Tracking and understanding the mental state of populations around the world is important for understanding how we are faring as a collective humanity and determining the trajectories of change, so that we can better respond to arising challenges and gain insight into contributing demographic, social, economic, technological and environmental factors.
Here we compare three large scale global initiatives that measure various facets of our mental state – the Global Mind Project (GMP), the World Mental Health Survey Initiative (WMHS) and the World Happiness Report (WHR) – outlining their similarities and differences in terms of their measurement outcomes and the methodologies they use. While the GMP measures mental wellbeing, the WMHS measures mental health disorders and the WHR measures ‘happiness’. We often talk about these interchangeably. However, there are fundamental differences, as we outline below, that depend on how they are defined and the semantics of what is queried.
Summary of each initiative:
| Global Mind Project (GMP) | World Mental Health Survey Initiative (WMHS) | World Happiness Report (WHR) |
| The Global Mind Project was established by Sapien Labs in 2020 and aims to track and understand the evolving relationship between our collective mental wellbeing and our changing social, technological, and cultural environments. It uses an online assessment of mental wellbeing and its life context drivers (the MHQ) and currently operates in over 70 countries. To date, it has collected data from over 1.4 million people around the globe. | The World Mental Health Survey Initiative is led by the World Health Organization and was started over 20 years ago to obtain accurate cross-national information about the prevalence and correlates of mental, substance, and behavioral disorders. It uses the CIDI assessment/interview guidelines, and to date, in-person psychiatric epidemiological surveys have been conducted in 30 countries, with data collected from over 166,000 respondents. | The World Happiness Report has been running since 2012 and is led by a partnership of Gallup, the Oxford Wellbeing Research Centre, the UN Sustainable Development Solutions Network, and the WHR’s Editorial Board. It uses life evaluation data, obtained through Gallup’s World Poll, together with metrics of positive and negative emotion and a selection of life context factors. The latest 2023 report spans 130 countries and includes data from over 100,000 respondents. |
Differences in measures and methodologies
Each of these initiatives uses different questions, tools and approaches to measure people’s mental state as described below. In addition, Table 1 summarizes the main similarities and differences between them.
GMP and the MHQ score
The GMP measures mental wellbeing, which is defined by the World Health Organization (WHO) as a state “that enables people to cope with the stresses of life, realize their abilities, learn well and work well, and contribute to their community”1. To accomplish this, it uses an assessment called the MHQ, that includes 47 aspects of cognitive, social, emotional, behavioral and physical functioning that span the breadth of mental functioning, rated on a 1-9 life impact scale2,3. These items were derived from an analysis of questions asked in 126 commonly used mental health assessment tools spanning 10 major disorders (including the CIDI used by the WMHS), as well as items taken from RDoC and positive aspects of mental functioning4. Ratings of these 47 MHQ items are aggregated into a composite mental wellbeing score – the MHQ score – which spans the spectrum from Distressed to Thriving (Figure 1). The score has been developed to reflect function and has been shown to relate linearly to functional productivity across the scale as well as reflect clinical burden3,5.
Figure 1: MHQ score range
WMHS and disorder diagnoses
The WMHS initiative uses the WMH-CIDI (Composite International Diagnostic Interview) clinical interview (most recently the CIDI-5). This interview is composed of multiple modules, with each module asking about symptoms associated with a particular mental health disorder definition. The modules and questions align with disorder classifications and criteria outlined in the Diagnostic and Statistical Manual of Mental Disorders (DSM) and the International Classification of Mental and Behavioral Disorders (ICD). It therefore provides outcome measures relating to the prevalence of clinical disorders as defined by the DSM and ICD across different countries and populations. More details on the method can be found here.
WHR and life evaluation
In the case of the WHR, ‘happiness’ is defined as the position on a core question that uses the Cantril Ladder which asks people to evaluate their present and future life using the image of a ladder with a scale ranging from 0 (worst possible life) to 10 (best possible life) (Figure 2). Country rankings of happiness are based on their average life evaluation ratings over the past 3 years, with scores falling in either the Suffering, Struggling or Thriving ranges. Other measures also include positive (laughter, enjoyment, and interest) and negative (worry, sadness, and anger) emotions, assessed through the question Did you experience the following feelings during a lot of the day yesterday? Y/N. More details on the Gallup’s World Poll methods can be found here.
Figure 2: The Cantril Ladder Life Evaluation Questions used in the WHR
Differences between outcome measures
While each initiative measures some aspect of people’s mental state, there are fundamental differences between them. For example, like many traditional mental health measures4, the CIDI used in the WMHS is structured around individual mental health disorders which focus on symptoms (i.e. what’s wrong with the individual) and not positive mental assets. Thus, it provides prevalence of individual disorders or symptoms but does not provide a view of mental wellbeing along the spectrum from distressed to thriving or a transdiagnostic/ aggregated metric of mental health burden as in the GMP. Since the GMP questions span symptoms across 10 major mental health disorders, approximate diagnostic indications can be arrived at, but with the caveat that the GMP uses a life impact scale rather than frequency or severity scales (see6). Thus, the GMP does not offer strict disorder prevalence statistics as defined in the DSM or ICD.
In contrast, in the WHR the measures of mental health and wellbeing are limited to the ratings of current and future life (used for country rankings), as well as whether positive and negative emotions were experienced the previous day. It therefore does not provide a comprehensive view of mental functioning or mental health burden, but instead interprets happiness as people’s perception of their current and future lives from worst possible to best possible, something that is more akin to a life satisfaction measure. It remains to be understood how closely such life evaluation relates to the assessment of one’s internal mental state versus one’s external environment or circumstance.
Table 1: Similarities and differences between the GMP, WMHS and WHR
| Global Mind Project (GMP) | World Mental Health Survey Initiative (WMHS) | World Happiness Report (WHR) | |
| Survey Instrument/Tool | MHQ | CIDI | Cantril Ladder |
| Mental Aspects measured | Life impact of 47 aspects of cognitive, emotional and social functioning, that include symptoms spanning 10 major mental health disorders as well as positive assets | Frequency/severity/duration of symptoms of 20+ mental health disorders including anxiety, mood, impulse control and substance disorders | Life evaluation position on a Cantril Ladder; presence of positive affect (laughter, enjoyment, and interest) and negative affect (worry, sadness, and anger) through Yes/No ratings |
| Life context factors | Multiple factors that can be rapidly reconfigured including but not limited to the following themes: demographics, exercise, diet, social behavior and relationships, technology use, faith, traumas & adversities, physical health conditions | Demographics, employment, family, social networks, childhood, physical health conditions, treatment seeking | Multiple factors across themes including: demographics, GDP per capita, healthy life expectancy, generosity, social support, freedom to make life choices, and perception of corruption |
| Outcome measures | (1) Aggregate metric of mental wellbeing linearly associated with productive function (MHQ) (2) 6 sub-metrics (Mood & Outlook, Social Self, Adaptability & Resilience, Drive & Motivation, Mind-Body Connection, Cognition) (3) Mental Health Burden (% Distressed/ Struggling) (4) Ratings of 47 elements | (1) Disorder diagnosis based on core diagnostic assessment (Part 1) (2) Correlates and diagnosis of disorders of secondary interest (Part 2; subsample only) | (1) Life evaluation rating on Cantril Ladder (2) Scores of positive and negative affect. |
| Survey medium | Online assessment | In-person interviews by trained lay interviewers | Telephone surveys & face to face interviews |
| Sampling strategy | Anonymous recruitment using paid advertisements placed on Google Ads, and Facebook using a dynamically adjustable quota-based recruitment strategy which systematically targets pre-defined age-sex groups across a series of selected geographies using a broad range of interest criteria/keywords with the goal of robust representation of the general population in each age-sex band for different countries of interest. Country metrics use a post-stratification weighting to match age-sex proportions in the sample to national population statistics | Typically a multi-stage clustered area probability sample of households in the population and then selecting one/two respondents from each sampled household using probability methods without replacement. Part 2 sample weighted to match population geographic/socio-demographic distributions and to adjust for under-sampling of Part I non-cases | Random-digit-dial (RDD) method or a nationally representative list of phone numbers for telephone surveys. Area frame design for face-to-face interviewing in randomly selected households. Samples weighted to correct for unequal selection probability, nonresponse, and double coverage of landline and cellphone users when using both cellphone and landline frames. Gallup also weights its final samples to match the national demographics of each selected country |
| Number of countries | 70 | 30 | 130 |
| Data reporting | Real time, insights aggregated on an annual basis. Latest report in 2024 from 2023 annual data | Different countries targeted in different years. Data ranges from 2001 from early country cohorts (e.g. US) to 2022 (Qatar) | Annual data collection. Annual report data based on past 3 years (e.g. 2023 report used data from 2020-2022) |
| Sample sizes | Annual sample: 500k (2023) (1000-50,000 per country/ year). Total Sample: 1.4 million since 2019 | Total Sample: 166,000 since 2001 (~1000-13,000 per country) | Annual sample: 100,000+ (~1000/country/year)
|
| Age range | 18+ (13+ in the US) | 18+ | 15+ |
Sampling differences
These initiatives also differ in terms of their sampling and data collection methodologies. For example, the GMP captures data only from the Internet-enabled population and therefore compares individuals of a more similar range of education and means across countries. In countries such as the United States where Internet penetration is 90%+ the GMP data closely mirrors a representative national sample7. However, for counties like Tanzania and India where internet penetration is lower (20% and 43% respectively), the GMP sample isn’t aligned with sampling from the Gallup World Poll or WMHS which, given the goal of representative sampling and methodology of in-person interviews, would likely include a substantial percentage of respondents of lower income, lower literacy and without a digital footprint. Thus, a direct comparison of country numbers across these surveys is not relevant.
A second dimension on which the sampling differs is on the scale and timing. The GMP collects data from all countries on an ongoing basis, acquiring data from over 500,000 people each year. Thus, data is available in real time and the scale of data allows the reliable tracking of country level trends annually, and in some cases quarterly, making it more sensitive to rapid changes in the environment such as a pandemic or war. The WHR acquires data from over 100,000 people each year and similarly reports data annually but based on data acquired over the previous 3 years. In contrast, the resource-intensive nature of WMHS means that data isn’t collected on an ongoing or annual basis from all countries but rather only from certain countries each year making it more of a point survey than one designed to compare countries or track trends. As such, data from some countries can be more than a decade old.
Summary
In summary, the 3 initiatives each capture different aspects of people’s mental state from different populations and on different timescales and thus will each provide differing perspectives on our collective mental state and trajectory.
References:
- World Health Organisation. World Mental Health Report: Transforming Mental Health for All. (2022).
- Newson, J. J. & Thiagarajan, T. C. Assessment of Population Well-Being With the Mental Health Quotient (MHQ): Development and Usability Study. JMIR Ment Health 7, e17935 (2020).
- Newson, J. J., Pastukh, V. & Thiagarajan, T. C. Assessment of Population Well-being With the Mental Health Quotient: Validation Study. JMIR Ment Health 9, e34105 (2022).
- Newson, J. J., Hunter, D. & Thiagarajan, T. C. The Heterogeneity of Mental Health Assessment. Front. Psychiatry 11, 76 (2020).
- Newson, J. J., Sukhoi, O. & Thiagarajan, T. MHQ: Constructing an aggregate metric of mental wellbeing. (2023).
- Newson, J. J., Parameshwaran, D. & Thiagarajan, T. PHQ-9, Life Impact, and Breadth of Symptomatic Experience. (2024) doi:10.31219/osf.io/p7q8x.
- Taylor, J., Sukhoi, O., Newson, J. & Thiagarajan, T. Representativeness of the Global Mind Project Data for the United States. (2023).
