The questionnaire sample was selected following a self-selection sampling system and administered through the Internet. The survey objective and rationale were publicised together with the survey access link via the School of Management system to all postgraduate master students. The questionnaire(available in the Appendix) was designed using a survey design software which allowed to obtain easier and more user-friendly navigation and an attractive graphical layout.
The questionnaire had a completion rate of 86% and a response rate of 9%, the 42 respondents demographic characteristics were very close to the school of management population where 61% of the students are female and 39% male (Figure 5.2). Also the age composition reflects quite well the student population in the school of management where 73% of the students are aged 21-25, 19% between 26-30 and 8% of the students are over 30. Looking at the nationality composition, UK and EU students are re-spectively 8% and 6% of the student population where in the survey theyrepresent 14% and 26% of the sample Far-east Asia students which count for64% of the student population represent 49% of the surveyed student (Figure5.1). Nine out of the fifteen Masters offered by the school are represented,the proportions are shown in Figure 5.3. http://www.ukassignment.org/marketing/ The type of data variables that were collected through the question-naire were behavioural, opinion based and attribute based (Dillman, 2007in Sounders et al., 2009) and its clearness and effectiveness was tested in apilot study conducted with some students. The author also provided personal contact details so that the studentswho wished to could contact the researcher, this has provided very usefulfeedback form some students which shared their ideas on the survey and onits clearness.
The survey reliability was implemented firstly through the very natureof the research parallel mixed method strategy, which allows to answer thesame research questions using to different methodologies and therefore guar-anteeing that the results are not dependent on the technique used. Sec-ondly, through the inclusion in the survey of check questions as suggested byMitchell (1996, in Sounders et al., 2009, p.374). |