Reference no: EM133076756

Directions: Consider the following scenario: You have been awarded a contract to analyze sales data for a snack food company. They have introduced a new flavor of potato chips (crisps). As part of your analysis, you will need to conduct a hypothesis test on the results of a simple linear regression to determine if there is a relationship between price and amount of product sold. Here is the data you need to analyze:

Price per Bag of Crisps	Number of Truck Loads Sold
$2.00	14
$1.90	18
$1.75	24
$1.50	30
$1.25	35
$1.00	41
$0.90	46
$0.75	52
$0.60	57
$0.50	64

Part 1: Set up a hypothesis test (using all six steps) and then conduct the test. You can assume that the data is okay to run a simple linear regression model.  So, the first step in the test is already done!

1. Choose the right test.

We will do a simple linear regression.  (DONE!)

2. Write down your hypothesis.

The hypothesis is our idea. So, we can formally state it as:

Sales by the truckload will be predicted by ________________________________.

3. Decide on your risk of being wrong.

Our statistical software automatically computes this as equal to 0.05, and this is a standard for academic studies. We'll stick with that.  (DONE!)

4. Run the statistical analysis and get the initial results.

We will need to use some statistical software to do this. Click on the link and enter the list of X (price per bag of chips) and Y (number of truckloads sold) values on the screen.

Once you do that, take a look at the output.

5. Figure out the probability if you are right or wrong.

The chance that our idea is wrong given this data is the probability value, which has been abbreviated here as P Value. The value is _________.

6. Decide whether your idea is supported by the data and test

The software tells us that the results are either significant or not significant on the "Deviation from horizontal?" line. Are your results significant? 

How well did the number of days visiting the coffee shop predict the sales? It explained about _____% of the variation in sales, according to the R square calculation. Is this a good result in your opinion? Why or why not?

Part 2: Let's suppose a manager wanted to raise price to $2.25 a bag. Take a look at the graph the calculator generated. If you could extend the line up to $2.25, can you predict about how many truckloads would be sold?

Part 3: What would happen if a very similar flavor from another company came on the market and was a good substitute for the chips? Would consumers be more or less sensitive to the price of your product?