# regression analysis

regression analysis

Paper details:

Chapter 10:

• Thank you for characterizing the application of the concept of regression analysis, an important component of business statistics, in designing solutions to real-business dilemmas. The rationale behind the regression analysis concept is for business researchers being able to estimate the value of an dependent variable (Y) in terms of a selected value of the independent variable (X)[McClave, Benson, & Sincich, 2011). According to Lind, Marchal and Wathen (2008), the technique researchers use to develop the equation-relationship between dependent and independent variables that results in estimates of values is statistically referred to as regressional analysis.

As an example, suppose Nick & Peter, Inc. tracked the numbers of sales-related calls its sales staff make in a quarter and the actual units of its product sold in that same quarter. Further, assume that a sample of 45 sales staff out of the company’s 85 population of sales representatives is selected for the study. In order to equip management with factual tools for purposes of decision-making for efficiency and effectiveness, a business researcher may want to construct a linear equation-relationship between the number of sales calls in a quarter (dependent variable, Y) on the basis of units of product sold (independent variable, X). Based on this example, the line a business researcher uses to develop an estimate value of variable Y (sales calls) on the basis of X (units of a product sold) is the regression analysis.

Regression analysis is closely associated with quantitative research methodology as a research technique for estimating relationships between dependent and independent variables. Fundamental to the concept of regressional analysis, like many others such as null hypothesis testing (Norusis, 2006), is to design meaningful and profitable solutions to operational dilemmas that constantly challenge business managers and executives.

Regressional analysis, like any other technique, needs to be carefully applied in order to yield credible results. If properly applied and with consistence in terms of dependent and independent variables, it serves as a guide in presenting study findings in chapter 4. In chapter 5, usually a continuation of chapter 4, the discussion of study findings is a reflection of presentation made in the previous chapter.

Question: Using your organization as an example, would you recall any incident when regressional analysis yielded misleading results, and what was the effect?
References

Lind, D. A., Marchal, W. G., & Wathen, S. A. (2008). Statistical techniques in business and economics (13th ed.). Boston, MA: McGraw-Hill.

McClave, J. T., Benson, P. G., & Sincich, T. (2011). Statistics for business and economics (11thed.). Boston, MA: Prentice Hall.

Norusis, M. J. (2006). SPSS 15.0 Guide to Data analysis. Upper Saddle River, NJ: Prentice Hall.

• Thank you both for your contribution to our understanding of how to apply regression analysis in designing practical solutions for critical business decisions. Central to the application of regression analysis, as the case is with null hypothesis testing, is to develop credible estimates that guide business managers and executives in decision-making practices (Lind, Marchal, & Wathen, 2011).

The application of regression analysis in applied business research covers a broader spectrum in business beyond determining efficiency and effectiveness in operations. For example, Schumert, Brown, Gysler, and Branchinger (1999) used the concept to determine truth from stereotyping regarding whether women are really more risk-averse than men in financial decision-making matters.

In today’s increasingly challenging and highly competitive business environment, precision in decision-making is more of a necessity rather than an application of convenience. Business managers’ desire to gain competitive advantage over their rivals and a favorable balance between cost containment and profit maximization (Gardner, Mills, & Cooperman, 2005) makes the application of regression analysis and null hypothesis testing a matter of success and failure.

Regression analysis, like any other statistical approach in designing solutions to management dilemma, depends on how its results are interpreted. As a scholar-practitioner, caution is needed to ensure that there is consistence in determining dependent and independent variables and how they are linked to the investigated subject.

Question: What would you do, as a scholar-practitioner, to avoid errors when using regression analysis to design a worthwhile solution to an operational dilemma in your organization?
References:

Lind, D. A., Marchal, W. G., & Wathen, S. A. (2011). Statistical techniques in business and economics (7th ed.). Boston, MA: McGraw-Hill/Irwin.

Gardner, M. J., Mills, D. L., & Cooperman, E. S. (2005). Managing financial institutions (15th ed.). Mason, OH: Thomson South-Western.

Schumbert, B., Brown, M., Gysler, M., & Branchinger, H. W. (1999). Gender and economic transactions financial decision-making: Are women really more risk-averse? The American Economic Review, 381.

• Thank you for this contribution regarding the importance of regression analysis in designing strategic solutions to business dilemma.
The purpose of regression analysis, an important component of business statistics, is being able to equip business researchers with the necessary statistically tested solutions for critically real-business challenges. By using regression analysis concept, researchers estimate the value of a dependent variable (Y) on the basis of a selected value of the independent variable (X). This statistical technique researchers use to develop equation-relationship of a variable in terms of another is what Lind, Marchal, and Wathen (2011) refer to as regression analysis.

Central to the application of regression analysis, like the case would be with null hypothesis testing (Norusis, 2006), is developing and applying statistically tested solutions to business solutions. The challenges that constantly plague business managers and executives demand carefully thought solutions that enhance efficiency and effectiveness reflected in sustainable profitability and growth over time.

Identification and classification of variables (dependent and independent) is crucial to meaningful application of the concept of regression analysis in quantitatively designed research methodologies. It is important to remain consistent throughout the study, because in lapse may materially affect the study findings. Credibility of study findings depend, among other factors, on proper identification and classification of variables under study. This, in turn, affects data collection techniques, analysis, interpretation and reporting. Chapter three of every scholarly research sets forth in details the description of variables, data collection techniques and strategies, and data sources. The presentation of findings in chapter four depends, and is influenced by, how clearly variables are classified and identified basis of which data are collected. The final discussion of study findings in chapter 5 then becomes a continuation of chapter 4. In all these stages, consistence is the key to yielding credible results.

Question: What can you do, as a scholar-practitioner, to safeguard against errors when applying regression analysis to design solutions to management dilemmas in your organization?
References
Lind, D. A., Marchal, W. G., & Wathen, S. A. (2011). Statistical techniques in business and economics 7th ed.). Boston, MA: McGraw-Hill/Irwin.

Norusis, M. J. (2006). SPSS 15.0 Guide to Data Analysis. Upper Saddle River, NJ: Prentice Hall.

• Thank you for advancing our discussion on the importance and effective application of regression analysis when designing statistically tested solutions to management dilemmas in organizations. In order for regression analysis to yield credible results Lind, Marchal, and Wathen (2011) caution on the importance of consistence. Independent variables should consistently be classified throughout the study, while depended variables should not be confused with independent variables.
The purpose of using regression analysis or even null hypothesis testing in decision-making, is ensuring that critical business issues receive tested solutions. Stephan and Rogers (1985) identified several conceptual advantages of using regression analysis for effective decision-making in business operations.
(I) Defined procedure for interpolation: Regression analysis, if properly applied, provides procedures of effect to untested variables or concentrations. This allows estimation of values of a known variable (independent) on the basis of another variable (independent), unlike hypothesis testing that provides quantitative information concerning already tested variables.
(ii) Causal-Effect relationship: Regression analysis fits a statistical equation for the relationship the cause of a management dilemma and the effect. Business scholars are able to draw statistically tested conclusions between causal-effect relationship between dependent and independent variables.
In every study, quantitative or qualitative, the objective is to produce credible findings that address management dilemma. Regression analysis, a statistical component, offers the measurement tools that aide in critical business decisions.

Question: Based on what you now know about the importance of regression analysis in designing meaningful solutions to management dilemma, can you recall any instances when regression analysis may lead to wrong interpretations?
References
Lind, D. A., Marchal, W. G., & Wathen, S. A. (2011). Statistical techniques in business and economics (7th ed.). Boston, MA: McGraw-Hill/Irwin.

Stephan, C. E., & Rogers, J. W. (1985). Advantages of using regression analysis to calculate results of chronic toxlelty tests. Aquatic Toxicology and Hazard Assessment: eith Symposium. ASTM STP 891, R.C. Bahaner and D. J. Hansen, Eds., American Society for Testing and Materials, 328-338.

Chapter 19:
– Thank you both for furthering our discussion with a distinction between independent and dependent samples. In research, independent variables are those constructs that can be manipulated by the research, which manipulation affects another research variable called the dependent variable (Cooper & Schindler, 2011). In real-business situation, and independent variable such as employees’ disposable income (independent variable) can be manipulated to affect employees’ spending behavior (dependent variable). Malvena raised an important point: Similarly, independent samples should be subjected to appropriate analysis techniques for the generation of precise and reliable findings. An independent sample may be comprised of a sample of males and another for males in a given department, for example. This implies that if the management of a company seeking to provide a solution to a human resource problem affected by differences in gender, they could only generate an appropriate solution using an independent sample comprised of women, and another of men (Bagot, May 12).
Researchers are interested in the relationship between independent and dependent variables and how one construct influences the other (Nurosis, 2006). When presenting research findings in chapter four of quantitative studies, the relationship between independent and dependent variables is stated, but not discussed until in chapter five of the study.

Cooper and Schindler (2011) identified some synonyms of independent and dependent variables:

Independent variable Dependent variable

Preditor Criterion
Presumed cause Presumed effect
Stimulus Response
Antecendent Consequence
Manipulated Measured outcome

When designing a solution to a management dilemma, it is important to identify independent and dependent variables and how the former influences the latter. The relationship between the two variables forms the basis of findings presentation (in chapter 4) and discussion for recommendation (in chapter 5).

Question: Using your organization as an example, identify an independent and dependent sample and how they can be applied to solve a known management dilemma.
References
Cooper, D. R., & Schindler, P. S. (2011). Business research methods (9th ed.). New York, NY: McGraw-Hill/Irwin.

Nurosis, M. J. (2006). SPSS 15.0 Guide to Data Analysis. Upper Saddle River, NJ: Prentice