Sign up

Positioning of international air passenger carriers using multidimensional scaling and correspondence analysis.
Subject:
Airlines (Case studies)
Airlines (Market research)
Travel industry (Case studies)
Consumer preferences (Case studies)
Web sites (Case studies)
Authors:
Wen, Chieh-hua
Yeh, Wen-ya
Pub Date:
01/01/2010
Publication:
Name: Transportation Journal Publisher: American Society of Transportation and Logistics, Inc. Audience: Academic; Trade Format: Newsletter Subject: Business; Transportation industry Copyright: COPYRIGHT 2010 American Society of Transportation and Logistics, Inc. ISSN: 0041-1612
Issue:
Date: Wntr, 2010 Source Volume: 49 Source Issue: 1
Topic:
Event Code: 240 Marketing procedures Advertising Code: 34 Research Findings Computer Subject: Marketing research; Company Web site/Web page
Product:
Product Code: 4500000 Air Transportation; 7010100 Tourist Travel NAICS Code: 481 Air Transportation SIC Code: 4512 Air transportation, scheduled; 4513 Air courier services; 4522 Air transportation, nonscheduled
Organization:
Company Name: Cathay Pacific Airways Ltd.; China Airlines Company Ltd.; United Air Lines Inc.; Singapore Airlines Ltd.; Jetstar Asia Airways; EVA Airways Corp. Ticker Symbol: 00293
Geographic:
Geographic Scope: China; Taiwan; United States; Singapore

Accession Number:
224775355
Full Text:
Abstract

This article uses multidimensional scaling and correspondence analysis to explore how international airline passengers position various air carriers. The analysis has produced perceptual maps of relative positions of airlines and their service attributes. Empirical data were collected from Taiwanese air passengers who had flown from Taipei to Tokyo and to Singapore. Factor analysis was performed to reduce a large number of airline service attributes into a small set of underlying factors. The results of multidimensional scaling analyses reveal that some air carriers are closely competitive; this can be seen by their close proximity in the perceptual map. However, certain airlines possess a unique service quality profile and are not strongly associated with other airlines; such unique firms appear as isolated points on the perceptual map. Interestingly, both multidimensional scaling and correspondence analysis produce similar perceptual maps of the airlines' relative positions and service attributes. However, the advantage of correspondence analysis over multidimensional scaling is that data collection is relatively quick and easy because the analyst is not required to obtain respondents' ratings for each attribute. This study provides specific suggestions about how air carriers on each route can use the results to improve their performance or to build a differential edge.

**********

The position of a brand is the perception that it creates in the minds of consumers. A positioning strategy that establishes competitive advantage is critical to the success of a company or product brand. The airline passenger market appears to be extremely competitive due to the increasing availability of airlines and flight schedules. Airlines that are properly positioned relative to their rivals can gain a competitive edge by differentiating their services from the services of their rivals. Airlines can undertake positioning analysis to understand customers' perceptions and expectations. Further, airlines apply positioning strategies to provide differentiated service quality for their current and potential passengers (Surovitskikh and Lubbe 2008).

Despite a significant number of previous studies on aviation, very few articles have examined the content of airline positioning. For instance, Kaynak et al. (1994) examined consumer perceptions in the global air industry and provided evidence that airlines should recognize that passengers are diverse and have high expectations concerning service quality. Based on actual service performance of U.S. domestic airlines, Gursoy et al. (2005) investigated relative positioning of these air carriers and identified closely competing airlines and their critical differentiating attributes. Surovitskikh and Lubbe (2008) studied the positioning of Middle Eastern airlines in South Africa and used a 3D centroid plot of three factors (i.e., consistency of service, reliability, and augmented products) to examine airlines' performance. More recently, Wen et al. (2008) explored segmentation and positioning in the international air passenger market and provided valuable insights on airlines' competitive standings relative to their close competitors.

Positioning analysis often uses perceptual maps to identify the positions of competitive products and brands. Multivariate statistical approaches, such as multidimensional scaling (MDS) and correspondence analysis (CA), have been widely applied to create perceptual maps (Myers 1996). MDS uses similarity or preference data that are ratings or ordinal scales (Aaker et al. 2007). The main data format advantage of CA over MDS is that in a CA the respondent only needs to put a check mark by each attribute to indicate whether the attribute describes a given brand; another advantage is that the analyst is not required to obtain respondents' ratings or ordinal preferences for all attributes of each brand. To the best of our knowledge, most previous studies on airline positioning have used the CA approach (Kaynak et al. 1994; Gursoy et al. 2005; Wen et al. 2008).

Although CA is easy to use and generate perceptual maps, MDS is also a popular method for brand or product positioning (Kaul and Rao 1995) and has been used for subjects such as hospitality/tourism (e.g., Dev et al. 1995; Kim and Agrusa 2005) and marketing (Wilkes 1977; Wirier and Moore 1989; Cha et al. 2009). Our objective is to perform positioning analysis for certain international airline passengers and to identify the comparative positions of the airlines and their service attributes. This study uses both CA and MDS approaches to create perceptual maps that illustrate the relative positioning of airlines and key service attributes. These two methods can produce perceptual or positioning maps, but very few studies have compared the results obtained from two different approaches. In particular, this study aims to highlight important differences between findings obtained from these two methods.

A survey instrument was designed to collect data from face-to-face interviews with airline passengers. Travelers who had flown from Taipei to Tokyo and from Taipei to Singapore were chosen as interviewees. The respondents were interviewed at Taoyuan International Airport, the international airport from which the majority of Taiwanese flights depart, in 2007. Data analyses were performed using SPSS V13.0 for CA and SYSTAT V12.0 for MDS. This article concludes with managerial implications and directions for future research.

METHODOLOGY

Service Quality Variables

Any airline's market share and revenue depend on customer satisfaction and loyalty.

These, in turn, depend on customers' perceptions of service quality. Service attributes of airlines are the important determinants affecting air travelers' selection of airlines (Proussaloglou and Koppelman 1999). Thus airlines' service attributes were chosen as positioning variables. Positioning analysis can identify major strengths and weaknesses of each airline's services and can explain how each airline can differentiate its services relative to its competitors.

A large number of airlines' service attributes have been identified in past studies. For example, Aksoy et al. (2003) used factor analysis to reduce thirty-nine service quality attributes of domestic and foreign airlines into a few important factors. Park et al. (2004) used twenty-two airline service quality items, primarily based on the SERVQUAL scale, together with service expectation, service perception, service value, passenger satisfaction, and airline image to understand the effects of these factors on passengers' intentions. Chen and Chang (2005) designed seventeen service attributes of ground service (e.g., convenient flight schedules, service efficiency of reservation staff, convenient ticketing and check-in procedures) and fifteen service attributes of in-flight service (e.g., seat comfort, clean and pleasant interior, good cabin equipment conditions, cabin crew's ability to handle customer complaints). They examined airline service quality by importance-performance analysis and by gap analysis. Using factor analysis to reduce twelve service attributes, Liou and Tzeng (2007) developed a non-additive model to evaluate the service quality of international airlines, using variables that included employees' service, safety and reliability, on-board service, schedule, on-time performance, and frequent flier program. Park (2007) measured the service quality of airlines, using variables named in-flight service, reservation-related, airport service, reliability, employee service, flight availability, overall service quality, ticket price, value, passenger satisfaction, and airline image. To study passengers' expectations and perceptions of airlines' service quality, Pakdil and Aydin (2007) used weighted SERVQUAL scores, including employees, tangibles, responsiveness, reliability and assurance, flight patterns, availability, image, and empathy.

The airlines' service attributes used in this survey were chosen in light of previous studies. A pilot test was used to check the wordings, format, and reliability of measurement items before the full-scale survey launched. Table 1 presents a summary of service attributes used in the present study.

Multidimensional Sealing

Perceptual mapping is a marketing research tool that has been extensively used to position products and services for customers (Kohli and Leuthesser 1993). Products or services are displayed as points on a map. Perceptual mapping is often used to help management find a positioning strategy. MDS provides a visual representation of the objects in a common space; this visual representation can be easily explained and readily understood. MDS output usually describes brand relationships with plots; each plot has a few dimensions, depending on the degree of similarity or preference between respondents' perceptions of brands. The fit between the data and the final configuration can be indicated by the stress parameter (Kruskal and Wish 1978), which is a badness-of-fit measure. The appropriate number of dimensions can be determined by plotting the stress values against the number of dimensions.

MDS typically uses non-attribute data based on the similarities or preference ratings of the brands under consideration. For similarity data collection, respondents are asked to rate or rank the similarity of all pairs of brands (e.g., all pairs of airlines in the set (A, B, and C) are ranked in order to determine which airlines are the most similar to each other and which are the most dissimilar). Several measures are often used to obtain respondents' preference data (e.g., each respondent is required to rank the airlines from the most preferred to the least preferred, to make paired comparisons and indicate which airline in a pair is preferable). The similarity and preference data can be analyzed by metric or nonmetric methods. Metric MDS uses either rating or interval data (e.g., Likert-type scales), while nonmetric MDS requires ordinal ranking data. MDS can also incorporate the attribute data for the brands being studied. Attribute-based MDS requires respondents to rate the brands on the identified attributes using a semantic or Likert scale. The metric ratings of all airlines with regard to a set of service attributes were input to an MDS analysis; the output of the MDS analysis is a perceptual map that illustrates the relationships between airlines and service attributes.

MDS can include all attributes to generate perceptual maps. The initial MDS result used eighteen service attributes, but the perceptual map was difficult to interpret, partly because of highly correlated attributes. These eighteen attributes caused serious overlap in the perceptual map. Because this overlap obscured the service attributes of the airline, it was difficult to position each airline with respect to all eighteen attributes.

Alternatively, MDS can use factor analysis to reduce attributes into fewer factors and subsequently use those reduced factors to produce perceptual maps (e.g., Kim 1996; Kim 1998). Factor analysis is a supplemental technique to the MDS analyses. Thus, exploratory factor analysis was used to extract a small number of factors from a large set of service attributes. The final MDS result used a small number of factors to create the perceptual map. Principal component analysis with an orthogonal (VARIMAX) rotation extracted the factors. Eigenvalues were plotted against number of factors; the scree test determined the optimal number of factors by finding the lowest eigenvalue greater than or equal to one (Cattell 1966). Items were retained in the final result if their factor loadings were greater than 0.45. The coefficient alpha was used to assess the reliability of the factors and to assess the items within each factor for further refinement. Any factor that had a coefficient alpha greater than 0.6 was considered reliable (Nunnally 1978).

Correspondence Analysis

CA is a method for exhibiting the rows and columns of a two-way contingency table (i.e., a cross-tabulation of two categorical variables) as points in low-dimensional vector spaces (Greenacre 1984). A two-way contingency table for CA in the present study consists of rows for airlines and columns for service attributes. CA produced a perceptual map based on relationships existing between air carriers, between attributes, and between airlines and attributes.

CA does not need to obtain respondents' ratings or ordinal scales for each attribute; informants only need to select the excellent attributes of given brands. Respondents simply put a check mark by each attribute to indicate whether the brands were best characterized by each attribute. Respondents could put any number of brands for each attribute. Respondents chose attributes if they believed, from travel experience or from the airline's image, that the airline performed well in this attribute.

CA is used to find a lower and adequate dimension to explain the map. The decision about the number of dimensions to retain for explication is made from eigenvalues (or singular values) which indicate the relative contribution of each dimension in explaining the variance. The dimensions identified in CA can be interpreted by pinpointing the largest relative contributor to the variance as explained by the axis. The larger the amount of an attribute in a dimension, the more essential that attribute is in determining the latent structure of that dimension.

EMPIRICAL STUDY

Data

In order to assess the two alternative positioning approaches, fifty samples with twenty-two service attributes were initially collected in April 2007 at Taoyuan International Airport. After the pilot survey, some service attributes were excluded due to lack of reliability. The final analysis included eighteen airline service attributes, as shown in Table 1.

Empirical data were collected from air passengers who had flown from Taipei to Tokyo or to Singapore. The Taipei-Tokyo route was selected for its large number of air passengers and its long flight time (three and a half to four hours). Notably, the Taipei-Hong Kong and Taipei-Macau routes have the largest and second largest numbers of passengers annually; however, these flights are only one and a half to two hours long and it is very likely that service quality, particularly in-flight services, may not be critical. The Taipei-Singapore route was chosen for investigation because a low-cost carrier, Jetstar Asia Airways, operates on this route. The competition between full-service and low-cost carriers has received attention in recent years due to the emergence and popularity of low-cost carriers around the world (O'Connell and Williams 2005; Fourier and Lubbe 2006). Although international air routes are likely to have passengers with diverse profiles and market characteristics, only two international air routes were selected for empirical study. To explore a large number of air routes and competing carriers would require a very large-scale survey. Because we restricted our survey to a small number of routes, we were able to obtain sufficient samples and reliable analyses for each selected route.

Although MDS and CA require different data formats, this study designed a single survey instrument that had four sections. The second section collected data for MDS and the third section collected data for CA. The first section obtained information regarding respondents' previous international travel experiences. The second section measured the perceived importance of and satisfaction with airline attributes; it used a 7-point Likert scale with a range from 1 (least important or very dissatisfied) to 7 (most important or very satisfied). The respondents were asked to rate only their most recently chosen airline for their trips from Taipei to Tokyo or to Singapore. The next section asked the respondents to put check-marks to indicate which airlines, if any, are described by each attribute. The respondents were allowed to select any number of airlines for each attribute. The respondents included both frequent and infrequent air passengers. Less frequent passengers may have had limited experience with air travel. As such, they were only expected to assess a limited number of airlines and their attributes. However, even experienced passengers were not expected to evaluate all possible airlines and service attributes. Thus, respondents were not required to fill in all airlines and attributes; the answers were based on their experiences and perceptions of airline images. The fourth section obtained the socioeconomic and trip-related characteristics of respondents.

Face-to-face interviews were conducted with airline passengers at the Taoyuan International Airport, from which the majority of Taiwan's international flights depart. Samples of the two routes were collected at different time periods. The Tokyo samples were collected during weekdays and weekends in May 2007. Out of 451 distributed questionnaires, final analysis was based on 372 valid questionnaires, i.e., a response rate of 82 percent. The Singapore samples, collected in December 2007, included 432 respondents; 344 were valid; the response rate was 80 percent.

Table 2 summarizes the socio-economic and trip profiles for the Taipei-Tokyo and Taipei-Singapore routes. For the Taipei-Tokyo route, among the 372 respondents, females were the majority (58.6 percent). Two large age groups in the sample were the "under 30 years old" (39.0 percent) and the "31 to 40 years old" (34.1 percent). 56.2 percent of the respondents reported working in the business or service industry. 89.2 percent of the respondents possessed college or graduate education. 71.7 percent of the respondents had personal monthly incomes between NT$10,000 and NT$80,000. Most had gone to Tokyo up to 3 times (63.4 percent), and 78.8 percent of the respondents had bought the tickets through a travel agency. The highest proportion of economy-class flight allowances (89.7 percent), and 68.5 percent of all respondents, traveled for tourism.

For the Taipei-Singapore route, out of 344 respondents, males (56.4 percent) were the majority; 36.0 percent of respondents were 31 to 40 years old. 87.3 percent had college or graduate education. There were similar proportions of business (26.7 percent) and service industry (27.9 percent) travelers. Most had personal monthly incomes between NT$40,000 and NT$80,000 (46.5 percent). 61.6 percent of respondents had gone to Singapore up to 3 times; 64.2 percent of the respondents had purchased the tickets from a travel agency. Most respondents chose economy-class flight allowances (91.0 percent); 60.8 percent of respondents traveled for leisure and recreation.

Table 3 reports the average scores of each service attribute; the scores, based on a 7-point Likert scale with a range from 1 (least important) to 7 (most important), are ranked from greatest to least importance. Respondents perceive flight safety (V15) as the most important attribute. Travelers on the Taipei-Tokyo route emphasized customer service, including the cordiality and kindness of cabin crews (V14) and customer complaint handling (V16). These two items obtained high scores. Passengers on the Taipei-Singapore route gave high scores to the image of the airline (V18) and to the cordiality and kindness of cabin crews (V14). Respondents on both routes attached importance to on-time performance (V6), comfort and spaciousness of seats (V10), and cleanliness on board (V12). However, they rated price (V1) and flight frequency (V3) as less important, because most respondents who travel for leisure usually join travel tours, which offer discounted air fares and itineraries arranged by agents.

Results

Taipei-Tokyo route

For the MDS analysis, average ratings of the perceived satisfaction with airline attributes were calculated. Initially, we used eighteen attributes to produce a relatively complex perceptual map for MDS. As Fig. 1 indicates, the eighteen attributes overlap appreciably, thus the map hardly distinguished the place of each airline and service attribute. We could not clearly define the relative positions of airlines and key attributes associated with airlines in the perceptual map; therefore it was difficult to interpret the map. This problem may have been due to the fact that the MDS averaged eighteen attributes across airlines to produce perceptual maps, but the average score of each attribute was not significantly different across airlines. For instance, some items measured similar service attributes such as comfort and spaciousness of seats, food and beverages services on board, cleanliness on board, and entertainment facilities on board, which were associated with on-board services; the average scores of these attributes were also very close.

Therefore, factor analysis with the principal components method was performed to reduce the eighteen airline service attributes into a small number of factors. To test whether these factors were consistent and reliable, the Cronbach coefficient alpha was calculated. As reported in Table 4, the values of the coefficient alpha were between 0.80 and 0.85, indicating good reliabilities. The total percentage of variance for the present solution was greater than 60 percent, which was acceptable to represent all the service attributes. All factor loadings of the service attributes are greater than 0.45, which indicates the statistical significance for our sample sizes. The final analysis produced

Mr. Wen is associate professor, Department of Transportation Technology and Management, Feng Chia University, Taiwan; e-mail chwen@fcu.edu.tw. Ms. Yeh is a graduate student, Institute of Traffic and Transportation, National Chiao Tung University, Taiwan; e-mail d9481223@fcu.edu.tw.

four factor solutions: price and flights (Factor 1), ground service (Factor 2), on-board amenities (Factor 3), and safety and image (Factor 4). These four factors were found to properly describe the distinct sets of service attributes for the Taipei-Tokyo air passenger market.

[FIGURE 1 OMITTED]

Factor scores for each respondent were calculated and used in MDS. This produced a low-dimension map that revealed airlines' relative positions and competitive relationships. Figure 2 shows a two-dimensional perceptual map which identifies the relative positions of six airlines and key differentiating factors. Eva Airways (BR) and Cathay Pacific Airways (CX) are positioned in the upper-left quadrant and are located near each other, indicating closely competitive airlines. The closer the airline is to a service factor, the more related it is to that factor. In contrast, the farther the airline is from a service factor, the less associated it is with that attribute. Because respondents perceive similar service performance between these two airlines on the ground service and safety/image factors (F2 and F4), Eva and Cathay Pacific Airways are recognized as close competitors. However, of all the airlines in this survey, Eva Airways has the best performance on safety/image. Cathay Pacific Airways performs comparatively well on ground service. Japan Asia Airways (EG) and Air Nippon Airways (EL) comprise another competing group positioned in the lower-right quadrant, close to some factors making the two airlines closely competitive. Japan Asia Airways is closer to the F1, F2, and F3 factors than Air Nippon Airways. Thus it appears that Japan Asia Airways has a differential edge over Air Nippon Airways. Both China Airlines (CI) and United Airlines (UA) have unusual attribute profiles and are isolated from other airlines. China Airlines is more associated with price and flights (F1); United Airlines is strongly related to ground service.

Fig. 3 illustrates the result of CA with a two-dimensional perceptual map. To interpret the dimensions, the proportion of variance explained by each variable in relation to each principal axis was used. The larger the proportion of an attribute is in a dimension, the more essential that attribute is in determining the meaning of that dimension. The first dimension, safety and price, is the horizontal dimension of a plane; V15 (flight safety) is the major contributor, and V1 (price) is a second one, as these two attributes are extreme in terms of their position on this dimension. The second dimension, attitude and information, had cordiality and kindness of cabin crews (V14) as its highest point, and Web site service (V17) as its lowest point (note that the location of V15 and V1 is lower than V17; however, these two attributes have been used to explain the first dimension).

Two groups of airlines are located near each other and are closely competitive. The first group of airlines includes Eva and Cathay Pacific Airways, and the second group consists of Japan Asia and Air Nippon Airways. Eva and Cathay Pacific Airways are highly competitive because passengers in this cluster perceive similar performance between these two airlines with regard to service attributes such as image of airline (V18), comfort of seats on board (V10), entertainment facilities on board (V13), and Web site services (V17). Japan Asia Airways and Air Nippon Airways are closely competitive because they are both Japan carriers. Air Nippon Airways is very close to Japan Asia Airways for on-time performance (V6), cleanliness on board (V12), and customer complaint handling (V16). China Airlines and United Airlines are located far from other airlines; each of these airlines has unique service performance and is not closely related to other airlines. Even though three attributes, including frequency of flight (V3), convenience of reservation and ticketing (V4), and cordiality and kindness of cabin crews (V14), are positioned near China Airlines, the distance between the position of China Airlines and the three attributes is not short, indicating that China Airlines does not have a strong competitive edge on those attributes. Frequency of flights (V3) is the attribute closest to China Airlines. United Airlines had infrequent flights compared with other airlines, and most importantly it does not belong to the flag carrier of the arrival or the departure country. As such, United Airlines has a unique position, far from other airlines.

[FIGURE 2 OMITTED]

Taipei-Singapore route

The interpretation of MDS with eighteen attributes for the Taipei-Singapore samples is difficult. Like the Taipei-Tokyo dataset, the Taipei-Singapore dataset has the problem of overlapping attributes and blurrily located airlines and service attributes (Figure 4). Factor analysis with the principal components method was performed to reduce the eighteen airline service attributes. In Table 5, the values of the coefficient alpha varied between 0.62 and 0.88, which indicated a satisfactory reliability. The total percentage of variance for the present solution was greater than 60 percent, which was adequate to characterize all the service attributes. The final result of factor analysis obtained five factors: price and flights (Factor 1), on-time performance and reservation service (Factor 2), ground service (Factor 3), onboard amenities (Factor 4), and safety and image (Factor 5).

[FIGURE 3 OMITTED]

Figure 5 depicts airline positions as calculated by MDS in a two-dimensional perceptual map. Eva Airways (BR) and Singapore Airlines (SQ) can be seen to be highly competitive because they are positioned together in the upper-fight quadrant. The map shows that both airlines have positive on-time performance and reservation service (F2); they are also close with regard to on-board amenities (F4) and safety and image (F5). Jetstar Asia Airways (3K) and China Airlines (CI) are far away from other airlines. Jetstar Asia Airways, a low-cost carrier, is highly associated with price and flights (F1) in the map, demonstrating price advantage over its competitors.

Fig. 6 illustrates the results of CA for the Taipei-Singapore samples. Most points are clustered in the extreme right of the graph. Point V1 is the leftmost point on the map, and points V10, V15, and V18 are all on the opposite edge. Looking at extreme points along the horizontal axis, it can be seen that price (V1) is very low; queuing at check-in counter (V7) and service attitude at check-in counter (V8) take moderate values; comfort and spaciousness of seats (V10) and entertainment facilities on board (V13) are at the high end of the horizontal axis. Thus, the first dimension is labeled "price and on-board services." The notably low vertical value is flight safety (V15). Many other values are high on the vertical axis, such as frequency of flight (V3), convenience of reservation and ticketing (V4), and convenience of flight schedule (V2). Hence, the second dimension is labeled "safety and convenience of flights."

[FIGURE 4 OMITTED]

Eva Airways and Singapore Airlines are close to each other in Figure 6 because they have similar performance on on-time service and comfort and spaciousness of seats. In addition, Eva Airways has better performance on food and beverages services on board (V11), cleanliness on board (V12), entertainment facilities on board (V13), and customer complaint handling (V16). Singapore Airlines is particularly close to some specific attributes, namely Web site services and image of airline. China Airlines is isolated in the upper-left quadrant. China Airlines performs well on convenience of flight schedule (V2), frequency of flight (V3), convenience of reservation and ticketing (V4), service attitude of reservation staff (V5), and cordiality and kindness of cabin crews (V14). Jetstar Asia Airways stands in the lower-left quadrant and also has a fairly unique position, very close to price (V1), which shows that it has a competitive advantage as a low-cost carrier.

DISCUSSION AND IMPLICATIONS

To explore how international airline passengers position various air carriers, this study applies both CA and MDS approaches to generate perceptual maps that illustrate the relative positioning of airlines and key service attributes for two selected routes. Interestingly, the results of MDS and CA identified the same competitive clusters of airlines. For the Taipei-Tokyo route, both approaches identified two groups of airlines that share similar attribute profiles and two airlines that have unique attribute profiles. The first group of airlines includes Eva and Cathay Pacific Airways, and the second group consists of Japan Asia and Air Nippon Airways. The quality profiles of China Airlines and United Airlines do not resemble each other, nor are they strongly associated with other airlines. For the Taipei-Singapore route, both approaches identified one group of airlines that are closely competitive. Eva Airways and Singapore Airlines are located in close proximity to each other, and both have similar attribute profiles.

Although the results of MDS and CA are fairly similar in terms of the competing groups identified, the data requirements for CA and MDS differ to some extent. MDS requires respondents' rating measures (e.g., a Liken scale) for every attribute, and CA simply asks respondents to consider each attribute and to indicate whether or not the attribute describes a given brand. CA successfully incorporates all eighteen attributes to generate perceptual maps that illustrate the relationships between air carriers, between attributes, and between airlines and attributes. MDS can produce perceptual maps by averaging the scores of eighteen attributes by airline, but the position of airlines was not obvious and did not properly describe the relationship between airlines and their service attributes. To avoid this problem, factor analysis with the principal components method was performed to reduce the eighteen airline service attributes into a small number of factors. The final result of MDS created a low-dimension map that explicitly revealed airline relative positions and competitive relationships.

[FIGURE 5 OMITTED]

Passenger perceptions of airlines differ substantially across air routes; therefore the competition among air carriers varies from route to route. For each route, airlines' relative positioning and airlines' perceived competitive strengths and weaknesses are distinct. The findings show that, when researching the international air travel market, one must explore airline positioning by routes. Based on the results of MDS and CA, the study therefore provides specific recommendations about how air carriers on each route can improve performance or build a differential edge. However, to develop and implement effective positioning strategies, Kim et al. (2007) suggested that the positioned objects should focus on only a small number of attributes when they build brand images or change their positions.

For the Taipei-Tokyo route, Japan Asia Airways is well positioned in on-board amenities, but it is in the same group with Air Nippon Airways in both methods; customers cannot readily distinguish the two airlines. As such, Japan Asia Airways needs to maintain its current level of on-board amenities and improve the attributes that are far from its point on the map. Japan Asia Airways must move away from the competitive group and position its corporate image differently from that of Air Nippon Airways. Whenever two or more airlines are located close to each other, they are all strongly similar with regard to some attributes. Thus, if such an airline wants to become an independent competitor or to distinguish its service from that of its competitors, that airline must find dissimilar attributes and differentiate its services by means of those dissimilar attributes.

[FIGURE 6 OMITTED]

A similar observation applies to other airlines. As Cathay Pacific Airways is positioned as a leader in entertainment facilities and Web site services by CA, Eva Airways is positioned as a leader in image by both methods. Eva Airways is very close to Cathay Pacific Airways; the performance of Eva Airways in entertainment facilities and Web site services is perceived as satisfactory. Cathay Pacific Airways should strengthen its advantages by offering more electronic commerce services for tourists and business travelers. Lastly, Cathay Pacific Airways needs to reinforce its Web site performance and reduce queue times to increase efficiency.

United Airlines in the Taipei-Tokyo route has a position in on-board amenities, but this factor is not close to United Airlines on the map. United Airlines' flight options are not extensive and its Tokyo line is not a central operating line because Taipei is no longer a United Airlines hub. Hence, customers have few travel experiences with United Airlines. United Airlines does not have many exceptional factors or attributes, and it should improve its services to increase competitiveness.

For the Taipei-Singapore route, Eva Airways and Singapore Airlines compete with each other, as indicated by the MDS analyses; they have similar levels of on-time performance, on-board amenities, and flight safety/corporate image. The CA result further shows that while Eva Airways is positioned as a leader in onboard amenities, Singapore Airlines is positioned as a leader in image. Because it has acquired many airline and tourism awards, Singapore Airlines is a primary choice for customers choosing an international airline. Singapore Airlines needs to maintain its brand image and enhance the service attributes that are distant from its point on the map. The positioning of China Airlines is similar on both routes; it is associated with service attitude of employees, frequency of flights, and convenience of reservation and ticketing. Most important, China Airlines must immediately improve its flight safety.

Jetstar Asia Airways in the Taipei-Singapore route is a low-cost carrier, providing few services to support the low price. Jetstar Asia Airways is positioned as a leader in price, as expected. In the results of both methods, Jetstar Asia Airways received high satisfaction in price, and thus the map associates Jetstar Asia Airways with low prices. However, Jetstar Asia Airways is far from all factors except price. Jetstar Asia Airways can maintain its business model, but it may face competition from other companies that adopt low-price strategies.

CONCLUSIONS

To face an increasingly competitive environment, air carriers must position their brands in passengers' minds to gain competitive advantage. Service differentiation is a crucial element to determine victory in the marketplace. The study identified the relative positioning of airlines based on examination of the airlines' performances on eighteen service attributes. Using data collected from Taiwanese international air passengers who have flown to Tokyo and to Singapore, this study employs two multivariate statistical methods, MDS and CA, to create perceptual maps. This study explores competitive standings, key strengths, and key weaknesses of each airline.

Both MDS and CA can create perceptual maps that illustrate airlines' relative positions and relations with service attributes. The results of MDS and CA are not entirely identical, but they are fairly similar even though CA and MDS require different data. Moreover, passenger perceptions of airlines are substantially different across air routes, and hence the competition among air carriers varies from route to route. This highlights the importance of route-based positioning. Each carrier should allocate marketing resources to maximize advantages relative to potential competitors in each route.

The main limitation of this research is that only two international air routes were selected for empirical study. Since international air routes are likely to have diverse passenger profiles and market characteristics, one might question the generalization of our results to other air routes. Future research might include a large-scale survey to collect data from a variety of air routes and make further comparisons of CA and MDS. The survey samples in this study included only Taiwanese air travelers who had flown from Taipei to Tokyo and to Singapore; foreign travelers were not interviewed. It is possible that some respondents favor national carriers (China Airlines and Eva Airways), which leads to home carrier bias. In the future, it may be worthwhile to collect both Taiwanese and non-Taiwanese air travelers' preferences and perceptions of different air carriers. Perceptions of the airlines differ for individuals with various socioeconomic and trip characteristics. To explore the relationships among airlines, service attributes, and individual characteristics, future research can use a multiple correspondence analysis (Greenacre 2006), which is an extension of simple CA for more than two categorical variables. The result of multiple correspondence analysis can help air carriers gain insights into their competitors, their services, and their passengers. In addition, travel demand could vary among airline passengers. Segmentation analysis might make it easier for air carriers to offer proper services and to target core markets. Future research might perform segmentation analysis for international air travel markets first, and then might carry out separate positioning studies on each segment, to identify comparative positions of the airlines and their service attributes. Research using segmentation first with subsequent positioning on separate segments might develop more effective marketing and operating strategies.

REFERENCES

Aaker, D.A., V. Kumar, and G.S. Day, 2007. Marketing Research. New York: John Wiley & Son.

Aksoy, S., E. Atilgan, and S. Akinci, 2003. "Airline Services Marketing by Domestic and Foreign Firms: Differences from the Customers' Viewpoint." Journal of Air Transport Management 9, pp. 343-351.

Cattell, R.B., 1966. "The Scree Test for the Number of Factors." Multivariate Behavior Research 1, pp. 245-276.

Cha, J., S. Kim, and Y. Lee, 2009. "Application of Multidimensional Scaling for Marketing-Mix Modification: A Case Study on Mobile Phone Category." Expert Systems and Applications 36, pp. 4884-4890.

Chen, F.Y. and Y.H. Chang, 2005. "Examining Airline Service Quality from a Process Perspective." Journal of Air Transport Management 11, pp. 79-87.

Dev, C.S., M.S. Morgan, and S. Shoemaker, S., 1995. "A Positioning Analysis of Hotel Brands: Based on Travel-Manager Perceptions." The Cornell Hotel and Restaurant Administration Quarterly 36, pp. 48-55.

Fourier, C. and B. Lubbe, 2006. "Determinants of Selection of Full-Service Airlines and Low-Cost Carriers--A Note on Business Travelers in South Africa." Journal of Air Transport Management 12, pp. 98-102.

Greenacre, M., 1984. Theory and Applications of Correspondence Analysis. London: Academic Press.

Greenacre, M., 2006. "From Simple to Multiple Correspondence Analysis." In: Greenacre, M. and Blazius, J. (Eds.), Multiple Correspondence Analysis and Related Methods. London: Chapman & Hall.

Gursoy, D., M.-H. Chen, and H.J. Kim, 2005. "The U.S. Airlines Relative Positioning Based on Attributes of Service Quality." Tourism Management 26, pp. 57-67.

Kaul, A. and V.R. Rao, 1995. "Research for Product Positioning and Design Decisions: An Integrative Review." International Journal of Research in Marketing 12, pp. 293-320.

Kaynak, E., O. Kucukemiroglu, and A. Kara, 1994. "Consumer's Perceptions of Airlines: A Correspondence Analysis Approach in a Global Airline Industry." Management International Review 34, pp. 235-254.

Kim, D.J., W.G. Kim, and J.S. Han, 2007. "A Perceptual Mapping of Online Travel Agencies and Preference Attributes." Tourism Management 28, pp. 591-603.

Kim, H., 1996. "Perceptual Mapping of Attributes and Preferences: An Empirical Examination of Hotel F&B Products in Korea." International Journal of Hospitality Management 15, pp. 373-391.

Kim, H., 1998. "Perceived Attractiveness of Korean Destinations." Annals of Tourism Research 25, pp. 340-361.

Kim, S.S. and J. Agrusa, 2005. "The Positioning of Overseas Honeymoon Destinations." Annals of Tourism Research 32, pp. 887-904.

Kohli, C.S. and L. Leuthesser, 1993. "Product Positioning: A Comparison of Perceptual Mapping Techniques." Journal of Product & Brand Management 2, 10-20.

Kruskal, J.B., and M. Wish, 1978. Multidimensional Scaling. California: Sage.

Liou, J.J.H. and G.-H. Tzeng, 2007. "A Non-Additive Model for Evaluating Airline Service Quality." Journal of Air Transport Management 13, pp. 131-138.

Myers, J.H., 1996. Segmentation and Positioning for Strategic Marketing Decisions. Chicago: American Marketing Association.

Nunnally, J.C., 1978. Psychometric Theory. New York: McGraw-Hill.

O'Connell, J.F. and G. Williams, 2005. "Passengers' Perceptions of Low Cost Airlines and Full Service Carriers: A Case Study Involving Ryanair, Aer Lingus, Air Asia and Malaysia Airlines." Journal of Air Transport Management 11, pp. 259-272.

Pakdil, F. and O. Aydin, 2007. "Expectations and Perceptions in Airline Services: An Analysis Using Weighted SERVQUAL Scores." Journal of Air Transport Management 13, pp. 229-237.

Park, J.-W., R. Robertson, and C.-L. Wu, 2004. "The Effect of Airline Service Quality on Passengers' Behavioural Intentions: A Korean Case Study." Journal of Air Transport Management 10, pp. 435-439.

Park, J.-W., 2007. "Passenger Perceptions of Service Quality: Korean and Australian Case Studies." Journal of Air Transport Management 13, pp. 238-242.

Proussaloglou, K. and F.S. Koppelman, 1999. "The Choice of Air Carrier, Flight, and Fare Class." Journal of Air Transport Management 5, pp. 193-201.

Surovitskikh, S. and B. Lubbe, 2008. "Positioning of Selected Middle Eastern Airline in the South African Business and Leisure Travel Environment." Journal of Air Transport Management 14, pp. 75-81.

Wen, C.-H., S.-C. Lai, and W.-Y. Yeh, 2008. "Segmentation and Positioning Analysis for International Air Travel Market: The Case of Taipei-Tokyo Route." Transportation Research Record 2052, pp. 46-53.

Wilkes, R.E., 1977. "Product Positioning by Multidimensional Scaling." Journal of Advertising Research 17, pp. 15-19.

Winer, R.S. and W.L. Moore, 1989. "Evaluating the Effects of Marketing-Mix Variables on Brand Positioning." Journal of Advertising Research 29, pp. 39-45.
Table 1. Studies of Airline Service Quality

                                            Aksoy       Park
                                            et al.     et al.
Attributes                                  (2003)     (2004)

1. Price                                      *
2. Convenience of flight schedule                        *
3. Frequency of flights                       *
4. Convenience of reservation and             *          *
   ticketing
5. Service attitude of reservation staff      *          *
6. On-time performance                        *          *
7. Queuing at check-in counter                           *
8. Service attitude at check-in counters      *          *
9. Neat and clean appearance and                         *
   employees
10. Comfort and spaciousness of seats         *          *
11. Food and beverages services on            *          *
    board
12. Cleanliness on board                      *
13. Entertainment facilities on board         *          *
14. Cordiality and kindness of cabin          *          *
    crews
15. Flight safety                                        *
16. Customer complaint handling               *          *
17. Web site services                         *
18. Image of airline                          *

                                             Chen
                                             and
                                            Chang       Park
Attributes                                  (2005)     (2007)

1. Price                                                 *
2. Convenience of flight schedule             *          *
3. Frequency of flights
4. Convenience of reservation and             *          *
   ticketing
5. Service attitude of reservation staff                 *
6. On-time performance                                   *
7. Queuing at check-in counter                *          *
8. Service attitude at check-in counters                 *
9. Neat and clean appearance and              *          *
   employees
10. Comfort and spaciousness of seats         *          *
11. Food and beverages services on            *          *
    board
12. Cleanliness on board                      *
13. Entertainment facilities on board         *          *
14. Cordiality and kindness of cabin          *          *
    crews
15. Flight safety                                        *
16. Customer complaint handling               *          *
17. Web site services
18. Image of airline                                     *

                                             Liou      Pakdil
                                             and        and
                                            Tzeng      Aydin
Attributes                                  (2007)     (2007)

1. Price
2. Convenience of flight schedule                        *
3. Frequency of flights                                  *
4. Convenience of reservation and             *          *
   ticketing
5. Service attitude of reservation staff      *          *
6. On-time performance                        *          *
7. Queuing at check-in counter                *          *
8. Service attitude at check-in counters      *          *
9. Neat and clean appearance and              *          *
   employees
10. Comfort and spaciousness of seats         *
11. Food and beverages services on            *          *
    board
12. Cleanliness on board                                 *
13. Entertainment facilities on board         *          *
14. Cordiality and kindness of cabin          *          *
    crews
15. Flight safety                             *
16. Customer complaint handling               *          *
17. Web site services
18. Image of airline                                     *

Table 2. Demographics and Trip Profiles of Respondents

                                              Taipei-
Characteristics               Taipei-Tokyo   Singapore

Gender
  Male                            41.1          56.4
  Female                          58.6          43.6
Age
  Under 30 years                  39.0          33.2
  31-40 years                     34.1          36.0
  41-50 years                     13.2          20.9
  51-60 years                     12.1           6.4
  Over 60 years                    1.6           3.5
Occupation
  Government worker                3.8           2.9
  Student                         10.2           7.8
  Business                        25.3          26.7
  Agriculture                      0.8           0.0
  Service industry                30.9          27.9
  Industrial                       7.0           8.1
  Self-employed                    4.3          11.9
  Housewife                        7.8           5.8
  Retired                          3.0           2.9
  Other                            7.0           5.8
Education
  Elementary                       1.1           0.0
  Middle high school               1.1           0.9
  High school                      8.6          11.9
  University-Undergraduate        73.1          54.7
  University-Graduate             16.1          32.6
Monthly income
  Under NT$ 10,000                12.4           7.0
  10,001-40,000                   32.2          26.0
  40,001-80,000                   39.5          46.5
  Over 80,000                     15.9          20.5
Number of previous trips
  0                                8.1           2.9
  1-3                             63.4          61.6
  4-6                             17.2          24.7
  7-9                              4.8           7.0
  10+                              6.5           3.8
Travel purpose
  Visit friends                    7.8          15.1
  Tourism                         68.5          60.8
  Business                        16.7          21.8
  Study aboard                     2.2           2.3
  Other                            4.8           0.0
Ticket booked
  Internet (airline direct)       11.0          25.6
  Travel agency                   78.8          64.2
  Telephone                        6.7           7.3
  Other                            3.5           2.9
Flight allowance
  First class                      1.2           0.0
  Business class                   9.1           9.0
  Economy class                   89.7          91.0

Table 3. Mean Scores and Ranking of Service Attributes

                                                   Taipei-Tokyo

Attributes                                       Mean     Ranking

V1. Price                                        5.46        17
V2. Convenience of flight schedule               5.62        12
V3. Frequency of flights                         5.12        18
V4. Convenience of reservation and ticketing     5.52        15
V5. Service attitude of reservation staff        5.72        10
V6. On-time performance                          6.10        3
V7. Queuing at check-in counter                  5.55        14
V8. Service attitude at check-in counters        5.85        8
V9. Neat and clean appearance and employees      5.62        13
V10. Comfort and spaciousness of seats           5.89        6
V11. Food and beverages services on board        5.66        11
V12. Cleanliness on board                        6.02        5
V13. Entertainment facilities on board           5.49        16
V14. Cordiality and kindness of cabin crews      6.11        2
V15. Flight safety                               6.41        1
V16. Customer complaint handling                 6.04        4
V17. Web site services                           5.73        9
V18. Image of airline                            5.85        7

                                                 Taipei-Singapore

Attributes                                       Mean     Ranking

V1. Price                                        5.87        11
V2. Convenience of flight schedule               5.76        15
V3. Frequency of flights                         5.63        17
V4. Convenience of reservation and ticketing     5.81        14
V5. Service attitude of reservation staff        5.98        10
V6. On-time performance                          6.09        6
V7. Queuing at check-in counter                  5.74        16
V8. Service attitude at check-in counters        5.85        13
V9. Neat and clean appearance and employees      5.85        13
V10. Comfort and spaciousness of seats           6.14        5
V11. Food and beverages services on board        5.86        12
V12. Cleanliness on board                        6.19        4
V13. Entertainment facilities on board           6.00        8
V14. Cordiality and kindness of cabin crews      6.20        3
V15. Flight safety                               6.59        1
V16. Customer complaint handling                 5.99        9
V17. Web site services                           6.06        7
V18. Image of airline                            6.26        2

Table 4. Factor Analysis of Service Attributes for Taipei-Tokyo Route

                                                 Factor     Cronbach's
Items                                           loading       alpha

Factor 1: Price and flights                                    0.85

V1. Price                                        0.541
V2. Convenience of flight schedule               0.739
V6. On-time performance                          0.470
V3. Frequency of flights                         0.661

Factor 2: Ground service                                       0.80

V4. Convenience of reservation and ticketing     0.612
V5. Service attitude of reservation staff        0.649
V7. Queuing at check-in counter                  0.463
V8. Service attitude at check-in counter         0.673
V9. Neat and clean appearance of crews           0.727

Factor 3: On-board amenities                                   0.85

V10. Comfort and spaciousness of seats           0.792
V11. Food and beverages services on board        0.747
V12. Cleanliness on board                        0.730
V13. Entertainment facilities on board           0.660
V14. Cordiality and kindness of cabin crews      0.615

Factor 4: Safety and image                                     0.81

V15. Flight safety                               0.720
V16. Customer complaint handling                 0.719
V17. Web site services                           0.675
V18. Image of airline                            0.671

                                                            Percentage
                                                                of
                                                             variance
Items                                          Eigenvalue   explained

Factor 1: Price and flights                      6.878        38.212

V1. Price
V2. Convenience of flight schedule
V6. On-time performance
V3. Frequency of flights

Factor 2: Ground service                         1.519        8.437

V4. Convenience of reservation and ticketing
V5. Service attitude of reservation staff
V7. Queuing at check-in counter
V8. Service attitude at check-in counter
V9. Neat and clean appearance of crews

Factor 3: On-board amenities                     1.226        6.813

V10. Comfort and spaciousness of seats
V11. Food and beverages services on board
V12. Cleanliness on board
V13. Entertainment facilities on board
V14. Cordiality and kindness of cabin crews

Factor 4: Safety and image                       1.092        6.067

V15. Flight safety
V16. Customer complaint handling
V17. Web site services
V18. Image of airline

Table 5. Factor Analysis of Service Attributes for
Taipei-Singapore Route

                                                 Factor     Cronbach's
Items                                           loading       alpha

Factor 1: Price and flights                                    0.72

V1. Price                                        0.525
V2. Convenience of flight schedule               0.869
V3. Frequency of flights                         0.853

Factor 2: On-time performance and                              0.81
  reservation service

V4. Convenience of reservation and ticketing     0.571
V5. Service attitude of reservation staff        0.767
V6. On-time performance                          0.870
V16. Customer complaint handling                 0.487

Factor 3: Ground service                                       0.78

V7. Queuing at check-in counter                  0.597
V8. Service attitude at check-in counter         0.858
V9. Neat and clean appearance of crews           0.799

Factor 4: On-board amenities                                   0.88

V10. Comfort and spaciousness of seats           0.795
V11. Food and beverages services on board        0.789
V12. Cleanliness on board                        0.806
V13. Entertainment facilities on board           0.683
V14. Cordiality and kindness of cabin crews      0.711

Factor 5: Safety and image                                     0.62

V15. Flight safety                               0.599
V17. Web site services                           0.590
V18. Image of airline                            0.762

                                                            Percentage
                                                                of
                                                             variance
Items                                          Eigenvalue   explained

Factor 1: Price and flights                      2.242        12.454

V1. Price
V2. Convenience of flight schedule
V3. Frequency of flights

Factor 2: On-time performance and                1.288        7.156
  reservation service

V4. Convenience of reservation and ticketing
V5. Service attitude of reservation staff
V6. On-time performance
V16. Customer complaint handling

Factor 3: Ground service                         1.529        8.494

V7. Queuing at check-in counter
V8. Service attitude at check-in counter
V9. Neat and clean appearance of crews

Factor 4: On-board amenities                     6.374        35.413

V10. Comfort and spaciousness of seats
V11. Food and beverages services on board
V12. Cleanliness on board
V13. Entertainment facilities on board
V14. Cordiality and kindness of cabin crews

Factor 5: Safety and image                       1.051        5.837

V15. Flight safety
V17. Web site services
V18. Image of airline
Gale Copyright:
Copyright 2010 Gale, Cengage Learning. All rights reserved.