Qnt 561 - Inferential Statistics and Findings

Inferential Statistics and Findings

Kevin Woodbridge

QNT 561

12/15/2015

Santoshia Oggs

Inferential Statistics and Findings

Lately, there has been a vast decline in newspaper sales of the HotLanta Times.  Even though this paper has experienced 8 decades of sustained growth, over the past twenty years there has been a substantial decline in newspaper sales and subscriptions. The purpose of this study is to determine what factors are causing the decline in the sales of newspapers and subscriptions.  

Research Question

What effect does the advertising have on the decline of newspaper sales?

Hypothesis:

• H0:  Advertising does not have an effect on the decline of newspaper sales
• H1:  Advertising does have an effect on the decline of newspaper sales

Select a Level of Significance

        After the null and alternate hypotheses have been set up and the direction of the hypothesis has been established, we must select a level of significance.  For the purpose of our experiment, the significance level we will choose is .05.  This number will let us know how extreme experimental results must be in order to reject the null hypothesis of a significance test.  Since our alpha (significance level) is .05, level of confidence is determined by subtracting the alpha from 1, giving us the confidence level of 95% (1-.05 = .95).  The 95% confidence level that has been determined represents the probability that another sample will garner the same results.

Inferential Statistics

[pic 1]

Interpretation of Inferential Statistics

        For our experiment, the appropriate statistical tool to test the hypothesis based on the research question above is the t-Test: Two-Sample Assuming Unequal Variances.  Our sample size was narrowed down to 385 randomly chosen individuals within the Gwinnett and Fulton County districts that are the age of 65 and older.  After conducting the t-Test, it was determined that the p-value was .05 for a one-tailed test, therefore we reject the null hypothesis in favor of the alternative hypothesis.  Furthermore, since the p-value was equal to .05, and the null hypothesis was rejected, this determines that the outcome is statistically significant.  Based on the inferential statistics above, we conclude that advertisement does in fact cause a decline on newspaper sales.