Most programmes write a questionnaire, field it once, and learn at the analysis stage what it measures. A development turns that order round: the construct is written down and approved first, the items are authored to a rule set and reviewed blind before any screen runs, the pilot is analysed against the blueprint, and the final form arrives with its scoring key, its evidence and a one-page statement of what the scores can support. It takes constructs authored in English; a development in Montenegrin or a regional language, and the parallel-language add-on in those languages, runs from psihologija.me.
Three products, one question each
The pre-screen reads a draft before the pilot. The audit diagnoses an instrument you already run. The development builds the one you do not have. Start wherever you are.
Where what needs reviewing is a manuscript rather than an instrument, the entry product is the Pre-Submission Methods Review, a signed methods referee report before you submit, from 550 EUR ex VAT.
1. Blueprint
Weeks 1 to 2
A one-hour construct interview, then a written definition with up to six facets, the population, the response format and the target length. You approve it in writing before any item is written.
2. Item bank and blind review
Weeks 2 to 4
Sixty to ninety candidate items authored to a written rule set. Every stem is reviewed blind, with the call recorded, before a deterministic screen runs; its disagreements with the review are read in both directions and kept on record.
3. Pilot
Your clock
The surviving bank goes to up to 400 of your respondents in up to two groups, by CSV export from your own tool. Data preparation is included and recorded.
4. Analysis and final form
Four weeks from clean data
Item analysis, parallel analysis and factor evidence, reliability, a stated selection rule, and a final form of twenty to thirty items with scoring key, administration note, an APA technical note with reproducible scripts, a signed verdict page and a sixty-minute handover.
What you bring, what you receive
Instrument Development
One construct, one language, one pilot, one final form.
You bring
- A construct you can name and describe in a paragraph, or the framework document that stands in for one
- The population, the setting and the authoring language
- Access to respondents for the pilot, recruited by you from the population the instrument is for
You receive
- The approved construct blueprint
- The candidate bank, the blind review sheet with every call, and the screen record
- The pilot form and the data-preparation record
- The final form with scoring key and administration note
- A technical note to APA seventh edition, with the scripts that reproduce every number
- A signed verdict page and a sixty-minute handover call
9,500 EUR, ten to twelve weeks from the approved blueprint; four weeks from clean pilot data to the note. Up to six facets, one authoring language, 60 to 90 candidate items, one pilot of up to 400 respondents in up to two groups, a final form of 20 to 30 items, one measurement occasion.
Two add-ons and one shorter route in
Parallel-language version
A second language authored alongside the source, not translated after it. Each stem is composed in the second language against the facet definition and the source stem, back-translated by a second person as the quality check, and every divergence in meaning, intensity, idiom or polarity is resolved before the pilot. The pilot then runs in both languages. Authored directly in English, Montenegrin, Serbian, Croatian, Bosnian, French and Dutch.
3,500 EUR per language; the two language groups count as the two pilot groups.
Two-group invariance test
Is the instrument the same instrument in both of your groups? The configural-to-strict cascade across your two pilot groups on the final form, with partial invariance sought where a level fails and one replace-and-rerun step where an item fails: the item is dropped, the next-ranked reserve item takes its place, and the cascade is rerun. You receive the cascade table, the freed parameters with their substantive reading, and the group comparison where invariance permits it.
2,500 EUR. Two groups, one grouping variable, at least 150 complete cases per group.
Rebuild from an existing instrument
For an instrument an audit has shown should be rewritten rather than repaired. The development starts from the audit’s fix-list and the existing structure instead of a blank blueprint: the construct definition is revised rather than drawn from scratch, replacement and supplementary items are authored to a bank of forty to sixty, and any existing item the audit passed is kept. The pilot, analysis, final form and technical note are the same as the base product.
7,500 EUR. One instrument of up to eight subscales as audited; everything else as the base product.
Prices exclude VAT; Montenegrin VAT of 21 percent is added where it applies.
Where a development stops and a validation programme begins
The ceiling is the product. A development establishes structure and reliability on one pilot. Outside it, and quoted separately: criterion or predictive validity against an external outcome; norms, reference distributions, cut scores or standard setting; comparison across more than two groups, or a second grouping variable; longitudinal stability or a second measurement occasion; reports to individual respondents; a battery of several constructs, or a construct with more than six facets. The technical note names the follow-on evidence the instrument would need before its scores are used for anything the pilot did not test, so you can commission exactly that, or nothing.
Where the pilot runs
By default you administer the pilot form in your own survey tool and send the CSV export; it is the fastest route to clean data and keeps respondent contact with you. Where you have no tool, administration on our assessment platform with a customer-branded invite link is available as an option.
Sample documents
Both samples run the whole method on a synthetic workplace construct, Internal Service Quality, so that every stage is visible without a client’s material.
- Sample 1: construct blueprint and blind review sheet, with the screen record and every disagreement (PDF, 7 pages)
- Sample 2: technical note on the pilot, from item analysis to the final form, scoring key and two-site invariance test (PDF, 7 pages)
The construct is invented, every item was written for the demonstration, and the pilot data are simulated with a known answer built in so that the reader can see the method find it. A client receives the same documents for their own construct and respondents.
Start a development
Tell us what you need to measure, who will answer it and in which language, and we reply within two working days with a start date and the written scope.
Prefer to attach the file, or the form gives you trouble? Send it to milos@centerforpsychology.me.