Students perceived text input as superior for accuracy, control, and editability, with transcription errors a key voice drawback¶

Claim · All claims
Evidence · 3 studies · 2 qualitative r2, 1 associational r2 · q2
Currency · no dated study recorded · not yet checked against its sources · not reviewed by a person

Subclaims¶

Reading the evidence codes

q is the design tier, i the impact, n the sample, and the kind with its r the rigour judged within that kind: what each code means.

This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.

q2 i? The dominant qualitative category was input accuracy and control, heavily favouring text, with transcription errors widely cited. → accuracy-category q2 i? Editability of text prompts before submission was a major reported advantage; voice-based revision was perceived strictly negatively. → editability-category q2 i? Seventeen comment entries explicitly cited non-native English speaker status as a reason for preferring text input, 15 concerning accent recognition. → nnes

Evidence¶

accuracy-category¶

Kaitlin Riegel, Yan Cathy Hua, Paul Denny, Victor-Alexandru Pădurean, Juho Leinonen, James Prather, and Adish Singla. 2026. Say What? Examining Text and Voice Input Modalities for Prompt-Based Programming in Computing Education. Proceedings of the 31st ACM Conference on Innovation and Technology in Computer Science Education V. 1 (ITiCSE 2026). https://doi.org/10.1145/3803400.3809397

q2 · i? · associational · r2

Aspect-based sentiment analysis of open-ended reflection responses (890 comment pieces; inter-rater F1 = 0.65 on 156 independently annotated entries) found "the ability to input prompts with control and accuracy" was the dominant category, heavily favouring text; Table 2 shows 53 positive vs 2 negative text sentiments and 6 positive vs 51 negative voice sentiments in this category.

"The dominant category that emerged was the ability to input prompts with control and accuracy, heavily in favour of text-based prompting."

editability-category¶

q2 · i? · qualitative · r2

The ABSA analysis of reflection responses identified editability as a major advantage of text; the article reports that "voice-based revision was perceived strictly negatively" (Table 2: 47 positive vs 2 negative for text; 0 positive vs 12 negative for voice in this category).

"The ability to edit and review text prompts before submitting was a major reported advantage."

nnes¶

q2 · i? · qualitative · r2

A post-hoc analysis of all 890 comment pieces using keyword search, LLM categorisation, and human consolidation identified 17 entries that "explicitly mentioned NNES status as a reason for preferring text input"; 15 expressed concerns about their accent being accurately recognised.

"Using a combination of keyword search, LLM categorisation (pre-trained Qwen3.5-4B in non-thinking mode [38]), and human consolidation, we identified 17 entries that explic- itly mentioned NNES status as a reason for preferring text input."

Discussion¶

Educators report ASR features can be inaccurate for learners' diverse accents, dialects, and speech patterns — the general claim this finding bears on, with the studies behind it
Edited voice prompts did not significantly differ from text prompts in first-attempt success — related
Students reported mixed, task-dependent perceptions of the efficiency and speed of voice versus text input — related
Greater proportion of voice prompts used correlates moderately with greater stated preference for voice — related
Unedited voice prompts have lower odds of first-attempt success than text prompts on two of three Prompt Problems — related

Page metadata

Field	Value
`type`	`claim`
`id`	`text-perceived-accuracy-control-editability`
`title`	`Students perceived text input as superior for accuracy, control, and editability, with transcription errors a key voice drawback`
`description`	`Students perceived text input as superior for accuracy, control, and editability, with transcription errors a key voice drawback`
`status`	`review`
`evidence_strength`	`moderate`
`generated`	by `process:wiki-ingest` on `2026-10-12`
`trust`	unverified — no one has confirmed this page's content
`sources`	3 — `accuracy-category`, `editability-category`, `nnes`

Frontmatter is the page's machine-readable record — see the Schema & Guide.