New Wellbeing Scale for Young Adults Passes Its Toughest Statistical Test Yet

Measuring how well a person is really doing has always been one of psychology’s most stubborn challenges. Most of the instruments that dominate the field were built in Western, industrialized societies and then translated for use elsewhere, a practice that researchers have long criticized for importing assumptions about what a good life looks like. A new study published in Discover Mental Health takes a direct swing at that problem for Pakistan, reporting the successful statistical validation of the Holistic Scale of Wellbeing, a measure designed from the ground up for young adults in the country. The work, carried out by Zehra Keshf and Arif Nadeem of the Department of Applied Psychology at Government College University, Faisalabad, transforms an earlier preliminary version of the scale into a finished, psychometrically vetted instrument ready for research use.

The story begins with the Holistic Scale of Wellbeing-Preliminary, or HSW-P, a 29-item questionnaire developed as an indigenous wellbeing measure for Pakistani young adults. Indigenous in this context does not simply mean locally translated; it means that the very architecture of the scale, the domains it covers and the way it frames a flourishing life, was built around the cultural realities of the population it serves. That includes domains that mainstream Western scales often treat as peripheral, most notably religiosity and spirituality, which the final scale retains as one of its seven core dimensions alongside purpose, psychological health, vitality and satisfaction, physical health, finances, and relationships.

The central task of the new study was confirmatory factor analysis, a statistical technique that puts a proposed structure of a questionnaire through a formal trial. Unlike exploratory methods, which hunt for patterns in data after the fact, confirmatory factor analysis starts with a specific hypothesis about how many underlying factors exist and which items belong to which factor, then tests whether the observed responses from real participants fit that blueprint. In this case, the researchers hypothesized a second-order structure: seven first-order factors corresponding to the seven domains of wellbeing, which themselves load onto a single higher-order factor representing overall holistic wellbeing. This kind of hierarchical model is statistically demanding, because it requires the seven domains to be correlated with one another in a pattern that can be summarized by one overarching construct.

To run that test, the team administered the 29-item preliminary scale to 464 young adults recruited through convenient sampling from three cities in Punjab: Faisalabad, Rawalpindi and Bahawalpur. Sampling across multiple urban centers rather than a single campus matters, because a scale that only works for one narrow student population is of limited value as a national research tool. The participants completed the questionnaire, and their responses fed into the confirmatory models as well as a battery of classical reliability analyses. The study received ethics approval from the Advanced Studies and Research Board of Government College University, Faisalabad, in 2024, and all participants signed informed consent forms describing the purpose, risks and benefits of the research, with procedures conducted in accordance with the Declaration of Helsinki.

The reliability results were strikingly strong. Cronbach’s alpha, the most widely cited index of internal consistency, came in at 0.91, indicating that the items hang together remarkably coherently as measures of a common construct. Split-half reliability, which checks whether the two halves of the test produce equivalent scores, reached 0.87, and test-retest reliability, which examines whether people score similarly when assessed on separate occasions, stood at 0.77. For readers outside psychometrics, these numbers matter because a scale that is unreliable cannot be valid: if the instrument produces wildly different readings for the same person or mixes unrelated content into a single score, no downstream conclusion drawn from it can be trusted.

The confirmatory factor analysis itself delivered the study’s headline verdict. The model fit statistics told a nuanced but broadly favorable story. Indices such as the standardized root mean square residual, the root mean square error of approximation, and the ratio of chi-square to degrees of freedom all indicated good fit, meaning the hypothesized seven-factor structure reproduced the observed data closely. Other indices, including the goodness-of-fit index, the adjusted goodness-of-fit index, the comparative fit index and the incremental fit index, showed acceptable fit. Two indices, the normed fit index and the Tucker-Lewis index, fell just slightly below conventional acceptability thresholds, a detail the authors report transparently rather than gloss over. In practical terms, the evidence supports the scale’s proposed architecture while acknowledging that model fit is always a matter of degree rather than a binary pass or fail.

Beyond overall fit, the study examined the behavior of individual items and the quality of each factor. Every item showed positive factor loadings in the medium to strong range, which means each question pulled its weight as an indicator of its intended domain rather than measuring something off-target. The researchers also computed composite reliability and average variance extracted, two statistics favored in structural equation modeling circles. Composite reliability asks whether the items within a factor collectively measure it dependably, while average variance extracted asks how much of the variation in the items a factor actually explains; values for both were reported as reasonable, reinforcing the picture of a well-behaved measurement model.

One item did not survive the refinement process. The final Holistic Scale of Wellbeing contains 28 items rather than the preliminary version’s 29, organized into the seven subscales of purpose, psychological health, vitality and satisfaction, physical health, finances, relationships, and religiosity and spirituality. The retention of religiosity and spirituality as a full subscale is arguably the scale’s most distinctive feature from an international perspective. In Pakistan, where religious and spiritual practice is deeply woven into daily life and social identity, a wellbeing measure that omits this dimension risks missing a major component of how young people evaluate their own lives. The HSW treats it not as an afterthought but as a core pillar, on equal statistical footing with physical health or finances.

The study went one step further than many validation efforts by converting raw scores into interpretable categories. Using the mean and standard deviation of the distribution, the researchers established score ranges corresponding to five levels of wellbeing: Poor, labeled as needing immediate improvement; Below Average, needing improvement; Average, which should be improved; Good, which can be improved; and Exceptional, requiring no improvement. This kind of categorical framework is what turns a statistical instrument into a practical tool, giving clinicians, counselors and researchers a shared vocabulary for describing where an individual or a group stands. It also creates benchmarks that future studies can compare against, which is essential for building a cumulative literature on youth wellbeing in the country.

The broader significance of the work lies in what it unlocks. With a validated, culturally grounded measure in hand, researchers in Pakistan can now assess youth wellbeing with an instrument whose structure has been formally tested rather than assumed, and intervention programs can be evaluated with outcome measures that resonate with the population they serve. The authors position the HSW as a foundation for further indigenous work on wellbeing enhancement, and the open-access publication means the tool is available to the wider research community. As the global psychology establishment continues to grapple with the limits of exporting Western measures to the majority of the world’s population, studies like this one offer a concrete template: build the scale locally, then hold it to the same rigorous statistical standards that any international instrument must meet. By that standard, the Holistic Scale of Wellbeing has now earned its place in the toolkit.

Subject of Research: Confirmatory factor analytic validation of an indigenous holistic wellbeing scale for Pakistani young adults

Article Title: Refining the Holistic Scale of Wellbeing-Preliminary (HSW-P) into Holistic Scale of Wellbeing (HSW) through confirmatory factor analysis

Article References: Keshf, Z., & Nadeem, A. (2026). Refining the Holistic Scale of Wellbeing-Preliminary (HSW-P) into Holistic Scale of Wellbeing (HSW) through confirmatory factor analysis. Discover Mental Health. https://doi.org/10.1007/s44192-026-00600-7

Image Credits: AI Generated

DOI: 10.1007/s44192-026-00600-7

Keywords: wellbeing, psychometrics, confirmatory factor analysis, scale validation, young adults, Pakistan, reliability, model fit, mental health, indigenous psychology, quality of life, Discover Mental Health