Happiness, or the degree to which a person evaluates his\her life positively and favorably (Hart et al., 2018), is a widely and commonly used word, and -as an emotion- it has been identified by numerous authors as one of the basic humans affects (Helliwell & Aknin, 2018; Jebb et al., 2020). Despite some personal and cultural differences in the inclination to show and recognize happiness (Chei et al., 2018), people like to feel happy, and they seek happiness in major life milestones (e.g., marriage, job promotion, parenting) and often wish to experience it more frequently during their life (Ford et al., 2014; Sotgiu, 2016).

Happiness, according to its various definitions (Diener et al., 2020; Steptoe, 2019), includes aspects such as joy, life satisfaction, personal growth, well-being, or meaning of life, and appears to be clearly related to life outcomes (e.g., satisfaction, well-being, personal stress; Lu & Argyle, 1994), socio-demographic factors (e.g., age, social support; Argyle et al., 1995; Cheng & Furnham, 2003; Sotgiu et al., 2011), and psychological skills (e.g., resilience, coping, awareness, acceptance; Argyle & Lu, 1990; Chrostek, 2016; Shoaakazemi et al., 2013; Veenhoven, 2012). Furthermore, other studies have reported significant relationships between happiness and self-esteem (Cheng & Furnham, 2003).

In addition to the plethora of theories on its components (e.g., Kahneman, 1999; Seligman, 2002), happiness is a central topic in public health, research, and applications (e.g., Chrostek, 2016; Pretty & Barton, 2020). As a result, the importance of measuring happiness in various contexts (e.g., De Giorgio et al., 2023; Kumar et al., 2022) has triggered a theoretical and methodological discussion, yielding an astounding plethora of practical tools and approaches throughout time (e.g., Helliwell & Aknin, 2018). The corporate world, institutions, and education have all recently become much more interested in the topic of happiness, which has resulted in awards and certifications (e.g., Great Places to work®, The Awards of Happiness®; Lee & Yoon, 2020; Ravina Ripoll et al., 2022), as well as the development of interventions for these contexts like organizational mindfulness, applied positive psychology, and positive coaching psychology (e.g., Biswas-Diener, 2010; Barattucci et al., 2019; Ramaci et al., 2020). Despite the importance of happiness, currently, there is a paucity of short, flexible, and reliable psychometric measures of this construct, especially within the Italian context. Considering that the psychometric properties of a measure cannot be generalized to other populations (Messick, 1989), we conducted a correlational design research in an Italian sample to provide evidence of both the cross-cultural generalizability and the good psychometric properties of a widely-used brief measure of subjective happiness, namely the Oxford Happiness Questionnaire.

Measuring happiness

There are several contributions in the literature that highlight the substantial overlap in the definitions of happiness, well-being, and health, as well as in their constituent dimensions (as discussed by Argyle in 2001, and Medvedev & Landhuis in 2018). The conceptualization of happiness changes based on the different theoretical frameworks and the context in which it is used (Kaczmarek, 2017; Veenhoven, 2016). Additionally, the difficulty in accurately assessing the construct is linked to the fact that happiness can be seen as both a temporary or situational experience, and a natural predisposition or disposition (Dambrun et al., 2012). Beyond the theoretical frameworks, some authors have focused primarily on happiness experience, mindset, and inner states such as optimism, freedom, acceptance, peace, and cheerfulness (Delle Fave et al., 2011; Diener et al., 2017). Others have focused on life events and achievements as success, achievement of goals, richness, and coherence of the results in the various dimensions of life (Ng & Diener, 2014; Waterman et al., 2010). The multitude of approaches to the study and measurement of happiness has resulted in an impressive number of studies across various psychological fields. This has led to the measurement of happiness becoming a crucial issue in both institutions and research (De Giorgio et al., 2023). Over the years, the need for tools to measure different dimensions of happiness (Monacis et al., 2021), has conducted to the development of various instruments, such as: the Self-centeredness/Self-lessness Happiness Model (SSHM; Dambrun et al., 2012); the Authentic Happiness Inventory, based on the three routes to happiness theory (life of meaning, life of pleasure, and life of engagement (Peterson et al., 2007); the Subjective Happiness Scale (SHS-4) (Lyubomirsky & Lepper, 1999), a short measure with four items assessing the global assessment of one’s happiness.

Among these questionnaires, one of the most used to investigate happiness is the Oxford Happiness Inventory (OHI), developed by Argyle et al. (1989). The OHI consists of 29 items that evaluate the general level of well-being and domains of self-realization. The OHI was originally based on the idea that happiness is made up of three partially independent components: (i) the frequency and intensity of positive affect or joy, (ii) the average level of satisfaction, (iii) the absence of negative feelings such as depression and anxiety (Argyle & Crossland, 1987). However, the new 29-item did not produce optimal results with regards to verifying internal dimensions, factorial solutions, and cultural differences (Hills & Argyle, 2002). Despite the validation study of the original instrument (Argyle et al., 1989), its psychometric characteristics have been mostly investigated to date with homogeneous and limited-in-size samples from individual working sectors (Stewart et al., 2010). In the Italian context, only the 29-items original tool has been validated in a sample of adolescents (Meleddu et al., 2012).

To create an easier-to-administer questionnaire, Hills and Argyle (2002) developed a short 8-item version (Oxford Happiness Questionnaire – OHQ) which has been translated into Italian (De Giorgio et al., 2023). Thus, the present cross-sectional study aimed to examine the psychometric properties (i.e., the factorial validity and reliability) of the Italian version of the OHQ scale and its validity in measuring happiness.

There is broad interest in gender differences in happiness. Findings on this topic have been inconsistent, with a recent meta-analysis finding no significant differences in subjective well-being in an overall sample of more than one million individuals (Batz-Barbarich et al., 2018). To provide additional literature on this topic, we tested the measurement invariance of the scale across gender.

From the literature review, an evident positive relationship emerges between happiness and satisfaction (of life, work, personal, etc.), and a negative one with stress (e.g., Peterson et al., 2007; Schiffrin & Nelson, 2010), however, the underlying psychological mechanisms are not completely clear yet. Recently, some research has focused on the role of some psychological skills in the mechanisms of happiness; in particular, several indications seem to highlight the role of some emotional and cognitive regulation mechanisms (e.g., mindfulness, emotion regulation problems, acceptance, awareness; Hills & Argyle, 2001; Crowley et al., 2022; Quoidbach et al., 2010; Van et al. 2023). As such, we further examined the convergent and divergent validity of the measure, and basing on the above mentioned literature indications, we hypothesized that the OHQ total score will be positively associated with greater life satisfaction (e.g., Hills & Argyle, 2001) and mindful attention to the present (e.g., Crowley et al., 2022), and negatively correlated with perceived psychosocial stress (e.g., Argyle et al., 1995; Suh et al., 1996; Schiffrin & Nelson, 2010) and emotion regulation problems (e.g., Hills & Argyle, 2001; Quoidbach et al., 2010).

Additionally, we aimed to explore the relationship between happiness and socio-demographic factors, such as gender, age, relationship status, education level, number of children, work seniority, and salary. Literature has shown that these factors can play a role in determining subjective happiness (Cheng & Furnham, 2003; Clark, 2018; Diener et al., 2018). Thus, it is important to understand the role of these variables in determining happiness levels in order to design targeted interventions that can help individuals increase their happiness.

Method

Translation process

First, two Italian researchers in the field of psychology with good knowledge and proficiency in English translated the 8-item independently. A third researcher then compared the translations and produced a first draft. This draft was administered to a small pilot sample (N = 38). The intelligibility and semantic coherence of the draft were evaluated through interviews with a group of 12 adults. Following the improvement of some items’ wording, a native English speaker was consulted to perform a back-translation of the new version and to compare it with the original version and the validated Italian version (Meleddu et al., 2012). Finally, based on the results of the previous phases, the two principal investigators created the final version of the instrument.

Participants

A total of 917 Italian participants (mean age = 37.58 ± 12.69 years; 694 females, 75.7%) volunteered for this cross-sectional study. The eligibility criteria included being over the age of 18 and proficient in Italian. Among the participants, 42.5% (N = 385) held a high school diploma, and 38.5% (N = 353) held a university degree. Most of the participants were either engaged or married (69.5%, N = 637) and had no children (62.9%; N = 576). Furthermore, the sample was comprised mostly of working individuals (79.2%, N = 726) including office workers (32.3%, N = 296), professional freelances (20.4%, N = 187), and teachers (9.6%, N = 88). The average work seniority was 13.67 ± 11.52 years.

Measures

Happiness. According to Argyle (2001), happiness can be defined as a positive inner experience, the highest good, and the ultimate motivator for all human behaviours; it’s made up of three main components: frequent positive affect or joy, a high level of satisfaction over a period, and positive feelings. The Oxford Happiness Questionnaire (OHQ; Hills & Argyle, 2002) is an 8-item self-report measure (item example: “I am well satisfied about everything in my life”) developed as a multidimensional scale to measure happiness. The items are rated on a 6-point Likert-type scale (from 1 = strongly disagree, to 6 = strongly agree), with higher total scores indicating greater subjective well-being and happiness. The OHQ is the short form of the 29-item Oxford Happiness Inventory (OHI; Argyle et al., 1989; Meleddu et al., 2012). In the current study, only the OHI was administered, and the OHQ was derived from its longer version. The internal consistency of the OHI was excellent (ω = 0.92).

Life satisfaction was measured using five items (item example: “How satisfied are you with….?”) that were translated from an overall job satisfaction scale (Cammann et al., 1983; Mariani et al., 2015), and pertained to different aspects of the life experience; items were assessed on a five-point Likert scale, ranging from 0 = “totally dissatisfied” to 4 = “totally satisfied”. In the current study, the reliability was excellent (ω = 0.89).

Perceived Stress. The Italian version of the Perceived Stress Scale (PSS-10; Cohen & Williamson, 1988; Mondo et al., 2021) is a 10-item self-report measure of perceived stress or the degree to which respondents appraise events as stressful during the past month. Items are rated on a 5-point Likert-type scale (from 0 = never, to 4 = very often), and higher total scores indicate greater perceived stress (item example: “In the last month, how often have you been upset because of something that happened unexpectedly?”). In the current study, the reliability of this measure was excellent (ω = 0.88).

Emotion regulation problems. The Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004; Sighinolfi et al., 2010) is a 36-items self-report measure of individuals’ typical levels of emotion dysregulation (item example: “I am clear about my feeling”). Items are rated on a Likert-type scale ranging from 1 (almost never) to 5 (almost always), and higher scores on DERS indicate greater tendencies towards emotion dysregulation. In the current study, the reliability of this measure was excellent (ω = 0.77).

Mindful Attention. The Mindful Attention Awareness Scale (MAAS; Brown & Ryan, 2003; Rabitti et al., 2013) is a 15-item scale designed to assess a core characteristic of dispositional mindfulness, namely open or receptive awareness of and attention to what is taking place in the present. Each item is rated on a 6-point Likert scale (from 1 = almost always, to 6 = hardly ever), where higher scores in the MAAS reflect a greater presence on mindfulness (item example: “I find it di cult to stay focused on what is happening in the present”. In the present study, the reliability of these measures was good (ω = 0.75).

Additionally, the survey involved the collection of socio-demographic information, such as age, gender, marital status, number of children, work seniority, salary, and wage.

Procedure

Participants were recruited through notices placed in social networks and through a snowball procedure. Participants voluntarily filled out an online and completely anonymous survey that was following a short post regarding happiness, between September 2021 and January 2022.

After providing informed consent, they were redirected to a webpage that outlined the purposes of the research and provided the researchers’ contact information. Then, participants completed a sociodemographic questionnaire, and the above-mentioned measures, taking approximately 40 min to complete. The study was reviewed and approved by the academic Ethics Committee of eCampus University (11/12/2019, protocol nr. 022019).

Statistical analysis

We evaluated the internal validity of the OHQ through a Confirmatory Factor Analysis (CFA) with Maximum Likelihood estimation. The criteria for optimal model fit were selected based on the following: a root mean square error of approximation (RMSEA) of 0.06 or lower, a greater RMRSEA’s 90% confidence interval bound of 0.08 or less, a comparative fit index (CFI) and a Tucker-Lewis (TLI) of 0.95 or higher, and a standardized root mean squared residual (SRMR) of 0.05 or lower (Hu & Bentler, 1999). Finally, we used a cut-off of > |0.32| to select significant factor loadings (Tabachnick & Fidell, 2018).

As a second step, we examined the internal consistency of the OHQ total score using two methods: McDonald’s model-based composite reliability coefficient (i.e., McDonald’s Omega, which is considered fair when > 0.70, and good when > 0.80; Cho, 2022) and mean inter-item correlations (that -when comprised between 0.15 and 0.50- are indicative of the good internal consistency of a scale; Clark & Watson, 2019). We further tested the measurement invariance (MI) of the scale across gender through the following steps: (a) testing for invariance of number of factors (configural invariance); (b) testing for the equality of factor loadings (metric invariance); and (c) testing for the equality of indicator intercepts (scalar invariance). We then compared the nested models (configural vs. metric, and metric vs. scalar) using cut-off values of ΔCFI < 0.01 and ΔRMSEA < 0.015 (Cheung & Rensvold, 2002).

Finally, we assessed the convergent validity of OHQ by comparing it with sociodemographic variables, and other self-report measures of perceived stress, life satisfaction, emotion dysregulation, and mindfulness attention and awareness, through independent sample t-tests, and Pearson’s r correlation coefficients. The effect sizes were interpreted according to guidelines established by Cohen (1988): small (d = 0.20; .10 < r < .30), medium (d = 0.50; .30 < r < .50), and large (d > 0.80; r > .50).

Statistical analyses were carried out using Statistical Package for Social Sciences (SPSS) version 28 and MPLUS version 8.4. All statistical tests were two-tailed and statistical significance was set at a p-value of 0.05.

Results

Preliminary analyses

Before analyses, data were initially screened for univariate and multivariate outliers, and for distribution normality. There were no univariate outliers as participants’ z values in all variables were < |2|. However, 47 multivariate outliers were identified through Mahalanobis distance with a p < .001 (Tabachnick & Fidell, 2018), and were removed leaving to a total of 870 cases for subsequent analyses. All OHQ items were normally distributed (i.e., their skewness and kurtosis values were ≤ |1|). Descriptives for each OHQ item are reported in Table 1.

Table 1 Descriptive statistics for the items of the Oxford Happiness Questionnaire (OHQ)
Full size table

Socio-demographic

Gender differences emerged in the level of happiness, with women reporting significantly lower levels (M = 3.88, SD = 0.89) in the dependent variable than men (M = 4.09, SD = 0.89), t (915) = -2.97, p < .003, d = 0.23, with a small effect. Differences also emerged between individuals who were single and people who were married or in a relationship, where married and people in a relationship were happier (M = 4.09; SD = 0.87) than single people (M = 3.77, SD = 0.91), t (915) = -4.36, p < .001, d = 0.36, with a small effect. Additionally, several other sociodemographic variables (i.e., age, educational level, children, working seniority and salary), were found to be significantly and positively associated with happiness (Table 2), but these effects were trivial to small. That is, being older, having a greater education level, having children, having a greater wage, or work seniority were all associated with higher levels of happiness. Among all socio-demographic variables, wage was the most highly correlated with happiness (Thomson et al., 2022).

Table 2 Zero-order correlations among the Oxford Happiness Questionnaire (OHQ) and socio-demographic variables
Full size table

Main analyses

The factor structure of the Italian version of the OHQ was evaluated through a CFA. The initial model evidenced a fair fit to the data (N = 870): χ2 (20) = 152.632, p < .001, RMSEA = 0.087 (90% CI: 0.075–0.100), SRMR = 0.042, TLI = 0.872, CFI = 0.908, despite all items having factor loadings > 0.43 with their latent dimension (see Fig. 1). Thus, we revised the solution by examining the modification indices and adding a covariance between two pairs of similarly worded items (i.e., items 1 and 2, and items 1 and 5; Table 1). The revised model had an excellent fit to the data (N = 870): χ2 (18) = 90.255, p < .001, RMSEA = 0.068 (90% CI: 0.054–0.082), SRMR = 0.032, TLI = 0.922, CFI = 0.950.

Fig. 1
figure 1

Standardized weights, error variances and correlations among errors of the Oxford Happiness Questionnaire (OHQ). Note. N = 870

Full size image

We then evaluated the generalizability of the model across males and females through a multigroup CFA, where an unconstrained model with factor loadings free to vary between subgroups was compared with a constrained model, in which the factor loadings were held constant across subgroups. Before conducting the multigroup analysis, we performed a CFA separately for males and females, whose results indicated a fair to good fit of the data for each subgroup. As evidenced in Table 3, the scale was invariant across gender.

Table 3 Measurement Invariance of the Oxford Happiness Questionnaire (OHQ) by sex (females = 694; males = 223)
Full size table

As for the internal consistency, the Italian version of the OHQ evidenced fair reliability, with a McDonald’s Omega of .78, and a mean inter-item correlation of .31.

Finally, means, standard deviations, and zero-order correlations between OHQ and all other psychological measures used in this study are reported in Table 4. The OHQ total score had a significant positive correlation with OHI, with large effects. This suggested that the OHQ led to almost overlapping results with its long version, providing evidence for its good convergent validity. Furthermore, the scale evidenced significant negative correlations with perceived stress and emotion dysregulation and significant positive correlations with life satisfaction and mindful attention to the present (Argyle et al., 1995; Schiffrin & Nelson, 2010), with medium-to-large effects.

Table 4 Descriptive statistics and zero-order correlations among all measured variables and the Oxford Happiness Questionnaire (OHQ; N = 870)
Full size table

Discussion

Happiness and well-being appear to be universally recognized concepts as less and less linked to the absence of illness, and more associated with the individual’s ability to experience positive emotions and fulfillment, both personally and professionally (e.g., Bolton, 2022). The study of happiness is cross-disciplinary and encompasses traditional and innovative approaches, leading to a growing number of interventions and theories that require proper indicators and measurement tools in different contexts. To validate a quickly administered tool for assessing happiness (the 8 items Oxford Happiness Questionnaire), a psychometric evaluation study was conducted on a sample of the Italian population.

This research aimed to investigate the psychometric characteristics of the short instrument widely used to measure happiness. Results of the CFA of the Italian OHQ were excellent in the terms of factor structure and validity (Hills & Argyle, 2002; Meleddu et al., 2012), with the measure being also gender-invariant. Additionally, the reliability of the instrument in the present sample was very good. Overall, our results confirm the construct validity, factor structure, and suitability for use in Italian contexts. Therefore, this research provides valuable insights into the study of happiness and offers potential avenues for future research and reflection.

The results also provided evidence of the convergent and discriminant validity of the OHQ measure. The scores were highly correlated with life satisfaction and (in the reverse direction) with perceived stress, which supports previous findings and the close relationship between these constructs (Argyle et al., 1995; Hart et al., 2018; Ruggeri et al., 2020).

The high correlations between happiness, mindfulness, and emotion regulation problems are particularly noteworthy. The results seem to confirm that there is a strong relationship between psychological skills, such as the ability to be present and focused on the present and to cope with their emotions, and happiness levels (Crowley et al., 2022; Quoidbach et al., 2010). On the one hand, happiness is composed of dimensions that have to do with the absence of negative feelings and the presence of positive feelings (e.g., Argyle, 2001); on the other hand, mindfulness practices act through processes of decentralization and acceptance of inner experiences and, like the processes of emotional and attentional self-regulation, they facilitate changes in the perspective of the self that are related to higher levels of happiness (e.g., Van Gordon et al., 2023).

In addition to the aspects of psychometric validity and easiness, this research also offers insights with regard to the socio-demographic variables. The results confirmed the relationship between happiness levels and gender (females are less happy than men; e.g., Weech-Maldonado et al., 2017; Sotgiu, 2019), age (older people are happier; e.g., Steptoe & Lassale, 2018; Sotgiu et al., 2011), marital status (individuals who are single are less happy than people who are married or in relationships; e.g., Kamp Dush et al., 2008; Steptoe & Lassale, 2018), having children (people with children are happier than those without children; e.g., Steptoe & Lassale, 2018; Verrastro et al., 2020), educational level (happiness rises as education level does; e.g., Hart et al., 2018). Among the sociodemographic characteristics, the salary was that most strongly correlated variable with happiness levels, which is consistent with prior findings (e.g., Cheng & Furnham, 2003; Kahneman & Deaton, 2010; Lu, 2020; Piff & Moskowitz, 2018).

Despite its strengths, there are some limitations to be considered in this research. First, the sample did not include very young or older people (subjects were mostly workers), thus the psychometric properties of the OHQ within these two populations are still under-investigated, and this may partly limit the generalizability of the results. Second, we modeled covariances between two pairs of item residuals, and this may have partially affected the replicability and interpretation of our findings (Bandalos, 2021; Ferrando et al., 2022); third, we did not assess any health problems or conditions, nor sedentary factors, both of which are very-well known causes of lower happiness and well-being (Burns & Crisp, 2022; De Neve & Oswald, 2012; Hart et al., 2018). Moreover, it should be noted that, in addition to satisfaction, an assessment of well-being would have been appropriate. Finally, a diverse set of socio-demographic factors, relationships, life situations, and socio-economic status, seems to be related to different dimensions of happiness (e.g., children and relationships for affective well-being; wage for evaluative well-being; psychological skills for eudaimonia; Cheng & Furnham, 2003; Steptoe, 2019) should be considered making it more difficult to appreciate the specific contribution of psychological variables. Therefore, further research should use more systemic theoretical models capable of simultaneously evaluating the contribution of each variable to the different dimensions that comprise happiness as a construct.

It is worthwhile to consider what factors generate happiness, what life choices we can favor becoming more happier, and what interventions and public policies we can implement to improve collective well-being. In any case, a question can be raised: isn’t happiness just a term we use for convenience to independently aggregate all these factors (feeling of satisfaction, positive evaluation, pleasure, etc.) that contribute to subjective well-being? All nuances are important in capturing the right sense of happiness: however, people usually tend to use a single more impactful term both on the individual and collective level (Helliwell & Aknin, 2018).

Happiness has become a central construct in many theoretical approaches, including applied positive psychology, coaching psychology, and third-generation cognitive behavioral therapies (CBT; Jongbloed & Andres, 2015; Seligman et al., 2005). As a result, the need for agile measures capable of capturing all these aspects is undeniable, both for research and for the progress of theoretical frameworks (integrated models that grasp the role of each psychological variable between determining factor, moderator/mediator, and outcome) and psychological skills interventions (improving the skills that help to live better and overcome life events more positively (Ingelström & van der Deijl, 2021; Suardi et al., 2016).