A cross-section survey method was used. Data were collected using Survey Monkey, an online survey system was used to administer the instrument. Using Survey Monkey provided ease of access to respondents from different locations.
Since the study was concerned with mobile banking, all individuals with a bank account or a cell phone potentially constituted the sampling frame. Given the challenges of drawing a random or other probabilistic sample, the snowballing technique was utilised. Such a technique minimised sampling bias and at the same time maximised the sample’s representativeness. To target as many consumers as possible from a wide range of backgrounds and taste, the online questionnaire system, Survey Monkey, was used. Researchers sent emails with the Survey Monkey link to 15 banking customers (10 in Gauteng, 3 in Limpopo, and 2 in Mpumalanga) informing them of the study and the questionnaire. In turn these participants notified potential respondents on their email lists, and they consecutively forwarded the link to other banking customers. This process was repeated numerous times between August and October 2010, in which 152 respondents completed the questionnaire. Of these 124 questionnaires were fully completed and usable for analysis. The remaining instruments were either spoilt or had too many missing values. Table 2 provides further details pertaining to the participants’ demographic characteristics.
The instrument used in Brown et al, was adapted for this study. Table 1 details the adapted questionnaire. The questionnaire was pre-tested on 10 undergraduate and postgraduate students to ensure that the questions would be understood by the target audience. The final questionnaire included a cover letter, instructions for completion, and four sections of structured questions. A likert scale (1= Strongly Disagree and 5 = Strongly Agree) was used.
Consistent with research objectives and methodology, a structured questionnaire was used to collect quantitative data. Therefore, statistical analysis was appropriate. PASW (SPSS) v.18 was used to perform the following analyses:
o Descriptive measurements to describe the data and sample;
o Reliability analysis using Cronbach’s alpha coefficient to measure internal consistency of the various sub-scales in the questionnaire; o Pearson’s chi-square test to measure relationships between various variables; o Factor Analysis to reduce items into latent dimensions existing in the scales o Multiple regression and Logistic regression to test hypotheses and answer research questions.
Table 1. Questionnaire Adaptation
|Original Brown et al Questionnaire||Adaptation||Rationale|
|Section A:Included questions about WAP and WIG features on respondents’ cell phones||WAP-enabled and WIG-enabled questions were removed.||o Current technology of CB no longer requires WIG-enabled and WAP-enabled cell phones to support CB.o The USSD technology similar to that used to recharge airtime is now in use|
|Section B:Included several questions about banking products and services||o Combined some banking products- e.g. current and cheque account were collapsed into one category o Included a list of current banking institutions||o To be consistent with currentbanking trends and terminology, as advised by banking experts o To provide a realistic list of banks|
|Section C:||Included more services that CB is now handling e.g. paying traffic fines and municipality bills||o To present a realistic account of CB.|
|Section D:Income quoted on an annual basis||o Modified monthly income to a monthly basis o Included a question on respondents’ religion||o To be consistent with many South African studies o The account for the importance of religion, as it is increasingly being considered an important predictor in consumer behaviour|
Table 2. Demographic Profile
|Demographic Characteristics||Per cent (%)|
|Age||Up to 20 years||1|
|21 – 24 years||17|
|25 – 29 years||19|
|30 – 34 years||16|
|University Diploma/ Degree||47|
|Monthly Income||< R5, 000||13|
|R 5, 001 – R10, 000||25|
|R10, 001 – R15, 000||6|
|R15, 001 – R20, 000||11|