IT 223: Data Analysis
Winter 2006
Assignment 3
Due Friday
January 27 before 11:30pm by DL Web Submission
Overview
For this assignment, you will be studying the relationship between the
position of a hyptertext link in a list and the time needed to click
on that link.
Plotting the data
Several years ago, I conducted a Web navigation study where I asked
participants to find various retail goods in a Web site. They
searched for items like coffee grinders, electric mixers and
chopsticks. One item was a birdbath. When they reached the page with
this item, they had to scan a list of 11 links. The link labeled
"Birdbath" was randomly positioned among the 11 links. I recorded its
link position (numbered 1 through 11) and the time it took for them to
select this link. The file for this data is menu.xls. Each row represents one navigation
result. The first column is the position in the list where the
desired link ("Birdbath") was located. The second column is the time
in milliseconds needed to scan the links and click on the desired
link. Assuming that people scan lists from top to bottom, a low
target position should have a faster click time.
Using Excel plot the data set. Have the link
position be the independent variable (x-axis) and the time be
the dependent variable (y-axis). To make your calculations easier,
use Excel to report the average and standard deviation for both
variables.
You do not need to turn in your entire Excel spreadsheet.
Calculations and analysis
- Report the average (mean) and standard
deviation (sample version) of both variables.
- Report standard scores for both variables, their products and the
correlation coefficient. Note that this is the general process for
calculating the correlation coefficient.
- Use Excel to draw the regression line and report the equation of the regression line. Paste the scatterplot in your report.
- What is the slope of the regression line?
- What is the y-intercept of the regression line?
- Name the one point that is intersected by both the standard
deviation line and the regression line.
- What is the slope of the standard deviation line?
- Using the equation for the regression line, how much time
it would take a user to scan the menu and click on the fourth item?
- Using your equation for the regression line, how much time does an
average user require to scan each link? That is, how much time does it
take to consider one link before moving on to consider the next link
(Hint: think about the task and what would be reasonable)
- Using your equation for the regression line, how much time does it
take an average user to click on a link (after the user has scanned and
found the needed link)?
- To what extent did these participants scan the list from top to
bottom? Discuss (approximately 1 paragraph).
Submission
Turn in a softcopy (contained in one document) of your answers and
graph using the dlweb site.