and for leisure. Moreover, the target group is very likely to travel extensively during their 40+
years careers. Understanding needs of this target group would help carriers to tailor their services
for this new generation and thus ensure customer satisfaction and loyalty in a longer run as well.
Hence, this survey aims to help a carrier to understand one of the customer groups and to make
sure that it attracts and retains only profitable customers, which the company is capable of
serving well. The survey will show what this particular customer segment values in service.
Majority of the respondents were or are studying at Aalto University School of Economics
or/and are members of CEMS Master International Management double degree program. In
other cases, the respondents are studying at or have graduated from other Finnish/European
universities. However, the survey did not make any distinction of the respondents’ nationality.
However, nationality may affect perception of service quality (as found by Gilbert & Wong in
2003), which could be taken into consideration in further research.
4.3.
Survey Scales
Seven-point Likert scale is used to measure service quality attributes in the survey. Seven-point
scale is the maximum scale to make it easy for people to identify their opinion (Likert Scale &
Surveys, Best Practices, 2007). On the other hand, five-point scale is considered the simplest, but
in this case, it is insufficient for mapping the results on Marilla and James’s (1977) scale.
Furthermore, Matell & Jacoby (1972) state that if the number of scale steps decreases, the use of
mid-point category decreases as well. The authors recommend to use either no neutral scale at all
or to use a scale with many points. However, Garland (1991) argues that whether or not neutral
point is used, the results will be distorted, and also Worcester & Burns (1975) suggest that in
case of not using neutral point, the results will be distorted towards positive side of scale.
Based on this, the scale used for the survey is balanced, i.e. neutral point is used for the
importance scale (measuring service quality attributes, found in part 2 of the survey) and the
scale points are taken from Vagias (2006). Agreement and frequency scale (questions 7 and 8
respectively) were kept at four points for simplicity (Dataguru.org). The summary of the survey
scales can be found in Table 3:
5. RESEARCH RESULTS
This chapter focuses on presenting empirical findings of the survey. The objective of the
empirical research is to test the service quality attributes and dimensions on a defined target
group and to assess what features of airline service process are most important for the target
group.
The survey was available for answering for two weeks, and 79 answers were received. In the
analysis, firstly, the respondents were profiled based on their gender and age as well as on their
profile as airline passengers. Second, the attributes of service quality were discussed and
analyzed. Last, the most interesting information from open questions was summarized. The
empirical research was carried out with an online survey using www.WebropolSurveys.com.
Some of the answers were collected using a printed version of the survey during the Logistics
Master’s Thesis Seminar at Aalto University School of Economics, to increase a response rate.
Out of many methods of collecting the information, a survey was chosen due to its relative
simplicity, the ability to collect qualitative data and ease of access to wide audience located in
different countries.
An analysis of the survey data was performed with Microsoft Excel as well as with Webropol
Insight statistical analysis tool. Spearman correlation was used in analysis, as it is more
appropriate for measures that are taken from ordinal scales such as Likert scale (discussed in
more details in chapter 4.3). Spearman correlation coefficient shows statistical dependence of
two variables, and statistical correlation is significant if coefficient is more than 0.5. Moreover, if
p-value is low (under 0.05), it means that the correlation actually exists. In case p-value is high,
it cannot be determined that the correlation actually exists. Furthermore, to measure differences
between respondent groups (such as female vs. male respondent group), Chi 2 test was chosen as
this test is best suited since it compares differences between the answers of two or more
independent groups (The Chi Squared Statistics, 2012). In Chi2 test, the results with p-value less
than 0.05 were considered statistically significant. In these cases the probability that the
differences were caused by chance alone are small. However, it is important to note that
mathematical correlation does not necessarily indicate any cause-and-effect relationship.
As it can be seen from Figure 8, majority of the respondents (55, or 70%) were from 25-29 years
age group. That well corresponds to the target group to be young professionals/graduate students
on their last years of study. As the majority of the respondents are Finns, average starting age for
university education is 21 years old (HS.fi, 2008), and Master of Science level education
normally takes five years. However, 30% of respondents belong to other age groups.
Furthermore, Figure 9 shows that the main reason to use air transportation for the survey
respondents is leisure flights. However, it does not deny the fact that 85% of respondents (or 67
out of 79) who chose leisure or personal reasons/commuting do not fly on business at all. Most
of them are very likely to combine more than one reason to use air transportation to some extent.
Out of 12 respondents who mostly fly on business, 11 were male (32.35% of all male
respondents) and only one was female (2.22% of all female respondents). On leisure flights,
differences were significant as well: 50% of male respondents and 77.78% of female respondents
fly mostly on leisure (data can be found in Appendix 3).
Somewhat predictable were the results of the question which airline the respondents use the most
(respondents could select up to three options). Finnair and Blue1 were the leaders, with 32 and
27 answers respectively, due to the survey was carried out in Finland. Low cost airlines tend to
be popular as well, though geographical access and a number of destinations (using smaller
airports and point-to-point transit system, which limits its reach) limit them. Ryanair scored 19
answers and Norwegian 20 answers (though Norwegian cannot be classified as pure low cost
airline as it actively implements hub-and-spoke model with major national airports; Ryanair on
the other hand is a pure example of low cost carrier).
Furthermore, the survey shows that 24 out of 79 respondents (or 30%) are not members of any
frequent flyer program. Finnair Plus again had the largest number of respondents (29, or 37%),
which well corresponds to Finnair being the most popular among the respondents. Only 19
respondents (24%) indicated that the membership affects their selection of a carrier. It is
especially common in business travelers as it was found out from discussions with some of
survey respondents.
the travel frequency among the respondents was divided rather equally (Figure 10): 41
respondents (52%) travel from 1 to 5 times a year, whereas 38 respondents (48%) travel more
than 6 times a year, and can be considered frequent flyers.