Quantitative Methods and Analysis
Abstract
Analyzing variables straight from raw data might be difficult to decipher, unless a technique in remodeling the data would be helpful in better understanding statistical data and that is through the use of regression analysis. This paper will show how to derived variables using methods applied on regression analysis. A good example of data that can be extracted with the use of regression analysis is the relation between job satisfaction among employees and their benefits. Measuring correlating variables would allow you to understand changes in dependent variable when either of the independent variables showed varied results. The principles of regression analysis are crucial for forecasting and prediction. When a company is baffled by the issues of employees not performing well with their job and employee survey showed their dissatisfaction with the job, the company can use the methods of regression to identify connection of the issue with benefits. When properly utilized, regression can help an organization to make better decision by looking into the variable results from regression analysis. This paper will explain the methods and the quantitative examples of how regression calculates variables.
Introduction
One of the goals of regression analysis is to identify parameters within a function and determine cause for the other function to best fit the observations. Linear regression implies linear equation being measured where linear of the straight line represents the function. A good example is when to calculate the value of a car and was decreasing constantly every year which in other terms is called depreciation. In order to get the exact value of the car, linear function can be used together with the number of miles the car has driven to obtain the depreciation and current value.
There are other regression methods that that claims to be nonlinear but a large number of techniques have been developed to carry out regression analysis. Other popular methods are parametric and ordinary least squares. The form of which data is being generated in general is still unknown because regression analysis depends on the extent of making assumptions pertaining to a process. These assumptions are to be tested optimally to extract probability. In most applications of regression analysis, determining the small effects or question derived from a large amount data often becomes misleading. In order to avoid errors it is detrimental to establish causality to gain optimal results.
The early method of regression was published in 1805 by Legendre and Gauss using least square method. They both applied the methods to identify problems often asked in astronomical subjects, observations and the planetary orbital around the sun. In 1821 the ores established method of regression was published by Gauss with more develop techniques on least squares theory creating the Gauss-Markov theorem.
Quantitative Methods and Analysis
The application of regression model varies in every field of discipline. If applied to economic analysis, the point of variables as independent is shows as family’s consumption plus expenditure (Elsa.berkeley.edu N.D.). It can also be amount of family income, size of the family and all factors that would determine the amount of consumption in the family. Another major use of regression analysis is to find out the employee’s job satisfaction with their given benefits. It can be determined using the AIU data set containing variable from intrinsic, extrinsic and overall variables. By plotting the set of data into a variable model, the relationship between job satisfaction and benefits can be obtained.
In analyzing the variables expect that when benefits will increase the rest of the variables will follow. The main determinant of the variable is the value set by benefits column in a data sheet. The rational behind it is that when an employee’s benefits increases so as his job satisfaction, it can be in a form of wages or incentives therefore to increase satisfaction level benefits has to be increased as well.
1. To determine relationship between benefits and intrinsic variables a model can be presented setting the intrinsic variable as the dependent variable while on the other hand, benefits variable will be set as the independent variable. When an estimated model takes the form Yi = B0 + B1X, Y would be the intrinsic and X is the benefits. Given that the B1 value is positive and that benefits variables increases, it is expected that the intrinsic variable will also increase following the pattern of benefits variables (Brown, Carolyn N.D.). Below is an example of excel output summary;
BENEFITS
2. In analyzing regression for benefits versus extrinsic, the extrinsic is set as the dependent variable while benefits is set as the independent variable. This model will take the estimate of the equation form Yi = B0 + B1X showing Y as extrinsic and Y is benefits (Nadia Sajjad Hafiza,et al N.D.). Below is an example of excel output summary;
In conclusion, an employee’s benefit determines his ability to work harder because of satisfaction. When an employee is well benefiting from his job, he is more likely to show enthusiasm and interest to the work he is doing. Satisfaction follows because the employee will feel that he is well compensated for the amount of work load he was assigned with. The regression analysis methodology performed on this paper showed variables of benefits influence dependent variables in other words, when benefit variables increases the rest of the dependent variables will also increase.
In practical application of the analysis, a manager can decide whether or not to increase an employee’s benefit rates if the analysis showed that the employee is not satisfied in general. Other factors can also be measured in the question of work satisfaction by using regression analysis. The method overall helps managers and administrative executives in making decisions and be able to predict the outcome based on the data extracted from the regression analysis.
Elsa.berkeley.edu (N.D.) Regression Analysis Web Retrieved February 5, 2012 from http://elsa.berkeley.edu/sst/regression.html
Brown, Carolyn. (N.D.) collective Regression analysis: Web Retrieved February 5, 2012 from http://www.upublish.info/Article/Regression-analysis-/329019
Nadia Sajjad Hafiza, et al (N.D.). Relationship between rewards and employee’s motivation in the non--profit organizations of Pakistan Web Retrieved February 5, 2012 from http://www.saycocorporativo.com/saycoUK/BIJ/journal/Vol4No2/Article_11.pdf
Ali, Nazim. (N.D.) Factors Affecting Overall Job Satisfaction and Turnover Intention: Web Retrieved February 5, 2012 from http://www.qurtuba.edu.pk/jms/default_files/JMS/2_2/05_qadar_nazim.pdf
