Generally, frequency is used for looking at detailed information on nominal (category) data and describing the results. Table 1 show the study sample comprises of 220 respondents, which vary on characteristics such as gender, age, educational level, marital status, ethnicity, employment status, income level, airline website that have ever visited and number of online purchases in airasia. com in last one year. The following table 4. 1 summarizes the socio demographic profiles of the respondents of this study.
Table 4. 1 Demographic profiles of respondents
Category
Frequency
Percentage (%)
Gender
Male
71
32. 3
Female
149
67. 7
Age
Below 20
19
8. 6
20-29
181
82. 3
30-39
14
6. 4
40-49
4
1. 8
Above 50
2
0. 9
Educational Level
High School Certificate
51
23. 2
Diploma
18
8. 2
Bachelor Degree
138
62. 7
Master Degree
13
5. 9
PhD (Doctorate)
0
0
Continue…
Continue…
Marital Status
Single
192
87. 3
Married
28
12. 7
Ethnicity
Malay
81
36. 8
Chinese
117
53. 2
Indian
13
5. 9
Others
9
4. 1
Employment status
Employed
140
63. 6
Students
80
36. 4
Income Level
Below RM1500
88
40. 0
RM1501-RM3000
77
35. 0
RM3001-RM4500
31
14. 1
RM4501-RM6000
7
3. 2
RM6001-RM7500
6
2. 7
RM7501-RM9000
1
0. 5
RM9001-RM10500
1
0. 5
Above RM10500
9
4. 1
Airline website that you have ever visited.
Airasia. com
220
100. 0
Malaysiaairlines. com
49
22. 3
fireflyz. com
10
4. 5
Others
9
4. 1
Number of online purchases in airasia. com in last one year.
1-3 times
172
78. 2
4-6 times
29
13. 2
7-9 times
9
4. 1
10-12 times
5
2. 3
13-15 times
1
0. 5
Above 16 times
4
1. 8
In term of gender, the sample indicates that female is more than male, 149(67. 7%) is made up of females and the rest 71(32. 3%) are males. The imbalance number of respondents between male and female could be due to the sampling process and females are more willing to answer surveys compared to males.
All of the respondents are above the age of 18 years old and they are predominantly young people with age of 20 to 29 years old (82. 3%). The second biggest number of respondents comprised of individuals in below 20 (8. 6%) age group. With respect to of educational level, more than 60 % of the respondents hold at least bachelor degree which has 62. 7%, followed by high school certificate 23. 2%, 8. 2% are diploma and 5. 9% are master degree.
Concerning the marital status of the respondents the majority 192(87. 3%) of the respondents are single and only 28(12. 7%) of them are married. This can be due to fact that the majority of survey respondents are young people who are aged between 20 to 29(70. 2%) years old.
With respect to the ethnic groups of the survey sample, the majority 117(53. 2%) of respondents is made up of Chinese, followed by 81(36. 8%) Malay, 13(5. 9%) Indians and 9(4. 1%) is others which consist of foreigners. The high Chinese respondents reported are slightly imbalanced, however it can be considered as a representative of Malaysian population composition.
Meanwhile, with regard to the employment status of the respondents more than 60% of them are employed. Working respondents consists of 63. 6% followed with the student 36. 4%. It is notable that household income which is below RM1500 is 40%, 35% are between RM1501 to RM3000, 14. 1% are between RM3001-RM4500, 3. 2% are RM 4501 to RM6000, 2. 7% are RM6001 to RM7500 and only 0. 5% are RM7501 to RM9000 and RM9001 to RM10500 while the remaining 4. 1% of respondents earned more than RM10501 a month. This is could be due to the fact that most of the respondents are young adults such as student.
While for the airline website that respondents have ever visited, all of the respondents had experience of visited airasia. com. This is due to the questionnaires only distributed to respondents who had online shopping experiences at airasia. com. 49 respondents had experiences of visit website of Malaysia Airline, 10 respondents had visit fireflyz. com and only 9 respondents visit other airline website. This is because there are 4. 1% of respondents who are non Malaysian and they had experience of online shopping with other airline website in their origin country.
Lastly, for the number of online purchases in airasia. com in last one year, large majority of 78. 2% of respondents purchase 1 to 3 times Air Asia airline ticket in last one year. 13. 2% of them purchase 4 to 6 times in airasia. com, 4. 1% of them purchase 7 to 9 times while the remaining 4. 6% of respondents purchase more than 10 times airline tickets.
4. 2 IDENTIFY THE UNDERLYING FACTORS
Factor analysis is used for data reduction and examine how underlying that the questions in the questionnaires asked is relating to the construct that need to study. There are two purposes in doing exploratory factor analysis; one is to identify the representative variables to create new variables the other purpose is to determine factors that would account maximum variance in the study that can be use for multivariate analysis (Quester & Lim 2003).
Factor analysis was conducted for all variables. In this study, two variable had been tested which were internet apprehensiveness and website quality. Variables of internet apprehensiveness contain 14 items while for website quality has 29 items. All the variables included in this study were tested by previous studies. Two separate sets of factor analyses were conducted so that factors were correlated that can produce the conceptual similarities within the two set of scale items (Norusis, 1993).
The eigenvalues setting at 1. 0 which mean factor with a variance greater than 1. 0 are retained. Kaiser (1974) recommends that the KMO accepting values greater than 0. 5 as acceptable. Furthermore, values between 0. 5 and 0. 7 are mediocre, values between 0. 7 and 0. 8 are good, values between 0. 8 and 0. 9 are great and values above 0. 9 are superb. The KMO value must be 0. 50 and above to retained to obtain 80% of significant level. Items with KMO below than 0. 5 were deleted (Hair et al 1998). Factor and item retention in this study were based on:
Items not display when the cross-loadings greater than 0. 40 with other factors.
Items exhibiting principal factor loadings approximating 0. 5 or above
A reliability coefficient for the aggregated scale of 0. 70 or greater.
4. 2. 1 Factor Analysis on Internet Apprehensiveness
Table 4. 2 shows the factor analysis performed on 14 items resulted in two factors explaining 66. 6% of the overall variance. In this study, the KMO is 0. 913, which fall into the range of superb indicating that all the variables for internet apprehensiveness were interrelated. So the factor analysis is appropriate for these data. For these data, Bartlett’s test is highly significant which is p= 0. 000 and therefore factor analysis is appropriate. Two factors were structured in Internet Apprehensiveness which was Transactional Internet Apprehensiveness and General Internet Apprehensiveness.
Table 4. 2 Factor analysis on internet apprehensiveness
Variables
IA1
IA2
Factor 1: Transactional Internet Apprehensiveness
I have fear of using the internet to make on-line purchases
0. 893
I am afraid to make on-line purchases at times
0. 890
I have fear of making on-line purchases.
0. 877
I am worried when using the internet to purchase products or services.
0. 846
I dislike using the internet to make online purchases.
0. 844
Ordinarily, I am not calm when making on-line purchases.
0. 842
I feel uncomfortable using the internet to make on-line purchases under RM1, 000
0. 793
I feel uncomfortable using the internet to make on-line purchases over RM1, 000
0. 673
The security of my credit card for use with on-line purchases concerns me.
0. 560
Factor 2: General Internet Apprehensiveness
I dislike using the internet for a variety of reasons.
0. 881
I am usually not calm while using the internet.
0. 862
Communicating with the internet usually makes me uncomfortable.
0. 851
Generally, I am uncomfortable using the internet to gather information.
0. 725
I would not use the internet to purchase airline tickets, book hotel rooms, or other travel-related service.
0. 520
Eigenvalue
6. 55
2. 773
Percentage of Variance
46. 786
19. 811
Cronbach Alpha(Reliability)
0. 934
0. 827
Bartlett’s Test of Sphericity-sig. 0. 000
Percentage of Cumulative Variance: 66. 597
KMO: 0. 913
4. 2. 2 Factor Analysis on Website Quality
Table 4. 3 Factor analysis on Website Quality
Variables
WQ1
WQ2
WQ3
WQ4
The information provided at airasia. com is clear to me.
. 822
The interactions with airasia. com are clear and understandable.
. 805
airasia. com is user friendly.
. 803
airasia. com is easy to use.
. 775
The information provided at airasia. com is reliable.
. 771
The information on airasia. com is complete for my purchase decisions.
. 762
I feel happy when I use airasia. com.
. 761
I believe the airasia. com provides accurate information to potential customers like me.
. 758
airasia. com is flexible to interact with.
. 745
The information in airasia. com is relevant.
. 711
The information provided on airasia. com is easily understandable.
. 601
I find it easy to obtain information from airasia. com.
. 590
. 528
I can easily find what I need on airasia. com.
. 531
. 503
airasia. com uses good color combinations.
. 829
I like the color combination of airasia. com
. 806
airasia. com is creative in design.
. 786
I like the layout of airasia. com.
. 773
I found it easy to move around in airasia. com
. 556
The start page leads me easily to the information I need.
. 536
. 527
The website and all of its linked pages work well.
Search engine provides accurate results
. 765
Is easily accessed via search engines
. 596
Can be accessed from a variety of other related websites
. 536
I can find all the detailed information I need.
. 525
Continue…
Continue…
The start page tell me immediately where I can find the information I am looking for.
. 503
airasia. com loads quickly
. 823
When I use airasia. com there is very little waiting time between my actions and the website’s response.
. 788
All my business can be completed via airasia. com
. 621
Most business processes can be completed via airasia. com
. 537
Eigenvalue
14. 729
2. 705
1. 271
1. 134
Percentage of Variance
50. 790
9. 329
4. 384
3. 910
Cronbach Alpha(Reliability)
0. 959
0. 913
0. 897
0. 831
Bartlett’s Test of Sphericity-sig. 0. 000
Percentage of Cumulative Variance: 68. 412
KMO: 0. 938
Table 4. 3 shows the factor analysis results on 29 items that produced only four distinct factors that explaine 56. 71% of the overall variance. One item has been discarded from further analyses because of factor loadings of that item are below 0. 50 which were not acceptable according to Lee and Crompton (1992). Items that are cross loadings were deleted, so in this study three items were deleted and not included in further analysis. The Kaiser-Meyer-Olkin(KMO) was 0. 938 and the Bartlett’s Test of Sphericity were significant which is 0. 000 and shows that the items were appropriate in factor analysis (Hair et al. 1998). For the remaining items, the factor analysis was between 0. 503 and 0. 829. The eigenvalues for component 1, component 2, component 3 and components 4 were 14. 729, 2. 705, 1. 271 and 1. 134 respectively.
Table 4. 4 Factor analysis on website quality re-run with the remaining 24 items
Variables
WQ1
WQ2
WQ3
WQ4
Factor 1: Website Usability
The information provided at airasia. com is clear to me.
0. 826
airasia. com is user friendly.
0. 812
The interactions with airasia. com are clear and understandable.
0. 811
airasia. com is easy to use.
0. 787
Continue…
Continue…
The information provided at airasia. com is reliable.
0. 772
The information on airasia. com is complete for my purchase decisions.
0. 771
I believe the airasia. com provides accurate information to potential customers like me.
0. 769
I feel happy when I use airasia. com.
0. 758
airasia. com is flexible to interact with.
0. 750
The information in airasia. com is relevant.
0. 711
The information provided on airasia. com is easily understandable.
0. 614
Factor 2: Website Design
airasia. com uses good color combinations.
0. 840
I like the color combination of airasia. com
0. 818
airasia. com is creative in design.
0. 774
I like the layout of airasia. com.
0. 767
I found it easy to move around in airasia. com
0. 553
Factor 3: Accessibility
Search engine provides accurate results
0. 763
Is easily accessed via search engines
0. 638
Can be accessed from a variety of other related websites
0. 590
I can find all the detailed information I need.
0. 515
Factor 4: Transactional Capabilities
airasia. com loads quickly
0. 830
When I use airasia. com there is very little waiting time between my actions and the website’s response.
0. 799
All my business can be completed via airasia. com
0. 619
Most business processes can be completed via airasia. com
0. 521
Eigenvalue
13. 055
2. 604
1. 209
1. 117
Percentage of Variance
50. 213
10. 015
4. 652
4. 298
Cronbach Alpha(Reliability)
0. 956
0. 904
0. 771
0. 831
Bartlett’s Test of Sphericity-sig. 0. 000
Percentage of Cumulative Variance: 69. 177
KMO: 0. 930
The first run of factor analysis did not produce a clean factor structure and there are three items were cross loadings, factor analysis had to re-run again in order to get better result. Table 4. 4 shows the factor analysis results on website quality construct resulted in four factor explaining 69. 17% of the total variance explained. The second run of factor loadings on the remaining 24 items were between 0. 515 and 0. 840. The Kaiser-Meyer-Olkin (KMO) was at a superb value of 0. 930 and the Bartlett’s Test of Sphericity was significant that is 0. 000 at the 5% level of significance. Therefore, the use of factor analysis was suitable in this study. The factor loading for each factor was high and stands alone in one factor. Therefore, the four dimensions in website quality were independently structured. For the first factor which is Website Usability contained 11 items with an eigenvalue of 13. 055 and has 50. 213 percent of the variance in the data. While for the second factor that is Website Design contained 5 items which has an eigenvalue of 2. 604 and explained 10. 015 percent of variance. The third factor Accessibility that has 4 items explaining 4. 652 percent of variance with an eigenvalue of 1. 209. Lastly, for the factor of Transactional Capabilities also has 4 items with an eigenvalue of 1. 117 with 4. 298 percent of variance.
4. 2. 3 Factor Analysis on Satisfaction
Table 4. 5 Factor analysis on satisfaction
Variables
Factor Loadings
My choice to visit airasia. com was a wise one.
0. 909
I am satisfied with my recent decision to purchase from airasia. com
0. 900
Overall, I was satisfied with airasia. com
0. 898
I think I did the right thing by visiting airasia. com.
0. 888
My choice to purchase from airasia. com was a wise one.
0. 872
I recommend airline. com to my colleagues.
0. 871
I recommend airasia. com to my friends.
0. 862
I have truly enjoyed purchasing from airasia. com
0. 858
I was satisfied with online buying when compared to offline buying.
0. 816
Eigenvalue
7. 45
Percentage of Variance
74. 497
Bartlett’s Test of Sphericity-sig. 0. 000
Percentage of Cumulative Variance: 74. 497
KMO: 0. 929
Table 4. 5 above summarized the result of factor analysis on satisfaction that produced only one factor. The result also shows that the ten statements explained 74. 50% of total variance and the value of KMO is 0. 929 which means this data is appropriate for factor analysis. All of the factor loadings in satisfaction were greater than 0. 80 that are between 0. 816 and 0. 909. The Bartlett’s Test of Sphericity is significant which 0. 000 at the 5% level which means the items reject the null hypothesis.
4. 3 RELIABILITY ANALYSIS
Table 4. 6 Result of reliability test
Variable
Number of items
Cronbach’s Alpha
General Internet Apprehensiveness
5
0. 827
Transactional internet apprehensiveness
9
0. 934
Website Usability
11
0. 956
Website Design
5
0. 904
Accessibility
4
0. 771
Transactional Capabilities
4
0. 831
Satisfaction
10
0. 961
The result of reliability analysis is presented in table 4. 6 above. A Conbach’s alpha reliability coefficient was used to check the reliability of the questionnaires (Cronbach, 1951). The Cronbach alpha values were conducted as 0. 827, 0. 934, 0956, 0. 904, 0. 771 0. 831 and 0. 961 for Transactional Internet Apprehensiveness (IA1), General internet apprehensiveness (IA2), Website Usability (WQ1) , Website Design (WQ2), Accessibility (WQ3, Transactional Capabilities (WQ4) and Satisfaction (SAT) respectively.
All of the Cronbach’s Alpha values are greater than 0. 7 which are exceed the acceptable value of 0. 7 and above suggested by (Hair et al. 1998). None of the above variable will be deleted according to Nunally (1978) which suggested that the Cronbach’s Alpha less than 0. 50 will be deleted. The result of reliability test in this study revealed that all the seven factors fulfilled the requirements.
4. 4 REGRESSION
Linear Regression was used in this study to measure the significance of the relationship between scales of Transactional Internet Apprehensiveness, General internet apprehensiveness, Website Usability, Website Design, Accessibility and Transactional Capabilities, which were identified in literature review. The results from the regression analysis are presented in the table 4. 7 below.
Table 4. 7 Regression analysis
Independent
Unstd. B
Std. Beta
T
Sig.(P value)
Constant
. 370
General Internet Apprehensiveness
. 002
. 002
. 041
0. 967
Transactional internet apprehensiveness
-. 059
-. 075
-1. 571
0. 118
Website Usability
. 437
. 393
5. 712
0. 000
Website Design
. 157
. 158
2. 364
0. 019
Accessibility
. 249
. 219
3. 310
0. 001
Transactional Capabilities
. 110
. 109
1. 734
0. 967
R= 0. 753 R Square= 0. 567
Sig. F= 0. 000
From the result obtained from analysis of linear regression in table 4. 7 above shows that three attributes which was Website Usability (0. 000), Website Design (0. 019) and Accessibility (0. 001) are all significance and positive relationship with satisfaction at the significant level of p < 0. 05. Therefore it can be concluded that three attributes are significantly associated with the satisfaction of air passengers towards airasia. com. Thus, null hypothesis is rejected, and it is concluded that there is significant relationship between independents variables and dependent variable.
In the other hand, three attributes which was General Internet Apprehensiveness, Transactional internet apprehensiveness and Transactional Capabilities have no effect on customer satisfaction.
The R-Square is 0. 567 explains 56. 7% of the dependent variable is explained by the independent variable and another 43. 3% is not explained. The output from the regression analysis shows that Website Usability has the largest absolute value of standardized beta coefficient (β= 0. 393) emerges as the most important factor that effect air passengers satisfaction towards Air Asia’s website. Website Usability has the strongest effect and become the important role of air travelers’ satisfaction which accounted 39. 3%. The data also indicated that Accessibility is the second most important element driving customers’ satisfaction (β= 0. 249). Accessibility is followed by Website Design (β= 0. 157), Transactional Capabilities (β= 0. 110) and General Internet Apprehensiveness (β= 0. 002). The results also indicate that there is a significant negative relationship between Transactional internet apprehensiveness (β=-0. 059) and the satisfaction towards airline website.
4. 5 DIFFERENCE IN AIR PASSENGERS’ SATISFACTION BASED ON SOCIO DEMOGRAPHIC
Independent T-test and One way ANOVA analysis were conducted to assess whether air passengers’ satisfaction towards airasia. com will differ according to their socio-demographic characteristics.
4. 5. 1 T -Test
T-Test is normally used when there are only two values in a variable. In this study, Independent T-Test was used to compare means among gender, marital status and employment status to know is there any difference between demographic profile of respondents and factors.
In T-test, items that have a probability that is less than 0. 05, null hypothesis of equal variance will be reject and the value of T of items are based on equal variance not assumed will be utilized. Therefore, if the items have a probability that is more than 0. 05 confidence level and for the value of T, those items are based on the equal variance assumed will be utilized and indicates that the variance of the two samples is approximately equal.
Table 4. 8 T-Test result between gender and factors that influence customers’ satisfaction.
I tem
Male
Female
T
Sig
General Internet Apprehensiveness
2. 18
2. 26
-. 648
. 227
Transactional internet apprehensiveness
3. 24
3. 92
-4. 452
. 549
Website Usability
4. 37
4. 19
1. 630
. 472
Website Design
3. 98
4. 14
-1. 303
. 448
Accessibility
4. 20
4. 11
. 820
500
Transactional Capabilities
3. 73
3. 73
. 048
. 325
Table 4. 8 above shows the result between gender and six dimensions that effect customers’ satisfaction towards airasia. com. Table 4. 8 indicates that none of the values give a probability that is less than the significant level of 0. 05. Thus, the null hypothesis of equal mean is not rejected. In other word, differences between gender and factors were not found.
Table 4. 9 T-Test result by marital status and factors that influence customers’ satisfaction.
I tem
Single
Married
T
Sig
General Internet Apprehensiveness
2. 21
2. 41
-1. 041
0. 12
Transactional internet apprehensiveness
3. 69
3. 77
-0. 381
0. 88
Website Usability
4. 24
4. 33
-0. 776
0. 035
Website Design
4. 08
4. 15
-0. 393
0. 082
Accessibility
4. 14
4. 11
0. 218
0. 1
Transactional Capabilities
3. 72
3. 76
-0. 202
0. 847
The result of t-test between marital status and factors influencing consumer satisfaction towards airline website is presented in table 4. 9 above indicates that the value of T for Website Usability is -0. 776 and gives a probability of 0. 035 which shows that website usability was significantly difference at 0. 05 confidence level. Thus, the null hypothesis of equal mean is rejected. The remaining five dimensions are not significant difference meaning that null hypothesis for those factors is not rejected.
Table 4. 10 T-Test by Test result by employment status and factors that influence customers’ satisfaction.
I tem
Employed
Student
T
Sig
General Internet Apprehensiveness
2. 13
2. 42
-2. 150
0. 008
Transactional internet apprehensiveness
3. 54
3. 98
-2. 850
0. 503
Website Usability
4. 24
4. 27
-. 280
0. 694
Website Design
4. 05
4. 16
-. 941
0. 993
Accessibility
4. 15
4. 11
. 352
0. 286
Transactional Capabilities
3. 67
3. 84
-1. 437
0. 342
In general, the results in table 4. 10 indicated only General Internet apprehensiveness was significant differences that gives the probability of 0. 008 and the value of T is -0. 2150. The other five dimensions were found that no significantly difference between the working people and students.
4. 5. 2 One Way ANOVA
One Way ANOVA is used when there are three or more values in a variable and used to compares mean of one group or more groups based on independent variable. In this study, One way ANOVA is used to analyze between the factors that affect air travelers’ satisfaction towards an airline website with respondents’ demographic profile such as age, education level, ethnicity, household income and Numbers of online Shopping Experience at airasia. com.
Table 4. 11 One Way ANOVA between age group
Variable
Age (Mean)
F
Sig.
< 20
20-29
30-39
40-49
> 50
General Internet Apprehensiveness
2. 68
2. 16
2. 39
2. 50
3. 30
2. 283
0. 61
Transactional internet apprehensiveness
3. 94
3. 67
3. 66
3. 69
3. 89
0. 265
0. 90
Website Usability
3. 83
4. 27
4. 48
4. 25
4. 73
1. 954
0. 103
Website Design
3. 88
4. 08
4. 40
4. 50
4. 00
0. 958
0. 431
Accessibility
3. 92
4. 15
4. 18
4. 25
4. 50
0. 539
0. 707
Transactional Capabilities
3. 49
3. 77
3. 46
3. 88
3. 88
0. 863
0. 487
Table 4. 11 displayed above shows that there is no significant level of 0. 05 for the different age groups. Thus, the null hypothesis of equal mean is accepted
Table 4. 12 One Way ANOVA between educational levels.
Variable
Educational level (Mean)
F
Sig.
High school
Diploma
Bachelor Degree
Master Degree
General Internet Apprehensiveness
2. 47
2. 60
2. 15
1. 71
3. 981
0. 009
Transactional internet apprehensiveness
4. 10
3. 85
3. 59
3. 09
4. 259
0. 006
Website Usability
4. 08
4. 20
4. 29
4. 61
1. 881
0. 134
Website Design
4. 02
4. 00
4. 09
4. 55
1. 452
0. 229
Accessibility
3. 88
4. 04
4. 20
4. 62
4. 304
0. 006
Transactional Capabilities
3. 71
3. 57
3. 88
3. 73
0. 375
0. 771
Table 4. 12 above shows the F value for Transactional internet apprehensiveness and Accessibility are 4. 259 and 4. 304 respectively. Both variables result a probability of 0. 006 which is less than a significant level of 0. 05. Besides that, that, General Internet Apprehensiveness also shows a significant difference that has a probability of 0. 009 and the value of F was 3. 981. Therefore, the null hypothesis is rejected and this proposes that there are significant differences among different education levels and Transactional internet apprehensiveness, General Internet Apprehensiveness and Accessibility. The remaining three variable shows that there is no significance differences among different education levels and Website Usability, Website Design and Transactional Capabilities.
Table