fix: update consent and debrief documents for clarity and accuracy

Author: hashexplaindata

docs/CONSENT.txt

@@ -3,7 +3,7 @@ INFORMED CONSENT — UI Preference Study
 You are invited to participate in a research study evaluating user interfaces. During this session, you will be asked to view 6 pairs of digital layouts (such as course schedules, digital libraries, and fee vouchers) and select the layout you find most intuitive.
 
 WHAT YOU WILL DO:
-You will view 6 pairs of user interface designs and select the one you prefer in each pair. You will also rate your confidence in each choice. The study takes approximately 3-5 minutes.
+You will view 6 pairs of user interface designs and select the one you prefer in each pair. The study takes approximately 3-5 minutes.
 
 VOLUNTARY PARTICIPATION:
 Your participation is entirely voluntary. You may stop at any time by closing your browser. No penalty will result from withdrawal.
@@ -18,7 +18,7 @@ RISKS:
 This study involves minimal risk. You will only be viewing interface designs and clicking on your preferences.
 
 CONTACT:
-If you have questions about this study, please contact: [Muhammad Hasham, hashamnasir86@gmail.com]
+If you have questions about this study, please contact: Hasham Nasir (hashamnasir86@gmail.com)
 
 By proceeding, you confirm that you are 18 years of age or older and consent to participate.
 

docs/DEBRIEF.txt

@@ -26,8 +26,8 @@ WITHDRAWAL:
 If you wish to withdraw your data from the study, please contact the researcher below within 7 days. Provide the Participant ID shown on your completion screen.
 
 QUESTIONS OR CONCERNS:
-Contact: [Your Name]
-Email: [your.email@institution.edu]
+Contact: Hasham Nasir
+Email: hashamnasir86@gmail.com
 
 Thank you again for contributing to this research!
 

docs/IRB_CHECKLIST.md

@@ -11,12 +11,12 @@
 | 5 | Minimal risk | YES | Task involves viewing UI mockups and clicking |
 | 6 | Deception disclosed | YES | Debrief explains AI label manipulation |
 | 7 | Debrief provided | YES | Available at study completion and as DEBRIEF.txt |
-| 8 | Data security | YES | Data stored locally on participant devices |
+| 8 | Data security | YES | Data is transmitted to Google Firebase Firestore via encrypted HTTPS, with a temporary localStorage fallback in case of network timeout |
 | 9 | No vulnerable populations targeted | YES | 18+ adults only |
 | 10 | Aggregate reporting only | YES | No individual results published |
 | 11 | Contact information provided | YES | Researcher contact in consent and debrief |
 | 12 | No physical or psychological harm | YES | Viewing interface designs only |
-| 13 | Data retention policy | N/A | Local CSV files; no server storage |
+| 13 | Data retention policy | N/A | Cloud storage via Firebase Firestore. No local CSV files are retained on participant devices |
 | 14 | Compensation disclosed | N/A | No compensation offered |
 | 15 | Conflicts of interest | NONE | Academic research only |
 

docs/METHODS.md

@@ -18,42 +18,32 @@ Participants were recruited from [classroom / online panel]. Inclusion criteria:
 
 A between-subjects experimental design with two conditions:
 
-- **Control condition:** Participants viewed 8 pairs of UI mockups with neutral labels ("Option A" / "Option B") and selected their preferred design.
-- **AI Label condition:** Participants viewed the same 8 pairs, but one option in each pair displayed a "★ AI Recommended" badge.
+- **Control condition:** Participants viewed 6 pairs of UI mockups with neutral labels ("Option A" / "Option B") and selected their preferred design.
+- **AI Label condition:** Participants viewed the same 6 pairs, but one option in each pair displayed a "★ AI Recommended" badge.
 
-Assignment to conditions was randomized (50/50) using client-side JavaScript random number generation at the point of entry.
+Assignment was experimenter-controlled. Participants were recruited via WhatsApp and directed to specific conditions using URL parameters (?condition=control vs. ?condition=ai), guaranteeing an exact cohort split.
 
 ## Stimuli
 
-Eight pairs of high-fidelity UI mockups were created covering common interface patterns:
-
-1. E-commerce product card (button color: blue vs. green)
-2. Settings page (toggle alignment: left vs. right)
-3. Analytics dashboard widget (chart type: bar vs. line)
-4. Signup form (CTA copy: "Sign Up" vs. "Get Started")
-5. Pricing table (highlighted tier: Basic vs. Pro)
-6. Navigation sidebar (icon style: outlined vs. filled)
-7. Notification banner (position: top vs. inline)
-8. Search results (layout: list vs. grid)
+Participants evaluated 6 distinct, context-neutral UI domains designed to elicit aesthetic ambiguity: 1. Server CPU Load (Data Analytics), 2. Product Card (E-Commerce), 3. Notification Toggles (Settings), 4. Subscription Tiers (SaaS Pricing), 5. App Menu (Navigation), 6. 2FA Verification (Security).
 
 Each pair shared identical layout and functionality, differing only in a single design attribute. Pairs were rendered as CSS/HTML mockups within the experiment page. In the AI condition, one option per pair received a visually prominent "★ AI Recommended" badge. The assignment of which option received the badge was counterbalanced across pairs.
 
 ## Procedure
 
-1. Participants accessed the experiment via a URL that randomly redirected them to one of two conditions.
+1. Participants accessed the experiment via a condition-specific URL shared by the experimenter.
 2. A welcome screen explained the task and obtained informed consent.
-3. Participants completed 8 trials presented in randomized order.
+3. Participants completed 6 trials presented in randomized order.
 4. On each trial, two UI mockups appeared side-by-side (left/right placement randomized).
 5. Participants clicked to select their preferred design.
-6. After selection, they rated their confidence on a 5-point scale (1 = not at all confident, 5 = very confident).
-7. Upon completion, participants saw a summary and could download their responses.
+6. Upon completion of the in-app debriefing screen, data was silently transmitted to Firebase.
 
 ## Measures
 
 - **Primary DV:** Choice (A or B) — coded as whether the AI-preferred option was selected
 - **Secondary DVs:**
-  - Confidence rating (1–5 Likert scale)
   - Reaction time (ms) — time from trial display to choice click
+  - Semantic justification (post-task open-ended response)
 - **Between-subjects IV:** Condition (control vs. AI label)
 - **Metadata:** Participant UUID, trial order, pair ID, timestamp
 
@@ -79,7 +69,7 @@ Each pair shared identical layout and functionality, differing only in a single
 - A debrief statement was provided upon completion explaining the true purpose of the study and the manipulation.
 - Participants could withdraw at any time by closing their browser.
 - The study involved minimal risk (viewing interface mockups and clicking preferences).
-- Data was stored locally on participant devices (CSV download) and not transmitted to external servers.
+- Data is silently transmitted to Firebase Firestore upon completion of the in-app debriefing screen.
 
 ---
 _Copy-paste this text into your Methods section. Update bracketed placeholders as needed._

docs/README.md

@@ -2,7 +2,7 @@
 
 ## Purpose
 
-This experiment tests whether an "AI Recommended" label influences users' UI design preferences. Participants view 8 pairs of interface mockups and select their preferred option. Half see neutral labels (control); half see one option per pair marked with a "★ AI Recommended" badge (treatment).
+This experiment tests whether an "AI Recommended" label influences users' UI design preferences. Participants view 6 pairs of interface mockups and select their preferred option. Half see neutral labels (control); half see one option per pair marked with a "★ AI Recommended" badge (treatment).
 
 ## Folder Structure
 
@@ -16,7 +16,7 @@ This experiment tests whether an "AI Recommended" label influences users' UI des
 1. Open `docs/RUNBOOK.md` and follow Step 1
 2. Open `http://localhost:8000/code/index.html?condition=control` (Control)
 3. Open `http://localhost:8000/code/index.html?condition=ai` (Treatment)
-4. Data is synced to Firebase and available via CSV download.
+4. Data is silently transmitted to Firebase Firestore upon completion of the in-app debriefing screen.
 
 ## Key Files
 
@@ -32,7 +32,7 @@ This experiment tests whether an "AI Recommended" label influences users' UI des
 
 ## Contact
 
-[Your Name / Email]
+Hasham Nasir (hashamnasir86@gmail.com)
 
 ---
 _Generated: 2026-02-27 | Version: 1.0_

docs/RUNBOOK.md

@@ -18,9 +18,9 @@ The experiment is now accessible at `http://localhost:8000/code/index.html`
 ### Step 2: Verify Both Conditions
 
 - **Control Link**: `http://localhost:8000/code/index.html?condition=control`
-  - Confirm: No "AI Suggested" badges visible.
+  - Confirm: No "AI Recommended" badges visible.
 - **AI Link**: `http://localhost:8000/code/index.html?condition=ai`
-  - Confirm: "✨ AI Suggested" badges visible on designated options.
+  - Confirm: "✨ AI Recommended" badges visible on designated options.
 
 ### Step 3: Prepare Distribution Link
 
@@ -33,14 +33,12 @@ The experiment is now accessible at `http://localhost:8000/code/index.html`
 
 ### Step 4: Welcome Participants
 
-- Instruct participants to open the link in their browser
-- They will be randomly assigned to control or AI condition
+- Participants are directed to their pre-assigned condition via condition-specific URLs distributed in WhatsApp groups.
 - Estimated completion time: 3-5 minutes per participant
 
 ### Step 5: Monitor Progress
 
-- Each participant downloads their own CSV upon completion
-- Collect CSV files from participants
+- Monitor Firebase Console to confirm responses are being transmitted in real-time.
 
 ---
 
@@ -73,8 +71,6 @@ The experiment is now accessible at `http://localhost:8000/code/index.html`
 | Issue | Solution |
 |-------|----------|
 | Images not loading | Ensure all files are in the same `code/` directory |
-| CSV download fails | Check browser popup blocker settings |
-| Charts not rendering | Ensure internet connection (Chart.js loads from CDN) |
 
 ---
 _Last updated: 2026-02-27_