Environmental Studies, Box 92006, Auckland 0627, New Zealand. E-mail address:
[email protected] (T.A. Taylor). 1 State Contentment Measure
http://dx.doi.org/10.1016/j.paid.2017.07.010 0191-8869/© 2017 Elsevier Ltd. All rights reserved.
investigating states of meditation (Cahn & Polich, 2006; Williams, 2009). On the other hand, contentment is also viewed as a trait with longer-term adaptive functions (Fredrickson, 1998) and is often used interchangeably with the term ‘life satisfaction’ (Lavallee, Hatch, Michalos, & McKinley, 2007). According to Izard (1991), an emotion is a highly complex phenomenon that primarily activates neural, cognitive, and motor responses with each emotion-type having distinct properties and actions. In comparison, a trait is the way an individual may behave emotionally across all situations and across their lifespan (Berkowitz, 2000; Izard, 1991). A trait may be more stable and enduring while an emotion-state is thought to be more easily induced, manipulated and measured (Berkowitz, 2000; Izard, 1991). Importantly, although a trait may not be observed directly, as a predisposition, it may influence an individual's present state (Buss, 1989; Epstein, 1984). Accordingly, if traits and states theoretically share the same latent construct (Medvedev et al., 2017), this would suggest that people higher on trait contentment would be predicted to experience state contentment more frequently. As such, a scale measuring state contentment should include both transient and stable components of contentment. As an affective state, Berenbaum (2002) defined contentment as the satisfaction derived from achieving the required resource levels associated with feelings of satisfaction in consuming behaviour, or in bodily care, which is experienced when a person's current resources match or exceed the level of need. The definition was based on an investigation
T.A. Taylor et al. / Personality and Individual Differences 119 (2017) 152–159
into whether different types of joy-related pleasurable feelings were associated with different kinds of activities. 162 college students listed and rated the five activities they engaged in most frequently that gave them the most pleasure. Factor analysis revealed that state contentment was associated with activities that were nurturant: “activities that involve taking care of or helping people or things”, spiritual: “activities that relate to or are concerned with religion or other forms of spirituality”, and physical activities: “activities that primarily focus on the use or maintenance of the body”. Further, contentment was primarily associated with being ‘satisfied’ or ‘fulfilled’. One limitation of this study was that it lacked Confirmatory Factor Analysis (CFA) or other psychometric testing, therefore potentially limiting validity. Descriptions suggestive of the state and trait ‘contented’ or ‘calm’ emotion-construct have emerged from studies investigating the neurophysiology of meditation. In their review, Cahn and Polich (2006) identified state changes via meditation as including: “a deep sense of calm peacefulness, a cessation or slowing of the mind's internal dialogue, and experiences of perceptual clarity and conscious awareness merging completely with the object of meditation, regardless of whether a mantra, image, or the whole of phenomenal experience is the focal point” (p.182). Trait changes were defined on the other hand as, “a deepened sense of calmness, increased sense of comfort, heightened awareness of the sensory field, and a shift in the relationship to thoughts, feelings, and experience of self” (p.182). The authors acknowledged the limitations in identifying neurophysiological changes induced from meditative practice however, due to the difficulty in quantifying selfexperience (Cahn et al.). Physiological correlates of state contentment were identified in a study investigating the autonomic specificity of basic emotions (Stephens et al., 2010). The autonomic reactions of 49 undergraduates were measured while they were listening to state emotion-inducing music and films. The measures included heart rate variability, peripheral vascular activity, systolic time intervals, and electro-dermal activity. Pattern classification and cluster analysis classified the contentmentemotion state with a significantly greater than chance level (z = 6.94, p b 0.001). Added to this, the overall self-report questionnaire classification hit rate for all emotions was 60.6%. The contentment classification condition was significantly greater than chance (z = 13.06, p b 0.001) (Stephens et al.). This study gives evidence for the presence of the transient and physiological state of contentment. These aforementioned definitions and studies are important for developing the conceptualisation of contentment by usefully identifying the nuances between the transient and stable components of contentment particularly when as a state, contentment may have a subtle physical manifestation (compared to a more pronounced emotion such as happiness). Ekman's (1993) formative investigations supported this notion that contentment should be categorised as one of the ‘basic’ emotions, despite it not being overtly apparent. Additionally, contentment is considered to have no real action tendency in comparison for example, to fear which may enable the fight-or-flight action; instead, contentment is considered as a mindful emotion involving awareness of, and openness to momentary experiences (Fredrickson, 1998). On the other hand, contentment has been suggested to have a more stable and adaptive utility, enabling a person to think more broadly in the longer term, to savour their current circumstance, and to integrate this information into forming new views of the world around them (Fredrickson, 1998). This is part of the ‘Broaden and Build’ theory of positive emotions, (Fredrickson & Branigan, 2005). This theory suggests that positive emotions increase social bonding and reciprocity, facilitate flexibility of mind, increase ability to problem solve and the ability to override standard, habitual, and other uncreative modes of thought. In one study investigating the undoing effects of positive emotions (on negative emotions), contentment-eliciting and ‘amusing’ films were found to induce faster cardiovascular recovery including heart rate, pulse amplitudes and transmissions to the finger and blood pressure, compared to neutral or sad films (Fredrickson, Mancuso,
153
Branigan, & Tugade, 2000). The study involved participants (n = 170), who first, viewed a fear-eliciting film, that consequently activated anxiety-induced cardiovascular reactivity. Following this, they viewed a second clip that elicited ether contentment, amusement, neutrality, or sadness. The effects of the subsequently induced positive emotion states were theorised as allowing participants to move from a narrowed thought–action state (caused by the fear-inducing clip) into a state involving a more broadened array of thoughts and actions. 1.1. Existing scales Presently, there are no published measures specific to the state contentment construct. This is a major limitation for existing research including, for example, the aforementioned Fredrickson et al. (2000) and the Stephens et al. (2010), studies. There are however, state and trait measures constructed to encapsulate the meditation ‘experience’ that, according to the neuropsychological literature, elicits ‘calm’ emotions (Cahn & Polich, 2006; Williams, 2009). Conversely, available instruments focus on the underlying phenomena as either, ‘states of mindfulness’, ‘transcendence’ or ‘altered consciousness’ (e.g., Dittrich, 1998; Friedman, 1983; Friedman & MacDonald, 1997; Vaitl et al., 2005) consequently overlooking associated emotions. This exclusion effectively limits the capacity to identify self-reported contentment levels induced by meditation (or other contentment-eliciting activities). There are, on the other hand, scales that measure constructs related to contentment including, state happiness, discontentment, general wellbeing, life satisfaction and contentment with life (eg., Hills & Argyle, 2002; Hudson & Proctor, 1977; Diener, Emmons, Larsen, & Griffin, 1985; Lavallee et al., 2007). Most relevantly, Lavelle and colleagues developed a self-report measure of general life Contentment (Contentment with Life Assessment scale: CLAS) that comprises of items combining affective responses and cognitive evaluations. Items measuring daily contentment levels considered as more closely reflective of perceived global contentment were also included. With this measure, the authors addressed the limitation identified in satisfaction with life scales suggested as capturing more of an objective and cognitive evaluation of what a person might have rather than how they feel about what they have. In other words, individuals may respond to the more traditional life satisfaction items by evaluating what they think they should be satisfied with rather than people who feel satisfied because they have the life they want (Lavallee et al., 2007). The CLAS, now enabled the measurement of a general feeling of contentment in life. Importantly, subjective evaluation of life events has been shown in previous research to influence emotional responses (LeDoux, 2000; Medvedev, Shepherd, & Hautus, 2015). Theoretically, an individual who takes a mindfulness class may show lower than expected levels of change in self-reported contentment if they had an underlying lower life satisfaction due to relationship issues (as an example). In comparison, another individual taking the same class might exhibit a higher state contentment change due to a higher satisfaction with their life in general. The CLAS would be useful in picking up these more stable trait-like contentment influences, however, transient cognitive and physical changes would additionally need to be measured to allow for a more complete picture of an individual's present state of contentment. In sum, existing happiness/contentment or meditation-experience scales are not designed to, and therefore not able to, capture momentary changes in state contentment. Scales measuring either contentment or satisfaction tend to be focussed on perceived life satisfaction with questionnaire items revolving around how an individual views their life overall and leave an incomplete picture of the transient state of contentment. Scales measuring the meditation phenomena are also limited in that they omit the associated emotions. The present study brings together the present literature to conceptualise state contentment as including stable and transient cognitive- affective and physiological components.
154
T.A. Taylor et al. / Personality and Individual Differences 119 (2017) 152–159
2. Study 1. Design and development of the State Contentment Measure 2.1. Procedure Forty-two items relating to state, stable, transient and physical correlates of contentment were initially generated. Ten items were adapted from existing measure items and 32 were generated by a team of postgraduate psychology students (n = 6) who were part of an existing academic psychology group within the university, and one academic and clinical psychologist. Items adapted from existing questionnaires included six items from The Contentment with Life Assessment Scale (CLAS: Lavallee et al., 2007) (permission was granted from the corresponding author), and four items from the Generalised Contentment Scale (GCS: Hudson & Proctor, 1977) (corresponding author deceased). Thirty-two cognitive/affective and physiological items were constructed based on feedback from six University of Auckland psychology post-graduate students. They were asked to report on how contentment felt and what the opposite to being content is, including physical symptoms. Both positively (n = 25) and negatively (n = 17) phrased items were included to reduce response bias when answering. The pool of items fell into two groups based on the theoretical definition of contentment (discussed in the introduction). These groups included: (a) Cognitive-affect - items related to perceptions of contentment (31 items), e.g. “I feel content” (b) The physical state condition of contentment (11 items), e.g. “My shoulders are relaxed” 3. Exploratory factor analysis
were removed that correlated too highly with other items (over 0.7), indicating they were semantically too similar and therefore redundant or that did not correlate (below 0.30). The remaining 24 items then underwent EFA using PAF in order to identify the underlying structure of the contentment scale. Inspection of the unrotated factor matrix and consideration of the Kaiser criterion suggested there were three factors having eigenvalues above 1 that might be retained. On the third factor no items showed loadings above 0.40. In addition, the scree plot suggested two clear factors above the elbow of the plot. It was therefore decided to drop the third factor and extract and rotate only the first two factors. Items on each factor were then carefully evaluated and systematically eliminated if, a) they had high cross-loadings, b) did not seem to fit the theoretical construct, c) had low primary loadings, or d) with each internal consistency check, the removal of which improved reliability. PAF with direct Oblimin rotation was used because it allows factors to correlate, with the present analysis indicating a correlation of 0.63 between the two factors. Table 1 presents the item loadings on the unrotated first factor and two-factor solution together with item-tototal correlations for the total scale. The two factors together accounted for 64% of the total variance in the correlation matrix. Both loadings on the first factor (range 0.51–0.84) and item to total correlation (range 0.49–0.77) were high and supported an overarching latent contentment factor underlying the construct. Two theoretically distinct factors have emerged: a) a cognitive-affective factor of contentment and, b) a physiological-affective factor of contentment. The final rotation indicated that the first factor accounted for 51% of the variance and the second factor accounted for 13% of the variance with a total of 64% of the variance explained by both. With the theory in mind, the two factors that emerged were labelled a) Cognitive Appraisal, and b) Physiological Appraisal. These labels reflected the underlying constructs of the SCM.
3.1. Participants and procedure 3.3. Discussion The items in the SCM were developed by linking together a cognitive and physiological underpinning of state contentment. Given this, a Confirmatory Factor Analysis (CFA) may have been considered sufficient to confirm the underlying structure. However, in this instance, it was important to explore the factor structure of the underlying variables first because this is a new measure in the development stage and there were items that were newly generated with no prior psychometric testing. Furthermore, some of the items from existing measures were adapted and may therefore perform differently compared to the original measures they were derived from and populations tested on. An Exploratory Factor Analysis (EFA) using Principal Axis Factoring (PAF) was conducted to investigate interrelationships among the items generated for the SCM. Thirty six items were analysed after a face-validity assessment eliminated six items due to having low face validity; specifically, these items were removed due to verbose or awkward sentence construction and not appearing to represent the construct at ‘face value’. The items were headed by the question, “please indicate whether you agree with each item on the list for you at this moment on a scale from 1 to 7”. Responses were made on a 7-point scale with anchors from 1 (strongly disagree) to 7 (strongly agree). The participants consisted of undergraduate and post-graduate students (n = 352) between 17 and 59 years of age (M = 21.96 SD = 6.60) from the university of Auckland). The sample consisted of 78% females and was predominantly European (n = 188), with the remainder being of Asian (n = 78); Indian (n = 21); Māori (n = 18); Pacific Island (n = 9), or ‘other’ ethnicity (n = 46). Participants completed online questionnaires and went into the draw for a $100 gift voucher. 3.2. Results The inter-item correlation matrix was tested using Bartlett's test of sphericity and Kaiser's measure of sampling adequacy detecting no problems that would preclude the use of factor analysis. Twelve items
Study 1 produced a 10-item self-report SCM with two distinct subscales (Table 1) that showed adequate internal consistency of the full scale (Cronbach's alpha of 0.89) (score range, 7–70) and subscales (the cognitive appraisal of 0.88 and the physiological-appraisal of 0.80). Items are scored on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). Either total-items or subscale-item scores are summed to create one score with higher scores indicating greater state contentment. There are four negatively worded items (items 2, 5, 7 and 10) included in the measure. The first factor revealed by the exploratory factor analysis is a cognitive-affective appraisal factor and the second factor revealed a physiological-affective appraisal factor. This result indicated that there were two distinct but related subscales measuring two factors of contentment. The first factor was conceptualised to be a subjective appraisal of one's momentary state of contentment and the second factor is a physiology-based appraisal of one's present contentment levels. Both subscales correlated significantly with each other (r = 0.54, p b 0.01), at a level that showed that they were significantly related and yet have distinct underlying factors making up state contentment. The resulting instrument encompasses the cognitive-affective factor of (state) contentment (score range, 7–28) and, b) a physiological-affective factor of (state) contentment (score range, 7–42). 4. Study 2. Confirmatory Factor Analysis 4.1. Overview A Confirmatory Factor Analysis (CFA) was conducted to compare the fit between the two EFA-extracted factors and a single factor solution using an independent sample, because all item-loadings on the single factor were above 0.51 and it explained over 50% of variance in the data. The CFA fit indices reflect to what extent the hypothesised factor structure is explained by the covariances between the items, which
T.A. Taylor et al. / Personality and Individual Differences 119 (2017) 152–159
155
Table 1 Item-to-total correlations and loadings on the unrotated first factor (1st FA) for the two factor solution of the SCM using Direct Oblimin rotation including State Cognitive Appraisal (CA) and State Physical Appraisal (PA) (n = 352).
6 1 5 2 3 4 10 7 8 9
Items
Item mean (SD*)
All items FA
CA
I feel content At this time I feel in control of my life I feel unfulfilled with what I am achieving in my life unhappy because other things that I would like to be doinga I feel love towards myself All of my needs for contentment have been taken care of My body feels tense I feel stressed out I am calm My shoulders are relaxed
4.46 (1.53) 4.41 (1.58) 3.50 (1.78) 4.08 (1.82) 4.79 (1.55) 3.87 (1.57) 3.72 (1.72) 4.30 (1.74) 4.96 (1.47) 3.76 (1.66)
0.84 0.76 0.68 0.66 0.73 0.68 0.59 0.64 0.61 0.51
0.83 0.78 0.76 0.74 0.72 0.68
PA
Item-total
0.75 0.73 0.7 0.66
0.77 0.71 0.63 0.61 0.67 0.63 0.57 0.62 0.58 0.49
*Standard deviation. a I am unhappy because there are other things in my life I would like to be doing.
refers to the model fit. Four fit indices were used in the current analyses including: the root mean square error of approximation (RMSEA; Marsh, Balla, & Hau, 1996), the comparative fit index (CFA; Bentler, 1990), standardized root-mean-square residual (SRMR; Bentler, 1995), and a chi-square test for deviation of the data from the model. Goodness of fit was assessed against the following criteria proposed for acceptable fits: RMSEA b 0.080, CFI N 0.95, and SRMR b 0.080 (Hu and Bentler, 1999). Due to the high sensitivity of the chi-square statistic to sample size, it is no longer applied to evaluate model fit, but it is included here for a more accurate comparison between two factor models. The maximum likelihood estimation method was used, because of its good applicability to a wide range of analytical conditions (Hu, Bentler, & Kano, 1992). 4.2. Participants and procedure 4.2.1. Participants The participants of the study consisted of adult members from a community in Auckland and students from the University of Auckland (n = 206). The sample were between the ages of 17 and 59 years of age (M = 35.38 SD = 14.69). The sample consisted of 83% females. They were predominantly European (n = 150), with the remainder being Pacific (n = 16), Māori (n = 15); Indian (n = 14); Asian (n = 11); or ‘other’ ethnicity (n = 18). Participants completed the SCM online and went into the draw for $50 movie tickets. 4.3. Study 2. Results and discussion A CFA was conducted to test the two-factor and single factor solutions and in both cases the fit was poor as indicated by the four principle fit indices (Table 2). A poor fit may be associated with residual error variances introduced by individual items. It also refers to noise that can be reduced by combining items into parcels. This has several advantages extensively argued elsewhere (Little, Cunningham, Shahar, and Widamon, 2002; Rushton, Brainerd, and Pressley, 1983). Item parcels Table 2 Goodness-of-fit indices (RMSEA, CFI, and SRMR) of confirmatory factor analyses for one and two factor solutions: Model A and B. Model Model A 1 Factor initial 1 Factor parcelled Model B 2 Factors initial 2 Factors parcelled
RMSEA
CFI
SRMR
Chi Square/df
p-Value
0.178a 0.051
0.750 0.997
0.099 0.019
291.491/35 3.197/2
0.000 0.202
0.141 0.065
0.847 0.993
0.082 0.018
190.750 7.902/4
0.000 0.095
have a higher reliability compared to individual items in that they increase scale points which improve the accuracy of measurement, and reduce the risk of spurious correlations. Moreover, more accurate estimates of latent structures can be obtained using item parcels compared to individual items. Therefore, items were combined in parcels randomly by using a procedure similar to Baer, Smith, Hopkins, Krietemeyer, and Toney (2006). An advantage of this method is that parcels are more stable indicators of a latent construct (Little et al., 2002). The criteria for acceptable fits were RMSEA b 0.080, CFI N 0.95, and SRMR b0.080 (Hu and Bentler, 1999). The fit indices for both the single and the two factor parcelled solutions are presented in Table 2. In this case, the best fit was achieved for both solutions with slightly better fit indices of the single factor solution (Table 2, in bold). The single factor solution indicated neither evidence for collinearity between items (all values b 0.8) nor for multidimensionality with only one eigenvalue above 1. All items displayed significant loadings on one factor with moderate to large magnitude ranging from 0.58 to 0.87 and squared multiple correlations in acceptable range from 0.33 to 0.76. All fit indices satisfied criteria for the good model fit that also included non-significant chi-square test. In the two factor solution there was no evidence for collinearity between items. However, the single eigenvalue above 1(3.09) indicated that a large proportion of total variance was explained by a single factor. Additionally, standardized factor loadings on underlying factors were in the acceptable range from 0.64 to 0.85, only 3 out of 5 factor loadings were statistically significant. Although both models demonstrated acceptable fit indices, the single factor solution appeared to be more adequate compared to the two-factor model. The squared multiple correlations for the two-factor solution ranged from 0.41 to 0.85, which is slightly higher compared to the single factor solution. Overall, the results of the CFA were consistent with the EFA as it indicates a good fit for both a single factor as well as for a two factor solution. In particular, for the EFA all loadings on the first factor without rotation were also relatively higher. Both studies indicate that either the total score of the scale may be used and in situations where either the cognitive or physiological measurement of contentment is required, the subscales may be used individually. 5. Study 3. Convergent and divergent validity 5.1. Participants and procedure
1 Factor parcelled: P1 (items 1 + 4 + 8), P2 (items 2 + 5), P3 (items 3 + 6) and P4 (items 7 + 9 + 10). 2 Factors parcelled; Factor1: P1(items 1 + 4), P2(items 2 + 5), P3(items 3 + 6); Factor 2: P4(items 8 + 10), P5(items 7 + 9). a To be able to distinguish between the goodness of fit of the two factor solutions, goodness-of-fit indices are shown to three decimal places.
To explore convergent and divergent validity the SCM was correlated with measures that were expected to show significant positive and negative correlations with it depending on whether they were negatively or positively valenced. Validity refers to the question of whether the measure is truly measuring the underlying construct it has been designed to measure (Field, 2013). In this case, convergent or divergent validity refers to if the SCM relates in either a positive direction (convergent) or negative direction (divergent) as theoretically expected. The
156
T.A. Taylor et al. / Personality and Individual Differences 119 (2017) 152–159
scores on the Depression, Anxiety and Stress Scale (DASS-21; Henry & Crawford, 2005) were expected to exhibit a significant inverse correlation with the SCM scores. Although it may be possible to feel sad and happy at the same time (Larsen, McGraw, & Cacioppo, 2001), individuals scoring highly in negative affect, would be less likely to score highly in positive affect (Green, Susan, & Peter, 1993). Conversely, the Mindfulness Attention Awareness scale (MAAS: Brown and Ryan, 2003), the Oxford Happiness (OHQ-short form: Hills & Argyle, 2002), and the Satisfaction with Life (SWLS: Diener et al., 1985) scores were posited to show convergent validity. They were expected to correlate positively with the SCM scores given that a) contentment has been reported as one of the emotions induced by mindfulness activities and b) happiness and contentment, although not completely synonymous, are both on the spectrum of emotions identified as positive (Coleman, 2015). Scores on the Satisfaction with Life Scale (SWLS; Diener et al., 1985) were predicted to show convergent validity with the SCM due to the theory that daily contentment levels may reflect perceived global contentment with Life as suggested by Lavallee et al. (2007). A sample (n = 83) of undergraduate and post-graduate students from the University of Auckland were recruited. They were between the ages of 19 and 61 years of age (M = 27.79 SD = 7.41). They completed an online questionnaire package and went into a draw for $100. The individual questionnaires in the questionnaire package were arranged in the following order: 1) SCM, 2) MAAS, 3) DASS, 4) OHQshort form, 5) SWLS. The sample consisted of 81% females and was predominantly European (n = 53), with the remainder being of Asian (n = 11); Indian (n = 2); Maori (n = 5); Pacific Island (n = 3), or other ethnicity (n = 14). 5.2. Measures 5.2.1. Mindfulness The Mindfulness Awareness Attention Scale (MAAS: Brown et al., 2003) is a subjective self-report scale designed to assess dispositional Mindfulness. The MAAS has 15-items with ratings ranging from almost always (1) to almost never (6). Dispositional Mindfulness is defined as having an open or receptive awareness of and attention to what is taking place in the present. The MAAS was designed to measure individual differences in the frequency of mindful states over time and is focussed on the presence or absence of attention to and awareness of what is occurring in the present. To score the measure, the 15 items are summed and the mean calculated. Higher scores indicate higher levels of dispositional mindfulness. Brown et al. (2003) evaluated the psychometric properties of the MAAS using data collected from five studies involving university and community adult samples. One of the community adult experience-sampling studies found that the MAAS alpha was 0.86. Cronbach's alpha reliability score for the MAAS in the present sample was 0.89. 5.2.2. Happiness The Oxford Happiness Questionnaire Short Form (OHQ-short form: Hills & Argyle, 2002) is a subjective self-report measure created with the aim of measuring current levels of personal happiness. The OHQshort form scores items using a scale from 1 (strongly disagree) to 7 (strongly agree). The questionnaire contains five negatively and positively worded items each and the items are designed so they can be intermingled with other items in a construction of personality questionnaires rendering them less susceptible to questionnaire and respondent bias (Hills & Argyle, 2002). Item scores are summed to give a measure of happiness with higher scores indicating higher happiness scores. The OHQ-short form was tested by Hills and Argyle (2002) using a stepwise discriminant analysis that showed the eight-item questionnaire could successfully predict group membership with the full and shorter versions being significantly and strongly correlated, r(168) = 0.93, p b
0.001. In the present study Cronbach's alpha reliability score for the OHQ-short form was 0.76. 5.2.3. Life satisfaction The Satisfaction With Life Scale (SWLS; Diener et al., 1985) is a subjective self-report questionnaire that measures global satisfaction with life. The SWLS includes 5-items, with items scored on a 7-point scale ranging from 1 (strongly disagree) to 7 (strongly agree). Item scores are summed to create one score and higher scores indicate greater life satisfaction. There are no specific domains named within the scale and items are general in their nature enabling the respondent to consider the personal life domains and affective components which they feel contributes to their subjective experience of happiness. Validity for the SWLS was indicated in Pavot, Diener, Colvin, and Sandvik (1991) who found that the Satisfaction with Life Scale converged with the Life Satisfaction Index-A at r = 0.81. The scale is short but effective enough to use in a larger questionnaire package. In the present study Cronbach's alpha reliability score for the SWLS was 0.76. 5.2.4. Negative affect The Depression Anxiety and Stress Scale - 21 (DASS-21; Henry & Crawford, 2005) is a subjective self-report short- version of the DASS and has three subscales which measure three negative emotional states of depression, anxiety and stress. The DASS-21 includes 21 items, with ratings ranging from 0 (did not apply to me) to 3 (applied to me very much, or most of the time). The DASS-21 is a dimensional-state tool and was not designed to be used specifically as a clinical assessment of depression, anxiety or stress; rather, it was designed to clarify the locus of emotional disturbance, as part of the broader task of clinical assessment. Total scores are calculated by summing the scores for each subscale and for the DASS-21, it is then multiplied by two to create the final score. Higher scores indicate higher levels of depression, anxiety, or stress. The DASS has been used widely in clinical and non-clinical populations (Parkitny et al., 2012; Henry & Crawford, 2005; Myint, Choy, Su, & Lam, 2011). Evaluation of the construct validity and normative data of the DASS-21 in a non-clinical sample involving 1794 adults in the UK population indicated that the internal consistency for the total and subscales were 0.93 for the total 21-item DASS-21, 0.88 for the Depression subscale, 0.82 for the Anxiety subscale, and 0.90 for the Stress scale (Henry & Crawford, 2005). In the present study, Cronbach's alpha reliability score was 0.91 for the Depression subscale, 0.74 for the Anxiety subscale, and 0.85 for the Stress subscale. 5.3. Study 3. Results and discussion As expected, the full SCM and the two individual subscales correlated significantly with all measures. The subscales of the DASS (Depression, Anxiety and Stress) and the SCM were moderately and inversely correlated indicating divergent validity while the OHQ, MAAS and SWLS were all positively correlated with the SCM, showing convergent validity. Table 3 shows these relationships. 6. Study 4. Temporal reliability of the SCM The main psychometric criterion, currently used to distinguish state from trait measures are test-retest reliability scores, which are expected to be lower for a state measure. Test-retest reliability estimates for the SCM were computed to explore how the measure works as a state measure over 1 and 2 weeks. Specifically, good trait measures are expected to have test-retest (1 or 2 weeks) of 0.8 and above and in comparison, a valid state measure should be relatively unreliable over time meaning that test-retest reliability scores should be below 0.80, which is a cutoff point for a stable trait measure (Ramanaiah, Franzen, & Schill, 1983; Spielberger, Gorsuch, & Lushene, 1970; Spielberger, 1999).
T.A. Taylor et al. / Personality and Individual Differences 119 (2017) 152–159
157
Table 3 Convergent and divergent correlations between measures. Measures
Mean (SDd)
Depression
Anxiety
Stress
OHQb
MAASc
SWLSd
SCMa SCM(Cognitive)b SCM(Physical)c Mean(SD) Score ranges
46.02 (11.06) 28.60(7.08) 17.43(5.11)
−0.651⁎⁎ −721⁎⁎ −0.406⁎⁎
−0.446⁎⁎ −0.405⁎⁎ −0.4023⁎⁎ 4.25(5.34) 0–28
−0.633⁎⁎ −0.584⁎⁎ −0.555⁎⁎ 11.58(9.17) 0–52
0.728⁎⁎ 0.725⁎⁎ 0.565⁎⁎ 33.67(6.28) 16–47
0.541⁎⁎ 0.464⁎⁎ 0.519⁎⁎ 3.84(0.72) 1.80–5.53
0.689⁎⁎ 0.734⁎⁎ 0.466⁎⁎
7.38(9.76) 0–52
22.89(6.36) 9–35
Scores for the total SCM ranged from 19 to 70, Cognitive appraisal subscale = 13–42, physiological subscale = 6–28. a State Contentment Measure (total score). b Cognitive appraisal subscale. c Physiological appraisal subscale. d Standard deviation. ⁎⁎ p ≤0.001.
6.1. Participants and procedure
6.2. Study 4. Results and discussion
Participants in sample 1 consisted of women who belonged to a commercial gym in Auckland (n = 70) between the ages of 20 and 66 years of age (M = 41.74, SD = 11.97). The sample was predominantly European (n = 41), with the remainder being of Pacific (n = 10); Indian (n = 6); and Māori (n = 4) and not specified (n = 7). The participants in sample 2 were a combination of community members from three central Auckland suburbs and students from the University of Auckland (n = 51) who had completed a five day mindfulness course (20 min per day), between the first test and re-test of the SCM. They were between the ages of 17 and 71 years of age (M = 35.33., SD = 16.10). The sample consisted of 70.59% females. The sample was predominantly European (n = 41), with the remainder being of Māori (n = 2), Asian (n = 5), and not specified (n = 5). The participants in sample 3 were a combination of community members from three Auckland suburbs and students from the University of Auckland (n = 47) who had completed a five day physical activity course (20 min per day), between the first test and re-test of the SCM. They were between the ages of 18 and 62 years of age (M = 33.26, SD = 12.26). The sample consisted of 65.96% females. The sample was predominantly European (n = 33), with the remainder being of Indian (n = 5), Asian (n = 5) and Māori (n = 2) and not specified (n = 5). The participants completed the SCM online and went into the draw for either movie tickets or $100 vouchers.
As noted above test retest reliability for state measures should be consistently lower than for trait measures if they are valid indices of a state rather than a trait. Results of the test-retest reliability estimated for the SCM subscales and the total scale (Table 4) support construct validity of the instrument as a state measure with all coefficients in the expected range (0.50–0.72). This figure is comparable to test-retest results from other state measures, (e.g., Ramanaiah et al., 1983; Spielberger, 1999; Spielberger et al., 1970). Overall, the test-retest coefficients were lower for the physiological subscale suggesting that this subscale is more sensitive to state changes. This result would be expected from a theoretical perspective given that changes in one's physical state may show variability in change over a short time period (Daley & Welch, 2007; Herring & O'Connor, 2009). The test-retest reliability was computed for individual SCM items between baseline and at 1 and 2 weeks (Table 5) providing further support for ‘stateness’ of the SCM items. All test-retest scores were lower than 0.70 with the exception of one item, ‘I feel unfulfilled with what I am achieving in my life’, in sample 3 over one week. Conversely, this relationship was not replicated with this item for the other two samples which were 0.52 over two weeks (sample 1) and 0.64 over one week (sample 2). To investigate the change sensitivity of the individual items, the proportion of agreement for test-retest differences was applied to the individual items of the scale using the method proposed by Nevill, Lane,
Table 4 Means, standard deviations (SD), internal and test-retest reliability estimates for the State Contentment Measure. Measurement
State Contentment Measure Mean (SD) Cronbach's alpha Test-retest (r) Cognitive Subscale Mean (SD) Cronbach's alpha Test-retest (r) Physiological Subscale Mean (SD) Cronbach's alpha Test-retest (r)
2 Weeks community gym sample (n = 65) (sample 1)
1 Week pre/post meditation (n = 51) (sample 2)
1 week pre/post physical activity (n = 47) (sample 3)
Baseline
2 Weeks
Baseline
1 Week
Baseline
1 Week
45.74(9.92) 0.84
48.26 (9.86) 0.85 0.72⁎⁎
41.47 (11.38) 0.9
46.94(10.90) 0.9 0.63⁎⁎
41.72(11.74) 0.92
44.65(10.10) 0.9 0.70⁎⁎
27.8(6.69) 0.82
29.55(6.49) 0.84 0.72⁎⁎
25.78(7.19) 0.9
28.61(7.27) 0.9 0.70⁎⁎
24.7(7.33) 0.89
25.97(6.92) 0.83 0.70⁎⁎
17.94(5.16) 0.84
18.71(5.08) 0.84 0.63⁎⁎
15.69(5.17) 0.84
18.33(4.71) 0.89 0.50⁎
17.02(5.33) 0.85
18.68(4.45) 0.89 0.62⁎⁎
Means are significantly different (p = 0.01); Test-retest Spearmans correlations were computed between the baseline scores and scores after a 2 weeks interval. For sample 1, total SCM scores ranged from 25 to 65 at baseline and 28 to 70 at Week 2. Cognitive appraisal subscale scores ranged from 13 to 42 at baseline and 15 to 42 at Week 2. Physiological appraisal subscale scores ranged from 6 to 26 at baseline and 10 to 28 at Week 2. For sample 2, total SCM scores ranged from 22 to 69 for baseline and 26 to 70 for 1 week. Cognitive appraisal subscale scores ranged from 14 to 41 for baseline and 13 to 42 for week 1. Physiological appraisal subscale scores ranged from 6 to 28 for baseline and 7 to 28 for 1 week. For sample 3, total SCM scores ranged from 13 to 64 for baseline and 25 to 63 for 1 week. Cognitive appraisal subscale scores ranged from 8 to 41 at baseline and 13–38 for Week 1. Physiological appraisal subscale scores ranged from 5 to 25 at baseline and 8–26 for Week 1. ⁎ p = 0.05. ⁎⁎ p = 0.01.
158
T.A. Taylor et al. / Personality and Individual Differences 119 (2017) 152–159
Table 5 Test-retest Spearman correlations between individual SCM items between baseline and at 1 and 2 weeks. 2 Weeks community gym sample (n = 65) (sample 1)
1 Week pre/post meditation sessions (n=51) (sample 2)
1 Week pre/post physical activity (n = 40) (sample 3)
Baseline
Baseline
Baseline
2 Weeks 0.65⁎⁎ 0.61⁎⁎ 0.58⁎⁎ 0.52⁎⁎ 0.52⁎⁎ 0.52⁎⁎ 0.52⁎⁎ 0.51⁎⁎ 0.47⁎⁎ 0.36⁎⁎
I feel love towards myself I feel content At this time I feel in control of my life needs for contentment have been taken care ofa I feel unfulfilled with what I am achieving in my life other things I would like to be doing in my lifeb My shoulders are relaxed I feel stressed out My body feels tense I am calm
1 Week 0.56⁎⁎ 0.58⁎⁎ 0.61⁎⁎ 0.57⁎⁎ 0.64⁎⁎ 0.48⁎⁎ 0.44⁎⁎ 0.53⁎⁎ 0.39⁎⁎ 0.32⁎⁎
1 Week 0.50⁎⁎ 0.57⁎⁎ 0.57⁎⁎ 0.65⁎⁎ 0.78⁎⁎ 0.61⁎⁎ 0.52⁎⁎ 0.41⁎⁎ 0.45⁎⁎ 0.39⁎⁎
a
All of my needs for contentment have been taken care of. I am unhappy because there are other things in my life I would like to be doing. ⁎⁎ p = 0.01. b
Kilgour, Bowes, and Whyte (2001). To increase sample size for the purpose of this analysis, both 1-week pre/post meditation and physical activities independent samples were combined producing a sample of n = 91. Table 6 shows that individual items display different stability across time points. For example, the item ‘I am calm’ in particular shows a high sensitivity to change compared to the item ‘I feel unfulfilled with what I am achieving in life’ which showed a higher degree of stability expected of an item relating to life satisfaction. However, all items consistently displayed stability patterns expected for a valid state measure.
7. Discussion, limitations and future directions In the present age when positive emotions are the focus of increasing interest for their adaptive benefits in improving wellbeing, the results of the four studies reported here suggest that the SCM is a promising new measure for the assessment of state contentment. The two-factor solution suggested that the contentment construct involves cognitive-affective and physical appraisal of present contentment levels with high internal reliability of both the total measure and the two individual subscales. The Confirmatory Factor Analysis indicated that acceptable fit was achieved for both the one and two factor solutions, but that there was a slightly better fit for the single factor solution that supported the SCM measures state contentment as influenced by emotion- related cognitions. Test-retest results from study 4 gave strong evidence that the SCM behaves as expected in terms of it being a state measure of contentment (compared to a life satisfaction or a trait measure). This indicated that the SCM could be reliable to use in gauging differences in an individual's state pre/post an experiment or intervention. Table 6 Confidence intervals for the proportion of agreement between two test administrations with one week interval (n = 91). SCM Items
Confidence Interval
I feel love for myself I feel content At this time I feel in control of my life All of my needs for contentment have been taken care of I feel unfulfilled with what I am achieving in life I am unhappy because there are other things I would like to be doing in life My shoulders are relaxed I feel stressed out My body feels tense I am calm
0.30–0.50 0.28–0.58 0.34–0.54 0.25–0.45 0.36–0.56 0.27–0.47 0.33–0.54 0.22–0.41 0.26–0.46 0.16–0.34
All four studies used the general public and university students. The university student sample may introduce a potential bias in that this population is more highly educated, from higher socio-economic backgrounds, and are younger than a representative community sample. At the same time, the results from the community sample (in study 4) also supported the psychometric properties of the measure giving support for generalisability. Generalisability may be limited on the other hand, in that the participants were predominantly female, and of Caucasian ethnicity, who reside in Auckland. Future studies should test the measure on more diverse populations with the potential for adapting it to suit specific cultural and socio-economic groups. Relatedly, the questionnaire items did not undergo a face-validity assessment with groups from a range of different educational backgrounds and as such, future adaptions of the measure should consider this at the design phase. Additionally, a limitation for validation was that the individual questionnaires were not counter-balanced which may have introduced order and sequence effects, potentially compromising significant correlations. Future validation assessments of the questionnaire should use counter-balancing. A further limitation of the present study is that the SCM's validity was tested in the context of other self-report measures. This is a limitation considering that it was designed partly, to measure physical symptoms of contentment. Future studies could further investigate whether this measure converges with physiological outcomes (e.g. EEG, heart rate, blood pressure or skin conductance) relevant to the state of contentment to give further validation along with self-report. Other ideas include the testing of the SCM in the clinical setting. Completed as part of counselling sessions, the SCM could potentially give the therapist an idea of where their client was at in terms of their relative contentment levels. Normative data may also be determined in future to assess population norms. Another avenue of investigation is the potential to create a trait-based contentment measure based on the ‘contented personality’ type. Finally, future research should explore the relationships between the cognitive, affective, and physiological elements of contentment to ascertain how these interact and produce their positive adaptive effects, and in comparison to other emotions. In conclusion, the present study created a measure of state contentment based on a conceptualisation of contentment that includes transient, stable and physiological underpinnings as identified by the existing literature. The SCM scale items can be summed together as a total score of state contentment, or the two subscales can be utilised separately, reflecting distinct cognitive and physiological factors. Finally, the psychometric testing in the present study confirmed that the items satisfied all theoretical and empirical criteria for a state measure and researchers can now use the SCM as a reliable and valid tool to measure state contentment induced through positive-emotion eliciting activities.
T.A. Taylor et al. / Personality and Individual Differences 119 (2017) 152–159
References Baer, R., Smith, G., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). Using self-report assessment methods to explore facets of mindfulness. Assessment, 13(1), 27–45. Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238–246. Bentler, P. M. (1995). EQS structural equations program manual. Encino, CA: Multivariate Software. Berenbaum, H. (2002). Varieties of joy-related pleasurable activities and feelings. Cognition & Emotion, 16(4), 473–494. Berkowitz, L. (2000). Causes and consequences of feelings. Cambridge University Press. Brown, K. W., & Ryan, R. M. (2003). The benefits of being present: Mindfulness and its role in psychological well-being. Journal of Personality and Social Psychology, 84(4), 822. Buss, A. H. (1989). Personality as traits. American Psychologist, 44, 1378–1388. Cahn, B. R., & Polich, J. (2006). Meditation states and traits: Eeg, erp, and neuroimaging studies. Psychological Bulletin, 132, 180–211. Coleman, A. M. (2015). Oxford psychology reference dictionary (4th ed.). Oxford University Press. Daley, A. J., & Welch, A. (2007). The effects of 15 min and 30 min of exercise on affective response both during and after exercise. Journal of Sports Sciences, 22(7), 621–628. Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale. Journal of Personality Assessment, 49(1), 71–75. Dittrich, A. (1998). The standardized psychometric assessment of altered states of consciousness (ASCs) in humans. Pharmacopsychiatry, 31(Suppl. 2), 80–84. Ekman, P. (1993). Facial expression and emotion. American Psychologist, 48(4), 384. Ellsworth, P. C., & Smith, C. A. (1988). Shades of joy: Patterns of appraisal differentiating pleasant emotions. Cognition and Emotion, 2, 301–331. Epstein, S. (1984). Trait theory as personality theory: Can a part be as great as the whole? Psychological Inquiry, 5, 120–122. Field, A. (2013). Discovering statistics using SPSS. London: Sage Publications. Fredrickson, B. L. (1998). What good are positive emotions. Review of General Psychology, 2(3), 300–319 (Sep 1998). Fredrickson, B. L., & Branigan, C. (2005). Positive emotions broaden the scope of attention and thought-action repertoires. Cognition and Emotion, 19(3), 313–332. Fredrickson, B. L., Mancuso, R. A., Branigan, C., & Tugade, M. M. (2000). The undoing effect of positive emotions. Motivation and Emotion, 24(4), 237–258. Friedman, H. L. (1983). The self-expansiveness level form: A conceptualization and measurement of a transpersonal construct. Journal of Transpersonal Psychology, 15, 37–50. Friedman, H. L., & MacDonald, D. A. (1997). Toward a working definition of transpersonal assessment. Journal of Transpersonal Psychology, 29, 105–122. Green, D. P., Susan, L. G., & Peter, S. (1993). Measurement error masks bipolarity in affect ratings. Journal of Personality and Social Psychology, 64, 1029–1041. Henry, J. D., & Crawford, J. R. (2005). The short-form version of the depression anxiety stress scales (DASS-21): Construct validity and normative data in a large non-clinical sample. British Journal of Clinical Psychology, 44(2), 227–239. Herring, M. P., & O'Connor, P. J. (2009). The effect of acute resistance exercise on feelings of energy and fatigue. Journal of Sports Sciences, 27(7), 701–709. Hills, P., & Argyle, M. (2002). The oxford happiness questionnaire: A compact scale for the measurement of psychological well-being. Personality and Individual Differences, 33(7), 1073–1082. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. Hu, L. T., Bentler, P. M., & Kano, Y. (1992). Can Test Statistics in Covariance Structure Analysis Be Trusted? Psychological Bulletin, 112(2), 351–362. Hudson, W. W., & Proctor (1977). Assessment of depressive affect in clinical practice. Journal of Consulting and Clinical Psychology, 45(6), 1206–1207.
159
Huppert, F. A., & So, T. T. C. (2013). Flourishing across Europe: Application of a new conceptual framework for defining well-being. Social Indicators Research, 110(3), 837–861. Izard, C. E. (1991). The Psychology of Emotions. New York: Plenum Press. Larsen, J. T., McGraw, A. P., & Cacioppo, J. T. (2001). Journal of Personality and Social Psychology, 81(4), 684–696. Lavallee, L. F., Hatch, P. M., Michalos, A. C., & McKinley, T. (2007). Development of the contentment with life assessment scale (CLAS): Using daily life experiences to verify levels of self-reported life satisfaction. Social Indicators Research, 83(2), 201–244. LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155–184. Little, T. D., Cunningham, W. A., Shahar, G., & Widamon, K. F. (2002). To parcel or not to parcel: Exploring the question, weighting the merits. Structural Equation Modeling, 9(2), 151–173. Marsh, H. W., Balla, J. R., & Hau, K. -T. (1996). An evaluation of incremental fit indices: A clarification of mathematical and empirical properties. In G. A. Marcoulides, & R. E. Schumacker (Eds.), Advanced structural equation modeling: Issues and tech-niques. Mahwah, NJ: Erlbaum. Medvedev, O. N., Shepherd, D., & Hautus, M. J. (2015). The restorative potential of soundscapes: A physiological investigation. Applied Acoustics, 96, 20–26. Medvedev, O. N., Krägeloh, C. U., Narayanan, A., & Siegert, R. J. (2017). Measuring mindfulness: Applying Generalizability theory to distinguish between state and trait mindfulness, 8(4), 1036–1046. Myint, K., Choy, K., Su, T., & Lam, S. (2011). The effect of short-term practice of mindfulness meditation in alleviating stress in university students. Biomedical Research, 22(2) (India). Nevill, A. M., Lane, A. M., Kilgour, L. J., Bowes, N., & Whyte, G. P. (2001). Stability of psychometric questionnaires. Journal of Sports Sciences, 19(4), 273–278. Parkitny, L., McAuley, J. H., Walton, D., Pena Costa, L. O., Refshauge, K. M., Wand, B. M., & Moseley, G. L. (2012). Rasch analysis supports the use of the depression, anxiety, and stress scales to measure mood in groups but not in individuals with chronic low back pain. Journal of Clinical Epidemiology, 65(2), 189–198. Pavot, W., Diener, E., Colvin, C. R., & Sandvik, E. (1991). Response artifacts in the measurement of subjective well-being. Social Indicators Research, 24, 35–56. Ramanaiah, N. V., Franzen, M., & Schill (1983). A psychometric study of the state-trait anxiety inventory. Journal of Personality Assessment, 47, 531–535. Rushton, J. P., Brainerd, C. J., & Pressley, M. (1983). Behavioral development and construct validity: The principle of aggregation. Psychological Bulletin, 94(1), 18–38. Sin, N. L., & Lyubomirsky, S. (2009). Enhancing well-being and alleviating depressive symptoms with positive psychology interventions: A practice-friendly meta-analysis. Journal of Clinical Psychology: In Session, 65(5), 467–487. Spielberger, C. D. (1999). Manual for the state-trait anger expression inventory-2. Odessa, FL: Psychological Assessment Resources. Spielberger, C. D., Gorsuch, R. L., & Lushene, R. (1970). Test manual for the state-trait anxiety inventory. Palo Alto, CA: Consulting Psychologists Press. Stephens, C. L., Christie, I. C., & Friedman, B. H. (2010). Autonomic specificity of basic emotions: Evidence from pattern classification and cluster analysis. Biological Psychology, 84(3), 463–473. Stevenson, A. (2010). Oxford dictionary of English (3rd ed.). Oxford University Press. Vaitl, D., Birbaumer, N., Gruzelier, J., Jamieson, G. A., Kotchoubey, B., Kubler, A., ... Weiss, T. (2005). Psychobiology of altered states of consciousness. Psychological Bulletin, 131, 98–127. Weiss, L. A., Westerhof, G. J., & Bohlmeijer, E. T. (2016). Can we increase psychological well-being? The effects of interventions on psychological well-being: A meta-analysis of randomized controlled trials. PloS One, 11(6). Williams, B. (2009). A glimpse into the meditating brain. University of New Mexico.
