Analysis Plan

Purpose

This document outlines the planned approach for processing and analysing responses to the Hi-AUDiO participant survey. The survey is designed as a companion instrument to the practice-based evaluation report and is intended to complement, refine, and contextualise the findings already presented in the repository.

The analysis is exploratory and descriptive in nature. It is not intended to validate a standardised instrument or support broad generalisation beyond the present study context.

Study Context

Researcher Positionality — Respondent Overlap

Nela Brown, who designed this questionnaire and authored the practice-based evaluation memo (README.md) against which survey findings are triangulated, is also a respondent in the dataset: she is P09 (submitted 2026-07-03). This is disclosed for transparency and is not treated as disqualifying — practice-based research expects the lead researcher to be embedded in the activity under study — but it has concrete implications for this analysis:

Analysis Goals

The survey analysis will be used to:

Quantitative Analysis

Closed questions will be analysed descriptively.

Background and Context

These items provide sample/context description rather than evaluative outcomes:

Planned outputs:

Core Evaluative Items

These items provide the main evaluative indicators:

Planned outputs:

Note: the published CSV/XLSX store these items as text (“Strongly disagree” … “Strongly agree”), not numbers — recode to 1–7 in a local working copy before computing means (see AGENTS.md Step 1b for the exact mapping). The published files intentionally keep the text labels for legibility; the recoded numeric copy is not committed.

Because the sample is expected to be small, interpretation will remain descriptive and cautious. Inferential statistics are not planned unless a later sample size clearly justifies them.

Q16/Q17 data correction (already applied to the hosted CSV/XLSX). The raw JotForm export originally split Q16 and Q17 into 4 sub-columns each, suffixed >> Service Quality, >> Cleanliness, >> Responsiveness, >> Friendliness. This text was never part of the actual form — confirmed directly in the live JotForm builder, which shows Q16 and Q17 each have exactly one matrix row with a blank row-name field (JotForm’s own “Type Row Name” placeholder hint, not real content). The 4-way split was a JotForm export defect. Before correction, all 4 sub-column values were identical for every one of the 9 respondents (e.g. P01: Q16 = “Strongly disagree” ×4), confirming no information was lost by collapsing them.

docs/participant_survey_responses.csv and .xlsx have already been corrected: each of Q16/Q17 is now a single column with the plain question title as its header, holding the one real value per respondent. No preprocessing is needed — treat Q16/Q17 exactly like any other single 7-point rating column (e.g. alongside Q18–Q20). Q16 is the headline suitability metric, Q17 the supporting artistic-intention metric.

Qualitative Analysis

Open-text responses will be analysed through thematic coding.

Open Questions

Note: Q17 sits at the boundary between quantitative and qualitative analysis. Its mean should be reported alongside Q16 and Q18–Q20, and responses should also be read in conjunction with Q21 and Q22 when interpreting artistic fit findings.

Coding Approach

Coding will begin with a hybrid approach:

Initial deductive frames:

Planned outputs:

Triangulation With the Memo

The survey findings will be compared against the practice-based evaluation memo in order to identify:

This triangulation is central to the study. The survey is not a separate standalone evaluation, but a secondary layer of evidence that helps interpret the memo findings.

Reporting Principles

When writing up results:

Survey Item Mapping

The table below maps each survey item to the corresponding section of the evaluation report and the type of metric it produces.

Survey item(s) Report section Metric
Q1 §2.1 User Profile Primary musical role — characterises evaluator perspective
Q2 §2.1 User Profile Prior browser-based tool experience — contextualises adoption findings
Q3 §7 Platform Fit Competitive awareness — qualitative context
Q4 §2.1 User Profile DAW/software literacy — contextualises technical findings
Q5 §2.2 Use Case Audio interface usage — contextualises recording quality findings
Q6 §2.1 User Profile Instruments and proficiency level — characterises performer expertise
Q7 §3 Methodology Track/instrument contribution — connects survey to broadcast deliverable
Q8 §3 Methodology Recording/upload method distribution — frequency counts
Q9, Q10 §3 Methodology Latency measurement frequency/applicability and reasons for skipping
Q11, Q12 §3 Methodology Session structure and duration distribution
Q13, Q14 §2.2 Use Case OS and browser distribution — frequency counts
Q15 §2.2 Use Case Headphone type breakdown
Q16 §7 Platform Fit Overall professional/artistic suitability — headline metric; mean
Q17 §7 Artistic Fit Artistic intention support — mean
Q18 §4.2 Ease of Use First-use usability dimensions (clarity, effort, ease, control) — mean per item
Q19 §7 Artistic Fit Hedonic/experiential quality (enjoyment, pleasantness, fun) — mean per item
Q20 §7 Platform Fit Future use intent and perceived usefulness — mean per item
Q21 §4 Platform Strengths / §7 Platform Fit Participant-reported strengths — thematic coding; complements Q22 challenge taxonomy
Q22 §5, §6 Design Implications Challenges — thematic coding against report challenge taxonomy
Q23 §7 Platform Fit Applicability scenarios — thematic coding

Matrix scales Q18–Q20: This questionnaire was developed for the present study and informed by established technology acceptance and usability evaluation approaches. Q18 adapts pragmatic usability dimensions (clarity, mental effort, ease, controllability) commonly used in usability questionnaires. Q19 captures hedonic/experiential quality (enjoyment, pleasantness, fun) using common UX evaluation dimensions. Q20 adapts perceived usefulness and behavioural intention constructs commonly associated with technology acceptance literature.

Open-text questions (Q21–Q23) are suited for thematic coding. Q22 (challenges) and Q21 (strengths) form a complementary pair — the challenge areas identified in §5 of the evaluation report (visual feedback, transport controls, editing, file format, etc.) can serve as the coding frame for Q22, while §4 (Platform Strengths) provides the frame for Q21.

Planned Deliverables