Annals of Tourism Research, Vol. 20, pp. 289-301, 1993 0160-7383193 $6.00 + .OO Printed in the USA. All rights reserved. Copyright 0 1993 Pergamon Press Ltd. CONCEPTUALIZATION OF MULTI- DESTINATION PLEASURE TRIPS Chi-Chuan Lue John L. Crompton Texas A&M University, USA Daniel R. Fesenmaier University of Illinois, USA Abstract: A four-cell typology of pleasure trips is developed based on number of destinations visited and the trip purpose. or benefits sought. Five distinctive spatial patterns evolving from this typology are identified that may be adopted by pleasure travelers: single destination, en route, base camp, regional tour, and trip chaining. Four of these involve visiting multiple destinations. Multi-destination visits are a rational behavior pat- tern and reasons that account for this behavior are discussed. It is sug- gested that the concept of cumulative attraction offers a conceptual under- pinning for explaining such trips. Benefits accruing from a better understanding of multi-destination trip behavior are discussed. Key- words: forecasting, multi-destination, tourism, pleasure trips, spatial pat- terns. R&urn& La conceptualisation des voyages d���agrCment B plusieurs desti- nations. L���article prtsente une typologie g quatre cellules pour discuter les voyages d���agrement, selon le nombre de destinations et le propos du voyage ou le genre de btnCfices que l���on cherche. A partir de cette typolo- gie, on identifie cinq modkles d���organisation spatiale: destination simple, escale en route, camp de base, voyage rCgional et enchainement de voy- ages. Pour quatre de ces modcles, il s���agit de destinations multiples. On discute les raisons pour lesquelles on entreprendrait un voyage 5 plusieurs destinations. On suggcre que le concept de l���attraction cumulative pour- rait &tre utile pour expliquer ces voyages. On discute les b&&ices que l���on peut tirer d���une meilleure ComprChension du comportement du voyage 5 destinations multiples. Mots-cl&: p&visions, destinations multiples, tourisme, voyages d���agrkment, modtles d���organisation spatiale. INTRODUCTION Substantial attention has been focused on developing destination demand models in outdoor recreation and tourism (Lieber and Fesen- maier 1984 Witt and Witt 1989). However, to this point, the models Chi-Chuan Lue recently received his Ph.D. degree from the Department of Recre- ation, Park and Tourism Sciences at Texas A&M University (College Station TX 77843-2261, USA). John L. Crompton is a professor in the same department. Daniel Fesenmaier is in the Department of Leisure Studies, University of Illinois. Their research interests include spatial and social-psychological aspects of tourism. 289

290 MULTI-DESTINATION PLEASURE TRIPS that have appeared in the literature have in large part been predicated on the assumption that when travelers leave home, they go to a single destination. In many cases, this assumption is fallacious. According to Wall, ���Recreation sites do not exist in isolation. They are found within a context of competing and complementary facilities. Thus, given the mobility of most pleasure travelers, it is likely that they will frequent more than one site or destination on a single trip��� (1978:35). Indeed, it has been reported that between 30% and 50% of all trips are multi- destination (Hanson 1980 O���Kelly 1982). The single destination as- sumption has been widely retained because it simplifies the modeling task, but its adoption is likely to substantially reduce the accuracy and efficiency of forecasting models (Hanson 1980). Early attempts to forecast tourism and outdoor recreation demand using spatial models focused on the gravity model (Cheung 1972 McAllister and Klett 1976 Van Doren 1967 Wilson 1971). Over time, the general form of the gravity model has been extended to encompass a variety of origin-destination specific influences on spatial interaction (Baxter and Ewing 1986). However, gravity models have a number of limitations (Fotheringham 1981, 1983). Among them is a lack of sensitivity toward spatial relations between the preferred desti- nation and other alternative destinations (Cesario 1973 Ewing 1980 Fotheringham 1983). Although the importance of multi-destination trips has been recog- nized in other fields (Thill and Thomas 1987 Timmermans and Golledge 1990), evidence of this recognition in the tourism and out- door recreation literature has been limited to a few pioneering studies (Baxter and Ewing 1981 Ewing and Baxter 1981 Leiper 1989 Wall 1978). Part of the reason more emphasis has not been given to model- ing multi-destination trips may be attributable to the failure of tourism organizations to collect data that provides insight into this phenome- non and that can be used to calibrate and test these models. According to Leiper, ���not much information is readily available about patterns of multi-destination tourism��� (Leiper 1989:531). He notes this problem in the context of international tourism and suggests the reason for it his comments are likely to be equally applicable in a domestic context: Each country���s tourism industry and its public sector supportive or- ganizations primarily want to know how many tourists visit that particular country, and want information about those visitors and their visit within that hosting country. Tourism interests in each country make only sideways glances at how many of their visitors travel to other countries (Leiper 1989:532). Wall (1978) suggested that single and multi-destination trips should be differentiated and should be modeled separately. In a study at Mammoth Cave National Park and at Carter Caves State Resort Park, both in Kentucky (USA), Wall identified different prevailing spatial patterns among visitors. At Mammoth Cave, a majority of visits were part of multi-destination trips, while at Carter Creek most visits were single destination trips. Another rare example of a study in which multi-destination data were collected was reported by Ewing and Baxter (1981). They asked