QNT 275 Linear Regression

Regression is a tool used in statistics to show the dependency of a variable of one or more explanatory variables.  Regression can be described by a mathematical model which helps individuals to describe or predict the relationship between dependent and independent variables. Researchers understand that assumptions and predictions are part of the regression model because of the given possibilities. “The investigator also typically assesses the “statistical significance” of the estimated relationships, that is, the degree of confidence that the true relationship is close to the estimated relationship” (Sykes, 1987).  Giving misleading results from regression models must be carefully avoided by the tester because assumptions and predictions are used often in testing.


Sekaran, U. (2003). Research methods for business: A skill-building approach (4th ed.). New York, NY: John Wiley & Sons. 


I think this is a really poorly understood point in society concerning variables. The simplest way to demonstrate this is to consider a measure of agreement – disagreement. If we scored it so that “strongly agree” is 5 and “strongly disagree” is 1, we would get one value of the correlation. If we reverse-scored it, we would get exactly the same value, but with the opposite sign. The strength of the correlation is the same, but the direction of the relation has switched. A zero correlation indicates that there is no relationship between the variables. A correlation of –1 indicates a perfect negative correlation, meaning that as one variable goes up, the other goes down. A correlation of +1 indicates a perfect positive correlation, meaning that both variables move in the same direction together.


One way of using regression analysis would be to measure the relationship between policy changes in government with public opinion poles.  Using regression analysis to model individual survey responses as a function of policy change, a comparative analysis of surveys and polls will reveal the quantifiable relationship between the dependent variable and independent variables. This statistical analysis is appropriate because it attempts to quantify the relationship between public opinion and two or more of the independent variables (Kastellec 2010). For example: public opinion (the dependent variable) is related to gender, race, issue salience, and geographic location (the independent variables).

Leave a Reply

Your email address will not be published.