Single-item perceived stress measure (Elo tradition, SISQ)
Identity
Version: A one-item global stress question. In the Elo tradition the item asks about stress understood as tension, restlessness, nervousness, anxiety or sleeplessness owing to worry, with a preceding short definition of stress, typically rated on a 5-point scale. Later occupational studies (for example Houdmont) use a one-item global job-stressfulness rating.
Structure: 1
Original citation: Elo AL, Leppanen A, Jahkola A (2003). Validity of a single-item measure of stress symptoms. Scandinavian Journal of Work, Environment and Health 29(6):444-451. doi:10.5271/sjweh.752
Steward / publisher: Originates from the Finnish Institute of Occupational Health (Tyoterveyslaitos, TTL); described in the peer-reviewed literature. There is no commercial test publisher and no proprietary distribution product for the single item.
Constructs claimed
Perceived psychological stress / stress symptoms, that is subjectively experienced tension, restlessness, nervousness, anxiety or difficulty sleeping attributed to worry, captured in a single global self-rating. In occupational variants, global job stressfulness.
Evidence
Structural validity Not applicable (category difference)untestedevidence form: canonical
direct category error for a single item
Not applicable by construction; a single item has no internal factor structure. Elo et al. did use factor analysis at the study level to position the stress item among other well-being items, but this concerns the item's placement in a wider correlation structure, not its own dimensionality (Elo, Leppanen & Jahkola 2003).
Convergent and discriminant validity Moderatewell-establishedevidence form: mixed
indirect core evidence is Finnish and Swedish occupational samples; UK evidence exists for the job-stressfulness variant (Houdmont) but the Elo item itself is validated in Nordic populations
The best-evidenced property. In the anchoring validation across four independent Nordic and Finnish datasets, the single stress-symptoms item converged with items on psychological symptoms and sleep disturbance and with validated scales of exhaustion, mental health, sleep, vitality and optimism, and behaved consistently with a validated emotional-exhaustion scale across gender, age and industry groups (Elo, Leppanen & Jahkola 2003). In a Swedish primary-health-care validation of the SMS-delivered single-item stress question (SISQ), the item correlated significantly with subscales of the Job Demand-Control-Support and Effort-Reward Imbalance models and with depression, exhaustion and sleep scales, supporting convergent validity (Arapovic-Johansson et al. 2017). Littman et al. 2006 similarly reported acceptable validity for two single-item psychosocial stress measures against longer instruments. Qualitative cognitive-interview work shows respondents anchor a global job-stressfulness item mainly on psychosocial working conditions, with wellbeing a secondary frame, which supports content validity (Houdmont et al. 2019).
Criterion validity: reference standard Lowthinevidence form: mixed
indirect the GHQ-12 criterion evidence is UK occupational groups but concentrated in high-stress occupations (prison, police) rather than the general UK working population
Some genuine reference-standard evidence exists for the occupational single-item stress form, which is unusual for this class. Houdmont et al. 2021 tested a single-item job-stressfulness measure against the 12-item General Health Questionnaire (GHQ-12) as a criterion for common mental disorder across seven occupational groups (N=20,658) and found acceptable sensitivity and specificity (both >=70%) in high-strain occupations such as prison officers and public-protection police, with discriminatory power optimal at a >=4 cut-off on a 5-point scale. This is screener-against-standard evidence rather than diagnostic equivalence, and performance depended strongly on base rate. Elo et al. framed criterion validity in terms of plausible associations with health and work characteristics rather than a diagnostic standard (Elo, Leppanen & Jahkola 2003).
Criterion validity: organisational Lowthinevidence form: mixed
indirect for hard outcomes (the strongest prospective evidence is Finnish forest workers and a specific injury outcome); the UK evidence (Ravalier) is cross-sectional and self-reported
Present and, for one hard outcome, prospective. Salminen et al. 2014 followed 16,385 Finnish forest-industry employees over eight years and found that high stress on the validated single-item measure was associated with about a 40% increased risk of subsequent hospitalised injury, robust to adjustment for age, sex, marital and occupational status, education and physical work environment. Ravalier 2018 linked single-item-assessed stress to sickness presenteeism, low job satisfaction and turnover intentions in 1,333 UK social workers, though cross-sectionally. Evidence against recorded sickness absence and actual turnover for the single item specifically remains limited, so the grade is held at Low to Moderate.
Internal consistency Not applicable (category difference)untestedevidence form: canonical
direct category error for a single item
Not applicable; a single item has no inter-item covariance and no alpha or omega can be computed.
Test-retest reliability Lowthin
indirect a single Swedish occupational sample over a very short (approximately 1 day) interval; the weighted-kappa values are appropriate for a fluctuating state but do not establish longer-interval stability, and no UK retest data were located
| Coefficient | Type | Interval | Sample | Population | Evidence form |
|---|---|---|---|---|---|
| 0.804 | kappa | consecutive days (approximately 1 day) | 99 | Swedish primary health care employees (SISQ delivered by SMS), whole retest sample | canonical |
| 0.868 | kappa | consecutive days (approximately 1 day) | 52 | Swedish primary health care employees (SISQ delivered by SMS), subgroup selected as reporting no meaningful day-to-day change | canonical |
Direct test-retest evidence exists but is limited to one study. Arapovic-Johansson et al. 2017 reported weighted kappa (squared weights) of 0.804 across the full retest sample (N=99) and 0.868 in a subgroup selecting days reported as not meaningfully different (N=52), over a roughly one-day interval, for the SMS-delivered single-item stress question. Because stress is a fluctuating state, the authors deliberately used a short interval; this supports short-term stability but says nothing about week-to-week or month-to-month reliability, and there is no UK retest evidence. Elo et al. 2003 did not report test-retest as their four datasets were cross-sectional.
Measurement invariance Not applicable (category difference)untestedevidence form: canonical
direct formal invariance modelling is a category error for a single item; substantive comparability across groups is untested
Not applicable in the multi-group confirmatory-factor sense for a single item. Elo et al. showed the item discriminated as expected between gender, age and industry groups, which is consistent with sensible cross-group behaviour but is not formal invariance evidence (Elo, Leppanen & Jahkola 2003). No study establishing formal single-item equivalence (for example differential item functioning) across sex, age, occupation or language was located in the latest review pass.
Responsiveness and MIC Absent (a finding about the literature)untestedevidence form: canonical
direct
No responsiveness or minimal-important-change evidence for the single stress item was located in the latest review pass. Predictive-validity evidence (Salminen et al. 2014; Houdmont et al. 2021) speaks to prognostic value, not to sensitivity to within-person change or an anchored minimal important change.
Populations, languages and norms
Validated and used mainly in Nordic occupational populations (Finnish and Swedish), with the job-stressfulness variant applied in UK occupational groups by Houdmont and colleagues (Elo, Leppanen & Jahkola 2003; Arapovic-Johansson et al. 2017; Houdmont et al. 2019; Houdmont et al. 2021). No canonical UK working-population norm table exists; the Salminen prospective evidence is Finnish forest workers. The short-form Perceived Stress Scale has UK normative data but is a different (multi-item) instrument (Warttig et al. 2013).
Criticisms and controversies
The general single-item critique applies (no internal reliability, no content breadth). Specific to stress, the construct is inherently fluctuating, so temporal-stability expectations must be modest and short-interval; a single global rating also conflates the many referents respondents actually draw on. Cognitive-interview work shows people answering a global job-stressfulness item mix working conditions and health referents, so the item is not clean with respect to any one of them (Houdmont et al. 2019). As a screener its performance is strongly base-rate dependent, working well only where high stress is common (Houdmont et al. 2021). The word 'stress' itself carries different meanings across respondents and languages, a known threat to comparability.
References (10)
- Elo; Leppänen; Jahkola (2003). Validity of a single-item measure of stress symptoms https://doi.org/10.5271/sjweh.752
- Arapovic-Johansson; Wåhlin; Kwak et al. (2017). Work-related stress assessed by a text message single-item stress question https://doi.org/10.1093/occmed/kqx111
- Littman; White; Satia et al. (2006). Reliability and Validity of 2 Single-Item Measures of Psychosocial Stress https://doi.org/10.1097/01.ede.0000219721.89552.51
- Houdmont; Jachens; Randall et al. (2019). What Does a Single-Item Measure of Job Stressfulness Assess? https://doi.org/10.3390/ijerph16091480
- Houdmont; Randall; Kinman et al. (2021). Can a single-item measure of job stressfulness identify common mental disorder? https://doi.org/10.1037/str0000231
- Salminen; Kouvonen; Koskinen et al. (2014). Is a single item stress measure independently associated with subsequent severe injury: a prospective cohort study of 16,385 forest industry employees https://doi.org/10.1186/1471-2458-14-543
- Ravalier (2018). Psycho-Social Working Conditions and Stress in UK Social Workers https://doi.org/10.1093/bjsw/bcy023
- Warttig; Forshaw; South et al. (2013). New, normative, English-sample data for the Short Form Perceived Stress Scale (PSS-4) https://doi.org/10.1177/1359105313508346
- Fisher; Matthews; Gibbons (2016). Developing and investigating the use of single-item measures in organizational research. https://doi.org/10.1037/a0039139
- Diamantopoulos; Sarstedt; Fuchs et al. (2012). Guidelines for choosing between multi-item and single-item scales for construct measurement: a predictive validity perspective https://doi.org/10.1007/s11747-011-0300-3
Record notes
Overall confidence: convergent validity Moderate and well-established; a genuine reference-standard screener finding against GHQ-12 (Houdmont et al. 2021) and a strong prospective organisational finding (Salminen et al. 2014, 40% increased injury risk) lift this above most single items, but both criterion grades stay Low/thin because evidence is concentrated in specific Nordic or high-stress-occupation samples. Test-retest is present but rests on a single short-interval Swedish study (weighted kappa 0.804 to 0.868). The v0.2 structured test-retest object worked well here, letting the two Arapovic-Johansson coefficients be recorded with their different N and selection conditions rather than averaged. Population indirectness is the main limit for a UK audience: the Elo item is Nordic-validated and there is no UK general-working-population norm.