Beer Sales and Youth
Student Worksheet
The purpose of this experiment is to show how much beer/alcohol young adults consume based on a yearly trend.
Part 1:
The independent variable on this graph is the Year because it is independent, and does not rely on any other data. The dependent variable on this graph is Beer Sales Per Capita, because it relies on the year to have any relevant data.

There is a trend of beer sales per capita declining, but we need to filter the data to get more accurate results. Graph 1 contains the data between the years 1974 and 1992. Graph 2 Contains the data between the years 1992 and 2012.

Graph 1 Filtered
This scatter plot shows the amount of beer sold per capita between the years 1975 and 1992. The Independent variable “Year” is graphed on the x-axis and represents the predictor. The scatter plot has a negative, non-linear relationship. It has a strong relationship, with constant scatter. There seems to be a trend of beer sales declining per capita at a rate of 1.23 litres per year.

Graph 2 Filtered: This scatter plot shows the amount of beer sold per capita between the years 1992 and 2012. The Independent Variable “Year” represents the predictor and is graphed on the x-axis. The Dependent Variable “Beer Sales Per Capita” is graphed on the y-axis and represents the response. The scatter plot has a non-linear trend, with a negative association that shows as one variable gets smaller, so does the other. It has a weak relationship with constant scatter. This graph shows that there is a trend of Beer Sales Per Capita declining at a rate of 0.31 litres per year.

Part 2:
The independent variable is “Year” because it does not rely on any other data and the dependent variable is the population aged 20-24 as a percent because it relies on the year.
The relationship shown in this scatter plot is a strong, linear correlation with a negative association. It is also not constantly scattered.

The similarities are both scatter plots have a negative linear correlation. There is a slight dip in the scatter plot between the years 1997 and 1999. From the years 2000 to 2005 both graphs go up and down in a wave form.
The differences: Graph 1(Beer Sales Per Capita) has a large gap in the graph between 1970 and 1980 we can assume that beer sales were a lot lower that year. Graph 2(Pop 20-24 as a percent) has a much steeper drop between the years 1985 and 1995.
Part 3:
It makes more sense to place Pop_20_24 as Percent on the x-axis because it is the independent variable, and Beer sales rely on the population. It is impossible for the population to be dependent on beer, so it has to be graphed with Beer on the y-axis.
The equation of the line of best fit in y=mx + b is y=5.71x + 37.
The value of the correlation coefficient is: r= 0.968. This means that the correlation between the two variables is very strong.

The value of the correlation coefficient is: r= 0.968. This means that the correlation between the two variables is very strong.

The rate of change is 5.71. In terms of the data it means that every 1% of the population, beer sales increase by 5.71 litres annually. I have found that when higher percentage of the population is between the ages of 20 and 24, beer sales per capita are much higher. These two attributes have a very strong correlation with an r2value of 0.94. Also, when the population of 20-24 year olds is less then 9.0 %, there is a moderate correlation with an r2value of 0.52. Therefore we can decide that beer sales per capita is dependent on the population of 20-24 year olds. As a marketing specialist, I would use my findings to decide who my target audience for advertising would be. In this case, the advertisements would be geared towards young adults between the ages of 20 and 24. By increasing sales to this age bracket, the company will then become more popular causing the value of the beer to rise.