Research Fieldwork and Qualitative Data Collection in Afghan Communities

Bilingual interviewers, focus groups and transcription in Pashto and Dari, with interviews in 22 more Afghan languages

We recruit Afghan participants, conduct interviews and focus groups in their own language and dialect, and deliver verbatim Pashto and Dari transcripts with English translations your analysis can rest on. Interviewers are matched to each participant by language, dialect and gender, and every step runs under your ethics approval.

For universities, research institutes, public health agencies, foundations, NGOs and evaluators in all 50 states, Canada, Europe and Australia.

See what your team receives

24
Afghan languages, matched to the participant’s dialect and region
Women
interview women: gender-matched interviewers and moderators
2
linguists on every transcript: a transcriber and an independent reviewer
50
U.S. states, plus Canada, Europe and Australia
0
data paths through channels controlled by Afghanistan’s de facto authorities

Session: IDI-07
Language: Pashto, southern
Interviewer: Woman, native speaker
Format: Verbatim, then English

00:14:32 · INT

کله چې دلته راغلئ، تر ټولو سخته ورځ کومه وه؟

When you arrived here, which day was the hardest?

00:14:41 · P07

هغه ورځ چې له کوره هېڅ خبر نه و راغلی. زړه مې ډېر تنګ و.

The day no news came from home. My heart was very tight.

Translator’s note: “zṛə me tang wo,” literally “my heart was tight,” is a common Pashto expression of sadness and longing.

00:14:58 · P07

[چوپتیا، ۴ ثانیې] خو ماشومانو ته مې څه ونه ویل.

[silence, 4 seconds] But I did not tell the children anything.

Illustrative transcript excerptWritten for this page. Not from a real participant.

جگرم خون شد

Dari · jigaram khun shud

Literally “my liver became blood.”

Meaning: I was devastated with grief.²³

An empty office meeting room with a long wooden table, pink chairs and floor-to-ceiling windows

What is research fieldwork and qualitative data collection in Afghan communities?

It is the field side of a qualitative study: recruiting Afghan participants, obtaining informed consent, conducting interviews and focus groups in Pashto, Dari or another Afghan language, and turning the recordings into verbatim transcripts and English translations ready for analysis. Ariana Nexus, a consulting and professional services firm in Washington, D.C., runs each of these steps with its own researchers, interviewers and linguists.

What you receive

What your team receives

00:14:41

P07

هغه ورځ چې له کوره هېڅ خبر نه و راغلی. زړه مې ډېر تنګ و.

The day no news came from home. My heart was very tight.

Translator’s note: “zṛə me tang wo,” literally “my heart was tight,” is a common Pashto expression of sadness and longing.

Family in Afghanistan

Second linguist

1
Time stamp
Side-by-side bilingual transcripts with time stamps and speaker labels
2
Speaker label
Names and identifying details are removed or coded before files leave our secure environment.
3
Verbatim source
Verbatim and intelligent verbatim transcription
4
English translation
English translations, side by side with the source
5
Translator’s note
Translator’s notes for idioms and culturally specific terms
6
Code, optional analysis support
Codebook development and coding
7
Second-linguist review
Independent second-linguist review and audio audits

The files you receive

/your-study/

01
Field plan and dated schedule
02
Translated consent and recruitment materials, with certificates of translation
03
Interviewer training and calibration record
04
De-identified recruitment and screening log
05
Encrypted audio recordings, if your protocol allows
06
Verbatim transcripts in Pashto and Dari script
07
English translations, side by side with the source
08
Field notes and debrief memos
09
Glossary of idioms and culturally specific terms
10
Quality record: second-linguist review and audio audits
11
Files ready for NVivo, ATLAS.ti, MAXQDA or Dedoose
12
Methods appendix with COREQ items
13
Certificate of destruction at close-out

Why studies stall

Where qualitative research with Afghan communities loses data

Most studies with Afghan people do not fail at analysis. They fail earlier, in the field, in ways a reviewer cannot see until the transcripts arrive.

L01
The wrong language, or the wrong Dari
27% Farsi · 26% Dari
A Kandahari Pashto speaker and a Kabuli Dari speaker do not share a first language. Screeners that offer only “Dari” also miss people who call their language “Farsi”: in U.S. census data, 27% of Afghan immigrants answer Farsi and 26% answer Dari.²
L02
Women who are never reached
6 of 32
Many Afghan women will not speak openly with a male interviewer, or in front of a male note-taker, so samples skew male. One 2025 study of Afghan evacuees interviewed 6 women and 26 men.¹¹
L03
Fear of being identified
Participants with family in Afghanistan, or with immigration cases pending, weigh every question against the risk of being recognized. Without visible protections, they decline or give safe answers.
L04
Answers filtered through an ad hoc interpreter
Three-way interviews through a relative or a community volunteer compress answers, soften what is painful and break confidentiality. The researcher hears a summary, not the participant.
L05
Machine transcription that is not Pashto
Under 1%
In a 2026 benchmark on read Pashto speech, Whisper models produced Pashto script in under 1% of utterances, with word error rates of 90% to 297%. Interviews are harder than read speech.¹
L06
Transcripts no one can analyze
Summary translations, missing speaker labels, no time stamps and idioms translated word for word leave coders guessing. In Dari, “jigaram khun shud” (جگرم خون شد) is not “my liver became blood.” It is grief.²³

Evidence

What the data says about fieldwork with Afghan communities

Margin figure 1

Machine transcription is not ready for Pashto

No Whisper speech model produced Pashto script in more than 0.8% of utterances, and the best model tested still had a word error rate near 40%, on read speech.¹

Pashto-script output from any Whisper model: at most 0.8% of utterances; MMS-1B, SeamlessM4T and OmniASR each above 93%.

Whisper models, zero-shot (lowest)90% WER
Whisper models, zero-shot (highest)297% WER
Best zero-shot model, SeamlessM4T39.7% WER
Fine-tuned models on new data32.5–59% WER

Word error rate above 100% means the system inserted more wrong words than the audio contained. All tests used read speech; interviews are harder.

Margin figure 2

Afghan immigrants name their language in more than one way

Screeners that list only “Dari” miss the 27% who answer “Farsi.”²

Pashto32%
Farsi27%
Dari26%
All other answers15%

Language spoken at home by Afghan immigrants in the United States, 2022 American Community Survey. Many Afghans call Dari “Farsi,” so screeners should offer both names. 56% of Afghan immigrants in the United States speak English less than very well.

Margin figure 3

Women who are never reached

One 2025 study of Afghan evacuees interviewed 6 women and 26 men.¹¹

Women, 6 ·

Men, 26

What we deliver

Fieldwork services, from recruitment to analysis-ready data

Commission the full study or only the parts you need. Every part is produced by our own people, under one engagement and one point of accountability.

Study design and ethics readiness

Field plans and ethics materials your IRB, REB or ethics committee can approve.

Fieldwork plan, sampling frame and quotas
Screeners that offer Pashto, Dari and “Farsi”
Interview and focus group guide adaptation
Consent, assent and short-form consent translation
IRB, REB, REC and HREC submission support
Certificate of Confidentiality and data management plans
Letters of support and budget justifications for grant proposals
Paperwork Reduction Act planning for federally sponsored studies

Participant recruitment and community access

Afghan participants reached through trusted community networks, never paid panels.

Purposive, snowball and respondent-driven sampling
Afghan refugees, evacuees, SIV holders, asylum seekers and long-settled families
Women, elders, youth and recent arrivals
Incentive management under your protocol
Scheduling, reminders, transport and childcare arrangements
Community advisory boards for participatory research
Private, accessible interview venues

Bilingual interviewers

Native-speaker interviewers, matched by language, dialect and gender.

In-depth interviews (IDIs)
Key informant interviews (KIIs)
Semi-structured and narrative interviews
Life-history and oral history interviews
Cognitive interviews
Dyadic and family interviews
Ethnographic observation and field notes
In person, or by phone, WhatsApp, Zoom or Microsoft Teams

Focus groups

Focus group discussions moderated in the participants’ own language.

Single-gender focus groups in Pashto, Dari and other Afghan languages
Moderator and note-taker pairs
In-person and online focus groups
Live English interpretation for observing researchers
Participatory methods: photovoice, community mapping, free-listing and ranking
Deliberative and co-design workshops

Transcription and translation

Verbatim transcripts in the original language, then English, reviewed by a second linguist.

Verbatim and intelligent verbatim transcription
Transcripts in Pashto and Dari script, or direct English translation
Side-by-side bilingual transcripts with time stamps and speaker labels
Translator’s notes for idioms and culturally specific terms
Independent second-linguist review and audio audits
Transcription of recordings you have already collected
Translation into French, German, Spanish, Swedish and other languages
Files ready for NVivo, ATLAS.ti, MAXQDA and Dedoose

Analysis support

Coding and interpretation with the cultural context your team may not have.

Codebook development and coding
Thematic, framework and rapid qualitative analysis
Inter-coder reliability
Saturation tracking
Cultural interpretation sessions with your team
Methods appendix with COREQ items
Findings returned to participants in Pashto and Dari

Data protection and close-out

Protection designed for participants whose families may be at risk.

Encrypted recording, transfer and storage
De-identification and redaction
Data use and confidentiality agreements
Retention under your protocol
Certificate of destruction at close-out

Mixed-methods studies can add phone surveys in Pashto and Dari, and survey instruments translated and tested through our survey translation service.

Three adults reviewing charts and documents together at an office meeting table — Transcription and second-linguist review

A quiet room with a long wooden table and chairs beside tall windows — Focus group preparation

Why Ariana Nexus

Fieldwork run by scholars from the communities we interview

There is no U.S. certification for research interviewers in Pashto, Dari or any Afghan language. We own that part: we train, test and calibrate our own interviewers, moderators and transcriptionists to a written standard, and we train interpreters and students in Afghan languages beyond our own team.

01
Scholars, not bilinguals
Interviewers and moderators hold degrees, complete human subjects research training and pass our interview calibration. The work is led by public health researchers and scholars from Cornell, Brown, the University of Chicago and the University of British Columbia.
02
Matched to every participant
Language, dialect, region and gender. Women interview women. A Kandahari elder meets an interviewer who speaks his Pashto, not a Kabuli Dari speaker reading a translation.
03
Access that rests on trust
We recruit through community relationships our team has built over years across all 50 states, Canada, Europe and Australia, not through paid panels.
04
Verbatim first, then English
Every interview is transcribed in its original language before translation, and a second linguist reviews transcript and translation against the audio. Nothing is summarized unless you ask for it.
05
Protection built for Afghan participants
Many participants have family under Taliban rule. We de-identify early, keep identities apart from transcripts, and never route data through channels controlled by the de facto authorities.
06
Ready for analysis on day one
Transcripts arrive with speaker labels, time stamps, translator’s notes and metadata, formatted for NVivo, ATLAS.ti, MAXQDA or Dedoose, with a methods appendix your manuscript can cite.
07
One team, one point of accountability
Our own researchers, never another firm’s. The people who scope your study run it, and the Cultural Compliance Bureau reviews every guide, consent form and a sample of translated transcripts.
08
Human-led, with AI kept in its place
Interviews, transcription and translation are done by people. Any software that assists runs inside our secure environment, and your data never trains an AI model.

How we deliver

How a fieldwork engagement runs

Seven stages, each with a named owner and a deliverable you can check. Timings are typical for a 30-interview study and are set with you at scoping.

Weeks 1–2
Stage 01
Scope and ethics
A scoping call with a senior researcher; field plan, quotas and timeline; inputs for your IRB, REB or ethics committee; translated consent materials.
Deliverable: Field plan and ethics pack · Owner: Engagement manager
Weeks 2–3
Stage 02
Guides and team
Interview and focus group guides adapted and reviewed by the Cultural Compliance Bureau; interviewers selected by language, dialect and gender; protocol training and mock interviews.
Deliverable: Final guides and training record · Owner: Senior researcher and Cultural Compliance Bureau
Weeks 3–6
Stage 03
Recruitment and consent
Community outreach, screening against your criteria, scheduling, and informed consent in the participant’s language.
Deliverable: De-identified recruitment log · Owner: Field coordinator
Weeks 4–8
Stage 04
Interviews and focus groups
Sessions in person or remote, field notes after every session, daily debriefs, and a distress protocol with referral options.
Deliverable: Audio and field notes · Owner: Interviewers and moderators
Rolling, from week 5
Stage 05
Transcription and translation
Verbatim source-language transcripts, English translations with translator’s notes, second-linguist review and audio audits, delivered in batches.
Deliverable: Transcripts and translations · Owner: Linguists and reviewer
Optional
Stage 06
Analysis support
Codebook, coding, inter-coder reliability, a saturation memo and a cultural interpretation session with your team.
Deliverable: Coded data and memos · Owner: Analysts
Final week
Stage 07
Handover and close-out
Data package, methods appendix with COREQ items, a findings summary for participants in Pashto and Dari, and secure destruction under your protocol.
Deliverable: Data package and destruction certificate · Owner: Engagement manager

A typical 30-interview study runs 8 to 10 weeks from ethics approval to final transcripts.

Our standard

The fieldwork standard every study receives

S01
Every interviewer is a native speaker of the language they interview in.
S02
Every interviewer holds a degree and has completed human subjects research training.
S03
Women participants are interviewed by women; groups are single-gender unless your protocol says otherwise.
S04
Consent is explained in the participant’s language, and no one is recorded without agreement.
S05
Every session has a distress protocol and referral options in the participant’s language.
S06
Transcripts are verbatim in the source language before translation, unless you choose direct translation.
S07
A second linguist reviews every transcript and translation, and a sample is audited against the audio.
S08
Idioms and culturally specific terms keep the original word, with a translator’s note.
S09
Names and identifying details are removed or coded before files leave our secure environment.
S10
Recordings are kept only as long as your protocol allows, and destruction is certified.

Security and compliance

Built around the rules your ethics board applies

We work inside your approval and your jurisdiction’s rules. These are the instruments this service is built around; no certification is implied where none exists.

AreaInstrumentsWhat it means for your fieldwork

Area

U.S. human subjects

Instruments

Common Rule (45 CFR 46), IRB review, short-form consent (45 CFR 46.117)

What it means

Consent in the participant’s language, documented the way your IRB approved.

Area

U.S. health and education data

Instruments

HIPAA Privacy Rule; FERPA and PPRA

What it means

Health and student information handled only as your agreements allow.

Area

U.S. confidentiality

Instruments

NIH Certificates of Confidentiality; NIH Data Management and Sharing Policy

What it means

Identifiable data protected against compelled disclosure; data plans written to NIH expectations.

Area

U.S. federal studies

Instruments

Paperwork Reduction Act

What it means

Federally sponsored information collections planned with OMB clearance in mind.

Area

Canada

Instruments

TCPS 2 (2022), Research Ethics Boards, PIPEDA and provincial privacy law

What it means

Consent, privacy and community engagement under Canada’s Tri-Council Policy.

Area

United Kingdom

Instruments

UK GDPR and Data Protection Act 2018; ESRC Framework for Research Ethics

What it means

Lawful basis, participant information sheets and data minimization.

Area

European Union

Instruments

GDPR Articles 9 and 89; ethics committees

What it means

Special-category data, such as health, ethnicity and religion, processed with research safeguards.

Area

Australia

Instruments

National Statement on Ethical Conduct in Human Research (2025); HRECs; Privacy Act 1988

What it means

Ethics review and privacy under Australia’s updated National Statement.

Area

Professional codes

Instruments

AAPOR Code; ICC/ESOMAR International Code; ISO 20252; ISO 17100

What it means

Research practice and translation process aligned with recognized standards.

Area

Reporting

Instruments

COREQ and SRQR

What it means

Methods documented so your manuscript can report them.

Area

Safety

Instruments

WHO ethical and safety recommendations on sexual violence research; SAMHSA trauma-informed principles

What it means

Interviews on violence and trauma conducted safely, with referral options ready.

Area

Information security

Instruments

Encryption in transit and at rest; multi-factor authentication; role-based access

What it means

Only named team members can open your data.

Area

Afghanistan

Instruments

No channel through the de facto authorities

What it means

No data, documents or inquiries routed through systems they control.

We do not route data, documents or inquiries through channels controlled by the de facto authorities in Afghanistan.

Registered in SAM.gov · UEI M2UDMUDFXGL9 · CAGE 1Z3A3

The team behind the work

Who runs your fieldwork

Fieldwork is led by public health researchers and scholars from Cornell, Brown, the University of Chicago and the University of British Columbia, and carried out by interviewers from the communities they interview. Every member comes from the Afghan community and brings lived experience of the places and events participants describe.

PersonOversightMethodologyFieldworkData protectionAnalysisQuality reviewStudy management

Hassan UkashaManaging Partner

Oversight

Tamana GhaznawiSenior Partner

Methodology

Fieldwork

Zeba HaqbaniSenior Partner

Data protection

Shukria SakhiEngagement Manager

Fieldwork

Study management

Diana AyubiEngagement Manager

Methodology

Fieldwork

Hussain AhmadAnalyst

Analysis

Wasil PerozPrincipal · Germany and the EU

Data protection

Maryam SafiPrincipal

Quality review

Program oversight

Hassan Ukasha

Managing Partner

Oversees the firm’s operations and this program. Sets the fieldwork standard and is the final point of escalation on every engagement.

B.S., Global and Public Health Sciences, Cornell University
M.P.H., Infectious Diseases, Cornell University

Grew up in Herat. Languages: Pashto, Dari, English, Urdu and Hindi; working Arabic.

ORCID · Google Scholar · LinkedIn

Tamana Ghaznawi

Senior Partner

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

Leads health research engagements: study design, consent and distress protocols, and fieldwork with patients and families.

Lived in Kabul.

Zeba Haqbani

Senior Partner

B.Sc., University of British Columbia

Builds and runs the secure systems behind fieldwork: encrypted capture, transcript pipelines and de-identification.

Lived in Kabul.

Shukria Sakhi

Engagement Manager

B.S., Brown University
M.P.H., Brown University

Manages studies day to day: recruitment, quotas, scheduling and field logistics.

Diana Ayubi

Engagement Manager

B.A., Cornell University
Psy.D., West Chester University

Sets trauma-informed interviewing and distress protocols, and leads interviewer debriefing and well-being.

Lived in Kabul.

Hussain Ahmad

Analyst

M.Eng., Cornell University
Ph.D., University of Chicago

Leads qualitative data management and analysis: codebooks, coding frameworks and inter-coder reliability.

Wasil Peroz

Principal

B.A., Milli University
M.Sc., Otto von Guericke University Magdeburg

Leads data protection and ethics requirements for studies in Germany and the European Union.

Maryam Safi

Principal

B.A., Cornell University

Leads fieldwork quality: interviewer calibration, audio audits and transcript review.

Field interviewers and focus group moderators
Native speakers of Pashto, Dari, Uzbeki, Turkmeni, Balochi, Pashayi and Aimaq, women and men, trained in your protocol.
Transcription and translation linguists
Degree-holding linguists who transcribe in the original script and translate into English and 13 other languages.
Cultural Compliance Bureau
Reviews every guide, consent form and a sample of translated transcripts for cultural, religious and safety risks.
Human Intelligence Collective
The community network through which we reach participants across all 50 states and abroad.
AI Data Factory
The secure environment where recordings, transcripts and analysis files are kept, de-identified and destroyed.

Languages

Pashto, Dari and 22 more Afghan languages

Every interview is conducted by a native speaker of the participant’s language. We publish what each language supports, because a flat promise across all 24 Afghan languages is not one a careful researcher should accept.

Pashtoپښتو

Dariدری

Full fieldwork

Interviews, focus groups, transcription and translation

Pashto پښتو
Dari دری

Interviews and transcription

In-depth and key informant interviews, transcription and translation

Uzbeki
Turkmeni
Balochi
Pashayi
Aimaq

By arrangement

Specialist interviews with advance notice

Ormuri
Parachi
Wakhi
Shughni
Sanglechi
Ishkashimi
Munji
Yidgha
Kyrgyz
Gawarbati
Tirahi
Nuristani (Ashkun group)
Kati
Prasun
Waigali
Tregami
Brahui

Pashto dialects we match: Southern (Kandahari), central (Paktia and Wardak), and northern and eastern varieties, including Yusufzai.
Dari dialects we match: Kabuli, Herati, Hazaragi and Badakhshi.

Also English, plus French, German, Spanish, Italian, Hebrew, Urdu, Arabic, Hindi, Portuguese, Chinese, Russian, Turkish and Swedish, for studies in host countries and for the translations your team needs.

Where we work

Fieldwork in all 50 states, Canada, Europe and Australia

In person wherever Afghan people live, and remotely where they cannot be reached in person. Studies outside the United States follow the host country’s ethics and privacy rules.

United States

Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming

And Washington, D.C., where we are headquartered.

Canada, Europe and Australia

Canada

Ontario, Quebec, British Columbia, Alberta, Manitoba, Nova Scotia

Europe

Germany, United Kingdom, France, the Netherlands, Sweden, Austria, Denmark, Norway, Belgium, Switzerland, Italy, Spain, Ireland, Finland

Australia

New South Wales, Victoria, Queensland, South Australia, Western Australia

Remote interviews with participants in Pakistan, Türkiye, the Gulf, Central Asia and inside Afghanistan are run from outside the country, after a safety assessment, with no data path through systems controlled by the de facto authorities.

Who we work with

Built for research teams, funders and programs

Universities and academic medical centers
Principal investigators and research teams in public health, medicine, nursing, social work, psychology, education, anthropology, sociology and migration studies.
Research institutes and survey organizations
An Afghan-language field team for studies you lead, as a subcontractor or teaming partner.
Public health agencies and health systems
Community research, needs assessments and program evaluation with Afghan patients and families.
Foundations and philanthropy
Evidence for grantmaking, and evaluations of Afghan-focused programs.
UN agencies, NGOs and resettlement agencies
Needs assessments, accountability to affected people and program evaluation in diaspora and regional settings.
Government research and evaluation offices
Federal, state, provincial and European studies of resettlement, integration and public services.
Think tanks, archives and oral history programs
Policy research, and oral histories of the Afghan war, displacement and resettlement.
Market, UX and AI product research
Consumer, user and product research with Pashto and Dari speakers.

Topics we field

Refugee resettlement and integration
Afghan evacuees and SIV holders
Mental health, trauma and PTSD
Moral injury (sometimes called moral trauma)
Afghan war veterans, interpreters and allies
Maternal, reproductive and child health
Health care access and patient experience
Women’s rights and gender
Gender-based and intimate partner violence
Newcomer students and schooling
Afghan girls’ education
Higher education pathways
Employment, livelihoods and remittances
Housing, food security and public benefits
Returns from Iran and Pakistan
Migration, displacement and family separation
Climate change and drought
Digital inclusion and misinformation
AI and language technology
Civic participation and citizenship
Oral histories of the Afghan war

Ways to work with us

Engagement models

E01
Full study
We design and run fieldwork from ethics to analysis-ready data.
Fixed fee by phase
E02
Field team for your protocol
Our interviewers and moderators run the study your team designed and your IRB approved.
Per completed interview or focus group
E03
Transcription and translation
Verbatim transcripts and English translations of recordings you already have.
Per audio hour
E04
Grant partner
Written into your proposal as a consultant or subrecipient, with a letter of support and a budget justification.
Budgeted with you before submission
E05
Longitudinal and multi-site
Repeat interviews over months or years, across cities and countries, with the same interviewers where possible.
Annual plan by wave

Fixed quotes, in writing, before work starts.
Recruitment billed only for interviews that take place.
Participant incentives paid under your protocol.
Registered in SAM.gov for federal awards and subawards: UEI M2UDMUDFXGL9, CAGE 1Z3A3.

Terms

Fieldwork terms we use

T01 In-depth interview (IDI): A one-to-one interview, usually 45 to 90 minutes, that explores a person’s experience in their own words.
T02 Key informant interview (KII): An interview with someone whose role gives them knowledge of a community or system, such as a community leader, teacher, caseworker or clinician.
T03 Focus group discussion (FGD): A moderated discussion with six to eight participants who share a key characteristic, such as language and gender.
T04 Verbatim transcription: A word-for-word record of everything said, including hesitations, repetitions and non-verbal events.
T05 Intelligent verbatim: A faithful transcript with fillers and false starts removed, used when content matters more than speech patterns.
T06 Code saturation: The point at which new interviews stop producing new themes; in published tests, about nine interviews in a homogeneous group.
T07 Meaning saturation: The point at which each theme is fully understood; published tests put it at 16 to 24 interviews.
T08 Inter-coder reliability: Agreement between two coders applying the same codebook, often reported as Cohen’s kappa.
T09 Certificate of Confidentiality: A U.S. protection against compelled disclosure of identifiable, sensitive research information, issued automatically for NIH-funded studies.
T10 Short-form consent: A U.S. consent method in which the elements of consent are presented orally in the participant’s language, with a witness, and documented on a short written form.
T11 Moral injury: Lasting distress after acting, failing to act or witnessing acts that violate one’s moral beliefs; sometimes called moral trauma.
T12 Idiom of distress: A culturally specific way of expressing suffering, such as the Dari “jigar khun” (جگر خون), that a literal translation misses.

Questions

Questions research teams ask

What does qualitative research fieldwork with Afghan participants include?

It covers everything between an approved protocol and analysis-ready data: recruitment and screening, informed consent in the participant’s language, in-depth and key informant interviews, focus groups, field notes, and verbatim transcription and translation. You can commission the full study or only the parts you need, such as interviewers for your protocol or transcription of recordings you already have.

Can you provide female interviewers and moderators for Afghan women?

Yes. Women participants are interviewed by women, and focus groups are single-gender unless your protocol calls for something else. Where a woman prefers it, we also keep male team members out of the room and off the call, including note-takers and observers.

Which Afghan languages and dialects do your interviewers speak?

We run full fieldwork in Pashto and Dari, interviews and transcription in Uzbeki, Turkmeni, Balochi, Pashayi and Aimaq, and specialist interviews in the other 17 Afghan languages by arrangement. Within Pashto we match southern, central, northern and eastern varieties; within Dari we match Kabuli, Herati, Hazaragi and Badakhshi speakers.

Do you transcribe Pashto and Dari interviews verbatim and translate them into English?

Yes. We transcribe in the original script first, then translate into English, and a second linguist reviews both against the audio. Transcripts carry time stamps, speaker labels and translator’s notes for idioms. We can also translate directly into English, or into French, German, Swedish and other languages, and we transcribe recordings you have already collected.

Can AI or automatic transcription handle Pashto and Dari research interviews?

Not reliably. In a 2026 benchmark on read Pashto speech, Whisper models produced Pashto script in under 1% of utterances, with word error rates from 90% to 297%, and the best model tested still had a word error rate near 40%. Interviews are harder than read speech. Our transcripts are made by people, and your audio never trains an AI model.

How many interviews or focus groups does a study with Afghan participants need?

Published tests find saturation within 9 to 17 interviews or 4 to 8 focus groups for one homogeneous group, and 16 to 24 interviews to understand each theme fully. Afghan samples are rarely one group: separating Pashto and Dari speakers, women and men, or recent and earlier arrivals multiplies the groups to fill. We size the sample with you at scoping.

How do you recruit Afghan refugees, evacuees and diaspora participants?

Through relationships across Afghan communities rather than paid panels, using purposive, snowball or respondent-driven sampling as your design requires. We screen against your criteria in the participant’s language, offer both “Dari” and “Farsi” as answers, manage incentives under your protocol, and arrange transport or childcare when that is what makes participation possible.

How do you protect participants who have family in Afghanistan?

Consent explains the risks plainly, identities are kept apart from transcripts, and names and identifying details are removed before files leave our secure environment. Participants choose how and where they are interviewed. We never route data, documents or inquiries through channels controlled by the de facto authorities, and we support Certificates of Confidentiality where they apply.

Can your team work under our IRB, REB or ethics committee approval?

Yes. Our team completes human subjects research training and works under your approved protocol, as study personnel, through an individual investigator agreement, or under a reliance arrangement your institution chooses. We supply the training records, team descriptions and translated materials your submission or amendment needs.

Can Ariana Nexus be written into our grant proposal?

Yes. We join proposals as a consultant or subrecipient, provide a letter of support, a scope and a budget justification, and are registered in SAM.gov (UEI M2UDMUDFXGL9, CAGE 1Z3A3). Contact us as early as you can before the deadline, so the fieldwork plan in your proposal is one we have agreed to deliver.

Do you run online focus groups and remote interviews?

Yes. We run interviews and focus groups by phone, WhatsApp, Zoom or Microsoft Teams when your protocol allows it and the participant prefers it, and in person where communities live. Online groups stay single-gender, and observing researchers can follow in English through live interpretation on a separate channel.

Do you conduct research inside Afghanistan?

We have no offices or staff operating inside Afghanistan. Remote interviews with participants inside the country, and in Pakistan, Türkiye, the Gulf and Central Asia, are run from outside, only after a safety assessment, and with no data path through systems the de facto authorities control. Monitoring and surveys inside Afghanistan are a separate service.

Can you support studies in Canada, Europe and Australia?

Yes. We work under Canada’s TCPS 2, the GDPR and UK GDPR, and Australia’s National Statement, alongside your Research Ethics Board, ethics committee or HREC. Interviews can be transcribed and translated into the host country’s language, such as French, German or Swedish, as well as English.

Do you interview veterans, interpreters and Afghan allies about the war and moral injury?

Yes. We conduct trauma-informed interviews and oral histories on the Afghan war, displacement and moral injury, sometimes called moral trauma, with Afghan interpreters, allies, families and veterans. Interviewers follow a distress protocol, participants can pause or stop at any time, and referral options are ready in the participant’s language.

How is fieldwork priced?

Each study receives a fixed written quote before work begins. Full studies are priced by phase, field teams per completed interview or focus group, and transcription and translation per audio hour. Recruitment is billed only for interviews that take place, and participant incentives follow your protocol. We reply within two business days.

Is Dari the same as Farsi for research purposes?

Dari and Iranian Persian (Farsi) are closely related but differ in vocabulary, pronunciation and everyday expressions, and many Afghans call their language Farsi. Screeners should offer both names. Our interviewers and transcripts use Afghan Dari, and translator’s notes flag terms an Iranian Persian reader would take differently.

Start a study

Plan your fieldwork with a senior researcher

Tell us about your research question, population, languages and timeline on a short scoping call, or send your protocol ahead. We reply within two business days with a field plan, a dated schedule and a fixed quote.

Conversations are confidential. We sign your NDA or data use agreement before any material is shared.

Request a fieldwork proposal

Reviewed October 4, 2026 by Hassan Ukasha, Managing Partner. Figures on this page are sourced and dated.

Sources

23 sources, dated and linked

1.Benchmarking Multilingual Speech Models on Pashto: Zero-Shot ASR, Script Failure, and Cross-Domain Evaluation. arXiv:2604.04598 (2026). arxiv.org
2.Migration Policy Institute. Afghan Immigrants in the United States (2022 American Community Survey). migrationpolicy.org
3.U.S. Census Bureau. 2024 American Community Survey 1-year, tables B05006 and B04006. data.census.gov
4.U.S. Department of Homeland Security. Operation Allies Welcome announcement, February 19, 2022. dhs.gov
5.UNHCR. Global Trends: Forced Displacement in 2025 (June 2026). unhcr.org
6.Immigration, Refugees and Citizenship Canada. Afghanistan: key figures. canada.ca
7.UK Home Office. Afghan Resettlement Programme operational data, to June 30, 2026. gov.uk
8.Destatis. Foreign population by citizenship (2026). destatis.de
9.Australian Department of Home Affairs. Offshore Humanitarian Program 2024–25. homeaffairs.gov.au
10.UNESCO. Five years on, UNESCO calls for full restoration of the right to education for Afghan girls and women (2026). unesco.org
11.“Then, We Lost Everything”: Afghan Refugee Mental Health Challenges Post-2021 Evacuation. Journal of Immigrant and Minority Health 27(2):251–257 (2025). pmc.ncbi.nlm.nih.gov
12.Hennink M, Kaiser BN. Sample sizes for saturation in qualitative research: A systematic review of empirical tests. Social Science & Medicine 292:114523 (2022). pubmed.ncbi.nlm.nih.gov
13.Hennink MM, Kaiser BN, Marconi VC. Code saturation versus meaning saturation: How many interviews are enough? Qualitative Health Research (2017). pmc.ncbi.nlm.nih.gov
14.Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ). International Journal for Quality in Health Care 19(6):349–357 (2007). doi.org
15.National Institutes of Health. Certificates of Confidentiality, NOT-OD-17-109. grants.nih.gov
16.Electronic Code of Federal Regulations. 45 CFR Part 46, Protection of Human Subjects. ecfr.gov
17.Government of Canada. Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (TCPS 2, 2022). ethics.gc.ca
18.NHMRC. National Statement on Ethical Conduct in Human Research (2025). nhmrc.gov.au
19.Regulation (EU) 2016/679, General Data Protection Regulation. eur-lex.europa.eu
20.World Health Organization. Ethical and safety recommendations for researching, documenting and monitoring sexual violence in emergencies (2007). who.int
21.Litz BT, Stein N, Delaney E, et al. Moral injury and moral repair in war veterans. Clinical Psychology Review 29(8):695–706 (2009). doi.org
22.SAMHSA. Concept of Trauma and Guidance for a Trauma-Informed Approach, HHS Publication (SMA) 14-4884 (2014). library.samhsa.gov
23.Miller KE, Omidian P, Quraishy AS, et al. The Afghan Symptom Checklist: a culturally grounded approach to mental health assessment in a conflict zone. American Journal of Orthopsychiatry 76(4):423–433 (2006). doi.org

Research Fieldwork and Qualitative Data Collection in Afghan Communities

What is research fieldwork and qualitative data collection in Afghan communities?

What your team receives

Time stamp

Speaker label

Verbatim source

English translation

Translator’s note

Code, optional analysis support

Second-linguist review

The files you receive

Where qualitative research with Afghan communities loses data

The wrong language, or the wrong Dari

Women who are never reached

Fear of being identified

Answers filtered through an ad hoc interpreter

Machine transcription that is not Pashto

Transcripts no one can analyze

What the data says about fieldwork with Afghan communities

Machine transcription is not ready for Pashto

Afghan immigrants name their language in more than one way

Women who are never reached

Fieldwork services, from recruitment to analysis-ready data

Study design and ethics readiness

Participant recruitment and community access

Bilingual interviewers

Focus groups

Transcription and translation

Analysis support

Data protection and close-out

Fieldwork run by scholars from the communities we interview

Scholars, not bilinguals

Matched to every participant

Access that rests on trust

Verbatim first, then English

Protection built for Afghan participants

Ready for analysis on day one

One team, one point of accountability

Human-led, with AI kept in its place

How a fieldwork engagement runs

Scope and ethics

Guides and team

Recruitment and consent

Interviews and focus groups

Transcription and translation

Analysis support

Handover and close-out

The fieldwork standard every study receives

Built around the rules your ethics board applies

Who runs your fieldwork

Hassan Ukasha

Tamana Ghaznawi

Zeba Haqbani

Shukria Sakhi

Diana Ayubi

Hussain Ahmad

Wasil Peroz

Maryam Safi

Field interviewers and focus group moderators

Transcription and translation linguists

Cultural Compliance Bureau

Human Intelligence Collective

AI Data Factory

Pashto, Dari and 22 more Afghan languages

Full fieldwork

Interviews and transcription

By arrangement

Fieldwork in all 50 states, Canada, Europe and Australia

Canada

Europe

Australia

Built for research teams, funders and programs

Universities and academic medical centers

Research institutes and survey organizations

Public health agencies and health systems

Foundations and philanthropy

UN agencies, NGOs and resettlement agencies

Government research and evaluation offices

Think tanks, archives and oral history programs

Market, UX and AI product research