retaining the visitor, enhancing the experience: identifying attributes of choice in repeat museum...

International Journal of Nonprofit and Voluntary Sector MarketingInt. J. Nonprofit Volunt. Sect. Mark. 14: 21–34 (2009)Published online 2 December 2008 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/nvsm.351
Retaining the visitor, enhancing theexperience: identifying attributes ofchoice in repeat museum visitationChristine Burton*, Jordan Louviere and Louise Young

� H

*CorFacuPO BE-ma

Cop

ow people make choices in relation to cultural and leisure consumption has been

explored from the viewpoint of motivation, lifestyle segmentation, and lifecycle. Little is

known about the specific characteristics associated with choices to visit, re-visit, or not to

visit a museum. Understanding characteristics of choice, developing incentives, bundled

packages, and levels of pricing is an essential element in marketing strategies for

museums operating in a competitive leisure marketplace. However, determining what

reallymatters to cultural consumers is complex andmethodologies to assist in unraveling

such complexities are not easily identified. This study aimed to address ways in which

people respond to specific incentives as influences in choosing museum visitation. The

study was conducted in two major museums in Australia to determine how useful choice

modeling is in identifying features that matter to cultural consumers. The results suggest

that choice modeling has much to offer in relation to understanding the benefits people

are seeking from a museum experience as well as offering strategic insight into potential

collaborative ventures and re-combinations of existing museum products and services.

Copyright # 2008 John Wiley & Sons, Ltd.

Introduction

The research grew from an inquiry to deter-mine factors associated with museum visitordecline over a 10-year period in Australia. Thedecline in cultural consumption was wide-spread but particularly dramatic in relation tomuseum attendance where visitation fell from27% in 1991 to 19% in 1999 (Australian Bureauof Statistics, 1999). While this downward trendhas been slightly reversed, museum visitationhas tended to remain about 22% in Australia(Australian Bureau of Statistics, 2007).

respondence to: Christine Burton, Senior Lecturer,lty of Business, University of Technology, Sydneyox 222, Lindfield 2070 Australia.il: [email protected]

yright # 2008 John Wiley & Sons, Ltd. Int.

Visitor decline was not merely associatedwith decreased frequency of attendance butwas a real decline in visitor numbers. Visitornumber uncertainty is a serious challenge formuseums operating in a competitive leisureenvironment. The history ofmuseums suggeststhat they are slow to change and their brandsignifies that their appeal is for those in searchof high cultural consumption and education(Hood, 1992; Conforti, 1995; Hein, 2000;Hooper-Greenhill and Dodd, 2002). Whilemuseum management has attempted tobroaden their appeal through a re-focus onvisitor needs, (Crimp, 1985; Vergo, 1989b;Clifford, 1997; Griffin and Abraham, 1999;Griffin and Abraham, 2000; Rentschler, 2002)the core business of museums still resides in

J. Nonprofit Volunt. Sect. Mark., February 2009

DOI: 10.1002/nvsm

22 Christine Burton et al.

collection development, interpretation, anddisplay, however diversified this may be (Weil,2002).This paper presents the findings of a discrete

choice modeling project in two major Aus-tralian museums. The research contextsuggests that museums are operating withina leisure environment (as well as an edu-cational and cultural environment) but thatthey do not clearly understand the strategicimplications of this, particularly where visitornumbers appear to be in decline.The paper presents first an overview of

the theoretical underpinnings of the leisureenvironment in which museums operate,alongside the compounding factors of workpressure and time squeeze of traditionalmuseums visitors. It suggests that museumsare operating within a highly competitiveleisure environment but are poorly equippedto respond. It then backgrounds the reasonsfor applying choicemodeling as an appropriatemethodology in exploring how current visitorscould be made more frequent consumers ofmuseum products and services. Finally itoutlines the findings from the research andsuggests that choice modeling is an appro-priate methodology to use in understandingpreferred benefits for the museum consumerand proposes further extensive testing.

Museums within the leisureenvironment

Given that the frequent museum visitor ischaracterized by higher socio-economic statusand high educational attainment (Griffin, 1978;Bourdieu, 1984; Bennett, 1994; AustralianBureau of Statistics, 1999), it is difficult toexplain museum visitation decline in an eramarked by unprecedented access to highereducation and increasing standards of living.However the competition for visitors’ time

and money has also increased and is com-pounded by the growth of new commoditizedleisure products and services which offersimilar types of benefits to museums but

Copyright # 2008 John Wiley & Sons, Ltd. Int.

which may also be fast-tracked, more acces-sible, and less engaged (Oliver et al., 2002;Burton, 2003).

Museums have had an ambivalent attitude inplacing themselves within the field of leisure.They have variously positioned themselves aspart of the formal and informal education andlearning process, part of scholarship, partlycultural custodians preserving the ‘‘best’’ partof civilized society to elevate citizens (Hein,2000). While this last attitude has goneunderground somewhat, it is still not far fromwhat most professional museum workerswould see as their mission and is reflected intheir professional bodies’ missions (Inter-national Council Of Museums, 2001; Hooper-Greenhill and Dodd, 2002; Museums Australia,2002).

As a consequence, museums have resistedcompeting for leisure time and money on thebasis that what they do is not entertainment,activity, or restful but is cultural and ‘‘improv-ing’’—it is separate from leisure andmay resultin marketing myopia (Levitt, 1991). In a sensethis is congruent with what many people alsobelieve. It would appear that in choosingleisure time activities, many people do noteven consider a visit to a museum as a viablealternative to another activity—it often doesnot even register on the leisure landscape(Lynch et al., 2000).

Leisure itself is a complex concept coveringaspects of time, activity, attitudes, and isdifficult to define as one single notion (Vealand Lynch, 2001, pp. 18–23). In a reductionistsense however, a number of elements aboutleisure emerge which are relevant to adiscussion of arts, culture, and museums andhave an impact on how people make choices.These elements revolve around what Stebbinshas defined as ‘‘serious leisure’’ and whatRojek has defined as ‘‘fast leisure’’ or post-modern leisure (Rojek, 1995; Stebbins, 1999;Rojek, 2000).

The paradigm associatedwith serious leisuresuggests that people make choices aboutleisure activities based on the opportunitiesthose leisure activities present in relation tocareer advancement or life chances. This is


DOI: 10.1002/nvsm

Table 1. One scenario of 8

Scenario Option A Option B

Opening hoursSummer hours 9.30 am–7pm 9.30 am–6 pmWinter hours 10 am–6 pm 10 am–5pmSpecial opening for school holidaysand popular exhibitions

Tuesdays till 8 pm 9.30 am–5 pm

Packages with attractionsNational museumþ State museum $40 per person $40 per personNational museumþ IMAX No joint ticket $45 per personNational museumþ Sydney aquarium $50 per person $40 per personHarbor cruiseþmuseum entry No package available $75 per personMusicþ dinner at the museum $70 per person $70 per person

Packages with transportNational museumþmonorail $25 per person $35 per personNational museumþ light rail $25 per person $30 per personNational museumþ ferry $30 per person $30 per personNational museumþ 2 hours secure parking $30 per person No joint ticket

IncentivesThe more visits you make each year,the less you pay

�10% 2nd visit, �25% thereafter 30% off all subsequent visits

Bring a friend—bring your previousticket—your friend pays

You get a 25% discount You get 50% discount

Become a member the more youvisit—receive free membership after

Two visits per year Four visits in 1 year

Be a multi-museum member—visitany Sydney museum as often as youlike per year

$250 per person $200 per person

Express lane for members No express lane No express laneMuseum entryþ special incentives 1 free coffeeþ 10% discount

on meals and gifts1 free coffeeþ 10% discount

on meals and giftsStandard entry feeGeneral entry fee for museum Single adult $20, children/

concession $10, Family $45Single adult $10, children/concession $6, Family $25

Can you leave and re-enter on thesame pass?

No re-entry allowed Yes, in same week

1. The museum can only offer oneoption at a time, which do you prefer?

Option A Option B

Neither option2. Suppose the museum could onlyoffer Option A, would you be morelikely to

Visit more often Visit less often

Visit as you do now

Attributes of choice in repeat museum visitation 23

consonant with public policy developmentsthat place museums within a social inclusionagenda (Evans, 2001; Casey, 2002). However,Stebbins (1999) understands the pursuit ofserious leisure may have become marginal in asociety that condones and encourages casualleisure pursuits and may not be sustainable tothose traditional (and potentially high yield)visitors who are seeking a number of entertain-ment and educational benefits.


Rojek (1995) theorizes that contemporaryleisure is marked by the post-modern conditionor what he calls the struggle betweenModernity 1 and Modernity 2. This manifestsitself in distraction, fragmentation, lack ofcommitment, contingency, and fast leisurepursuits in the form of digital technologies andgames. In this scenario, the emotionality of lifesupplants or is seen as equally important asrational aspects of life (Rojek, 1995).


DOI: 10.1002/nvsm

Table 2. Significant choices for the state museum

Attribute Feature Estimatedutility

T-stat

Summer hours 9 am–5pm during Xmas holiday �0.145 �1.8239 am–6pm throughout January 0.171 2.138

Family evenings Pizza, soft drinkþ special exhibit 5.30 pm–8.30pm,adults $30; children $15

0.145 1.813

Pizza, soft drinkþ special exhibitþ special kidsactivities 5.30 pm–8.30 pm (two adultsþ two kids) $80

�0.146 �1.825

Chat with curator Talkþwineþ special exhibitionþ exhibition catalog6.30 pm–8.30 pm $55

�0.176 �2.200

After school programs No special programs �0.304 �3.800Day ticket on monorail No joint ticket �0.129 �1.613

Monorailþmuseum entryþ entry to special payingexhibitions: adult $22, child $12, family $59

0.186 2.325

Combined entry to IMAX Adult $23, child $12, family $60 (two adults andtwo children)

0.129 1.613

Joint museum pass All January adult $75, child $45, family $172(two adults and two children)

0.161 2.013

No joint pass �0.299 �3.738Guided walking tour Walking tourþmuseum general admission adult $35,

child $20, family $80 (two adults and two children)0.200 2.500

No package available �0.220 �2.750Re-entry on same pass Keep admission ticket and return within 3 months for free 0.191 2.388

Keep admission ticket and return within 6 months for free 0.168 2.100Single entry per admission �0.242 �3.025


In terms of these two positions, museumsmay well be caught in a ‘‘serious leisure’’paradigm offering traditional modes of increas-ing both cultural and social capital that isimportant for public policy social inclusionagendas, but may be less palatable for a post-modern, highly educated potential visitor,seeking fast leisure (Stebbins, 1999; Gleick,2000; Rojek, 2000). The contemporary highlyeducated visitor may also be less likely to seekonly high cultural experiences, but rather ascultural omnivores, seek converged and serialcultural experiences offered by cinema packages,home-based digital entertainment and restaurantgoing experiences (Bennett et al., 1999).

Work pressure, time squeeze, and

leisure choice

Compounding these theoretical positions ofnotions of serious and fast leisure is thereporting of contemporary time squeeze andwork pressures. It appears that demographic


trends and work pressures on professionalsand managers—traditional museum visitors—may exacerbate the decline in museum visita-tion (Bittman and Rice, 1999a; Bittman, 1999b;Yann et al., 1999; Lynch et al., 2000; AustralianBureau of Statistics, 2000a; Australian Bureauof Statistics, 2000b; Oliver et al., 2002). Whilethese reports indicate that work pressure andthe pace of work are creating an imbalancebetween work and family life, some unex-plained anomalies also arise in relation toleisure choices.

The effect of work pressure often resultsin people seeking out superficial and non-committed leisure engagement (Lynch et al.,2000). However, this is also characterized byeither a planned investment in time thatcombines serial leisure activities over a con-centrated time period or alternatively spon-taneously arranged leisure engagement withfriends via SMS or mobile phone. Thecommonality appears to be that a number ofserial activities occur to maximize the timeinvestment (Lynch et al., 2000).


DOI: 10.1002/nvsm


This kind of change in leisure patterns inrelation to work pressure and time availabilityshould mean, predictably, that museum visita-tion would suffer as a result of both theintellectual demands and the time commit-ment to this one activity. However, otheractivities which also demand concentrationand time commitment are increasingly popu-lar, for example, cinema-going (Australian FilmCommission, 2002). One explanation of theanomaly of why museums may be in declinewhile cinema visitation is increasing, lies inthe way cinemas have restructured in the past10 years through minimizing risk associatedwith time and cost by offering a number ofpackages and flexible ways to pay for film-going (Scott, 1999; Burton, 2003).While it is improbable that museums can

competewith commercial activities of cinemasparticularly in relation to distribution of theirproduct and intense marketing through hori-zontal and vertical integration, the example ofbundling and packaging products and servicespresented an interesting case for furtherexploration. Understanding risk minimizationand flexible packaging and bundling of aleisure activity became an abiding concern inhow people may make choices in a harriedleisure and work environment and how thiscould be applied to the museum sector andtheir visitors (Oliver et al., 2002).

Challenges for museums as leisure

providers

Howmuseums respond to challenges imposedby changes in leisure consumption maydepend not only on public and privateresource priorities but on developing a senseof entrepreneurship which seeks out partner-ships with competitor leisure providers andunderstanding the cultural consumption pat-terns and preferences of their visitors (Burton,2003). Previous research suggests thatmuseums are operating in a highly competitiveleisure environment, but it was still unclearhow people actually made choices aboutleisure consumption and about museum


visitation within this context. Marketingresearch tended to focus on motivation andlifecycle elements of the visitor (Todd andLawson, 2001; Hume et al., 2007; Slater, 2007).Added to this is that over the past decade, oneof the most compelling public policy devel-opments in relation to arts and culture hasconcentrated on demographically broadeningthe visitor base and increasing access tocultural provision (Roberston and Migliorno,1996; Prentice et al., 1997; Sandell, 1998;Saatchi, 2000; Evans, 2001; Museums Australia,2002; Parker et al., 2002). At the same timethat this public policy initiative has beendeveloped there has been a noticeabledecrease in the traditional visitor base to mostsubsidized cultural attractions particularly inAustralia, but to some extent mirrored in theUnited Kingdom (Australian Bureau of Stat-istics, 1999; MORI, 2001) prior to entry feeelimination. Fighting a battle to convert non-goers to attend cultural attractions whileexperiencing a decline in the traditionalcustomer base may result in a scattergunapproach to marketing and trying to be allthings to all people—a course of action thathas been identified as ‘‘stuck in the middle’’strategic suicide (Porter, 1985; Porter, 1996;Kotler and Kotler, 2000).Of the two marketing issues, arresting

decline seemed more urgent than attemptingto deploy scarce resources on a costly pursuitof non-visitors. The research therefore posedthe question of how to make those who areaware of your product and services morefrequent consumers by offering a range ofbenefits from which visitors could choose.Research conducted by Hume et al. (2007)

on repurchase intention in the performing artssuggests that perceived value and quality ofservice is important to the consumer alongsidethe cultural product itself. While those familiarwith the art product may be less inclined toseek high contact with the organization, in amuseum setting a more high contact agendamay be important, especially for a less frequentvisitor. This is especially so, because it is notalways evident to the visitor what they shouldor should not be seeing or experiencing in the


DOI: 10.1002/nvsm


museum itself (Falk and Dierking, 1992; Falk,2006). Regardless of high or low contact, asHume et al. (2007, p. 137) suggest, the ‘‘entireexperience and service provided’’ is whatmatters in repurchase intention investigation.Their research identified a number of factorsaround satisfaction, value, emotion, involve-ment, and appropriate supporting service aspart of the mix of repurchase intention. Using aqualitative approach, the researchers wereable to identify areas that matter generally toconsumers. They identified that there may bea disconnection between cultural managers’emphasis on the quality of the product whilevisitors are seeking ‘‘functional quality’’around the services they receive within thevenue (Hume et al., 2007, p. 146). However,their research does not drill down to thespecific factors of service and amenity that aresalient in the repurchase intention of thevisitor.Rentschler et al. (2007) survey of pricing

strategies in museummarketing is also relevantin this context. They suggest that museumshave been unsophisticated in understandingpricing strategies relative to their customersegment needs. Rather than adopting anentrepreneurial approach to pricing whereperceived value and customer utility arematched, museum marketers tend to adhereto a normative pricing policy with access andequity foremost in their minds.Taking the nascent research of Hume et al.

(2007) on repurchase intent based on wholeservices provision and Rentschler et al. (2007)on pricing strategies, we evaluated twomethodologies that could further elicit whatproducts and services matter to segmentedvisitors and howmuch theywould be preparedto pay for these services.

An evaluation of contingentvaluation and choice modelingapproaches

Contingent Valuation Methodology (CVM) orWillingness to Pay (WTP) has been usedextensively in the arts and was pioneered by


Throsby and Withers (Throsby and Withers,1979, 1984).

Research conducted by Throsby and With-ers was predicated on the concept of the arts asan example of market failure. The price peoplewere willing to pay to maintain or grow acultural product or service was seen as ameasure of value placed on cultural goodsthemselves. Throsby has identified some of thedrawbacks to this form of methodology(Throsby, 2003). These limitations primarilyrevolve around the level of information peoplemay or may not have in order to makejudgments or decisions; whether art andculture has intrinsic value; whether the valuecannot be determined by individuals person-ally but may be regarded as having externalvalue to a third party, collectively or individu-ally and over time; and whether there is animpossibility of economically valuing artbecause the products and services can beinherently unstable, lack value consensus, andare difficult to measure quantitatively andqualitatively (Throsby, 2003, pp. 278–280).

While Throsby (2003) and others (Thompsonet al, 2002; Noonan, 2003) have been primarilyengaged in the ‘‘big picture’’ of culturalprovision, still others have looked at particularmicro-applications of CVM and WTP includingcovering the costs associated with cleaningcathedrals, maintaining specific museum pro-vision, and determining entry fee to culturalvenues (Santagata and Signorello, 2000; Polli-cino and Maddison, 2001; Willis, 2002).

It has been claimed that CVM can go someway to making opportunity costs transparentand give a more rounded picture of thedemand for cultural provision and the intan-gible benefits that arts and cultural consump-tion can provide (Noonan, 2003). Some of theidentified down-sides of the application ofCVM have been associated with less thanrigorous application.

Unraveling complex factors, demographics,and psychographics associated with howpeople make choices through choice modelinghave been the subject of research in areasas diverse as ‘‘econometrics, transportation,marketing, decision science, and biostatics’’


DOI: 10.1002/nvsm


(Louviere et al., 2000, p. 1). While these areasseemingly have little in common, the search isfor a better understanding of how peopleactually make choices and to predict behavioron the basis of those stated choices—that is toidentify the preferred product or service, orcombination of products and services thatpeople state they will consume over others(McFadden and Train, 2002).The underpinnings of choice modeling rest

not so much with understanding the con-sumption of things but rather with under-standing what are the properties of things thatresult in additional utility for a consumer.While this approach has been used todetermine how consumers will react tochanges in price structures for goods andservices or changes to the products them-selves, choice modeling suggests that pro-ducts/services can be described in terms oftheir ‘‘characteristics’’ and further suggeststhat the ‘‘attractiveness’’ of these character-istics singly or in multiple combinations can betested through an application of random utilitytheory. The confounding factor in choice is theconsumer’s perception of personal utility, notwhether ‘‘objectively’’ (or from the supplier’spoint of view) utility has been created by re-combinations of services or products.In this way, paradigms of choice can be

developed which identify attributes associatedwith a product/service that can then be refinedinto additional features, such as add-ons orcross feature functions. How strongly aconsumer is prepared to take-up the attributeand/or its feature permutations can be testedthrough the relative value a consumer placeson both the attribute and/or its bundledfeatures. In other words, a consumer may befaced with a multiple set of attributes andfeatures of a product and asked to weigh therelative benefits they perceive from eachmutually exclusive choice set. This way theconsumer’s marginal utility and marginal rateof substitution between attributes and featurescan be measured.Choice processes and decision-making are

complex and involve multiple thought andaction stages. A consumer may start by


identifying a need, seeking out information,and identifying a utility function that a numberof actions might fulfill and from which somepreference may be identified. The consumercan then decide whether they will take up theaction immediately (or over a specified time-frame) through to various stages of delay tonever actualizing the choice.Implied however is a tension between what

people say theymay do (Stated Preference) andwhat they actually do (Revealed Preference). Anumber of researchers have attempted toresolve this dichotomy by separating Statedand Revealed Preference. This then re-con-nects what consumers say they will choose todo in future with what they have actuallychosen to do in the past (Adamowicz et al.,2003; Riddington et al., 2003). However inthese cases, researchers have argued that thereis little difference between Stated Preferenceand Revealed Preference and that the resourcesrequired in time, explanation and survey costs tofirst separate and then re-connect these modelsis not justifiable. They suggest, as others do, thatStated Preference alone yields valid and reliabledata on which to base predictions aboutbehavior (Louviere et al., 2000, p. 21). Usingmodels derived from random utility theory, theprobable relationship between utility andproduct/service selection can be measuredstatistically, by translating individual differ-ences into predicting general behavioral type.The implications of discrete choice model-

ing research for managers may involve adjust-ments to strategic planning and product/service offerings, re-positioning the pro-ducts/services to better fit with consumers’beliefs, evaluating alternative packages, andbundles as a utility function and evaluating theresulting demand or market share that mightaccrue to the organization as a result ofreconfiguration of products and services.Choice modeling has been used in a number

of contexts in testing the demand for existingand potential products and services and attri-butes of those products and services (Augeret al., 1999; Hanley et al., 2003; Crouch andLouviere, 2004). The alternative to choicemodeling is conjoint analysis. We discarded


DOI: 10.1002/nvsm


conjoint analysis on the basis that rating scalesand rankings were not a suitable way tounderstand decision-making. Rather, on thebasis of random utility theory, choice exper-iments are discrete choices—not simply listsof preferences—and estimate the benefit derivedto the individual (Louviere and Woodworth,1983). Research suggests that choice modelingis sufficiently robust to predict behavioraloutcomes on the basis of stated choice.Choice modeling methodology seemed to

offer a better way of understanding thepreferences of visitors to cultural attractionsand then to develop strategies that wouldmakethe traditional visitor a more frequent culturalconsumer. While contingent valuation methodsmay be appropriate for testing demand amongboth visitors and non-visitors our cases understudy required specificity and concentrationon visitors only in order to confidently makepredictions about what mix of museumproducts and services would increase culturalconsumption.

Methodology

Two major museums operating at the state(SM) and national (NM) level were partnersin the research. Though directed by diffe-rent government instrumentalities, thesemuseums had many similarities. They wereof similar size and close geographic proximity.Although each had specialized collections,their fundamental context was informed bysocial history. In addition these two institu-tions had a track record of collaborationparticularly in ways of enhancing customersatisfaction. This project presented ways inwhich their collaboration could be furtherexplored. The research consisted of threestages in each museum:

� A

Co

qualitative study to determine why peoplechoose to come to museums and whatinfluences their choices. These consistedof 40 face-to-face interviews divided demo-graphically in order to understand differentneeds among different life cycle cohorts(Adults with and without dependent chil-

pyright # 2008 John Wiley & Sons, Ltd. Int. J.

dren, young people 18–25, couples/singles25–35). The interviews were from 20–60minutes in duration and recorded withfull transcription to enable all possiblechoice factors to be identified for sub-sequent modeling.

� A
workshop with the Directors of the twomuseums to eliminate those identified fac-tors that would be impossible to include aspotential choice sets on the basis of govern-ment and workplace constraints.
� A
n online survey instrument for visitorsrecruited at the two museum sites andmuseum-goers in Sydney recruited on-line.This resulted in 82 respondents for the NMand 89 for the SM.
Visitors were approached within eachmuseum and asked a series of screening ques-tions (quota sampling based on age, gender,lifecycle), offered an incentive and taken to aroom for the in-depth interview. The lifecycledistinction in particular allowed us to testwhether children determined a significant pro-portion of leisure time activity and con-sequently what attributes parents seek for theirchildren in leisure experiences that may bedifferent from attributes sought in other life-cycle categories.

The interviewers were senior researchers(authors of this paper) with considerableexperience in-depth interviewing that allowedfor some deviation from the script in orderto capture unexpected data or insights bythe informant that could prove valuable. Theprotocol focused on what the visitor haddone that day; what decision-making processeswere involved in coming to the museum,including alternative activities; evaluation ofthe experience of that day’s museum visit(things liked/not liked); substitute culturalexperiences and why these are liked; theinfluence of others in decision-making; time,and financial constraints. The interview guide-line was qausi-standardized to incorporatedifferent modes of questioning to probe foropinions and preferences.

The depth interviews were analyzed using acombination of key word search (Seale et al.,

Nonprofit Volunt. Sect. Mark., February 2009

DOI: 10.1002/nvsm


2006), context and content analysis (Salminenand Jalkala, 2006) and thematic analysis (Braunand Clarke, 2006). Key word search involvedhighlighting all possible choice attributes.Context analysis considered the frequencyand position of potential attributes—the con-texts in which theywere discussed andwhetherthese emerged prompted or unprompted.Content analysis analyzed the meanings ofthese attributes and commonalities or differ-ences. The relationship between the attributes(including those mentioned by informants andinferred by the analyst) was considered byusing thematic analysis.Analysis of the depth interviews identified

four clusters of factors that facilitated orimpeded informants’ leisure choices:

� l

Co

ogistics (time to reach destination, transpor-tation, parking),

� fi
nite time (alternative leisure commitments,a willingness to visit a museum only in aspecified pattern, from yearly to once perschool holidays, need for more flexible andcreative opening hours),
� c
ost (high value and reasonable cost ofmuseumentry but high cost of associated items suchas food, parking and the other activitiesundertaken to make a ‘‘day out’’) and
� f
ulfillment of museums generally and speci-fic attractions within them (educationalenrichment, cultural variety, pleasure) asthe important factors.
Themain components of the first three wereconsidered by museum directors and research-ers as suitable for inclusion in the preliminarychoice modeling experiment.An on-line questionnaire design was devel-

oped with scenarios consisting of differentattributes and levels identified in the qualita-tive research. Participants were chosen on thebasis of museum attendance—that is, potentialrespondents must have visited either the SM orthe NM at least once in the past 12months. Theon-line questionnaire consisted of 8 discretechoice scenarios. Each scenario listed two setsof features labeled Option A or Option B. Eachfeature had slightly different levels. The

pyright # 2008 John Wiley & Sons, Ltd. Int.

respondents were then asked to indicatewhether Option A or Option B would resultin more frequent visits, less frequent visits, orno change to visiting patterns (Table 1). Whilethe concentration required appeared at firstlaborious, respondents in the pilot projectreported a high interest in completing the task.While a free ticket to the museum was offeredas recognition of their time investment,another factor promoted as an incentive tocomplete the task was that it might revealinteresting personal preferences about therespondent him/herself.Somewhat different options were developed

for the two museums. The qualitative researchshowed thatwhile therewas considerable com-monality of choice factors in some instancesthe two publics valued different features.Although the scenarios remained constant:some features changed, but the attributes ofopening hours (Summer, Winter Extended),packages with attractions, packages withtransport, incentives and standard entry feesremained the same.The data from the choice surveys are

discrete choices. That is, the respondents tothe survey each received eight scenarios. Eachscenario consisted of similar attributes butdiffered slightly in relation to features. Respon-dents were asked to weigh the benefits of eachscenario to reflect which features matteredmost overall to each respondent. For each ofthe eight scenarios they then stated whetherthey would prefer museum Option A, museumOption B or neither option. Because this was anovel application to the museum sector of thisapproach that was primarily concerned withdemonstrating proof of concept, we under-took a relatively simple analysis of the data tobegin with. That is, the surveys are designedaccording to an experimental design in orderto allow us to vary the values of each featureindependently over the entire set of scenarios.Thus, the choice experiment constitutes alarge, incomplete, and sparse cross-tabulationtable, and so we began our analyses simply bycross-tabulating the choice responses againsteach of the features. This allowed us tovisualize the results so that we could anticipate


DOI: 10.1002/nvsm


what was likely to be significant, and also ifneed be, select appropriate statistical specifi-cations with which to model the data.Following the cross-tabulation analysis, we

estimated multinomial logit (MNL) choicemodels from the data. As noted by Louviereet al. (2000), MNL models are derived bymaking certain assumptions about the errorcomponents in consumer utility functions(statistical approximations to consumers’decision rules). In the case of the MNL model,the errors are assumed to be distributed asType 1 Extreme Value random variables withmean zero and constant variance, and it is thisassumption that results in the MNL model. Inthe present case, the MNL model should beviewed only as a reasonable first approxi-mation to what is likely to be a more complexchoice process that underlies the data.Due to resource limitations, the sample sizes

obtained in our surveys were relatively small,which in turn discouraged us from trying toestimate more complex statistical model formsthat would allow one to capture differences insensitivity to the features in the populationstudied. As a proof of concept demonstration,this is not a particularly worrisome limitation.The next section presents the results of theMNL model analyses.

Results and discussion

Not all attributes and features were statisticallysignificant although interestingly there appea-red some commonality between the twomuseums. Features that significantly suggestedpositive or negative choices for the StateMuseum are detailed in Table 2.While this gives an indication of preferred

ways in which a visitor might use the museumand in combination with other activities andbenefits it also gives an indication of what isstrongly not preferred. For example, althoughthere were no strong preferences for the speci-fic after school activities listed in the scenarios,there was a strong indication that after schoolactivities may be welcome. Similarly there wereindicators that no joint tickets, no combined


travel tickets with entry and single entry onlywere all negatively correlated. These strongnegative correlations warrant further investi-gation—what was suggested was not alwayswanted but not having the feature at all seemedequally unacceptable.

For the National Museum the followingfeatures were significant:

� e

J.

xtended Summer opening hours;
� j oint ticket with the Powerhouse (at $30);with Imax (at $35); the Aquarium (at $40);Harbor Cruise (at $75); Ferry ($15);
� b
ecome a member after three visits; � b ecome a multi-museum member at $200per annum;
� e
njoy the express lane for $60 membershipfee per annum;
� f
ee options of $20 per adult, $6 per child and$25 per family;
� r
e-entry in the same month.
There was a strong indication that visitorsshould be rewarded for frequency of visitwhere the estimated utility of single entryadmission was �0.242. One of the assump-tions in the research was that children’s needswere foremost in the minds of their carers inchoosing museums. This seems to be borneout by the results where child-centered activi-ties or added value for families were among themost attractive features.

Unlike many exercises in developing ‘‘lists’’of attributes for discrete choice modeling, the‘‘lists’’ of interest in this research are notfactors that drive visits to museums per se.Instead these are factors that museums mightconsider that would change the way peoplemake visitation choices, including repeatchoices. The objective in this research wasnot to explain choices but to ask what thingscan thosemuseums do that they do not do nowthat would have a positive impact on visitationchoices.

Conclusion and future research

The application of choice modeling to themuseum context was an innovative way to

Nonprofit Volunt. Sect. Mark., February 2009

DOI: 10.1002/nvsm


understand how additional value can becreated for the museum visitor. One of theconstraints of the research was that museumshave little capacity to increase value byinvesting additional resources to meet per-ceived customer demands within a competi-tive leisure environment. Attractive featuresassociated with increased benefit were ident-ified by visitors themselves. The permutationsof these features were constructed in ways thatminimized any additional cost to the museumsor if they were potentially costly then thesefeatures were eliminated from the choice sets.The implications of this research suggest thatthere are a number of ways that museums cancombine and re-combine their offerings inways that are attractive to visitors and whichinvolve little or no additional resources. It alsoindicates that museums may need to seek outstrategic alliances with other like attractionswithin their proximity and to cost jointpackaging offers to add value to the customerexperience and mutually benefit organizationsthat are otherwise potentially in competition.The conclusions that we drew from this

research tended to reinforce some of thefindings of earlier research. Visitors tomuseums tend to be cultural omnivores andtend to be actively engaged in social andcultural pursuits. They are often after serialleisure pursuit combing a number of activitieswithin a one-day outing (Burton and Scott,2003). Those organizations that can cater forthese sequential experiences through present-ing ready made packages or allowing flexibilityin terms of re-entry and extended openinghours at particular times of the year are likelyto benefit from increased visitor frequency.This means that museums can respond tovisitor demand by packaging and re-packagingtheir core purpose of experience, education,and social activity to meet needs either as asingle entity or in collaboration with otherleisure and cultural providers.One critical variable that was not tested in

this research was the relative attractivenessof permanent or temporary exhibitions. Infollow-up research this variable will be addedas a separate component of the research. That


is, we will be able to measure how attractiveparticular types of exhibitions are to particularsegments and to also identify what combi-nations of bundles and packages will enhancethe visitor experience. It is intended toconduct this research with six comparablemajor museums in the next phase.

Acknowledgements

The authors acknowledge the participation ofthe two museums in this research. They arealso grateful for the insightful comments of thereviewers of this paper.

Biographical notes

Christine Burton is a Senior Lecturer in theFaculty of Business, University of TechnologySydney.Jordan Louviere is a Professor of Marketingand is the Director of the Centre for the Studyof Choice (CenSoc), University of TechnologySydney.Louise Young is a Professor of Marketing at theUniversity of Western Sydney.

References

Adamowicz W, Swait J, Boxall PC, Louviere J. 2003.

Perceptions versus objective measures of

environmental quality in combined revealed

and stated preference models of environmental

valuation. In The New Economics of Outdoor

Recreation, Hanley N, Shaw WD, Wright RE

(eds). Edward Elgar: Cheltenham; 165–190.

Auger P, Devinney TM, Louviere JL. 1999. Wither

ethical consumerism: do consumers value ethical

attributes?Working Paper 99-006, AGSM, Sydney.

Australian Bureau of Statistics. 1999. Attendance at

Selected Cultural Venues, 4114.0. ABS: Can-

berra.

Australian Bureau of Statistics. 2000a. Labour Force

(Cat No. 6203.0). ABS: Canberra.

Australian Bureau of Statistics. 2000b. Wage and

Salary Earners (Cat No. 6248.0). ABS: Canberra.

Australian Bureau of Statistics. 2007. Attendance at

Selected Cultural Venues and Events, 2005–

2006: 4114.0. ABS: Canberra.


DOI: 10.1002/nvsm


Australian Film Commission. 2002. Get the Picture:

Essential Data on Australian Film, Television,

Video and Interactive Media, 6th edn. AFC:

Sydney.

Bennett T. 1994. The Reluctant Museum Visitor: A

Study of Non-Goers to History Museums and Art

Galleries. Australia Council: Sydney.

Bennett T, Emison M, Frow J. 1999. Accounting

for Tastes: Australian Everyday Cultures.

Cambridge University Press: Cambridge.

Bittman M. 1999b. The land of the lost weekend?

Trends in free time among working age Austra-

lians, 1974–1992. Society and Leisure 21(2):

435–454.

Bittman M, Rice J. 1999a. Are working hours

becoming more unsociable? SPRC Newsletter

74: 3–5.

Bourdieu P. 1984. Distinction: A Social Critique of

the Judgement of Taste, trans. R. Nice. Harvard

University Press: Cambridge, MA.

Braun V, Clarke V. 2006. Using thematic analysis in

psychology. Qualitative Research in Psychology

3: 77–101.

Burton C. 2003. Scoping the challenge: entre-

preneurial arts management in times of uncer-

tainty. Journal of Arts Management, Law and

Society 33(3): 185–195.

Burton C, Scott C. 2003. Museums: challenges for

the 21st century. International Journal of Arts

Management 5(2): 56–68.

Casey D. 2002. Museums as agents for social and

political change. Australasian Science 23(4):

18–19.

Clifford J. 1997. Routes: Travel and Translation in

the Late Twentieth Century. Harvard University

Press: Cambridge, MA.

Conforti M. 1995. Museums past and museums

present: some thoughts on institutional survival.

Museum Management and Curatorship 14(4):

339–355.

Crimp D. 1985. On the museum’s ruins. In Post-

modern Culture, Foster H (ed.). Pluto: London;

43–56.

Crouch GI, Louviere J. 2004. The determinants of

convention site selection: a logistic choice model

from experimental data. Journal of Travel

Research 43: 118–130.Evans M. 2001. Renaissance in the Regions: A New

Vision for England’s Museums. Resource: The

Council for Museums Archives and Libraries:

London.


Falk JH. 2006. An identity-centred approach to under-

standing museum learning. Curator 49(2): 151–

166.

Falk JH, Dierking LD. 1992. The Museum Experi-

ence. Whalesback Books: Washington.

Gleick J. 2000. Faster: The Acceleration of Just

About Everything. Vintage Books: New York.

Griffin DJG. 1978. The Australian Museum Visitor,

1976. The Australian Museum Trust: Sydney.

Griffin DJG, Abraham M. 1999. Management of

museums in the 1990s: governments and organ-

izational reform. In Management in Museums,

Moore K (ed.). The Athlone Press: London; 45–

92.

Griffin DJG, Abraham M. 2000. The effective man-

agement of museums: cohesive leadership and

visitor-focused programming. Museum Manage-

ment and Curatorship 18(4): 335–368.

Hanley N, Shaw WD, Wright RE. 2003. The New

Economics of Outdoor Recreation. Edward

Elgar: Cheltenham.

Hein HS. 2000. The Museum in Transition: A

Philosophical Perspective. Smithsonian Institu-

tion Press: Washington.

Hood M. 1992. After 70 years of audience research,

what have we learned? Who comes to museums,

who does not and why? Visitor Studies: Theory,

Research and Practice, Thompson D, Benefield

A, Bitgood S, Shettel H, Williams R (eds). Visitor

Studies Association: St Louis Missouri; Vol. 5, 16–

27.

Hooper-Greenhill E, Dodd J. 2002. Seeing the

Museum through the Visitors’ Eyes: The Evalu-

ation of the Education Challenge Fund.

Research Centre for Museums and Galleries, Uni-

versity of Leicester: London.

Hume M, Mort GS, Winzar H. 2007. Exploring

repurchase intention in a performing arts con-

text: who comes? And why do they come back?

Internatioanl Journal of Nonprofit and Volun-

tary Sector Marketing 2: 135–148.

International Council Of Museums. 2001. ICOM

Statutes (Online), 2002. Available: http://ICOM.

museum/statutes.html

Kotler N, Kotler P. 2000. Canmuseums be all things

to all people? Missions, goals and marketing’s

role. Museum Management and Curatorship

18(3): 271–287.

Levitt T. 1991. Marketing myopia. In Levitt on

Marketing, Levitt T (ed.). Harvard Business

Review: Cambridge, MA.


DOI: 10.1002/nvsm


Louviere J, Hensher DA, Swait JD. 2000. Stated

Choice Methods: Analysis and Application.

Cambridge University Press: NY.

Louviere J, Woodworth G. 1983. Design and

analysis of simulated consumer choice or allo-

cation experiments: an approach based on aggre-

gate data. Journal of Marketing Research 20:

350–367.

Lynch R, Burton C, Scott C,Wilson P, Smith P. 2000.

Leisure and Change: Implications for Museums

in the 21st Century. University of Technology,

Sydney and Powerhouse Museum: Sydney.

McFadden D, Train K. 2002. Hybrid choice models:

progress and challenges.Marketing Letters 13(3):

163–175.

MORI. 2001. Visitors to Museums and Galleries.

Resource: The Council for Museums, Archives

and Libraries: London.

Museums Australia. 2002. Museums Australia

(Online). Available: http://www.museumsaustralia.

org.au

Noonan D. 2003. Contingent valuation and cultural

resources: a meta-analytic review of the litera-

ture. Journal of Cultural Economics 27: 159–

176.

Oliver K, Burton C, Lynch R, Scott C. 2002. Choos-

ing Museums: Competition, Leisure Trends and

Decision-Making in the Free-Time Marketplace.

UTS/Powerhouse: Sydney.

Parker S, Waterston K, Michaluk G, Rickard L. 2002.

Neighbourhood Renewal and Social Inclusion:

The Role of Museums, Archives and Libraries.

Marketing Management Services International:

Glasgow.

Pollicino M, Maddison D. 2001. Valuing the benefits

of cleaning Lincoln cathedral. Journal of Cul-

tural Economics 25: 131–148.

Porter M. 1985. Competitive Advantage: Creating

and Sustaining Superior Performance. The Free

Press: New York.

Porter M. 1996.What is strategy?Harvard Business

Review 74(6): 61–79.

Prentice R, Davies A, Beeho A. 1997. Seeking

generic motivations for visiting and non-visiting

museums and like cultural attractions. Museum

Management and Curatorship 16(1): 45–70.

Rentschler R. 2002. The Entrepreneurial Arts Lea-

der: Cultural Policy, Change and Reinvention.

University of Queensland Press: Brisbane.


Rentschler R, Hede A-M, White TR. 2007. Museum

pricing: challenges to theory development and

practice. Internatioanl Journal of Nonprofit

and Voluntary Sector Marketing 12: 163–

173.

Riddington G, Sinclair C, Milne N. 2003. Modelling

choice and switching behaviour between Scot-

tish ski centres. In The New Economics of Out-

door Recreation, Hanley N, Shaw WD, Wright

RE (eds). Edward Elgar: Cheltenham; 123–136.

Roberston H, Migliorno P. 1996. Open Up! Guide-

lines for Cultural Diversity Visitor Studies. Aus-

tralia Council and Powerhouse Museum: Sydney.

Rojek C. 1995. Decentring Leisure. Sage: London.

Rojek C. 2000. Leisure and Culture. Macmillan

Press: London.

Saatchi S. 2000. Australians and the Arts: What Do

the Arts Mean to Australians? A Report to the

Australia Council by Saatchi & Saatchi. Austra-

lia Council: Surry Hills.

Salminen RT, Jalkala A. 2006. Content analysis of

customer reference descriptions in Ict companies,

AdvancingTheory,MaintainingRelevanceANZMAC,

QUT, http://www.anzmac.org/proceedings,

Brisbane, Qld.

Sandell R. 1998. Museums as agents of social

inclusion. Museum Management and Curator-

ship 17(4): 401–418.

Santagata W, Signorello G. 2000. Contingent valua-

tion of a cultural public good and policy design:

The case of ‘‘Napoli Musei Aperti’’. Journal of

Cultural Economics 24: 181–204.

Scott R. 1999. Full screen ahead Australian Leisure

Management no. April-May.

Seale C, Ziebland S, Charteris-Black J. 2006. Gender,

cancer and internet use: a comparative keyword

analysis of interviews and online cancer support

groups. Social Science and Medicine 62: 2577–

2590.

Slater A. 2007. Escaping to the gallery: understand-

ing the motivations of visitors to museums. Inter-

national Journal of Nonprofit and Voluntary

Sector Marketing 12: 149–162.

Stebbins RA. 1999. Serious leisure. In Leisure Stu-

dies: Prospects for the Twenty-First Century,

Jackson EL, Burton TL (eds). Venture Publishing:

Pennsylvania.

Thompson E, Berger M, Blomquist G, Allen S.

2002. Valuing the arts: a contingent valuation


DOI: 10.1002/nvsm


approach. Jounal of Cultural Economics 26: 87–

113.

Throsby D. 2003. Determining the value of cultural

goods: howmuch (or how little) does contingent

valuation tell us? Journal of Cultural Economics

27: 275–285.

Throsby D, Withers G. 1979. The Economics of the

Performing Arts. Edward Arnold: Melbourne.

Throsby D, Withers G. 1984. What Price Culture?

Australia Council: North Sydney.

Todd S, Lawson R. 2001. Lifestyle segementation

and museum/gallery visiting behaviour. Inter-

national Journal of Nonprofit and Voluntary

Sector Marketing 6(3): 269–277.


Veal AJ, Lynch R. 2001. Australian Leisure

(2nd edn). Longman: Frenchs Forest.

Vergo P. 1989b. The reticent object. In The New

Museology, Vergo P (ed.). Reaktion: London; 41–

59.

Weil SE. 2002. Making Museums Matter. Smithso-

nian Institution Press: Washington.

Willis KG. 2002. Iterative bids in contingent valua-

tion and the estimation of the revenue maximis-

ing price for a cultural good. Journal of Cultural

Economics 26: 307–324.

Yann Campbell, Hoare Wheeler. 1999. Employment

Security and Working Hours: A National Sur-

vey of Current Workplace Issues. ACTU: Sydney.


DOI: 10.1002/nvsm

retaining the visitor, enhancing the experience: identifying attributes of choice in repeat museum...

Documents