Fixed price, 48 hours

Questionnaire Pre-screen

Send a draft questionnaire for expert review before your pilot. Within two working days, receive a signed report with item-level concerns, suggested revisions and a provenance check. The second option adds translation-risk notes. Every item is reviewed by a named psychometrician; the review does not predict pilot statistics or establish cross-language equivalence.

Order a pre-screenRead the sample report

Why before the pilot

The cheapest hour in a survey is the one before fieldwork.

A weak item costs as much to field as a good one, and a pilot finds it only after you have paid for it. Most drafts carry a few: a double question, a stem that assumes a job the respondent does not have, an idiom no translator can carry. The pre-screen puts a second pair of eyes on the draft while it is still cheap to change.

How it works

Three passes, one report.

  1. 1You send the questionnaireAs docx, xlsx or csv. Item text is enough; construct labels, keying and the response format help. Name the target languages, if any.
  2. 2An AI first read, then an expert reviewA blind AI critique rates every item for rejection risk. Miloš Kankaraš then reads every item himself. The AI pass decides where he reads hardest, not what he concludes.
  3. 3The report, within 48 hoursSigned and in your inbox within two working days: the items to revise or cut with a critique and a suggested rewrite, the provenance check, and the translation-risk notes for each language.

Two options

One questionnaire, one price.

Both options cover one questionnaire of up to 40 items and end in the same signed report. The second adds a translation-risk pass for each target language you name.

Pre-screen, one language

Up to 40 items in the language they were written in.

You send

The draft questionnaire as docx, xlsx or csv
Construct labels, keying and the response format, if you have them
Who it is for: the population and the setting

You receive

Every item read by a named psychometrician, after an AI first read
One row per item to revise or cut: the critique and a suggested rewrite
A provenance check: items that resemble a published instrument, with its licence class and citation
Format notes on wording: length, reading level, negations, absolutes
A signed report with a plain-words limitations page
Turnaround
48 hours from receipt of the file, on working days.
Scope ceiling
One questionnaire, up to 40 items, one language.
Price
290 EUR
Order a pre-screen

Pre-screen, up to three target languages

The one-language pre-screen plus a translation-risk pass for each language you name.

You send

Everything in the one-language pre-screen
The target languages, up to three

You receive

Everything in the one-language pre-screen
A translation-risk table per language: idioms, locale-bound referents, gender agreement, address form, unsettled technical terms
Notes written for the translator, before translation starts
Turnaround
48 hours from receipt of the file, on working days.
Scope ceiling
One questionnaire, up to 40 items, up to three target languages.
Price
490 EUR
Order a pre-screen

Prices exclude VAT; Montenegrin VAT of 21 percent is added where it applies. Longer questionnaires and more languages are quoted on request.

What the AI pass is, and is not

A first read that agrees with an expert panel. Not a predictor.

We tested the AI critique before selling it. On the 332 items of a large international field trial, where a real expert review had dropped 140, its cut calls were items the review had dropped in 16 of 23 cases, against a base rate of 42 percent (odds ratio 2.78, 95 percent confidence interval 1.76 to 4.40). On 162 published adult self-report items the same pass did not predict which items were statistically weakest. So the claim is exactly this: the AI pass reads the way an expert panel reads, and it is a first read only. The reviewer's judgement is the product.

Two papers that set the boundary

Maeda, H. and Lu, Y. (2026). Automated item evaluation: Predicting item acceptance and rejection using LLM-generated critiques. arXiv:2608.06609. 52,759 items, 34 percent rejected from operational use; adding LLM critiques to the item text improved rejection prediction (fusion model AUC .80).

Why an AI first read is worth having.

arXiv:2608.06609

Lukauskas, M. and Sarkauskaite, V. (2026). Plausible but not valid: A psychometric audit of LLMs as synthetic survey respondents. arXiv:2608.14606. 37 models, every one with a strong acquiescence shift (+0.84 SD); not a drop-in replacement for human respondents.

Why the pre-screen never simulates a respondent.

arXiv:2608.14606

Limitations, in plain words

What a pre-screen does not do.

It does not predict item statistics. On published adult scales its flags were unrelated to which items were empirically weakest.
It does not simulate respondents and does not predict a response distribution.
It does not replace a pilot, cognitive interviews or a psychometric analysis of real responses.
The provenance check names instruments by name, licence class and citation. It quotes no catalogued item and is not a legal opinion.
The translation-risk notes are prompts for the translator, not a translation and not a guarantee of equivalence.

The same limitations page is printed in every report, so the people who read it after you know what it can and cannot carry.

Sample report

What lands in your inbox.

An anonymised pre-screen of a 75-item work-behaviour questionnaire with three target languages: the summary, the revise-and-cut rows with rewrites, the provenance and translation-risk tables, and the limitations page.

Open the sample report (PDF, 8 pages)

The sample is a real run with item ids and scale codes relabelled. A client receives the same document on their own questionnaire.

Order

Send the questionnaire. The 48 hours run from receipt.

Fill in the form and paste the items, or email the file. We confirm receipt and the price by return, and the report follows within 48 hours on working days.

Or send the questionnaire directly to

milos@adriamont.me
Measurement Quality AuditInstrument DevelopmentResearch, Evidence & EvaluationView all services