Business research methods, part III

Business Research Methods, Part III

Weijia Gao

QNT 561

Gary Saxena

Introduction

Correlation analysis is a group of techniques to measure the association between two variables and also it is the study of the relationship between two or more variables. Basically, correlation analysis functions to report the association between two variables. It helps in the determination of the equation of the line that best fits the data collected, then it uses the equation from the line to estimate the value of one independent variable based on another dependable one , it also measures the error in the research estimate, and then finally goes ahead to establish confidence and prediction intervals for the esatimate.

Research design

Data collected from sampling of the supermarkets in the two regions was recorded for analysis. By selecting 10 supermarkets to do our sampling, we may be able to draw conclusions from the entire population, hence generalization is possible. The elements of our sample will included the supermarket staff and shoppers in the supermarket. A big sample is needed and that is why we took to sample 20 supermarkets to have near accurate information (Lind et al,., 2010).  The hypotheses of this study were formulated and they will be as follows. The mean sales values of SalesPro products is generally higher in the North region as compared to the South region. For this study, the mean sales of the SalesPro products in the North will be calculated from the data obtained and also the same done for the values that will be obtained from the South. These mean values will be the ones that will be used for the comparison studies.

The difference in the means will be calculated and recorded. For this study, a level of significance of 0.01 is selected for the calculation of these values.

H_0: There is no difference between the sales of SalesPro products between the North and South regions.

H₁: The mean sales values of SalesPro products is generally higher in the North region as compared to the South region.

SPSS software was used to analyze the data. Focus was on the reasons for declining sales in the south. The second step involved research on why sales in the north are booming. The last step  defined the overall reasons for the two different behaviors

Control variables

A variable is defined as a characteristic that varies. Independent variables are those that the researcher has control over thus they can be manipulated over the existing variables or by introducing new variables. Hence the sales of SalesPro products constituted the independent variable. Dependent variables are those variables that show the effect of manipulating or introducing the independent variables. Sales representatives, purchasing power, product pricing and competition were among the dependent variables.Several variables can relate to decline in sales at the south. The variables can be those that do not influence the reasons for decline. These are behavior, and personality variables, while the economic and demographic variables may affect the decline (McNamura. Basic business research Methods).

Statistical Data Analysis

The data collected from the research study were analyzed using SPSS tools. This was to give the relation between the sales of SalesPro’s Products between the North and the South regions. An analysis of variance was done on the data collected. Then thereafter, a regression analysis was done. For this, multiple regression analysis was employed to find the relation of the independent variable, sales of the SalesPro products and the dependent variables including factors contributing to sales volumes.

Potential challenges to validity and reliability

Some of the threats to validity and reliability to the research question include:

1. Outside threat ( maybe something outside the research caused the result for example history threat, maturation threat),
2. Design threat (maybe something in the design caused the result for example testing threat, instrument threat, data threat),
3. Tangential threat (maybe something related to the research, but potentially out of your control caused the result for example mortality threat, regression threat) or
4. Multi-Group studies – same as single group studies but also have to worry about comparability of groups.

These challenges were minimized by making sure that the groups being studied are comparable prior to the research program. This will prevent the probability of selection threat or bias. The best procedure would therefore be to pick out comparable groups, basically achieved by selecting an entire group (the sample) and then randomly assigning them to a group.

References:

Lind D., Marchal W. and Wathen S. (2010). Statistical Techniques in Business and Economics. McGraw-hill Irwin, 2010.

Lind, D., Marchal, W. And Wathen, S. (2008). Statistical Techniques in Business and Economics, 13th Edition 14. Multiple Regressions and Correlation Analysis Text © The McGraw−Hill Companies.

McNamura. Basic business research Methods. Retrieved from: http://managementhelp.org/research/research.htm