AI Tools Improve Qualitative Analysis of Sports Officiating

According to the PLOS One evaluation (Grylls et al., 2026), the NTIS HPOP produced reliable improvements in self-efficacy and resilience for several participants, with data collected at four timepoints in February, May, August, and November 2021.

• The PLOS One study sampled five officials in 2021 and reported that three of five officials showed reliable improvements in self-efficacy and resilience by program end (published August 5, 2026).
• Official-level changes included a 37.5% increase in resilience for one participant (Jamie, RCI = 3.99) and a 67.57% increase in mentor-rated self-efficacy for another participant (Alex, RCI = 7.20), as reported in PLOS One on August 5, 2026.
• Data were collected at four fixed timepoints in February, May, August, and November 2021, enabling an idiographic longitudinal design emphasized by the PLOS One authors.

What Happened: The PLOS One study (Grylls et al., 2026) evaluated the 2021 NTIS High Performance Officiating Program using mixed-methods data from five officials collected across four timepoints in 2021.

How it was measured: The PLOS One authors combined validated psychometrics (for example the CD-RISC10 for resilience and the Referee Self-Efficacy Scale for self-efficacy), visual single-case graphs, Reliable Change Indices (RCI), and thematic analysis of open responses to assess individual-level change, as described in PLOS One (Grylls et al., 2026).

Constraints: The PLOS One report (Grylls et al., 2026) notes a small sample (n = 5), data collection disruptions due to Covid-19, and multimodal program content that limits causal attribution to specific HPOP elements.

Implication: The PLOS One study shows that mixed-methods idiographic designs can surface meaningful individual change even with small n, but they create heavy transcription, coding, and cross-case synthesis work for researchers.

Practical takeaways: The PLOS One authors used thematic analysis with an independent coder and Excel-based meaning units; the report recommends neutral interview protocols and independent data collection in future work to minimise bias (Grylls et al., 2026).

Analytical challenge: The PLOS One study (Grylls et al., 2026) reports many participant-level numerical RCIs and qualitative excerpts such as Alex’s comment, “I feel more confident in my abilities. I have been given opportunities to ref high level games with high level referees. Benchmarking against other referees has been good for me”, which require integrated coding and quick cross-referencing to draw program-level claims.

The PLOS One evaluation (Grylls et al., 2026) demonstrates how mixed-methods idiographic designs can detect reliable individual changes in psychological outcomes for sports officials, while creating a heavy analytic burden for researchers.

AI-enabled qualitative research tools can reduce that burden by automating transcription, suggesting codes, and linking quantitative indices to qualitative excerpts so teams can test hypotheses faster and produce audit-ready reports.

If you run program evaluations with small, rich cohorts and want to accelerate thematic synthesis while preserving idiographic rigor, Try Evidano for free.