Loizos S. Michael

Mr. Loizos S. Michael

Mesoyios College Cyprus, Limassol, Lemesós, Cyprus
orcid
PhD Can. | Political Sciences | University of Cyprus

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 September 2026 in Tourism and Hospitality
Verified
Journal Article
Open access
PDF file
Israr AhmadIsrar Ahmad
Israr Ahmad
Loizos S. MichaelLoizos S. Michael
Loizos S. Michael
SASaima Abdullah
Saima Abdullah
Guests weight negative and positive attribute performance asymmetrically, and destination and culture shape service expectations. Whether destination alters this asymmetry itself has received little direct testing. Extending stimulus–organism–response theory, we treat destination as an attribute-level boundary condition on penalty and reward effects. We analyze 12,018 English-language reviews of 75 luxury hotels in eleven destinations (22,144 Booking.com and TripAdvisor reviews in total), coded by keyword-based aspect detection and VADER (Valence Aware Dictionary and sEntiment Reasoner) sentiment scoring. Compared with two human coders, a documented re-implementation of this coding detected mentions and positive evaluations well but recovered only 57 percent of negative mentions, with no detectable destination difference in accuracy. Negative mentions carry larger rating penalties than positive mentions carry rewards, including under human coding. Conventional hotel-clustered tests show destination differences in this asymmetry for four of six attributes, but this result is fragile. A hotel-level permutation diagnostic is non-significant; multilevel models giving each hotel its own penalty and reward leave the difference detectable only for Price–Value and Amenities and Facilities; and results for individual attributes change with the keyword lists (two attributes) and with the treatment of mixed-valence mentions (three attributes). With eleven destinations, destination-level variance cannot be estimated precisely. The findings are preliminary evidence of attribute-specific, destination-organized heterogeneity, with much of it indistinguishable from hotel-level variation, not a confirmed destination effect.