consumer Review

1. Star Ratings of Products

a. Describe what criteria you personally use when assigning each rating to a product.

b. Read Amazon’s description of the product star ratings by doing the following: click on any product for sale and scroll down to the reviews. Then click on “Write a Customer Review”. You should see five empty stars that give a description when you put the mouse over them. Write Amazon’s criteria below.

c. How did Amazon’s criteria compare to yours?

d. What type of data is a star rating?

i. Qualitative or quantitative?

ii. Discrete or continuous?

iii. Nominal, Ordinal, Interval, or Ratio?

e. Based on the answers to the above question, what type of average is best used with the data—mean, median, or mode?

f. What type of average does Amazon use?

g. Suppose a product’s star ratings are 1, 1, 1, 1, 1, 1, 4, 4, 5, 5, 5, 5, 5.

i. Find the mean of these numbers.

ii. Find the median of these numbers.

iii. Find the mode of these numbers.

h. What sampling technique does Amazon use to get its star ratings—census, stratified, cluster, random, systematic, convenience, or none of the above? Explain your answer.

i. What, if any, bias exists in customer reviews?

j. Do you believe the review below contains bias?

k. It is possible that some 5 star reviews are written by the companies who sell the products rather than a customer. Find and paste into this assignment a 5 star review that you suspect may be written by someone paid to pretend to be a customer. Look for a review that has perfect grammar and spelling, is incredibly positive, and uses vocabulary that most people don’t use every day. The review will likely sound like an advertising brochure.

l. Find and paste into this assignment a 5 star review that you strongly believe was NOT written by a professional paid reviewer.

2. Frequency Distributions of Product Ratings

a. Look at several of these bar graphs. Are most skewed to the top, skewed to the bottom, uniform, or none of the above? Explain your choice.

b. Find two bar graphs that are significantly different but have roughly the same star rating. Take screen shots of these to include in this assignment.

c. Suppose a bar graph is fairly uniform shaped. What, if any, information can you get from the graph that would influence your buying decisions?

d. Suppose a bar graph is U shaped. What, if any, information can you get from the graph that would influence your buying decisions?

3. Sample size in Ratings

a. Suppose a product has 5 ratings that are all 5 stars. Suppose the next rating the product gets is a 1 star rating. What does this do to the mean of the product’s ratings on Amazon?

b. Suppose a product has 1000 ratings that are all 5 stars. Suppose the next rating the product gets is a 1 star rating. What does this do to the mean of the product’s ratings on Amazon?

c. How many, if any, total ratings does a product need before you personally feel comfortable trusting the mean of the star ratings? Explain.

4. Seller Ratings

a. Although consumers rate the sellers as 1, 2, 3, 4, or 5 stars; Amazon gives the percent of the ratings that are positive. 1 and 2 stars are classified as negative, 3 stars is classified as neutral, and 4 and 5 stars are classified as positive. True or False? If a seller has 100% positive ratings, it received all 5 star reviews. Explain your reasoning.

b. Suppose a seller received 3 1 star ratings, 2 2 star ratings, 0 3 star ratings, 4 4 star ratings, and 37 5 star ratings.

i. What is the mean star rating if we allow the number of stars to be considered quantitative? Show your calculations.

ii. What percent of the ratings are positive? Show your calculations.

iii. Which method of presenting the seller’s ratings is most advantageous to the seller? Explain why.

c. Search Amazon for “Texas Instruments TI-84 Plus Graphics Calculator, Black“. Click on the product. Under the price should be the phrase “Note: Available at a lower price from other sellers”. Click on “other sellers”. You should see a list of sellers, each of which has a rating on the right side. Look at the total ratings for each seller on the page.

i. What are the two highest and two lowest total number of ratings?

ii. What are the two highest and two lowest percent positive ratings?

iii. If you were to buy a Texas Instruments TI-84 Plus Graphics Calculator, Black from one of the sellers on the page, which one would you pick and why?

iv. Suppose a seller has 5 ratings that are all 5 stars. Suppose the next rating the seller gets is a 1 star rating. What does this do to the seller’s percentage of positive ratings?

v. Suppose a seller has 50 ratings that are all 5 stars. Suppose the next rating the seller gets is a 1 star rating. What does this do to the seller’s percentage of positive ratings?

vi. How many, if any, total ratings does a seller need before you personally feel comfortable trusting that the percent of positive reviews is accurate? Explain.

d. While seller feedback most certainly does not come from a random sample, for convenience we will assume it does for the calculations in this section. Suppose a seller has 90% positive ratings. Then out of 100 randomly selected buyers, it is reasonable to predict that 10 of those buyers will have a neutral or negative experience. Reducing 10 out of 100, we get 1 in 10 buyers not having a positive experience. Do the same calculations for the following:

i. 60%

ii. 85%

iii. 96%