Chapter 2- Data EveryWhere:
Types of Data * Qualitative/Categorical: objects being studied are grouped into categories based on some qualitative trait * Nonnumeric * “Dummy” or “Indicator” variables * Male/Female, Small/ Medium/ Large * Quantitative/Measurement: the objects being studied are “measured” based on some quantitative trait * The resulting data are a set of numbers * Height, Weight, Price, Unit Sales * Discrete (e.g. # of items in a basket) or Continuous (e.g. weight)
WHERE DO WE GET IT? * Primary: collected for the purpose of a specific research project * Survey/focus group * Experimentation * Observation * Actual behavior/action * Secondary: already exist, collected for reasons other than the specific project of interest * Collected in an uncontrolled environment
Syndicated Data is… * Collected, cleaned, and analyzed according to a standard procedure by a 3rd party and then sold to interested parties. * Store scanner data: POS data * Household scanner data (can include Wal-Mart) * Information Resources, Inc. (IRI): Consumer Network * Nielsen Homescan * Survey, house hold scanners, meters (setup box, satellite) * Consumer purchasing panel: 260,000 households worldwide3 * Demographically balanced to represent the household population
INTERNET * Clickstream: recording of what a computer user clicks on while Web browsing * IP address of computer * User name (if registered) * Time stamps * Click path through the website * Purchase/No Purchase * Product Evaluation * http://www.epinions.com/ * http://www.cnet.com/ * http://amazon.com/ : * Collaborative filtering (CF) * the process of filtering for information or patterns using techniques involving