ADA # 2: Final Group Project
Research Objective:
 To critically evaluate the affect of advertising variables i.e. News, Print, Broad and Outdoor on total sales of vodka industry.
 To identify factors that impact of price per unit on the total sales of the vodka company.
 To determine how tiers along with domestic versus nondomestic brands impact the vodka industry.
 To evaluate the impact of advertising dollars spent to predict sales of the company.
Research plan and Methodology:
It is important to analyze the approach you get to know the effects of one variable to another. This section of the report aids information about the procedure and test that has been used in the project. In order to find out the trends in the vodka industry, method of survey research has been used to find the affect of advertising variables on total sales of the vodka industry. Data has been collected through survey, and the findings will be calculated from the data that has been extracted from it. Furthermore, the data has been collected from 30 brands of vodka for around fifteen years. In addition to that, correlation and regression analysis have been done to find a relationship between one variable or another. The main function served by correlation analysis is to analyze whether the two variables are fluctuating with each other or not. In regression analysis, the dependence of one variable on another has been evaluated. Further, ANOVAs function in R has been used as well to analyze and compare the scores of two models or variables.
Various variables have been used to analyze the impact of the overall environment of the Vodka industry. These variables include total sales along with all advertising media variables i.e. Print, News, Broad and Outdoor. Furthermore, price per unit, GDP, domestic and nondomestic variables have been used to analyze their impact on total sales. Moreover, unit sales of tier1, tier2 and tier 3 have been used as well. Finally, dollar sales have been analyzed with all the advertising media.
Findings from the data:
Total sales predicted by Print, News, Broad and Outdoor:
Model Summary^{b} 

Model 
R 
R Square 
Adjusted R Square 
Std. Error of the Estimate 
DurbinWatson 
1 
.782^{a} 
.611 
.606 
1036.315 
.411 
a. Predictors: (Constant), media, broad, news, outdoor
b. Dependent Variable: total sales 
Figure 1 model summary
ANOVA^{b} 

Model 
Sum of Squares 
df 
Mean Square 
F 
Sig. 

1  Regression 
5.260E8 
4 
1.315E8 
122.445 
.000^{a} 
Residual 
3.351E8 
312 
1073949.331 



Total 
8.611E8 
316 




a. Predictors: (Constant), media, broad, news, outdoor
b. Dependent Variable: total sales

Figure 2 Anova
Coefficients^{a} 

Model 
Unstandardized Coefficients 
Standardized Coefficients 
t 
Sig. 

B 
Std. Error 
Beta 

1  (Constant) 
872.698 
66.047 

13.213 
.000 
news 
.198 
.125 
.061 
1.588 
.113 

outdoor 
.808 
.142 
.243 
5.704 
.000 

broad 
.651 
.042 
.578 
15.563 
.000 

media 
.067 
.011 
.260 
6.084 
.000 

a. Dependent Variable: total sales

Figure 3: coefficients
Residuals Statistics^{a} 


Minimum 
Maximum 
Mean 
Std. Deviation 
N 
Predicted Value 
743.68 
10026.21 
1420.71 
1290.178 
317 
Residual 
5963.447 
4229.452 
.000 
1029.735 
317 
Std. Predicted Value 
.525 
6.670 
.000 
1.000 
317 
Std. Residual 
5.754 
4.081 
.000 
.994 
317 
a. Dependent Variable: total sales 
Figure 4: Residuals statistics
After removing news advertising variable:
Model Summary^{b} 

Model 
R 
R Square 
Adjusted R Square 
Std. Error of the Estimate 
DurbinWatson 
1 
.780^{a} 
.608 
.604 
1038.830 
.415 
a. Predictors: (Constant), media, broad, outdoor
b. Dependent Variable: total sales 
Figure 5: model summary removing advertising variable
ANOVA^{b} 

Model 
Sum of Squares 
df 
Mean Square 
F 
Sig. 

1  Regression 
5.233E8 
3 
1.744E8 
161.635 
.000^{a} 
Residual 
3.378E8 
313 
1079167.687 



Total 
8.611E8 
316 




a. Predictors: (Constant), media, broad, outdoor
b. Dependent Variable: total sales

Figure 6: anova (after removing advertising variable)
Coefficients^{a} 

Model 
Unstandardized Coefficients 
Standardized Coefficients 
t 
Sig. 

B 
Std. Error 
Beta 

1  (Constant) 
857.511 
65.510 

13.090 
.000 
outdoor 
.788 
.141 
.238 
5.572 
.000 

broad 
.650 
.042 
.577 
15.496 
.000 

media 
.062 
.011 
.239 
5.864 
.000 

a. Dependent Variable: total sales

Figure 7: coefficients (after removing advertising variables)
Residuals Statistics^{a} 


Minimum 
Maximum 
Mean 
Std. Deviation 
N 
Predicted Value 
857.51 
9995.48 
1420.71 
1286.853 
317 
Residual 
5930.095 
4254.013 
.000 
1033.887 
317 
Std. Predicted Value 
.438 
6.663 
.000 
1.000 
317 
Std. Residual 
5.708 
4.095 
.000 
.995 
317 
a. Dependent Variable: total sales

Figure 8: residuals (after removing advertising variables)
Rsquare predicts the dependence of dependent variable over other independent variable. Regression analysis has been done to analyze the impact of advertising variables that are Print, News, Broad and Outdoor on total sales in the Vodka industry. Rsquare in the model summary is 0.611 that means 61% of the variables are correlated with each other. R square of 61.1% indicates that in case advertising media has changed by 61.1% then total sales will change by 61.1% as well.
From the Anova table, one can see that the connection between the predictors and dependent variable is very significant.
The four predictors used in the model ..................
This is just a sample partial case solution. Please place the order on the website to order your own originally done case solution.