Evidano home
Commentary on News

Beyond Publish or Perish: AI Qualitative Research Analysis

Evidano9 min read

Researchers, research funders, and university leaders need more than publication counts to understand why academics choose particular journals, especially when institutional incentives may shape perceived research quality. The November 18, 2026 study in PLOS One shows how interview-based qualitative analysis can uncover the context, mechanisms, and outcomes behind African researchers’ publication decisions. This article refracts those findings through AI qualitative research analysis, showing how research teams can organize interview evidence without losing participant voice, institutional context, or competing explanations. The practical payoff is a clearer workflow for analyzing transcripts, comparing themes, and tracing how funding, promotion rules, journal visibility, and prestige interact. AI can accelerate synthesis, but researchers must retain interpretive judgment, protect participant confidentiality, and validate emerging explanations with the people whose experiences produced the data.

Key Takeaways

According to the November 18, 2026 PLOS One study, “Theorising research publication practices among African university-based researchers”, publication training and funding can improve researchers’ technical skills while also intensifying pressure to publish, shifting attention from perceived quality toward publication quantity.

The study demonstrates why AI qualitative research analysis is useful for institutional research: structured coding can connect interview evidence to context, mechanisms, and outcomes, while human researchers remain responsible for interpretation, ethics, and validation.

  • The November 18, 2026 PLOS One study analyzed 19 interviews with African university-based health researchers drawn from 32 eligible participants across 3 research project teams.
  • The interviews were conducted between April 17, 2023, and December 8, 2023, and the researchers used an Indigenous realist evaluation approach with a Context-Mechanism-Outcome framework.
  • The study reported that one university’s read-and-publish agreements covered journals that were 98% Western, a condition associated with perceived credibility, trust, and incentives to publish outside Africa.
  • The study cited evidence that less than 10% of African-published journals were indexed in major international databases, limiting visibility, citations, and global reach.
  • The study concluded that universities should examine promotion policies, impact-factor use, and publishing agreements through a decolonial lens rather than treating publication counts as a complete measure of research value.

What the African Publishing Study Found

The November 18, 2026 PLOS One study found that African researchers’ publication practices were shaped by interacting individual, institutional, and publishing-system conditions rather than by training or funding alone.

The researchers interviewed principal investigators, collaborators, mentors, and supervisors from 3 African university-based health research teams. The study treated the interviews as one dataset because the goal was to build an explanatory theory of publication practices, not to compare the teams statistically.

The study’s Indigenous realist evaluation combined realist evaluation with principles including relationality, responsibility, reciprocity, reflexivity, respectful representation, and decolonization. The analysis linked context to mechanisms and outcomes using the formula C + M = O.

The researchers identified three main explanatory theories. Training and publication funding empowered researchers with technical skills but also activated perceived pressure where promotions depended on publication numbers. High-impact Western journals activated pride, prestige, and validation where impact factor was treated as a proxy for quality. Read-and-publish agreements and the low visibility of African journals activated perceived credibility, trust, and incentives to publish in Western journals.

One early-career participant described the tension directly: “It is that pressure that has driven some of us, if not all, to always leverage every collaboration or funding opportunity to ensure that we generate as many papers as possible.” The PLOS One study attributed this statement to an early-career principal investigator.

A senior researcher in the study summarized the resulting risk by saying that “the current obsession with generating many papers has inadvertently affected the quality of the outputs.” The quotation reflects a participant’s perception, not an independent assessment of the quality of the papers.

Findings Snapshot for AI Qualitative Research Analysis

Date or scopeMetricValueImplication for analysis
April 17, 2023 to December 8, 2023Interview fieldwork period19 completed interviewsA transcript-level workflow can preserve detailed participant explanations while making recurring mechanisms easier to compare.
Study sampleEligible versus participating researchers32 eligible, 19 interviewedResearchers should distinguish the eligible population from the analyzed sample when reporting qualitative evidence.
Study designResearch project teams and universities3 teams based at 3 African universitiesCross-segment analysis can compare institutional conditions without assuming that every team has identical experiences.
Study findingWestern journals in one read-and-publish agreement98%Coding can connect an institutional policy condition to perceived credibility, trust, incentives, and journal choice.
Background evidence cited by the studyAfrican-published journals indexed in major international databasesLess than 10%Frequency and document analysis can help research teams examine visibility and infrastructure themes alongside interview narratives.

Implications for Researchers and University Leaders

University leaders should use qualitative evidence to test whether publication policies reward scientific contribution, local relevance, and research integrity, or mainly reward journal prestige and output volume.

The November 18, 2026 PLOS One study shows that the same intervention can produce rival outcomes. Training and financial support can improve publication skills and reduce rejection, while the same support can intensify pressure and encourage quantity over quality when promotion depends heavily on publication counts.

Research teams should therefore code both positive and negative pathways instead of searching only for evidence that a capacity-building program worked. A useful coding framework can include context codes such as promotion rules, funding access, journal visibility, impact factor, and read-and-publish agreements; mechanism codes such as pressure, empowerment, pride, validation, trust, and perceived credibility; and outcome codes such as journal choice, submission volume, acceptance, reluctance to use African journals, and perceived changes in quality.

Researchers should also preserve disconfirming evidence. The PLOS One authors reported that their initial theory of empowerment was refuted and replaced with a more complex explanation in which empowerment and pressure operated simultaneously.

The PLOS One study included a two-hour sensemaking and validation workshop with 13 participants after the initial analysis. That step illustrates a practical safeguard for AI-assisted qualitative research: analysts can use AI to surface patterns, then return emerging themes and causal interpretations to participants or domain experts for critique.

AI-assisted analysis should not determine whether a journal, institution, or researcher is objectively high quality. The PLOS One authors explicitly noted that their study assessed participants’ conceptions and perceptions of quality rather than independently measuring the quality of individual publications.

How Evidano Helps with AI Qualitative Research Analysis

Problem: Large interview datasets slow down thematic synthesis

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interview transcripts, open-ended survey responses, papers, reports, and other documents.

Researchers can use Evidano to organize transcripts, identify recurring themes, and examine how concepts such as pressure, empowerment, prestige, trust, and credibility appear across a corpus. Thematic, content, frequency, and cross-segment analyses help teams move from scattered excerpts to a reviewable analytical structure.

Researchers can review Evidano’s outputs in context rather than treating automated themes as final findings. The workflow is especially relevant to studies that need to distinguish participant perceptions from researcher interpretations.

Problem: Context, mechanisms, and outcomes are difficult to connect

Evidano’s hierarchical codes and subcodes can represent a Context-Mechanism-Outcome framework, while co-occurrence networks can show which conditions and responses appear together in participant accounts.

For the African publishing case, a research team could code promotion requirements as context, perceived pressure as a mechanism, and publication quantity as an outcome. The team could separately code training as an intervention, technical skill as a mechanism, and fewer rejected papers as an outcome.

Researchers can learn more about Evidano’s qualitative analysis features before designing a transcript and coding workflow.

Problem: Multilingual or sensitive interview data require safeguards

Evidano supports transcription with custom dictionaries and PII redaction, which can help research teams prepare interview material for analysis while reducing avoidable exposure of personal information.

Evidano states that data are encrypted and never used to train third-party models. Research teams should still obtain informed consent, follow institutional review requirements, limit access, and confirm that AI-assisted processing matches the study protocol.

Researchers handling health-related interviews should treat AI outputs as research support rather than clinical or diagnostic conclusions.

FAQ: AI Qualitative Research Analysis

What did the 2026 study find about African researchers’ publication practices?

The November 18, 2026 PLOS One study found that publication training and funding could empower African researchers while also increasing perceived pressure to publish when promotion depended on publication counts.

The study also linked Western journal prestige, impact-factor perceptions, and institutional read-and-publish agreements to journal preferences, perceived credibility, and reluctance to publish in African journals.

How many interviews were included in the African publishing study?

The study included 19 interviews from 32 eligible researchers across 3 African university-based health research teams.

The interviews took place from April 17, 2023, through December 8, 2023, and the researchers analyzed the interviews as one dataset rather than conducting a statistical comparison across teams.

How can AI support thematic analysis of research interviews?

AI can support thematic analysis by helping researchers organize transcripts, retrieve relevant excerpts, identify recurring concepts, compare segments, and visualize relationships among codes.

Human researchers must still define the research question, assess context, evaluate alternative explanations, protect participants, and validate interpretations. The PLOS One study’s participant sensemaking workshop provides a useful model for checking emerging theories.

What is a Context-Mechanism-Outcome analysis?

A Context-Mechanism-Outcome analysis explains how pre-existing conditions interact with responses or processes to produce particular outcomes.

In the PLOS One study, promotion requirements and publish-or-perish culture formed part of the context, perceived pressure was a mechanism, and a shift toward publication quantity was an outcome. The framework helped the authors explain why the same training intervention produced both empowerment and pressure.

Can AI determine whether a journal or paper is high quality?

AI cannot determine research quality from journal prestige or impact factor alone.

The November 18, 2026 PLOS One study cautioned that the impact factor is not a measure of the quality of an individual article and that its own analysis captured participants’ perceptions rather than independently assessing publication quality.

What safeguards should researchers use with AI-assisted qualitative analysis?

Researchers should use informed consent, PII redaction, secure data handling, human review, transparent coding decisions, and participant or stakeholder validation when appropriate.

Research teams should also document where AI contributed to transcription, coding, retrieval, or synthesis and should not allow automated outputs to replace methodological judgment.

Conclusion and Next Steps for AI Qualitative Research Analysis

The November 18, 2026 PLOS One study shows that African researchers’ publication choices emerge from connected pressures, incentives, perceptions, and institutional structures, not from skills or funding alone.

AI qualitative research analysis can make those connections easier to examine across interview transcripts, but responsible research still depends on participant voice, reflexivity, transparent coding, and human validation.

Evidano helps research teams move from transcripts to thematic, frequency, content, and cross-segment analysis while keeping the analytical workflow reviewable. Start your next qualitative project with Try Evidano for free.

Topics

  • AI qualitative research analysis
  • qualitative analysis of research interviews
  • AI thematic analysis
  • African academic publishing
  • publish or perish research

Keep reading

Browse all articles