Regression analysis methodology and applications

Regression Analysis

The methodology for analyzing data is just as important as having a strong foundation for collecting. Within qualitative and quantitative research applications, the use of statistics is a common practice. One of the most used statistical analysis techniques is what is known as regression analysis. This current report explores the processes within regression analysis in what environments the methodology is appropriate for use in to analyze data.

Regression analysis uses statistical formulas in order to show a correlation between particular variables. Sykes (2012) posits that it is "a statistical tool for the investigation of relationship between variables." It is essentially a way to quantify the relationship between the variables and manipulate environmental changes to forecast what impact they would have on such variables (Sykes, 2012). Thus, this statistical process uses both a dependent and one or more independent variables. Changes in the independent variables can impact the nature of the dependent variable, which can be quantified by plotting data in progression through a statistical formula that focuses on the basic equation of Y = a0 + a1 X, or the formula that evaluates the distance between the points in order to show whether a relationship exists or not between the two variables (Dizikes, 2010). This method is also known as the "least squares" method (Dizikes, 2010). Today, many data processing software programs also include progression data analysis tools. Programs like Microsoft Excel have additional add-on data analysis tools that help the user conduct regression analysis and then provide the visual linear graph on which the relationship between the variables can be seen and evaluated.

The process can be used in a wide myriad of applications, both in terms of research and business contacts. It helps quantify more abstract data, putting specific numbers on a plot line that visually represents the environment and its impact on the dependent variable being studied or evaluated. Regression analysis continues to transcend industry boundaries, with more and more applications being implemented as it continues to prove a successful statistical tool for a valuation. It has even moved into the legal and political realms, which focus on its ability to forecast information based on similar or manipulated environmental conditions (Sykes, 2012). It is often used in quantitative research methodologies, but also conserve as a way to test for reliability in mixed methods and qualitative studies using triangulation, or more than one method to prove the same results hold true (Dizikes, 2010). In this, the research suggests that "the investigator assembles data on the underlying variables of interest and employees regression to estimate the quantitative effect of the casual variables upon the variable that they influence" (Sykes, 2012). The type of regression used in any particular situation depends on the number of variables. Simple linear regression focuses on one dependent and one independent variable. From there, more complicated forms of aggression include a greater number of independent variables, like that of multiple regression, which is a more complicated process using more than one independent variable (Sykes, 2012).