This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: 1. (a) A target unit would be a search for information on a webpage by a person. It is insufficient to just say “a person” or “a website”. The combination of a person with a website defines a unit. The target population would be all possible combinations of webpages, persons searching, and information searched for, in the present or future. (b) The response variate is the time taken to find the information. (c) Possible explanatory variates include: • Language used (HTML/Flash). • Age of the person searching. • Gender of the person searching. • Years of experience in browsing the web that the person has. • Employment of the person. • Website visited. • Information requested. (d) Let Y u be the time to find the price for unit u , and Z u be 1 if the page associated with unit u is a Flash page, and 0 otherwise. If a represents the attribute of interest, then a = ∑ all u Y u Z u ∑ all u Z u , Where the summation over “ all u ” is over all units in the target population. (e) The aspect is causative. We want to know whether Flash or HTML affects the time to find information on a webpage. 2. (a) The study population is all possible searches by the 40 employees at www.tiffany.com for the price of a particular men’s tie. It would be acceptable to replace the “40 employees” by “all coworkers of the guy conducting the study”, since the plan could be assumed to include the sampling of co-workers. (b) i. The gauge would be some time-measuring device, such as a stopwatch, clock, watch, or other timer. ii. Measurement error is likely to be of the order of a second, since it would mainly be due to starting and stopping the watch exactly when the search started and ended. Differences in search times could vary by half a minute from user to user....
View Full Document
This note was uploaded on 04/18/2010 for the course STAT 231 taught by Professor Cantremember during the Spring '08 term at Waterloo.
- Spring '08