Why pre-screen digital humanities researchers before the interview
Computational methods will find patterns in any corpus, including one that only contains what happened to be digitised. Survival bias, uneven cataloguing and optical recognition errors all shape the result, and none of them announce themselves in a visualisation. Researchers worth hiring interrogate the corpus first. A short screen asks what their corpus left out and how they accounted for it.
What actually matters when screening Digital Humanities Researcher candidates
- 01
Theoretical command
Check command of humanities method plus computational theory: TEI P5 encoding decisions, topic modelling assumptions, distant reading critiques, FAIR and Linked Open Data vocabularies such as CIDOC-CRM.
- 02
From theory to hardware or code
Ask what they built and shipped: digital editions in Omeka or Scalar, IIIF manifests, Python or R pipelines, OCR/HTR workflows in Transkribus, cleaned datasets on GitHub or Zenodo.
- 03
Research judgement
Probe how they choose corpora and scope: handling OCR noise, sampling bias in archives, copyright limits, deciding when a research question suits computation at all.
- 04
Explaining it to non-specialists
Look for translation work: grant narratives for NEH or AHRC, briefings for librarians and archivists, teaching workshops, public exhibits built from their data.
Pre-screening questions to ask Digital Humanities Researcher candidates
12 questions grouped by what they test. Ask the same set in every screen and score answers on a consistent scale, or send them as an async video screen and compare answers side by side.
Projects and publications
3 questions01Can you describe a project where you applied computational methods to humanities research?
Listen forA humanities question driving the method, with a finding that survived scrutiny by domain experts.
Methods applied to see what emerges, or findings that restate what the field already knew.
02Have you published research that combines digital methods with humanities work?
Listen forPublications they can explain in depth, with the methodological choices defended clearly.
Outputs limited to project websites, or publications they cannot discuss methodologically.
03Have you worked on collaborative projects, and what was your role?
Listen forTheir specific contribution described, with credit given properly across a mixed team.
Contributions described at project level, or technical collaborators uncredited.
Critical with tools
4 questions04What is your experience working with digital archives and collections?
Listen forCollections used with their cataloguing and digitisation gaps understood as shaping the analysis.
Archives treated as complete, or selection bias in what was digitised never considered.
05Which programming languages have you used, and how in your research?
Listen forCode written and adapted by them, with analysis steps reproducible from what they saved.
Analysis performed entirely in graphical tools, or steps that cannot be repeated.
06Have you used text mining or language processing in your research?
Listen forAwareness of how recognition errors and historical spelling affect results, with checks applied.
Text treated as clean, or transcription errors not examined before analysis.
07Do you have experience with digital mapping in humanities research?
Listen forHistorical geography handled carefully, with changing boundaries and uncertain locations both acknowledged.
Modern coordinates applied to historical places, or locational uncertainty not represented.
Data preserved
3 questions08What is your approach to managing and preserving research data?
Listen forData documented and deposited so others can reuse it, with formats chosen for longevity.
Data held on personal machines, or no deposit plan once a project finishes.
09How do you handle the problem of digital obsolescence in your projects?
Listen forOutputs designed to survive, with data and documentation preserved separately from any interface.
Projects built as bespoke websites, or past work already offline and unrecoverable.
10How do you balance traditional humanities methods with computational approaches?
Listen forClose reading used alongside computation, with each checking the other rather than replacing it.
Computation treated as superior, or results accepted without returning to the sources.
Reaches an audience
2 questions11What strategies do you use to reach a wider audience with your work?
Listen forOutputs made accessible to non-specialists, with the limits of the findings preserved.
Public engagement described as a website, or findings simplified past accuracy.
12What role do ethical considerations play in your projects?
Listen forConsent, cultural sensitivity and rights over material all considered, especially for community collections.
Ethics limited to copyright, or community-held material digitised without permission.
How to score responses
Score every candidate on the same four criteria immediately after the screen. At this stage you are shortlisting for panel interviews, not making the final call.
Theoretical command
35%5Names specific encoding or modelling choices, cites the scholarly debate behind them, and states where quantitative method breaks down interpretively.
From theory to hardware or code
30%5Points to live URLs, repositories with commit history, and documents their own role in code, encoding, or data curation.
Research judgement
20%5Describes abandoning or reframing a method after evidence, and explains sampling gaps in the archive rather than glossing over them.
Explaining it to non-specialists
15%5Explains a network graph or model output in plain language to curators and funders, with concrete examples of funded or adopted proposals.
Any method finds patterns in a corpus that only contains what was digitised. A one-way video screen asks what was missing.
Try it on HirevireScreening FAQ
Process basics
How long should a pre-screening round for this role take?
Fifteen minutes across eight to ten questions, answered async. Enough to establish projects and outputs, test their method and tooling, and check data management practice.
How much programming should I expect?
Enough to run and adapt an analysis rather than depending entirely on a graphical tool. Someone who cannot inspect the processing steps cannot judge whether the result is meaningful.
Evaluating answers
What is the strongest signal when screening this role?
What their corpus left out. Careful researchers describe digitisation bias and cataloguing gaps unprompted. Anyone treating a collection as complete will produce confident, misleading findings.
How do I judge their preservation practice?
Ask what happens to a project after funding ends. Real answers include archived data and documented methods. Anyone whose past projects are offline has produced work nobody can build on.
Go deeper on this role
Sanat has been hiring since 2012 and watching the recruitment industry change up close ever since, and turned that screening process into Hirevire's video screening platform. LinkedIn
