From the tables, it is evident that there is 204 observations in Brand A, 287 observations in brand B and 209 observation in Brand C. As such, Brand A the statistics for brand A that has been purchased are 58, 71 and 75 for zone 1, zone 2 and zone 3 respectively. Contrariwise, Brand B has received purchase amounting to 83, 112 and 92 respectively for each zone. Additionally, brand C has showcased a purchase amounting to 72, 55 and 82 respectively for each zones. From the 3rd table, the chi-square value is 9.26 (Chi-Square = 9.259) with an associated value p-value for the chi-square is 0.055, which is higher than 0.5. Hence, it can be concluded that the consumer’s favorite brand is not influenced by their location/zones. Hence:
H1: The consumer’s favorite brand is not influenced by their location/zones of the stores
Various confidence levels exist given that it is a range of likely values that can be used in a population parameter between 0-100 percent. Evidently, the chosen level of confidence will affect the CI width. Simply put, the CI width is an indication if it’s precision (that is the degree of random error that is linked to it), but is does not describe its accuracy (that is whether population parameters are included); influenced by the chosen level of confidence. To elaborate further, choosing a confidence level of 99% as opposed to the 95 % Cl, an expected increased in accuracy of Cl will be realized (that is a have a greater chance of being on of the aspects that will include the population parameters), but will consequently reduce its precision; simply put it will be wider in comparison to the 95 % CL)
Nonetheless, the incorporation of a confidence level of 95 % is mostly used which implies that 95 % of the time, the confidence levels should contain the actual value that is evidenced in the study population. If the confidence level is increased to say 99%, then the size of the size of the range of the estimate will increase as well.
Assessing the minimum observed and expected counts is imperative given that their frequencies are used in determining the statistical significance between two variables, hence create relationships in the cross tabulation. Evidently, various approaches are used in understanding the associations; where expected alleles frequencies can be calculated using column and rows created from the data. As such, the analysis of the presented data will define the degree of association through the probabilities obtained.