This chapter contains recommended subjective well-being modules to be integrated into official surveys. These include (1) a core module, which contains a concise set of measures that span the different components of subjective well-being (life evaluation, affect and eudaimonia), each with strong statistical properties and unique predictive power for policy-relevant outcomes; (2) extended modules for each component area, alongside a module for population mental health measures, and (3) a discussion of time use survey integration. Implementation instructions for each module are also provided. The modules are designed to be ready-for-use in official surveys, due to their brevity and explicit accompanying instructions on question phrasing, response scales and implementation details.
OECD Guidelines on Measuring Subjective Well‑being (2025 Update)
Abstract
This chapter presents a series of short subjective well-being modules that can be integrated into existing surveys. Each module is accompanied by a brief description of the measures contained therein, along with instructions for how to implement and (where relevant) how to interpret the resulting data. The following modules are included in this chapter: a core module of subjective well-being (Box 2.2); extended modules for life evaluation (Box 2.3), domain evaluation (Box 2.4), affect (Box 2.5), eudaimonia (Box 2.6) and population mental health (Box 2.7); and a discussion of time use survey modules (Box 2.8).
The purpose of creating a core module of subjective well-being measures is to provide official data producers with a short set of the strongest-performing measures, in terms of a range of selection criteria including statistical quality and policy relevance (Box 2.1). Adding questions to official surveys is resource intensive: fielding new questions requires time and additional expense and can contribute to respondent fatigue. The brevity of the core module enables data producers to incorporate a set of subjective well-being questions that span evaluative, affective and eudaimonic concepts into different types of household surveys, including general social surveys, community health surveys, and specialised surveys on specific topics such as social cohesion or trust, etc. The short nature of the module also enables it to be fielded more frequently (ideally at least annually).
For survey designers with sufficient resources and an interest in collecting more detailed subjective well-being data, more flexible extended modules for each dimension of subjective well-being are provided. The extended modules identify important conceptual areas to measure and, for each, select one measure that the evidence shows is most valid for measurement. Each module also identifies a priority short list of measures as well as additional extended options. Items in the extended modules that also appear in the core module are in bold.
Four criteria govern how measures are selected into the modules. These criteria are applied to the core as well as extended modules; however, they are applied more strictly to the core.
Statistical properties: Measures need to have strong evidence for their psychometric properties, including statistical validity and reliability, across population groups.
Brevity: Measures for a given concept should be concise and, ideally, single-item questions.
Unique policy relevance: Selected measures should be meaningful for policy-relevant outcomes – including, but not limited to, physical and mental health, financial, labour market, relational and civic engagement outcomes – justifying their inclusion in large-scale official data collections. Importantly, a measure should reflect meaningful outcomes for all populations, globally, to the degree that this is possible. Of particular consideration for the core module: measures should have unique and complementary value-add compared to others in the set, demonstrating the need to include all items.
Consistency: There is significant value in having generally agreed-upon statistical standards and potential harm in changing existing standards and disrupting long-running time series. When new measures are considered, consistency with other OECD measurement recommendations, as well as with existing national or international data collection practices, is prioritised.
Annex A provides detailed explanations for the decisions underpinning which concepts and measures to include in each of the modules.
Data producers may approach subjective well-being measurement from different starting points. For those who are not currently measuring subjective well-being, it is recommended to begin with the core module. Where survey space allows, those looking for a more in-depth understanding of the different components of subjective well-being can expand their efforts by including priority measures from the extended modules. Data producers particularly interested in increasing the global inclusivity of their measures may wish to explore the cross-cutting experimental module and question banks provided in Chapter 3. Figure 2.1 provides data producers with a flowchart of the modules contained in these guidelines to aid in their selection of subjective well-being measures for household (Panel A) and time use (Panel B) surveys.
Note: Panel A: Steps appear in order of priority. Data producers interested in fielding subjective well-being questions in household surveys should begin with the core module (Step 1); following this, if desired, data producers can include priority measures from the extended modules (Step 2), which are not presented in any order of importance given that all provide added value; for data producers with sufficient space and interest, the cross-cutting module (Step 3) can then also be used. The extended modules and cross-cutting module contain a short list of measures, as well as an extended list of measures should space allow. Modules in Step 1 and Step 2 appear in this chapter; the module in Step 3 appears in Chapter 3. Panel B: One time use module is recommended and appears in this chapter; the Annex provides details on alternate approaches, depending on the structure of the survey, technical resources and availability of space.
The core module contains three measures, with a single question for each of the subjective well-being components: a summary measure for life evaluation (Q1) and for eudaimonia (Q2) and one example measure of affect (Q3) (Box 2.2). The measures included have the strongest evidence for their validity, relevance and international comparability (refer to Annex A).
The following question asks how satisfied you feel, on a scale from 0 to 10. Zero means you feel "not at all satisfied" and 10 means you feel "completely satisfied".
Q1. Overall, how satisfied are you with life as a whole these days? [0-10]
The following question asks how worthwhile you feel the things you do in your life are, on a scale from 0 to 10. Zero means you feel the things you do in your life are "not at all worthwhile", and 10 means "completely worthwhile".
Q2. Overall, to what extent do you feel the things you do in your life are worthwhile? [0-10]
The following question asks how much pain you have felt, on a scale from 0 to 10, where 0 refers to no pain and 10 refers to the worst imaginable pain.
Q3. Over the past 7 days, how would you rate your pain on average? [0-10]
Implementation details:
The questions in the module should be fielded using the exact phrasing and answer formats provided. Questions should be asked in the order in which they appear.
When integrating this module into a survey, designers should ensure that it appears early on and ideally at the start of the survey immediately following demographic information. In particular, avoid placing this module immediately after questions that may elicit strong emotional responses in respondents or questions that may frame how respondents interpret this module (e.g. items on health or household finances).
Q1, an evaluative question asking respondents how satisfied they are with their life, is the primary measure of subjective well-being. If space only allows for a single item, it should be Q1.
For data producers already capturing a measure of pain in separate surveys (most often health surveys), it is still recommended to collect Q3, in particular if the other measure of pain focuses on “physical” or “bodily” pain, or assesses the functional limitation pain plays in performing daily activities. These questions are important in their own right but capture different information and have different policy applications. For a longer discussion on this, refer to Annex A. For a precise definition of pain, refer to the “Guidelines for interviewers” section below, as well as Annex A.
Origin:
Both Q1 and Q2 appear in the original 2013 OECD Guidelines on Measuring Subjective Well-being.
Q3 is a form of the Numerical Pain Rating Scale (NPRS), a single-item pain assessment tool that has been validated in clinical settings. This question appears in the PROMIS (patient-reported outcomes measurement information system) Adult Short Form survey and is also included in the OECD’s Patient Reported Indicator Survey (PaRIS) (refer to Annex A for details).
Time:
The module is expected to take about 60 seconds to complete.
Output:
Data on life satisfaction (Q1) and feelings about things in life being worthwhile (Q2) can be presented as the mean value of responses, excluding missing values. Summary measures of distribution (inequalities), such as the standard deviation of responses and the ratio of the top 20% of outcomes compared to the bottom 20%, should be reported (refer to Box 1.2 for details). The mean value and standard error of this estimate can be used to describe differences in outcomes between different population groups.
The percentage of the population reporting an outcome below a threshold (a score ≤ 4) should also be reported to identify those with “low life satisfaction” and those who feel that the things they do in in life are “not worthwhile”.
Data on pain (Q3) can be presented as the mean value of responses, excluding missing values, as for Q1 and Q2. In addition, the percentage of the population reporting “no pain” (a score of 0) or “mild” (a score ≥ 1 and ≤ 4), “moderate” (5-6) or “severe” (a score ≥ 7) pain can also be reported.
Guidelines for interviewers:
The primary question (Q1) deliberately focuses on how people are feeling “these days” rather than specifying a longer or shorter time period. The intent is not to obtain the current emotional state of the respondent, but to obtain a cognitive evaluation on their level of life satisfaction.
The eudaimonic question (Q2) concerns the extent to which people feel their activities in general (“the things you do in your life”) are worthwhile, in the sense that their activities contribute to attaining personally valued goals. If the concept of “worthwhile” is difficult to translate to the local language, translators should be advised that the question gets at the underlying construct of one’s life having a sense of meaning or purpose. The question is not intended to ask respondents whether they feel their life is worth living, and translations should take care to make this distinction. What is deemed meaningful and worthwhile is determined by the respondent; the question should not be asked in a context that could prompt respondents to think about achievements in particular domains of life (such as work). No specific time frame is supplied: respondents are invited to make an overall assessment.
The question on pain (Q3) refers to pain in general and makes no distinction between physical and psychological pain given there is reasonable consensus in the medical and psychological literatures advising against such a strict delineation and noting that pain by definition contains an emotional component (see Annex A for an extended discussion). The International Association for the Study of Pain defines pain as “An unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage” that “is always a personal experience that is influenced to varying degrees by biological, psychological and social factors” (Raja et al., 2020[1]). This definition can be used as a reference when translating this question to local languages, in particular if the word “pain” in the target language typically is associated only with physical pain. Enumerators can also use this definition to help prompt respondents if they express confusion as to the target construct of the question.
The following modules provide additional measurement recommendations for each component of subjective well-being. It is recommended that extended modules are fielded as a second-order priority, after the core module has been implemented in full. Whenever an item in an extended module also appears in the core module, it appears in bold and should be asked only once per survey.
When cleaning and reporting data from a longer series of items using the same response scale, it is recommended to screen for potential response sets. These can be visible when the respondent scores at the top or bottom of the scale for all indicators (regardless of the indicators’ direction – e.g. positive or negative valence). Response sets may indicate either a lack of understanding on the part of the respondent, boredom, disengagement or an unwillingness to respond meaningfully. The first best guard against response sets is to mitigate their risk through good survey design, and all of the modules included here are designed to keep the battery of items as short as possible, coupled with transition text that provides a break between items with similar response scales.
Life evaluation
There are two extended modules under the life evaluation heading – the first is an extended module for general assessments of life evaluation, which focus on life in general, or life overall. The second is a domain evaluation extended module, which asks respondents to reflect on their satisfaction with specific aspects of their lives.
The life evaluation extended module contains questions regarding a respondent’s cognitive judgements on how they evaluate their own lives (Box 2.3). The bolded question (Q1) also appears in the core module. Q2 captures information on a respondent’s perception of their life satisfaction in the past, and Q3 provides insight into a respondent’s perceptions of the future.
The following two questions ask how satisfied you feel, on a scale from 0 to 10. Zero means you feel "not at all satisfied" and 10 means you feel "completely satisfied".
Q1. Overall, how satisfied are you with life as a whole these days? [0-10]
Q2. Overall, how satisfied with your life were you one year ago? [0-10]
-----
Q3. Overall, how hopeful do you feel about your future, where 0 is [0-10]
”not at all hopeful”, and 10 is ”completely hopeful”?
Implementation details:
The questions in the module should be fielded using the exact phrasing and answer formats provided. Questions are designed be used as a set and asked in the order in which they appear. Where space limitations prevent adoption of the full module, Q2 can be prioritised over Q3. (Q1 is the primary measure of the core module and should always have priority over other measures.)
Q3 is a new question for this edition of the Guidelines; refer to Annex A for more details on its statistical properties and the policy relevance of hope.
Time:
The module is expected to take about 60 seconds to complete in its entirety.
Origin:
Q1 appears in the core module and also appeared in the original 2013 OECD Guidelines on Measuring Subjective Well-being.
An earlier version of Q2, with a 5-year rather than 1-year recall period, appeared in the original OECD Guidelines (see Annex A for more on the rationale for this change).
Q3 was introduced by the Office for National Statistics in the United Kingdom in 2023 (ONS, 2024[2]).
Output:
All indicators can be presented as the mean value of responses, excluding missing values. Summary measures of distribution (inequalities) should also be reported (see Box 1.2 for details). The mean value and standard error can be used to describe differences in outcomes between different population groups.
The percentage of the population reporting an outcome below a threshold (a score ≤ 4) should also be reported in order to identify those with “low life satisfaction” now (Q1) and one year ago (Q2) and those who feel “little hope” for their future (Q3).
Guidelines for interviewers:
These questions deliberately focus on how people are feeling about life as a whole rather than specifying a longer or shorter time period, and they ask the respondents for a reflective judgement rather than a statement of their current mood.
The question on hope (Q3) refers to both respondents’ positive outlook towards the future as well as their sense that they have the ability (or agency) to achieve good outcomes, which can be prompted if there is confusion on the respondent’s side. The word “hope” is used in English to distinguish from optimism, which is more associated with stable personality traits. However, this distinction may not translate across languages; in some languages, the word “optimism" may be a better fit to capture the intended meaning.
Domain evaluations collect respondents’ evaluative judgements on how well different specific aspects of their lives are going (Box 2.4), using the same framing as the satisfaction with life question included in the core module and the life evaluation extended module. The questions in the module capture people’s satisfaction with various domains of life but are distinct from self-reported measures of outcomes in these domains given the focus on satisfaction over self-assessment (e.g. satisfaction with health status vs. self-reported good vs. poor health). Domain evaluations can help to pinpoint particular areas of (dis)satisfaction in people’s lives and have in some cases been associated with behaviour (Kaiser and Oswald, 2022[3]): job dissatisfaction predicts job quits (Clark, 2001[4]; Clark, Georgellis and Sanfey, 1999[5]) and marital dissatisfaction predicts divorce (Powdthavee, 2009[6]), for example.
The following questions ask how satisfied you feel about specific aspects of your life, on a scale from 0 to 10. Zero means you feel "not at all satisfied" and 10 means you feel "completely satisfied".
Q1. How satisfied are you with your standard of living? [0-10]
Q2. How satisfied are you with your health? [0-10]
Q3. How satisfied are you with what you are achieving in life? [0-10]
Q4. How satisfied are you with your personal relationships? [0-10]
Q5. How satisfied are you with how safe you feel? [0-10]
Q6. How satisfied are you with feeling part of your community? [0-10]
Q7. How satisfied are you with your future security? [0-10]
Q8. How satisfied are you with the amount of time you have to do the things [0-10]
that you like doing?
Q9. How satisfied are you with the quality of your local environment? [0-10]
Only for respondents who are employed:
Q10. How satisfied are you with your job? [0-10]
Implementation details:
The questions in the module should be fielded using the exact phrasing and answer formats provided. Questions are designed to be used as a set and asked in the order in which they appear.
Since the questions in the module cover evaluative forms of subjective well-being, they can be asked immediately following questions in the life evaluation extended module (Box 2.3).
Time:
This module takes about 3 minutes to complete in its entirety.
Origin:
The full module appeared in the original 2013 OECD Guidelines on Measuring Subjective Well-being.
Questions 1-7 refer to items in the Personal Wellbeing Index (PWI) (International Wellbeing Group, 2024[7]).
Output:
Information on individual questions can be presented as the mean value of each response, omitting missing values. Summary measures of distribution (inequalities) should also be reported (see Box 1.2 for details). The mean value of responses, and the standard error of this estimate, can be used to describe differences between population groups.
The percentage of the population reporting scores below a “low satisfaction” threshold (a score ≤ 4) should also be reported.
Guidelines for interviewers:
Respondents are asked to evaluate their satisfaction with various domains of their life; each domain should be considered separately.
Affect
Affect refers to a person’s feelings or emotional states, measured with respect to a specific period of time. This module collects information on positive and negative emotional states experienced the day before the survey (Box 2.5). (Refer to Box A A.1 in the Annex for a discussion of the difference between affect and mental health measures and the policy applications of both.)
The following questions ask about how you felt yesterday on a scale from 0 to 10. Zero means you did not experience the feeling "at all" yesterday while 10 means you experienced the feeling "all of the time" yesterday. Here is a list of ways you might have felt yesterday.
Q1. How about happy? [0-10]
Q2. How about worried? [0-10]
Q3. How about calm? [0-10]
Q4. How about sad? [0-10]
-----
Q5. How about angry? [0-10]
Q6. How about joyful? [0-10]
Q7. How about tired? [0-10]
Q8. How about stressed? [0-10]
Q9. Did you smile or laugh a lot yesterday? [0-10]
Implementation details:
This module is designed for use in very large sample surveys only, so that the aggregate of experiences “yesterday” sum to a population-representative picture of affect, regardless of how (a)typical yesterday was for each individual respondent. It is not suitable for surveys of fewer than 2 000 respondents (and ideally should have a minimum of 5 000 respondents).
The questions in the module should be fielded using the exact phrasing and answer formats provided. Questions are designed be used as a set and asked in the order in which they appear. In case of limited space, only Q1-4 should be asked.
Q1, Q3, Q6 and Q9 capture aspects of positive affect; Q2, Q4, Q5, Q7 and Q8 capture aspects of negative affect. There are more negative than positive questions, reflecting evidence that negative affect is more multidimensional than positive affect (see Annex A).
Q1, Q2, Q5, Q6, Q8 and Q9 are measures of high arousal, while Q3, Q4 and Q7 are measures of low arousal.
Interested survey designers with sufficient space may be interested in supplementing this module with affect measures from the cross-cutting experimental module (see Box 3.1).
Time:
This module is expected to take about 3 minutes to complete in its entirety.
Origin:
Q1-5 and Q7-9 appeared in the original OECD Guidelines. Q3 was developed by the Gallup-WPE Global Wellbeing Initiative and continues to be fielded in Gallup’s annual World Poll (Lambert et al., 2020[8]).
Q6 has been slightly adapted from the original OECD Guidelines, changing “enjoyment” to “joyful” (see Annex A for the rationale).
Output:
It is recommended to present the answers to individual questions to provide information on particular emotional states.
Information on individual questions can be presented as the mean value of each response, omitting missing values. Summary measures of distribution (inequalities) should also be reported (see Box 1.2 for details). The mean value of responses, and the standard error of this estimate, can be used to describe differences between population groups.
For each individual question, the percentage of the population reporting a score exceeding a certain threshold (a score ≤ 4 for positive affect and a score ≥ 6 for negative affect) should also be reported: this designates the share of the population with “low levels” of each positive emotion or with “high levels” of each negative emotion on the day prior.
In addition to reporting outcomes for individual questions, it is also possible to calculate a composite measure of balance between positive and negative affects, which enables comparison across countries in that affect balance calculations smooth out cultural and regional differences in the propensity to report extreme emotions (both positive and negative).
A composite measure of positive affect can be calculated as the mean score for Q1, Q3, Q6 and Q9.
A composite measure of negative affect can be calculated as the mean score for Q2, Q4, Q5, Q7 and Q8.
Affect balance is then calculated as positive affect minus negative affect for each respondent, averaged across all respondents. This will give a value ranging from -10 to +10. It is recommended to report the share of the population with a negative affect balance (a score less than 0).
Note that if this module is implemented alongside the core module (Box 2.2), it is not recommended that the pain measure from the core module (Q3) be included in the affect balance calculation; the pain measure uses an intensity rather than a frequency framing; it uses a different labelling scheme for the response scale; and it uses a different recall period (past seven days), making the question less comparable to the set of affect questions in this module.
Guidelines for interviewers:
The aim of the module is to capture information on respondents’ feelings, states, moods or emotions on the previous day. The time frame is explicitly short, because the primary focus is the affective states that people actually experienced, which is most accurately recalled over short time periods. If a respondent indicates that the previous day was unusual in some respect (something particularly bad or good happened, or they were feeling unwell), they should still report how they felt that day, since this question is intended to capture the feelings that people have actually experienced, not how people feel on a “typical” day. These affect items are designed to be used in very large sample household surveys only: under these conditions, because several thousand people should be interviewed over a relatively long period of time over the course of a household survey, unusual events will not overly bias the aggregated statistics that are produced. More importantly, the reference to a specific day permits the data to be used to unravel day-of-week effects and responses to external events for which the dates are known.
Eudaimonia
Eudaimonia is broadly defined as a feeling that one is living well, encompassing a sense of meaning, purpose and autonomy in life, together with self-acceptance, feelings of competence, and personal growth. The questions in this module are designed to capture important constructs of eudaimonia, including: a sense that the things one does in life have worth, meaning and purpose (Q1); a sense of autonomy (Q2); a feeling of competence, accomplishment or environmental mastery (Q3); self-esteem or self-acceptance (Q4); and personal growth (Q5) (Box 2.6). (Relatedness, another construct of eudaimonia, is not included in this module given that it is collected as a part of social connections measurement efforts – refer to Box A A.2. for a discussion on this.)
The following question asks how worthwhile you feel the things you do in your life are, on a scale from 0 to 10. Zero means you feel the things you do in your life are "not at all worthwhile", and 10 means you feel they are "completely worthwhile".
Q1. Overall, to what extent do you feel the things you do in your life are worthwhile? [0-10]
For the following questions, please use a scale from 0 to 10 to indicate how you feel. Zero means you "disagree completely" and 10 means you "agree completely".
Q2. I am able to do things that I really want and value in life. [0-10]
Q3. Most days I get a sense of accomplishment from what I do. [0-10]
-----
Q4. In general, I feel very positive about myself. [0-10]
Q5. For me, life has been a continuous process of learning, changing and growth. [0-10]
Implementation details:
The questions in the module should be fielded using the exact phrasing and answer formats provided. Questions are designed to be used as a set and asked in the order in which they appear.
Given limited space, only Q1-Q3 can be implemented.
Q5 is a new measure; refer to Annex A for more details on its statistical properties and the policy relevance of a sense of personal growth.
Time:
This module is expected to take about 3 minutes to complete in its entirety.
Origin:
Q1 appears in the core module and appeared in the original OECD Guidelines, as did Q3-4.
Q2 has been used in the Finnish Citizen’s Pulse Survey led by the Prime Minister’s Office (Statistics Finland, 2024[9]), and it will also be included in the 2025 European Social Survey (Rutherford et al., 2024[10]). It has been adapted to use a 0-10 answer scale.
Q5 is a single item in the Shortened Psychological Wellbeing Scale (Ryff and Keyes, 1995[11]), and it too has been adapted to use a 0-10 answer scale.
Output:
Information on individual questions can be presented as the mean value of each response, omitting missing values. Summary measures of distribution (inequalities) should also be reported (see Box 1.2 for details). The mean value of responses, and the standard error of this estimate, can be used to describe differences between population groups.
The percentage of the population reporting scores below a threshold (a score ≤ 4) can also be reported.
Guidelines for interviewers:
The questions in this module assess the degree to which respondents agree or disagree with various statements about themselves. They are intended to capture how people see themselves rather than emotions or feelings they have experienced. As a result, the questions are about how people are now and do not refer to a specific time period.
Population mental health
Mental health is defined by the World Health Organization as “a state of mental well-being that enables people to cope with the stresses of life, realise their abilities, learn well and work well, and contribute to their community” (WHO, 2022[12]), a definition also adopted by existing OECD work in this space (OECD, 2021[13]; 2021[14]; 2023[15]). Mental health is more than the absence of a clinically diagnosed mental health condition; mental health can be conceptualised on a single continuum of experience (from severe mental ill-health to positive mental health), or, as is increasingly done, as a dual continuum in which mental ill-health and positive mental states are related, but distinct, experiences (refer to OECD (2023[15]) to for an extended discussion).
Population mental health measures capture the persistence of moods, emotions and affective states over an explicit reference period, often the past two or four weeks. Experiencing negative emotions is not indicative of a mental health condition – this is a normal and healthy part of life – however, the persistence of these emotions over a longer duration of time suggests the risk of a mental health condition. Therefore while the concepts included in these mental health scales overlap with emotional states that also appear in affect modules (Box 2.5), the question formulations are different: the reference period is longer – to capture persistent states – and many tools have been validated against clinical diagnoses of specific mental health conditions (in particular, major depressive disorder and generalised anxiety disorder, the two most commonly occurring mental health conditions). For this reason, the data obtained from these tools should be used and interpreted differently from the data resulting from a measure of affect balance. Refer to Box A A.1 in the Annex for a discussion of the differences between affect and mental health measurement and the policy applications of both.
The population mental health extended module includes a general overall assessment of the respondent’s mental health (Q1); a mental health screening tool to assess the risk for depression and anxiety, using the Patient Health Questionnaire-4 (Q2-5); and an assessment of positive mental health, using the WHO-5 Well-being Index (Q6-10) (Box 2.7). These recommendations appear in Measuring Population Mental Health (OECD, 2023[15]).
Q1. In general, how is your mental health? [Excellent; very good; good; fair; poor]
-----
[Patient Health Questionnaire-4 (PHQ-4) ]
Over the last two weeks, how often have you been bothered by any of the following problems. When thinking about “how often”, please choose from the following options: not at all; several days; more than half the days; nearly every day.
Q2. Feeling nervous, anxious or on edge
Q3. Not being able to stop or control worrying
Q4. Feeling down, depressed or hopeless
Q5. Little interest or pleasure in doing things
-----
[WHO-5 Well-being Index (WHO-5)]
Over the past two weeks, how often have you felt the following. When thinking about “how often”, please choose from the following options: all of the time; most of the time; more than half the time; less than half the time; some of the time; at no time.
Q6. I have felt cheerful and in good spirits
Q7. I have felt calm and relaxed
Q8. I have felt active and vigorous
Q9. I woke up feeling fresh and rested
Q10. My daily life has been filled with things that interest me
Implementation details:
The questions in the module should be fielded using the exact phrasing and answer formats provided. Questions are designed be used as a set and asked in the order in which they appear. Given limited space, only Q2-Q5 (the PHQ-4) can be asked. If implemented in full, it is recommended to include a brief transition statement between the PHQ-4 (Q2-Q4) and the WHO-5 (Q6-Q10).
The PHQ-4 screening tool (Q2-5) is a short, four-question tool to identify the presence and severity of core symptoms of both depression and anxiety, given that these are two of the most prevalent mental health illnesses among the general population and are often comorbid.
The PHQ-4 pulls the two, core depression-related questions from the longer Patient Health Questionnaire (which together are called the PHQ-2), plus two core anxiety-related questions from Generalized Anxiety Disorder (GAD)-7 (which are called the GAD-2).
For survey designers with sufficient space, it is recommended to field the full length, eight-item Patient Health Questionnaire (PHQ-8) survey and the full length seven-item Generalized Anxiety Disorder (GAD-7) survey. The full questionnaires and scoring conventions of both are outlined in detail in Annex 2.B of Measuring Population Mental Health (OECD, 2023[15]).
The World Health Organization Well-Being Index (WHO-5) is a short questionnaire of five items that focus on a respondent’s positive affect. The questionnaire was first used in a project on well-being measures in primary health care by the WHO Regional Office in Europe in 1998 and since then has been translated into more than 30 languages. The WHO-5 has been applied as a generic scale for well-being across a wide range of study fields and countries and also as a sensitive screening tool for depression in clinical trials. For more detail, refer to Annex 2.B of Measuring Population Mental Health (OECD, 2023[15]).
Time:
This module is expected to take about 4 minutes to complete in its entirety.
Origin:
This set of recommendations appears in (OECD, 2023[15]).
Q1 is fielded by Statistics Canada (Statistics Canada, 2024[16]).
Q2-4 make up the Patient Health Questionnaire-4 (Kroenke et al., 2009[17]). The full-length Patient Health Questionnaire (PHQ-8) and the WHO-5 are both included in the European Health Interview Survey, fielded in EU countries every five to six years. Detailed implementation instructions can be found in Eurostat (2024[18]). Additionally, the PHQ-8 and GAD-7 are included in on-going OECD Health Division work to monitor mental health system performance (OECD, forthcoming[19]).
Q6-10 make up the WHO-5 Wellbeing Index (Topp et al., 2015[20]). The WHO-5 Index has been translated into multiple languages; questionnaires in 29 different languages can be downloaded at: https://www.who.int/publications/m/item/WHO-UCN-MSD-MHE-2024.01.
Output:
Q1 can be reported as the share of the population reporting each answer option (“excellent”, “very good”, “good”, “fair”, “poor”), omitting missing values. In addition, rather than report a deprivation, the share of respondents who report having “excellent” or “very good” mental health should be reported jointly as a measure of positive mental health.
The Patient Health Questionnaire-4 should be scored as follows:
Each answer option is assigned the following numerical value: not at all = 0; several days = 1; more than half the days = 2; nearly every day = 3.
All items are added together to provide a total score of psychological distress ranging from 0-12, with higher scores indicating the presence of more symptomatology: 0-2 normal, 3-5 mild, 6-8 moderate and 9-12 severe.
Missing values should not be imputed. If one of the two questions comprising the depression component of the scale is missing (Q4 or Q5), the risk for depression should not be calculated; however, if both questions comprising the anxiety component of the scale are present (Q2 and Q3), the risk for anxiety can be calculated. The same is true vice versa, should both depression questions be present, but one anxiety item is missing.
A total score greater than or equal to 3 for the first two items (GAD-2) indicates that the respondent is at risk for generalised anxiety. A total score greater than or equal to 3 for the final two items (PHQ-2) indicates that the respondent is at risk for depression. Data producers can then report on the share of the population at risk for psychological distress, at risk for anxiety and at risk for depression, respectively.
The WHO-5 Well-being Index should be scored as follows:
Each answer option is assigned the following numerical value: all of the time = 5; most of the time = 4; more than half the time = 3; less than half the time = 2; some of the time = 1; and at no time = 0.
If any one of the five items is missing, the overall score cannot be calculated, and all values should be treated as missing.
All items are added together to provide a total score from 0 to 25, which is then multiplied by 4 to normalise to a score of 0 (worst possible well-being) to 100 (best possible well-being). Data producers can report the mean value; the mean value of responses, and the standard error of this estimate, can be used to describe differences between population groups.
The percentage of the population reporting scores below a threshold (a score < 50) can also be reported to indicate the share of the population experiencing “low levels of positive mental health”.
It is not recommended to present the answers to individual scale items to provide information on a particular emotional state. Data producers interested in doing so should instead refer to measures of affect (Box 2.5).
Guidelines for interviewers:
Q1 is intended to capture how people evaluate their mental health in general and does not refer to a specific time period.
Q2-10 refer to the previous two weeks and, as such, are designed to capture the persistence of each emotional state.
For Q2-Q10, if a respondent reports feeling the emotion only one day over the past two weeks, the output should be coded as “Not at all”, rather than “Several days”. If the respondent felt the emotion for two days, this should be coded as “Several days”. This aligns with protocols outlined by Eurostat (2024[18]).
Visual aids may be needed to remind respondents of the different answer options for each question set. For phone-based surveys, additional prompting and reminders may be necessary.
Time use surveys
The preceding survey modules in this chapter are designed to be integrated into household surveys. Time use surveys, on the other hand, provide data on how people spend their time, how long they spend on different types of activities and who they spend their time with. Increasingly, time use surveys are used to provide important insights into unpaid labour and care work (UNSD, 2025[21]; Tchipeva, Miceli and Ninka, 2024[22]), and they can also be very valuable in understanding relational components of well-being, such as time spent socialising (OECD, 2024[23]).
Subjective well-being data can, and have been, integrated into time use surveys and are especially well suited for collecting data on affect. Refer to Annex A for a discussion of how affect measures – and in particular those collected in time use diaries – are relevant for policy. Since measures of affect aim to capture information on how a respondent actually feels in a given moment, affect data collected via time use surveys are considered the gold standard in measuring directly experienced emotions and states through population surveys. Because countries have varied approaches to fielding time use surveys – and many are in the process of updating and digitising these surveys – further options, in addition to the one shown in the module below, are provided for interested data producers in Annex A. Practical examples of how national statistical offices have implemented the different approaches are also shown.
Time use diary measurement
Time use diary measurement captures information on how the respondent feels directly in the time use diary alongside information on the activities a respondent engaged in the day prior, whom they were with and where they were. This approach – shown in Box 2.8 – includes one question (Q1) asking respondents to rate how pleasant or unpleasant each specific activity was, which can be integrated into an existing time use diary (see Figure 2.2 for an example).
The primary question on overall satisfaction with life (Q1 in the core module, Box 2.2) should also be asked as part of the time use survey.
Q1. Was this moment pleasant or unpleasant? (see right column in figure below)
[from -3 (very unpleasant) to +3 (very pleasant)]
Source: Adapted, to include a subjective well-being question, from Eurostat (2020[24]), Harmonised European Time Use Surveys: 2018 Guidelines: Re-edition, 2020 edition, https://ec.europa.eu/eurostat/web/products-manuals-and-guidelines/-/ks-gq-20-011.
Implementation details:
Q1 should be answered as a part of the time use diary, so that respondents capture information on how pleasant/unpleasant an activity is, as soon after that activity as possible.
Time:
Q1 is expected to add an extra 5 minutes to the time it takes respondents to complete the time use diary (OECD, 2013[25]).
Origin:
Q1 was included in the original Guidelines, and is the formulation currently used by the Italian national statistical office. Additionally, the Finnish statistical office uses a version of this question, supplemented by a visual aid, and it has been used by the French statistical office in previous time use surveys (refer to the Annex for details).
Output:
Information derived from Q1 is conceptually similar to calculating an affect balance (see Box 2.5). Responses to Q1 can also be presented as the mean score for different activity types or the mean score for different demographic groups (e.g. sex, age groups, labour force status).
It is also possible to construct a U-index (unpleasantness index) from this data (Kahneman and Krueger, 2006[26]), calculated as the proportion of time when the net affect associated with an activity is positive (i.e. >= 0).
Guidelines for interviewers:
These questions relate to how the respondent felt during a specific episode identified from a time use diary. It is important that the respondent answers with respect to how they felt during the period of time covered by that episode rather than providing information on how they felt during the day as a whole or what the dominant emotion was during the day.
References
[4] Clark, A. (2001), “What really matters in a job? Hedonic measurement using quit data”, Labour Economics, Vol. 8/2, pp. 223-242, https://doi.org/10.1016/S0927-5371(01)00031-8.
[5] Clark, A., Y. Georgellis and P. Sanfey (1999), “Job satisfaction, wage changes and Quits: Evidence from Germany”, SSRN, https://papers.ssrn.com/abstract=132109.
[7] Cummins, R. (ed.) (2024), Personal Wellbeing Index Manual, Australian Centre on Quality of Life, School of Psychology, Deakin University – Melbourne Campus, https://www.acqol.com.au/uploads/pwi-a/pwi-a-english.pdf.
[18] Eurostat (2024), European Health Interview Survey wave 4 – Methodological manual – 2024 edition, European Commission, https://ec.europa.eu/eurostat/web/products-manuals-and-guidelines/w/ks-gq-24-018.
[24] Eurostat (2020), Harmonised European Time Use Surveys : 2018 Guidelines : Re-edition, 2020 edition, https://ec.europa.eu/eurostat/web/products-manuals-and-guidelines/-/ks-gq-20-011.
[26] Kahneman, D. and A. Krueger (2006), “Developments in the measurement of subjective well-being”, Journal of Economic Perspectives, Vol. 20/1, pp. 3–24, https://doi.org/10.1257/089533006776526030.
[3] Kaiser, C. and A. Oswald (2022), “The scientific value of numerical measures of human feelings”, PNAS, Vol. 119/42, https://doi.org/10.1073/pnas.2210412119.
[17] Kroenke, K. et al. (2009), “An ultra-brief screening scale for anxiety and depression: The PHQ-4”, Psychosomatics, Vol. 50/6, pp. 613-621, https://doi.org/10.1176/APPI.PSY.50.6.613.
[8] Lambert, L. et al. (2020), “Towards a greater global understanding of wellbeing: A proposal for a more inclusive measure”, International Journal of Wellbeing, Vol. 10/2, pp. 1-18, https://internationaljournalofwellbeing.org/index.php/ijow/article/view/1037/877.
[23] OECD (2024), How’s Life? 2024: Well-being and Resilience in Times of Crisis, OECD Publishing, Paris, https://doi.org/10.1787/90ba854a-en.
[15] OECD (2023), Measuring Population Mental Health, OECD Publishing, Paris, https://doi.org/10.1787/5171eef8-en.
[14] OECD (2021), A New Benchmark for Mental Health Systems: Tackling the Social and Economic Costs of Mental Ill-Health, OECD Health Policy Studies, OECD Publishing, Paris, https://doi.org/10.1787/4ed890f6-en.
[13] OECD (2021), Fitter Minds, Fitter Jobs: From Awareness to Change in Integrated Mental Health, Skills and Work Policies, Mental Health and Work, OECD Publishing, Paris, https://doi.org/10.1787/a0815d0f-en.
[25] OECD (2013), OECD Guidelines on Measuring Subjective Well-being, OECD Publishing, Paris, https://doi.org/10.1787/9789264191655-en.
[19] OECD (forthcoming), OECD Mental Health Data Pilot, 2025, OECD Publishing, Paris.
[2] ONS (2024), UK Measures of National Well-being Dashboard, Office for National Statistics (ONS), https://www.ons.gov.uk/peoplepopulationandcommunity/wellbeing/articles/ukmeasuresofnationalwellbeing/dashboard.
[6] Powdthavee, N. (2009), “I can’t smile without you: Spousal correlation in life satisfaction”, Journal of Economic Psychology, Vol. 30/4, pp. 675-689, https://doi.org/10.1016/J.JOEP.2009.06.005.
[1] Raja, S. et al. (2020), “The revised International Association for the Study of Pain definition of pain: concepts, challenges, and compromises”, Pain, Vol. 161/9, pp. 1976-1982, https://doi.org/10.1097/J.PAIN.0000000000001939.
[10] Rutherford, L. et al. (2024), European Social Survey Round 12 Cognitive Interviewing, National Centre for Social Research, https://europeansocialsurvey.org/sites/default/files/2025-04/natcen-ess12-cognitive-interviewing.pdf.
[11] Ryff, C. and C. Keyes (1995), “The structure of psychological well-being revisited.”, Journal of Personality and Social Psychology, Vol. 69/4, pp. 719-727, https://doi.org/10.1037/0022-3514.69.4.719.
[16] Statistics Canada (2024), Quality of life indicator: Self-rated mental health, Statistics Canada, https://www160.statcan.gc.ca/health-sante/mental-health-sante-mentale-eng.htm.
[9] Statistics Finland (2024), Citizens’ Pulse, Statistics Finland, https://stat.fi/en/surveys/pulssi.
[22] Tchipeva, T., M. Miceli and E. Ninka (2024), Time Use Surveys in Europe: State of play and a glimpse into the future, Eurostat, https://unstats.un.org/unsd/demographic-social/meetings/2024/tus-egm-20240130/Session-2B-c-Eurostat-Eniel-Ninka.pdf (accessed on 14 February 2025).
[20] Topp, C. et al. (2015), “The WHO-5 well-being index: A systematic review of the literature”, Psychotherapy and Psychosomatics, Vol. 84/3, pp. 167-176, https://doi.org/10.1159/000376585.
[21] UNSD (2025), Guide to Producing Statistics on Time Use, United Nations Statistics Division, https://unstats.un.org/UNSDWebsite/statcom/session_55/documents/BG-3k-Guide_to_Producing_Statistics_on_Time_Use-55UNSC_background-E.pdf.
[12] WHO (2022), Mental Health, World Health Organization, https://www.who.int/news-room/fact-sheets/detail/mental-health-strengthening-our-response.