Study Description

National Institute for Science Education Evaluation of
"The Why Files" World Wide Web Site

Abstract

Unique Identification Number:
BA-007-001-1-1-United States-DPLS-1997
Summary:
In February 1996, the National Institute for Science Education (NISE) created The Why Files (originally at http://whyfiles.news.wisc.edu, now at http://whyfiles.org/), a Web site designed to explain the science behind the news.  In order to understand the characteristics of the users of The Why Files, a brief questionnaire was incorporated into the site.  This survey was administered from March 20 through April 3, 1997.  The survey requested information about demographic characteristics, Web use frequency, and interest in topics related to science, math, engineering, and technology. 
 
An initial examination of audit-trail data, previously collected in three two-week censuses during late 1996 and early 1997, had indicated that most users encountered the home page at some point in their visit, so the survey was attached to all links on this page.  When a person who had visited The Why Files on at least two previous occasions clicked on a link on the home page he/she was asked to fill out the survey.   Therefore, the survey was limited to users who had previously visited the site at least twice in the two weeks prior to the survey or during the two-week survey period itself.  If the user accepted the invitation to participate in the survey, they then were sent to a Web page that served as a consent form.  After filling out the form, the user was forwarded to the online survey instrument.  Once the user completed the survey, the data were forwarded to a central database.
 
Each time a file was requested from The Why Files server, the server made a record of the time, the page requested, and so on.  The Why Files server also had an additional software module that made use of "cookies."  Cookies were used to set random identification numbers for visitors to The Why Files so that files requested from the same computer over time could be identified.  The first time a user requested a page of content from the The Why Files server, an identification number was stored on the user's hard drive and recorded by The Why Files server, in addition to information about the page requested.  Then this identification number was retrieved from the user's hard drive and stored on The Why Files server, along with information about subsequent page requests.
 
Methodology:
The overall response rate was 63%.  There were 37% non-respondents.  17% of the non-respondents refused to participate when asked and 20% indicated they would fill out the survey at a later time, but did not return to the site again within the survey period, or returned but again indicated that they would fill the survey out later.  In total, responses to the survey were collected from 399 repeat users of The Why Files.   The sampling error varies by item; it can be calculated by the user via the sample size. 
 
Technical Notes:
Two types of data have been collected from The Why Files World Wide Web Site.  1) Survey data collected by a questionnaire electronically distributed to repeat users of the site (survey1.por).   survey1.por contains 13 variables and has 399 valid records and 633 total records.   2) Computer-collected audit trail data representing the sequentially ordered file requests made by users (wk1arch.por, wk2arch.por, and wk3arch.por).   wk1arch.por contains eight variables and has 178,841 records.  wk2arch.por contains eight variables and has 126,410 records.  wk3arch.por contains 10 variables and has 173,842 records.  All of the datasets contain some string variables, the files are unweighted, and are available in SPSS portable export file format.
 
Data Format:
Each line of the data represents a page requested from the server.  These lines of data contain the name of the file requested, the page the user was on when this file was requested, the time of the request, as well as a cookie identification number and a session identification number.
Date of Study:
#1: November 21, 1996 - December 5, 1996 (wk1arch.por)
#2: December 20, 1996 - January 2, 1997 (wk2arch.por)
#3: March 20, 1997 - April 3, 1997 (wk3arch.por and this time period includes survey1.por, too)
 
Geographic Coverage:
The respondents of the survey and other data are from anywhere in the world.
 
Descriptors:
World Wide Web, science communication, Internet, new technology
 
Funding Statement:
These data were initially collected by Sharon Dunwoody and William Eveland with the support of a cooperative agreement between the National Science Foundation and the University of Wisconsin-Madison (Cooperative Agreement No. RED-9452971).  At UW-Madison, the National Institute for Science Education is housed in the Wisconsin Center for Education Research and is a collaborative effort of the College of Agricultural and Life Sciences, the School of Education, the College of Engineering and the College of Letters and Science.  The collaborative effort is also joined by the National Center for Improving Science Education, Washington D.C.  Opinions, findings, and conclusions based on these data do not necessarily reflect the views of the supporting agencies or those who initially collected the data.