Due: 9/10/14 (Section D), 9/11/14 (Sections A, B, and C)
In the past decade, we have observed a huge shift in consumer shopping patterns from traditional bricks and mortar retail outlets to the Internet, as typified by the dramatic success of Amazon.com and other online sales sites, as well as the failure of retailers (such as Borders Books) who failed to adapt sufficiently to these trends. Many have argued that Internet shopping is not only more convenient for consumers, but also offers them lower prices. The argument for lower prices is as follows: because users can easily check prices at alternative outlets, Internet retailers will be under more pressure than their counterpart retailers selling from physical locations to match the prices of their competitors. So-called shopbots reinforce such price competition. Shopbots search the Internet for the availability and price of a specific product (e.g., John Grishams latest bestseller) and then provide the user with a listing of alternative Internet outlets selling the desired product and the price charged at that outlet. A related hypothesis is that prices on the Internet will not only be lower but less variable. This lower variance hypothesis also follows from the competition argument. As Internet retailers compete to match their competitors prices, this will also tend to reduce the variability of prices across Internet outlets.
To test the lower price, lower variability, and related hypotheses, your first mini-project will involve the analysis of a data set (SMITH_BOOKS_FALL_2014.MTW) assembled by Heinz School faculty member Mike Smith and a collaborator from MIT. The dataset is comprised of the selling price and other information for a selected sample of book titles (e.g. The Perfect Storm) and can be downloaded from Blackboard. Prices for these books were assembled from major Internet retailers such as Amazon and from major physical ("bricks and mortar") retailers such as Barnes and Noble. (Note that the data set distinguishes the Internet and physical outlets of organizations such as Barnes and Noble that have both types of operations.) In total there are 800 records in the data set that you will be analyzing. Each record is comprised of the following variables:
Variable Description
retid retailer identification number (see table) Internet 1 if Internet retailer, 0 if physical retailer title id title identification number (see table) popular 1 if bestseller, 0 if not bestseller hardcover 1 if hardcover book, 0 if paperback fiction 1 if fiction, 0 if non-fiction listpr publishers suggested retail price storepr selling price of the book at that outlet
Because the data set is comprised of prices for a sample of book titles, each with a different list price, you should create a standardized price equal to 100*(storepr/listpr) for purposes of comparison. If the standardized price equals 100, the store price is 100% of the list price. If it equals 90%, this implies it is selling at a 10% discount. If it equals 107, this implies that it is selling for 7% more than the producers suggested list price. Use this standardized measure of price to examine what these data suggest about the following questions:
1) Do (standardized) prices at Internet outlets tend to be lower than at physical outlets? By how much? Are Internet prices less variable? By how much? The standard deviation and interquartile range are two useful summary statistics for discussing variability in a concrete and understandable way. Please consider both of these measures in your analysis, and also create a histogram comparing the distributions of physical and Internet prices.
2) Does the difference in price between Internet and physical outlets seem to depend on the publisher list price of the book? To perform this analysis, compare prices for Internet and physical outlets for two distinct sub-samples: the sample of books with list prices greater than or equal to the sample median, and the sample with list prices below the median. If you find that the difference in discount depends on price, speculate on why this might happen.
3) How does whether a book is in hardcover or paperback affect the difference in price between Internet and physical outlets? Speculate on why this might be the case.
4) How does whether a book is a best-seller affect the difference in price between Internet and physical outlets? Speculate on why this might be the case.
5) At the time when this data was collected, the three largest Internet book retailers were Amazon, Borders (Internet), and Barnes & Noble (Internet). Some have suggested that these Big Three book retailers had sufficient market power and prestige that they could sell at higher prices than their other Internet rivals. Do the data support this contention?
6) The sample includes two major "hybrids" that had both physical locations and Internet locations, Barnes & Noble and Borders. One strategy for maximizing profits is to price discriminate across outlet types based on consumers price sensitivity. Does each of these two retailers differently price between their Internet and physical outlets? Do your results give any hints as to why Borders might have gone bankrupt? Briefly, describe your findings and their implications.
Your response should be in the form of a brief 2-3 page discussion of your hypotheses and your findings. Attach any relevant Minitab output in support of your conclusions. Retid (Retailer identification code)
101 The Partner John Grisham 102 Hornet's Nest Patricia Cornwell 103 Cold Mountain Charles Frazier 104 The Winner David Baldacci 105 John Cameron's Titanic Ed W. Marsh 106 Under the Tucson Sun Frances Mayes 107 Midnight in the Garden of Good and Evil John Berendt 108 Angela's Ashes Frank McCourt 109 Don't Sweat the Small StuffAnd It's All Small Stuff Richard Carlson 110 Simple Abundance Sarah Ban Breathbach 111 The Street Lawyer John Grisham 112 Here on Earth Alice Hoffman 113 The 9 Steps to Financial Freedom Suze Orman 114 Tuesdays with Morrie Mitch Albom 115 The Perfect Storm Sebastian Junger 116 Divine Secrets of the Ya-Ya Sisterhood Rebecca Wells 117 What Looks Like Crazy on Any Ordinary Day Pearl Cleage 118 Bag of Bones Stephen King 119 The Loop Nicholas Evans 120 A Man in Full Tom Wolfe 121 For the Love of the Game Michael Jordan 122 Something More: Excavating Your Authentic Self Sarah Ban Breathnach 123 The Ghost Danielle Steel 125 The Greatest Generation Tom Brokaw 126 The Street Lawyer John Grisham 127 Southern Cross Patricia Cornwell 128 Dr. Atkins New Diet Revolution Richard C. Atkins 129 How to get what you want and want what you have John Gray 130 The Courage to be Rich Suze Orman 131 The Testament John Grisham 132 Vittorio, the Vampire Anne Rice 201 Travels Along the Edge David Noland 202 12 Lessons on Life I Learned from My Garden Vivian Elisabeth Glyck 203 The Pill Book Guide to over the Counter Medications Robert P. Rapp 204 It's Obvious You Won't Survive by your Wits Alone Scott Adams 205 No Fat Chicks Terry Poulton 206 A Superior Death Navada Barr 207 Gump & Co. Winston Groom 208 Our Sacred Honor William J. Bennett 209 The People's Choice Jeff Greenfield 210 The Center David Shobin 211 White Butterfly Walter Mosley 212 Season for Miracles Marilyn Pappano 213 Moon Music: A Novel Faye Kellerman 214 Half Asleep in Frog Pajamas Tom Robbins 215 Why Americans Hate Politics E.J. Dionne, Jr. 216 Israel's Secret Wars Ian Black, Benny Morris 217 Lincoln's Men William C. Davis