How an instrument enters this registry

Clarification, 7 September 2026. The stage-one description below has been corrected. The initial set does not establish which instruments UK employers and vendors most often field or support a field-wide conclusion about measurement evidence. The dated policy source is retained as published. Admission criteria, grading order and the independent-rater condition are unchanged.

Who may propose an instrument, what a proposal must contain, what the registry checks before a record exists, and what happens to proposals it declines. Grading is a separate step under the rubric; nothing here assigns a grade.

Who graded this. Every grade in this registry was assigned by one rater employed by the steward (1 rater, 0 independent), with AI assistance in literature retrieval and drafting. Grades and statuses are single-rater and frozen from first publication; they will not move until two named psychometric raters who are not employees of the steward have joined. Until then, automated sweeps add citations and flag cells for review; corrections of fact are made in public; no grade changes. Why, and how to volunteer.

Open Workplace Health Standard, instrument registry. Admission rule version 1.1, 2 September 2026, drafted with external review. Licence CC BY 4.0.

The registry's scope is the whole field of workplace health and wellbeing measurement. An instrument is in scope when it measures a construct in the OWHS construct-domain code list and has published measurement-property evidence in working-age populations, in any country and any language. The registry is built one record at a time, each added properly or not at all, and this page states how an instrument gets in, who may propose one, and what happens to proposals we do not accept.

Stages

The registry's scope is the whole field: every instrument with published measurement-property evidence in working-age populations, in any country, whose construct sits in the workplace health and wellbeing code list. It is built one record at a time, properly or not at all, and the stages are public.

Stage one, the initial instrument set. The initial set contains 27 instruments, including measures represented in the OWHS question bank. It is a bounded selection of the field, not a survey of UK employer usage or a census of measurement evidence.

Stage two, the occupational-epidemiology tradition and the single items. The instruments built to measure job conditions and validated against recorded outcomes in long cohorts: the Job Content Questionnaire and the Effort-Reward Imbalance questionnaire first, then the short demand-control forms and the organisational-justice scales. Alongside them, the well-evidenced single-item measures that have been validated against a multi-item instrument for the same construct, such as the single-item burnout and work-engagement measures.

After that, by proposal and by sweep. Anyone may propose an instrument. A proposal names the construct, cites two studies reporting a measurement property in working-age samples, and identifies the licence position. Verified proposals join the watchlist, where every property reads not assessed until a rater grades it. Proposals we decline are listed with the reason. The steward's own proposals go through the same public route.

Three doors

The question bank. Where the OWHS question bank draws items from an instrument, that instrument gets a registry record. This is automatic and needs no proposal: no published item should be without an evidence record behind it.

A proposal. Anyone may propose an instrument as a public issue on the repository's instrument proposal template. The steward proposes on the same template, in public, with the same interest declaration. Until the repository opens with the public release, proposals reach the steward by email at hello@openworkplacehealth.org and are published with the repository.

The monthly evidence sweep. Where the sweep surfaces an instrument that meets the threshold below, it is proposed on the same public template so that the proposal, the check and the decision are on the record in the same way as anyone else's. The sweep never adds a record itself.

What a proposal must contain

The instrument, named as its steward names it, with the version or form intended for the record.
The construct, mapped to a construct-domain code in the OWHS code list, or accompanied by a domain proposal where no code fits.
The evidence, cited by DOI: two peer-reviewed studies reporting a measurement property of this instrument in working-age samples, in any country and any language. Studies that use the instrument without reporting a measurement property of it do not count.
Use, where known: the countries, sectors or national surveys in which the instrument is fielded, and, where its construct family already has records, what this instrument adds. This informs grading order. It never decides admission.
The licence position: the steward's current terms page, or a statement that no steward and no instrument-level licence exist.
An interest declaration: any financial or professional interest the proposer has in the instrument. Instrument developers and copyright holders are welcome to propose their own instruments and are asked to say so.

Evidence that an instrument's scores relate to recorded organisational outcomes is welcome in a proposal and informs grading order. It never substitutes for the evidence in point 3: an instrument with a strong outcome literature and a thin psychometric base is admissible and will be graded as it is found.

A single-item measure is admissible on exactly the same terms. Single items have the measurement-property literature the rule asks for (test-retest, criterion and convergent evidence against a multi-item instrument); the properties that are category errors for a single item are recorded as not applicable, never as absent.

What we check, and what we do not

Our check at this stage is clerical. We confirm that the cited DOIs resolve, that the studies concern working-age samples, that they report a measurement property of the named instrument, and that the licence position can be established. We do not assess the quality of the evidence at admission, because that is grading, and grading follows the rubric and the rater conditions set out in the grading rubric.

A proposal that passes the check is admitted to the watchlist. It is not a finding about the instrument that we have admitted it, and it is not a finding that we have not yet graded it.

From proposal to graded record

Proposed. The proposal is a public issue, visible from the moment it is opened.

Verified. The clerical check has passed. The date is recorded on the record.

On the watchlist. The instrument has a record, and every property on it reads not assessed: we have not yet searched the literature, which says nothing about the literature. No grade of any kind appears, including provisional or indicative grades. The record restates why: grades are pending independent raters, not pending our opinion. Each watchlist record is also a specific ask for a rater, with the proposer's cited studies and nine ungraded properties on it.

Queued. Grading order is published and is organised by construct family, not by the order proposals arrived. Instruments the steward's own products select are graded last within their family. Where there is no scheduled grading date, the record says so.

Graded. A rater grades the record under a published rubric version, under the rater conditions in force at the time. The grades enter the matrix, and the record leaves the watchlist.

Counting

Graded instruments and watchlist instruments are counted separately wherever the registry states a number. They are not added together into a single total.

Capacity

The registry has one part-time rater and grades and statuses are frozen from first publication until two named independent raters have joined. Until the first grading cycle after the freeze lifts has run, the registry publishes no throughput figure and no admissions cap, because it would be a guess. Grading order runs by construct family, one or two instruments per family per cycle, so that breadth across the field grows evenly rather than depth in one corner. That is an ordering rule, not a limit on scope.

Proposals we decline

Declined proposals are listed publicly with the reason, from this set:

below-evidence-threshold: the cited studies do not report a measurement property of this instrument in working-age samples.
construct-out-of-scope: the construct is outside workplace health and wellbeing as the standard's code list defines it.
duplicate-of-existing-record: the instrument, or the form proposed, is already covered.
licence-position-undeterminable: the steward's terms could not be read and no positive finding of no-formal-licence could be made.
item-text-unobtainable: the instrument cannot be described without reproducing licensed item wording, which this registry does not do anywhere.

A decline is a decision about a proposal against a published threshold. It is not a judgement about the instrument, and it is not a judgement about whether the construct matters at work. Neither the country an instrument comes from nor the fact that its construct family already has records is a reason to decline: an in-scope instrument is admissible from anywhere, and a family with records gains from a further one.

Conflict of interest

The steward of this registry builds commercial products that may implement OWHS. Two things follow, and each can be checked rather than taken on trust. The steward proposes instruments on the same public template as anyone else, with the same interest declaration. Instruments the steward's products select are graded last within their construct family, so that proposal order and commercial interest cannot set the order of our work.

What the registry is, and is not, seeking

The registry grades evidence about instruments, never evidence about interventions. Evidence that an instrument's scores relate to outcomes an organisation records is in scope and has its own graded property. Evidence that using an instrument improves anyone's wellbeing is a different literature, about interventions, and this registry does not cover it.

The registry records that a single item has been shown to screen for a construct, with the published correspondence to the full instrument; it never says that an organisation should use the item instead of the instrument, and it never states a threshold for action. Those are the implementer's decisions and outside the registry.

Source: ADMISSION-v1.1.md in the dataset bundle. CC BY 4.0. Registry text; the construct-domain admission rule for the standard itself is decided by RFC under the governance page, not here.