Survey and Research Instrument Translation and Validation in Pashto and Dari (IRB and Back-Translation Ready)

Survey translation and validation produces a Pashto or Dari version of a questionnaire that measures the same thing as the original, not just a version that reads well. The work combines team translation, review and adjudication, cultural adaptation, testing with real respondents and full documentation. Back-translation can be added as evidence for an ethics board, but on its own it cannot show that a translated question works.

• 195,000
  Afghan-born immigrants lived in the United States in 2022¹⁷
• 56%
  of them speak English less than very well¹⁷
• 3.7 million
  Afghan refugees and other people in need of international protection, end of 2025³⁰
• 2.4 million
  Afghan girls kept out of secondary school³¹

IRB rules still vary. The University of Pittsburgh requires back-translation for expedited clinical and full-board studies, Tufts requires it unless a professional service translates, and Michigan Medicine accepts it as one route if the back-translator has had no contact with the original text.¹³¹⁵¹¹

So we do both. Your instrument goes through team translation and pretesting, and your IRB receives a blind back-translation with a reconciliation log showing every discrepancy and how it was resolved.

• af_01

  Afghan Dari, not Iranian Persian

  Vocabulary, spelling and everyday terms differ. When Minnesota’s health department adapted a child screening questionnaire, it found many words “more appropriate to an Iranian dialect and not appropriate for Afghan Dari speakers.”¹⁶
• af_02

  The Pashto your sample speaks

  Southern and northern Pashto differ in vocabulary and pronunciation. We translate into the variety your respondents speak and test with speakers of it.
• af_03

  Language labels in your screener

  Afghan immigrants in the U.S. report their home language as Pashto (32%), Farsi (27%) or Dari (26%). Offer all three labels and route Farsi and Dari speakers to Afghan Dari materials.¹⁷
• af_04

  Dates and the Solar Hijri calendar

  Many respondents know dates in the Solar Hijri calendar, with Afghan month names such as Wray and Zmaray in Pashto and Hamal and Asad in Dari. Someone who arrived in August 2021 may answer “1400.”
• af_05

  Numerals and response codes

  Respondents may see ۱ ۲ ۳ ۴ ۵ while your data must still store 1 2 3 4 5. We check that every code survives translation and programming.
• af_06

  Answer scales that hold

  We test whether five agreement or frequency steps keep their order and intended strength in Pashto and Dari, and replace scales that collapse.
• af_07

  Household and family

  An Afghan home can hold several related families. We define “household” inside the item so counts compare across languages.
• af_08

  Literacy and mode

  Literacy in Afghanistan is 37%, and 27% among women. We prepare interviewer-administered and audio versions where self-completion would exclude people.¹⁸
• af_09

  Safety on sensitive items

  Many respondents have family inside Afghanistan. We reword or flag items on politics, religion, income and legal status, and add distress and referral wording where needed.

Language spoken at home by Afghan immigrants in the U.S., 2022 American Community Survey. Many Afghans call Dari “Farsi,” so screeners should offer both labels.¹⁷

• out_01

  Translated instruments

  Pashto and Dari versions in Word and PDF, laid out to match your English.
  L1+
• out_02

  Blind back-translation

  Into English, by a linguist who has not seen your source, whenever your IRB asks for one.
  L1+
• out_03

  Reconciliation log

  Every discrepancy between source and back-translation, the decision taken and why.
  L1+
• out_04

  Certificate of translation

  Signed, naming translator and reviewer, with study title, document names and version dates matched to your approved English.
  L1+
• out_05

  Qualification statements

  Degrees, languages and varieties for everyone who translated or reviewed.
  L1+
• out_06

  Adjudication record

  Item-by-item decisions from the review and adjudication meeting.
  L2+
• out_07

  Cognitive interview report

  Who was interviewed by variety and gender, what they understood and what changed.
  L2+
• out_08

  Platform-ready files

  Qualtrics, REDCap, SurveyCTO, KoboToolbox or CommCare, checked right to left on phones.
  L2+
• out_09

  Amendment-ready versions

  Tracked Pashto and Dari changes and a new certificate whenever your English changes.
  L1+
• out_10

  Methods paragraph

  Wording for your protocol, grant or manuscript that states exactly what was done.
  L1+

1. step_1

   Review the source

   We check translatability, look for existing Pashto or Dari versions, confirm licensing with the instrument’s developer and agree target varieties and mode with you.

   Owner: Senior methodologist · Output: Translation brief and licensing record

2. step_2

   Translate twice

   Two translators produce independent forward translations in each language. Both are native speakers of the target variety and hold university degrees.

   Owner: Language specialist bench · Output: Two independent versions per language

3. step_3

   Review and adjudicate

   Translators, a reviewer and the adjudicator compare versions item by item. The adjudicator decides and records the reason.

   Owner: Reviewer and adjudicator · Output: Adjudication record

4. step_4

   Adapt for Afghan respondents

   The Cultural Compliance Bureau reviews variety, dates, numerals, examples, sensitive items and safety wording.

   Owner: Cultural Compliance Bureau · Output: Cultural Compliance Bureau adaptation notes

5. step_5

   Back-translate blind

   An independent linguist who has not seen your English translates back. We reconcile and log every discrepancy.

   Owner: Independent linguist · Output: Back-translation and reconciliation log

6. step_6

   Test with respondents

   Two rounds of cognitive interviews with Pashto and Dari speakers across varieties and genders, then revisions; changed items are back-translated again before certification. Level 3 adds a pilot and psychometric analysis.

   Owner: Interviewers and analysts · Output: Cognitive interview report

7. step_7

   Document, certify and deliver

   Certificate, logs, reports and platform files, versioned for every IRB amendment.

   Owner: Engagement manager · Output: Certificate of translation and delivery files

• Cf

  Informed consent forms

  Key information section first, as the revised Common Rule requires.

  Level 1

• Sf

  Short form consent

  Short form documents and the written summary for witnessed consent.

  Level 1

• Pp

  Parental permission

  Permission forms for research with children.

  Level 1

• As

  Child assent

  Assent scripts written at the child’s level.

  Level 1

• Rc

  Recruitment materials

  Flyers, emails, text messages and phone scripts.

  Level 1

• Sc

  Eligibility screeners

  Including language-of-interview questions that route correctly.

  Level 1–2

• Sv

  Survey questionnaires

  Web, phone and face-to-face modes.

  Level 2

• Sl

  Validated scales

  Psychometric measures, with the developer’s permission.

  Level 2–3

• Ig

  Interview guides

  Semi-structured guides with probes and prompts.

  Level 2

• Fg

  Focus group guides

  Moderator guides, consent scripts and handouts.

  Level 2

• Db

  Debrief and distress

  Debriefing scripts, distress and referral protocols.

  Level 1–2

• Pi

  Participant information

  Information sheets for UK, EU and Australian studies.

  Level 1

• Pn

  Privacy notices

  GDPR notices under Articles 13 and 14, in the plain language Article 12 requires.

  Level 1

• Ed

  School surveys

  School climate, parent and student surveys for education research.

  Level 2

• Cb

  Codebooks

  Value labels and data dictionaries in Pashto and Dari.

  Level 1

• Au

  Audio versions

  Recorded items and read-aloud scripts for respondents who do not read.

  Level 2

• Or

  Open-ended responses

  Translated back into English for analysis.

  Add-on to any level

• Tr

  Translation review

  Review of a Pashto or Dari translation you already have, with a change log and a certificate once it is accurate.

  Level 1

• Tx

  Interview transcripts

  Pashto and Dari interviews and focus groups transcribed and translated into English.

  Add-on to any level

• Im

  Interviewer materials

  Interviewer manuals, training slides and field protocols in Pashto and Dari.

  Level 2

• Rs

  Results summaries

  Plain-language summaries of findings for participants and communities.

  Add-on to any level

• Ct

  Certificates

  Certificate of translation for every document.

  Included at every level

• Qualtrics

  Its built-in language list includes Persian but not Pashto or Dari, so we configure and document the language setup and check every question right to left.

• REDCap

  Multi-language management for Pashto and Dari, with instruments and alerts translated and field codes unchanged.

• SurveyCTO

  Multi-language forms tested on Android, with plain-text choice labels because HTML is not allowed there.

• KoboToolbox and ODK

  XLSForm label::Pashto (ps) and label::Dari (prs) columns, with right-to-left display checked on devices.

• CommCare

  Bulk application translations, with Pashto and Dari checked on the devices your team uses.

• Paper and audio

  Print-ready PDFs, interviewer scripts and recorded audio for respondents who do not read.

• Pashto letters ټ ډ ړ ږ ښ ځ څ ڼ ګ ې ۍ render in the platform font
• Dari half-spaces (zero-width non-joiners) survive import
• Numerals display as intended while codes store as numbers
• Piped text and skip logic work in both languages
• Every screen checked on a phone and a tablet

Sources: institutional IRB policy pages, checked September 2026.^{91011121314152627}

A signed, version-matched certificate with named qualifications is built to meet the certificate requirements above, and a blind back-translation covers the IRBs that require one. Your IRB makes the final call.

• Confidentiality agreements

  Signed by every linguist and interviewer before any file is shared.

• Your data agreements

  Data use agreements, HIPAA business associate agreements when health information is involved, and GDPR data processing agreements for European partners.

• Certificates of Confidentiality

  We work within the disclosure limits of your NIH Certificate of Confidentiality and bind our team to them.

• No public AI on participant data

  Transcripts and open-ended responses never go into public AI tools, and client material is never used to train models.

• Access and deletion

  Encrypted transfer and storage, access limited to named team members, and a written deletion record at close.

• No routing through Afghanistan

  We never route documents, data or requests through channels controlled by the de facto authorities in Afghanistan.

• Safe pretesting

  Cognitive interviews run under your study’s IRB approval or determination, with our protocol and consent script, gender-matched interviewers, trauma-informed probes, no names in notes, recordings only with permission, and participants paid for their time.

• Data location by agreement

  Data stays where your agreement says, including U.S.-only handling when your IRB or funder requires it.

SAM.gov registered · UEI M2UDMUDFXGL9 · CAGE 1Z3A3 · Registered vendor in 37 states

Standards our method is aligned with

What this service does not do

• We do not act as your IRB or give legal advice.
• We do not certify instruments or issue seals.
• We do not translate copyrighted scales without the developer’s permission.
• We have no operations inside Afghanistan and never route documents, data or requests through channels controlled by the de facto authorities.
• We do not present machine translation as human translation.

Program oversight

Hassan Ukasha

Managing Partner

• B.S.Cornell University
• M.P.H.Cornell University

Oversees Ariana Nexus’s operations and its research translation and validation program. He approves each study’s translation and documentation plan and is the senior point of accountability for principal investigators, IRB offices and research sponsors.

• ORCID
• Google Scholar
• LinkedIn

Delivery team

• 

  Zeba Haqbani

  Senior Partner

  • B.Sc.University of British Columbia
  Survey platforms, data systems and AI checks: Qualtrics, REDCap, SurveyCTO and KoboToolbox builds and right-to-left testing.

  Lived in Kabul

• 

  Hussain Ahmad

  Analyst

  • M.Eng.Cornell University
  • Ph.D.University of Chicago
  Psychometric analysis: reliability, factor structure and measurement invariance across language versions.
• 

  Wasil Peroz

  Principal

  • B.A.Milli University
  • M.Sc.Otto von Guericke University Magdeburg
  Research ethics and data protection: IRB documentation, GDPR and data agreements, European ethics committees.
• 

  Maryam Safi

  Principal

  • B.A.Cornell University
  Engagement delivery and documentation: certificates, reconciliation logs and version control.

Research specialists

• 

  Tamana Ghaznawi

  Senior Partner

  • B.S.Cornell University
  • M.P.H.Cornell University
  Public health survey design and health research instruments.

  Lived in Kabul

• 

  Diana Ayubi

  Engagement Manager

  • B.A.Cornell University
  • Psy.D.West Chester University
  Psychological, trauma and moral injury measures, trauma-informed pretesting and interviewer well-being.

  Lived in Kabul

• 

  Shukria Sakhi

  Engagement Manager

  • B.S.Brown University
  • M.P.H.Brown University
  Program evaluation and cognitive interview fieldwork.

Units that review every engagement

• Cultural Compliance Bureau

  Reviews every adapted item for cultural fit, safety and dialect before sign-off.

• Human Intelligence Collective

  Our language specialist bench: native translators, reviewers and cognitive interviewers across Pashto varieties and Dari dialects, matched by gender.

• AI Data Factory

  Consistency, terminology and platform checks. Never the translation of record.

This service in other languages

• FrançaisTraduction, rétrotraduction et validation de questionnaires en pachto et en dari
• DeutschÜbersetzung, Rückübersetzung und Validierung von Fragebögen auf Paschtu und Dari
• EspañolTraducción, retrotraducción y validación de cuestionarios en pastún y dari
• ItalianoTraduzione, traduzione inversa e validazione di questionari in pashtu e dari
• PortuguêsTradução, retrotradução e validação de questionários em pashto e dari
• SvenskaÖversättning, återöversättning och validering av enkäter på pashto och dari
• TürkçePeştuca ve Darice anket çevirisi, geri çeviri ve geçerlik çalışması
• РусскийПеревод, обратный перевод и валидация опросников на пушту и дари
• 中文普什图语和达利语问卷翻译、回译与验证
• हिन्दीपश्तो और दरी में प्रश्नावली अनुवाद, बैक-ट्रांसलेशन और सत्यापन
• العربيةترجمة الاستبيانات وترجمتها العكسية والتحقق من صحتها باللغتين البشتوية والدارية
• اردوپشتو اور دری میں سوالنامے کا ترجمہ، بیک ٹرانسلیشن اور توثیق
• עבריתתרגום, תרגום חוזר ותיקוף של שאלונים לפשטו ולדארי

• United States

  IRBs. 45 CFR 46 and 21 CFR 50; FERPA and PPRA for school-based research.
  195,000Afghan-born immigrants (2022 American Community Survey)¹⁷

• Canada

  Research Ethics Boards. TCPS 2 (2022), with English and French source instruments.
  55,195Afghans arrived since August 2021 (to November 2024)¹⁹

• United Kingdom

  Research Ethics Committees. Participant information sheets and UK GDPR.
  39,058people relocated or resettled under UK Afghan schemes (to June 2026)²⁰

• Germany and the EU

  Ethikkommissionen and ethics committees. GDPR Article 12 plain-language information; German source questionnaires.
  449,780Afghan nationals living in Germany (end of 2025)²⁸

• Australia

  Human Research Ethics Committees. National Statement, 2025 edition, in force since 23 June 2026.
  27,500+offshore humanitarian visas granted to Afghans (2020–21 to 2024–25)²⁹