Sei sulla pagina 1di 13

DATA ANALYSIS AND FINDINGS

1.1

Introduction

This chapter presents the empirical results of the main study. It is divided into four sections. The first section describes the response rate and an overview of the respondents characteristics. The second section reports the results of exploratory factor analysis. Next, the results of the reliability test and descriptive statistics are discussed. The final section reports the results of correlation analysis and hypothesis testing through multiple regression analysis.

1.2

Profile of Respondents

Table 1.1 shows the demographic profile of the respondents. A total of 352 participants were included in the final sample. The analysis of the respondents information reveals that most of the respondents were female (78.1%) with the remainder made up of the man (21.9%). Approximately 20.4% of the respondents were in the age range of 21 to 22, 75.3% were aged under 24, 4.3% in the age range of 25 to 29. The ethnic background was predominantly Bumiputera, accounting for 38.1% of the samples. This was followed by the Chinese (32.7%), the Malays (27.3%), and Indians (2.0%). As for year of study, about 6.5% of the respondents were in the First year student, 8.5% were in the Second Year, and 84.4% of the respondents were Third year student, while the remaining 0.6% were in the Final year student. With regard to School of respondent, about 23.6% of the samples were from School of Education and Social Development (SPPS), 18.8% were from School of Business and Economics (SPE), 7.7% were from School of Art Studies

(SPS), 0.9% were from School of Social Science (SSS), 11.4% were from School of Psychology and Social Work (SPIKS), 0.3% were from School of Engineering and Information Technology (SKTM), 13.6% were from School of Science and Technology (SST), 14.8% were from School of Food Science and Nutrition (SSMP), 9.1% were from School of International Tropical Forestry (SPTA) and none of respondent from School of Medicine. Table 1.1 Profile of Respondents Demographic Variables Gender Age Ethnics

Categories Male Female 21-22 23-24 25-29 Malay Chinese Indian Bumiputera 1st Year 2nd Year 3rd Year Final Year SPPS SPE SPS SSS SPIKS SKTM SST SSMP SM SPTA

Frequency 77 275 72 265 15 96 115 7 134 23 30 297 2 83 66 27 3 40 1 48 52 0 32 382

Percentage 21.9 78.1 20.4 75.3 4.3 27.3 32.7 2.0 38.1 6.5 8.5 84.4 0.6 23.6 18.8 7.7 0.9 11.4 0.3 13.6 14.8 0 9.1 100.0

Years

School

1.3 As

Factor Analysis of Registration Study. mentioned earlier, the brand image construct in this study is a

multidimensional construct which consists of three main elements; namely university attributes, satisfaction towards UMS and follow up action. Exploratory

factor analysis was performed simultaneously on both university attributes, satisfaction towards UMS and follow up satisfaction variables. This step was undertaken to provide a comprehensive profile of the university attributes, satisfaction and follow up action as the researcher has adapted the

measurements of attributes, satisfaction and loyalty from several sources. Originally, there were thirty three items used to measure university attributes; programme (8 items), location (7 items), facility (9 items), cost (4 items) and reputation (5 items). Satisfactions towards UMS were represented by twelve items and satisfaction (8 items). While, the follow up action were represented by only one item; loyalty (5 items). A principal component method with a varimax rotation was utilized in order to reduce a large number of variables/items to a smaller numbers of factors. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.923 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the first item The University has excellent campus layout was removed due to cross loading on Factor 5 (i.e. Reputation). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.921 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5.

However, some items were cross-loaded on other factors or different dimensions. For example, the second item The University is clean was removed due to cross loading on Factor 2 (i.e. location). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.920 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the third item The University has excellent campus layout was removed due to cross loading on Factor 5 (i.e. Reputation). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.919 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the fourth item The University has the course that interests me was removed due to cross loading on Factor 5 (i.e. Reputation). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.920 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5.

However, some items were cross-loaded on other factors or different dimensions. For example, the fifth item The University is easy to enter was removed due to cross loading on Factor 2 (i.e. Loyalty). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.922 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the sixth item The university has course matches my expectations was removed due to cross loading on Factor 2 (i.e. Location). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.922 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the seventh item The University has flexible payment of fees was removed due to cross loading on Factor 2 (i.e. Location). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.924 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5.

However, some items were cross-loaded on other factors or different dimensions. For example, the eight items The University has wide range of course was removed due to cross loading on Factor 4 (i.e. Cost). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.925 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the ninth item The University never let me down was removed due to cross loading on Factor 5 (i.e. Reputation). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.921 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the tenth item The University has a good reputation was removed due to cross loading on Factor 3 (i.e. Facility). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.918 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5.

However, some items were cross-loaded on other factors or different dimensions. For example, the eleventh item The University has good transportation services was removed due to cross loading on Factor 7 (i.e. satisfaction). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.918 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the twelve items The University does a good job in satisfying my needs was removed due to cross loading on Factor 5 (i.e. reputation). After this item was removed, the factor analysis was run again. . The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.915 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the thirteen items I believe this university is the best choices was removed due to cross loading on Factor 5 (i.e. Reputation). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.910 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5.

However, some items were cross-loaded on other factors or different dimensions. For example, the fourteen items I am proud to be the student in this university was removed due to cross loading on Factor 5 (i.e. Reputation). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.904 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the fifteen items I will recommend this university to anyone who seeks my advice was removed due to cross loading on Factor 1 (i.e. Programme). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.899 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the sixteen items The University has offers high quality education was removed due to cross loading on Factor 3 (i.e. facility). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.893 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5.

However, some items were cross-loaded on other factors or different dimensions. For example, the seventeen items I will encourage friends and relative to study in this university was removed due to cross loading on Factor 1 (i.e. Programme). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.890 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the eighteen items The University has job placement was removed due to cross loading on Factor 7 (i.e. loyalty). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.891 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the nineteen items The University is prestigious was removed due to cross loading on Factor 6 (i.e. Satisfaction). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.889 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5.

However, some items were cross-loaded on other factors or different dimensions. For example, the twenty items I intend to continue my study in this university in future was removed due to cross loading on Factor 7 (i.e. loyalty). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.892 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the twenty one items The University do not have

overcrowded classes was removed due to cross loading on Factor 7 (i.e. loyalty). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.894 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the twenty two items I am very satisfied with my decision to attend this university was removed due to cross loading on Factor 6 (i.e. satisfaction). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.886 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5.

However, some items were cross-loaded on other factors or different dimensions. For example, the twenty three items The University has complete library was removed due to cross loading on Factor 7 (i.e. loyalty). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.882 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the twenty four items The University provides financial aids was removed due to cross loading on Factor 6 (i.e. satisfaction) Factor 7 (i.e. loyalty). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.881 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the twenty five items The University has good image was removed due to cross loading on Factor 3 (i.e. facility) Factor 4 (i.e. cost). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.875 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5.

However, some items were cross-loaded on other factors or different dimensions. For example, the twenty six items The University has specialized

programme was removed due to cross loading on Factor 1 (i.e. programme). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.873 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the twenty seven items The University has quality teaching was removed due to cross loading on factor 1 (i.e. programme). After this item was removed, the factor analysis was run again. The results of factor analysis on university attributes are presented in Table 1.2. The initial run of the factor analysis on 46 items of university attributes with eigenvalues greater than one. The Kaiser-Meyer-Oklin value was 0.859 and the Bartlett test of sphericity was significant at 0.00. Anti-image correlation for the entire university attributes variables were greater than 0.5. However, some items were cross-loaded on other factors or different dimensions. For example, the twenty eight items I will say positive things to other people about this university was removed due to cross loading on factor 2 (i.e. location). After this item was removed, the factor analysis was run again. The final run of factor analysis produced 5 factors with eigenvalues more than one, which explained 66.44% of the total variance. The Kaiser-Meyer-Oklin value was .859 and Bartlett Test of Sphericity was significant at 0.000. Antiimage correlation of the remaining 0.50. (Table 1.2) 18 items of university attributes exceeded

Factor Analysis 1 2 3 4 5 6
7 Items I am satisfied with my choice I am happy to study in this university I am very satisfied with my decision to attend this university I am sure it was the right thing to do study in this university The University has ideal location The University is strategically located The University is accessible The University is beautiful The University is convenient The university is close to my hometown The University has functioning facilities The University has good gym facilities The University has good teaching facilities The University has modern classroom and equipment The University is cheap The University has lower living cost The University has easy admission procedures 18.00 66.44 .859 2919.03 Factor Loadings
.879 .847 .841 .824 .815

.774 .763 .762 .731 .698 .816 .818 .757 .709 .620 .888 .882 .620

8 9 10 11 12 13 15 16 17 18

Eigenvalue Total Variance Explained Measure of Sampling Adequacy Bartletts test of Sphericity

Potrebbero piacerti anche