Every number, in one place
All of it on one page so you can check yourself. Nothing here is rounded or guessed.
Do not try to memorise all of it. Learn the twelve numbers in the first block. Everything else is here so you can check yourself, and so you know the figure exists if somebody asks for it.
Every number on this page came out of an analysis script or a test script run against real data. None of it is rounded for effect or estimated.
The twelve numbers to actually know
Reliability, before and after purification
| Construct | Sample | Initial alpha | Item removed | Final alpha |
|---|---|---|---|---|
| IV1 Real-time visibility | Tourist | 0.502 | Q7, at −0.304 | 0.789 |
| IV1 Real-time visibility | Organizer | 0.425 | Q7, at −0.338 | 0.691 |
| IV2 Pre-booking | Tourist | 0.802 | none | 0.802 |
| IV2 Pre-booking | Organizer | 0.665 | none | 0.665 |
| IV3 Centralized information | Tourist | 0.868 | none | 0.868 |
| IV3 Centralized information | Organizer | 0.758 | none | 0.758 |
| Outcome, E0 | Tourist | 0.540 | Q20, at −0.622 | 0.886 |
| Outcome, E0 | Organizer | 0.159 | Q20, at −0.623 | 0.690 |
Table 4.3. Only the two reverse-coded items ever crossed the 0.30 removal threshold, in either sample.
Means, correlations and betas together
| Construct | Tourist mean (SD) | Organizer mean (SD) | r with outcome | beta |
|---|---|---|---|---|
| IV1 Real-time visibility | 3.676 (0.721) | 3.456 (0.658) | .571 / .580 | .096 / .234 |
| IV2 Pre-booking | 3.672 (0.744) | 3.519 (0.697) | .677 / .633 | .118 / .192 |
| IV3 Centralized information | 3.812 (0.784) | 3.788 (0.706) | .809 / .683 | .665 / .461 |
| E0 Coordination Effectiveness | 3.945 (0.759) | 3.719 (0.677) | — | — |
In the paired columns, the first figure is the tourist sample and the second is the organiser sample.
The two regression models
Tourist model
n = 64
R² = 0.673
adjusted R² = 0.657
F(3, 60) = 41.181
p < 0.001
Durbin-Watson = 2.165
VIF: 2.058 / 2.959 / 2.236
Organizer model
n = 40
R² = 0.572
adjusted R² = 0.536
F(3, 36) = 16.038
p < 0.001
Durbin-Watson = 1.617
VIF: 1.996 / 2.462 / 1.562
All VIF values are below the conservative threshold of 5 and all tolerances exceed 0.10, so multicollinearity is within acceptable limits. Both Durbin-Watson statistics sit inside the acceptable 1.5 to 2.5 range, so independence of errors holds.
The point is that these were checked before interpreting anything, not afterwards to justify a result.
Who answered
The 40 organisers
Role. 45.0 percent event coordinator or events manager. 27.5 percent owner or general manager. 17.5 percent front desk or guest relations. 10.0 percent food and beverage.
Property. 35.0 percent boutique or guesthouse under 20 rooms. 30.0 percent mid-size, 20 to 60 rooms. 27.5 percent beach or holiday villa. 7.5 percent resort of 60 rooms or more. So 62.5 percent below mid-size.
Event frequency. 47.5 percent a couple of times a month, 30.0 percent weekly, 12.5 percent several times a week, 10.0 percent rarely. So 90.0 percent host regularly.
Staffing. 42.5 percent have dedicated event staff. 32.5 percent handle it alongside other duties. 22.5 percent outsource. 2.5 percent have nobody. So 55.0 percent have no dedicated event staff.
The 64 travellers
Age. 60.9 percent aged 25 to 34. 39.1 percent aged 18 to 24. Nobody outside that range.
Origin. 96.9 percent domestic Sri Lankan travellers. 3.1 percent international, which is 2 people. This is the biggest limitation in the study.
Visits. 46.9 percent had been 2 to 3 times including this trip. 25.0 percent more than 6 times. 17.2 percent first time. 10.9 percent 4 to 6 times.
Trip length. 40.6 percent 2 to 3 days. 39.1 percent a day trip. 12.5 percent more than a week. 7.8 percent 4 to 7 days. So 79.7 percent are there 3 days or fewer, which compresses the window in which discovery has to happen.
The themes
| Code | Theme | Group and question | Share |
|---|---|---|---|
| T1 | Information incomplete, unclear or outdated | Traveller, what went wrong | 47.2% |
| T2 | Forced to contact the property directly or ask locally | Traveller, what went wrong | 33.3% |
| T3 | Missed the event, or sold out after the delay | Traveller, what went wrong | 16.7% |
| T6 | Booking or payment mechanism failed | Traveller, what went wrong | 13.9% |
| T4 | Resource exhausted on arrival | Traveller, what went wrong | 8.3% |
| T5 | Conflicting information between channels | Traveller, what went wrong | 8.3% |
| T7 | One centralized platform as a single source | Traveller, what would you change | 52.5% |
| T9 | Complete event detail in one view | Traveller, what would you change | 50.0% |
| T10 | Direct, secure or instant booking | Traveller, what would you change | 35.0% |
| T8 | Live availability and capacity visibility | Traveller, what would you change | 27.5% |
| T11 | Mobile application form factor | Traveller, what would you change | 25.0% |
| O1 | Last-minute changes and coordination gaps | Organiser, what went wrong | 46.4% |
| O5 | Schedule slippage and delays | Organiser, what went wrong | 28.6% |
| O6 | Technical or preparation failure | Organiser, what went wrong | 21.4% |
| O3 | Turnout mismatch causing waste or cost overrun | Organiser, what went wrong | 21.4% |
| O4 | Promotion and reminder failure | Organiser, what went wrong | 17.9% |
| O2 | Event information scattered across channels | Organiser, what went wrong | 14.3% |
| O8 | Reliability and stability under pressure | Organiser, adoption condition | 66.7% |
| O7 | Ease of use | Organiser, adoption condition | 53.3% |
| O9 | Security and data protection | Organiser, adoption condition | 46.7% |
| O11 | Proof from other operators first | Organiser, adoption condition | 43.3% |
| O10 | Cost and pricing transparency | Organiser, adoption condition | 40.0% |
Percentages are of substantive responses, not of all respondents. Blank answers, answers under 15 characters and non-answers were excluded and reported separately. Substantive counts: traveller question 24 had 36, question 25 had 40. Organiser question 24 had 28, question 25 had 30. One answer can carry more than one theme.
The highest-scoring individual items
| Group | Item | What it said | Mean |
|---|---|---|---|
| Traveller | 22 | Would recommend an app like this to other travellers | 3.984 |
| Traveller | 23 | Trustworthiness and ease matter more than feature count | 3.969 |
| Traveller | 21 | Even a basic version with live listings and booking would ease trip planning | 3.938 |
| Organiser | 17 | A centralized view would improve decisions made during an event, not only while planning | 3.900 |
| Organiser | 22 | Would recommend the platform to other properties in the area | 3.900 |
| Organiser | 18 | Answering a guest question about inclusions currently means checking with someone else first | 3.875 |
| Organiser | 21 | Would try it for at least one upcoming event | 3.850 |
| Organiser | 16 | A single dashboard of bookings, stock and listings would save real time weekly | 3.800 |
Both describe an information failure happening in real time during an event, not a promotional failure happening before it.
That is what pointed the requirements towards a live operational dashboard rather than a marketing listing. If someone asks how you decided the organiser needed a live view, these two items are the answer.
Testing and measurement
| What | Result |
|---|---|
| Functional test cases | 10 of 10 passed |
| TC-05, concurrency | 20 requests on 10 units: 10 committed, 10 refused, remaining landed on 0, peak 20 in flight |
| TC-09, access control | 9 of 9 checks passed, issued directly against the API |
| NFR-01 browse page load | 501 ms p95, 400 ms p50, n = 30, target under 2000 ms |
| NFR-01 event detail load | 336 ms p95, 252 ms p50, n = 30, target under 2000 ms |
| NFR-02 live update latency | 580 ms p95, 265 ms p50, n = 11, target under 2000 ms |
| NFR-03 concurrent load | 40 concurrent requests, 40 succeeded, 0 failed |
| NFR-07 responsive layout | 39 screens clean at 375, 768 and 1440 pixels |
| NFR-06 usability | Not measured. Specified but unverified |
| Defects found and fixed during testing | 7 |
One of twelve latency samples never arrived inside the observation window and is excluded, leaving n = 11. This is stated in the thesis rather than dropped silently.
The artefact
| Thing | Figure | Note |
|---|---|---|
| Database tables | 6 | Deliberately small |
| Functional requirements | 14 | 4 marked critical: FR-04, FR-05, FR-08, FR-11 |
| Non-functional requirements | 9 | Derived from stated adoption conditions, not a generic checklist |
| Use cases | 11 | 6 specified in full in Chapter 4, 5 in Appendix C |
| Screens | 12 | All verified at 3 widths |
| Seeded properties | 6 | Across all 4 property types |
| Seeded events | 14 | 12 of them published |
| Seeded resource types | 42 | Partial availability, so screenshots look realistic |
| Seeded bookings | 288 | No real person's name, phone or email anywhere |
The thesis itself
| Part | Pages |
|---|---|
| Chapter 1, Introduction | 12 to 19 |
| Chapter 2, Literature Review | 20 to 33 |
| Chapter 3, Research Methodology | 34 to 45 |
| Chapter 4, System Requirement Specification | 46 to 86 |
| Chapter 5, Implementation and Designing | 87 to 103 |
| Chapter 6, Testing and Evaluation | 104 to 114 |
| Chapter 7, Concluding Remarks | 115 to 121 |
| References | 122 to 126 |
| Appendices A to E | 127 to 165 |
104 people surveyed, 64 travellers and 40 organisers.
0.665 and 0.461. The IV3 beta in each sample, the only significant predictor in either.
10 and 10. Twenty simultaneous booking requests against ten units, splitting exactly, with nothing oversold.
47.2 percent. The share of travellers whose problem was information that was incomplete, unclear or out of date, which is what makes this a problem of integrity rather than volume.