Impact Of Perceived Service Quality On Customer Satisfaction And Brand Loyalty
[A CASE STUDY OF HONEYWELL FLOUR MILLS PLC]

3.4    Population of Study and Sample size
The population for this study is the customers of Honeywell and Lister flour mills in Ibadan and Ilorin metropolis from the total population of this study. Based on this, a total of two hundred (200) samples were drawn for this study.  The Ibadan and Ilorin customers are the study population. However to study the entire population is not always feasible due to constraints such as time, cost and repetition of response. Hence, this research has been restricted to Two Hundred customers of Honeywell and Lister Flour mills.
3.5.     Sampling Technique
Bakers and dealers were randomly selected in both Ibadan and Ilorin and questionnaires were administered among the chosen sample. The division into stratum was based on the company and customer categories in order to capture all categories of customers of the two flour mills, after which random sampling was applied in selecting respondents from each stratum. This is expected to minimize the level of biasness associated with the simple random sampling technique. Each of the respondents was also interviewed to review and shed some light on some of the responses. The questions posed were designed to provide explanations on the major areas of interest to the attainment of the objectives of the study.
3.6     Method of Data Analysis
The data collected from the respondent was analysed and interpreted using statistical tools such as simple percentages and tables. For the purpose of testing three (3) hypotheses formulated for this study, a correlation technique was used. The correlation techniques measure the degree or the extent of relationship between two variables. The correlation technique is useful when dealing with rating scales as was used in this study (Likert’s Scale).
The main result of a correlation is called the correlation coefficient or ‘r’; it ranges from -1 to +1. The closer ‘r’ is to +1 or -1, the more closely the two variables are related.
However if ‘r’ is close to zero (0), it means there is no relationship between the variables. If ‘r’ is positive, it means that as one variable increases, the other also increases, but if ‘r’ is negative, it means that as one variable increases, the other one decreases (i.e. inverse correlation).