Pregled bibliografske jedinice broj: 820997
Assessing the reliability of percentage above cut scores in monitoring tourist satisfaction levels
Assessing the reliability of percentage above cut scores in monitoring tourist satisfaction levels // Proceedings of the ISCCRO - Internatinal Statistical Conference in Croatia / Dumičić, K ; Erjavec, N. ; Pejić-Bach, M. (ur.).
Zagreb: Hrvatsko statističko društvo, 2016. str. 214-220 (ostalo, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 820997 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Assessing the reliability of percentage above cut scores in monitoring tourist satisfaction levels
Autori
Sever, Ivan ; Marušić, Zrinka ; Čorak, Sanda
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the ISCCRO - Internatinal Statistical Conference in Croatia
/ Dumičić, K ; Erjavec, N. ; Pejić-Bach, M. - Zagreb : Hrvatsko statističko društvo, 2016, 214-220
Skup
Conference topic: "New Challenges of Official and Applied Statistics in European Union"
Mjesto i datum
Zagreb, Hrvatska, 05.05.2016. - 06.05.2016
Vrsta sudjelovanja
Ostalo
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
percentage above cut (PAC) score ; receiver operating characteristic (ROC) analysis ; tourist satisfaction
Sažetak
Statistical reporting practices commonly include percentage above cut (PAC) scores. An important issue in using PAC score is the choice of appropriate threshold that separates achievement categories. This is an arbitrary decision that could affect study outcomes, thus calling for measures to assess its reliability. This paper uses the receiver operating characteristic (ROC) method to select optimal threshold for separating tourist satisfaction from dissatisfaction in valuing performance of destination attributes in Croatian coastal counties, and to assess its reliability. The results indicate that optimal PAC threshold applied on the share of ratings of 4 and 5 on the 5-point satisfaction scale should be placed at 75%, instead of 60% that has been used in the official survey reports. The results also support the decision to lengthen the scale in 2014 survey, as the 7-point scale better discriminates between satisfaction levels.
Izvorni jezik
Engleski
Znanstvena područja
Interdisciplinarne društvene znanosti
POVEZANOST RADA
Ustanove:
Institut za turizam, Zagreb