Click here
|
Kaiser Foundation Releases Internet Filter Study 12/11/2002 The study, released December 10, should silence the American Library Association (ALA), the ACLU and other critics of filters once and for all. It shows that filters know it when they see it – they tell the difference between health information and smut. This is further proof that filtered Internet access in public schools and libraries provides useful and appropriate material without becoming dirty peep shows funded by taxpayers.
The study, released by the Kaiser Foundation, is titled, “See No Evil: How Internet Filters Affect the Search for Online Health Information,” by Paul Resnick, Ph.D., and Caroline Richardon, Ph.D., of the University of Michigan. According to Kaiser’s press release, “The Internet filters most frequently used by schools and libraries can effectively block pornography without significantly impeding access to online health information.”
The main focus of the study was to assess “whether Internet filters block young people’s access to non-pornographic health information.” According to the study, “When set at the least restrictive level of blocking (“pornography only”), filters block an average of 1.4% of all health sites” and “block an average of 87% of all pornographic sites.”
Research methodology:
1) Searches for health information on 24 health topics, on each of six different search engines that are popular with young people: Yahoo!, Google, America Online (AOL), Microsoft Network, Ask Jeeves, and Alta Vista.
2) The health topics were selected based on three factors: previous national surveys about the subjects most frequently researched online by teenagers, frequency data about commonly used search terms, and researchers’ decision to test a variety of subjects including health topics unrelated to sex, topics involving body parts but not related to sex, health topics related to sex, and controversial health topics.
3) Health information was defined as information about topics that would be discussed in a medical school or school of public health. Pornography was defined as any text or graphic of a sexual act or genitals designed to appeal to prurient interests [and] that was not of an educational or scientific nature. The health and pornography Web sites returned from those searches were then tested against a variety of commonly used blocking products.
4) The primary purpose of the study is said to be to test filtering products widely used in schools and libraries: Smartfilter, 8e6, Websense, Cyberpatrol, Symantec, and N2H2. Researchers also tested AOL’s Parental Controls, which is widely used in the home.
5) Each of the seven filters was tested at three levels of configuration:
A. Least restrictive, which blocked pornography, sex acts, adult/sexually explicit. (This configuration was based on a recent test commissioned by the U.S. Department of Justice.)
B. Intermediate, which blocked drugs, nudity, weapons, hate/discrimination.
C. Most restrictive, which blocked tobacco, profanity, swimsuits, jokes, auctions, games, dating.
6) Researchers also tested two settings for AOL Parental Controls: “Young Teens” and “Mature Teens,” but these settings were not comparable to those tested for the school and library products, according to researchers.
“In the end, a total of 3,053 health Web sites and 516 pornography sites were tested against the filtering products (2,467 of the health sites were from the simulated searches, and 586 were “recommended sites from the online directories).”
The percent of health-information URLs blocked:
Product Name
Least restrictive
Intermediate
Most restrictive
Smartfilter:
2.3
5.8
18.2
8e6:
1.1
4.5
15.1
Websense:
0.6
3.8
35.4
Cyberpatrol:
1.6
2.8
22.4
Symantec:
1.9
7.6
33.5
N2H2:
0.8
6.5
19.5
Average:
1.4
5.2
24.0
AOL PC:
---
3.2
16.1
The percent of pornography URLs blocked:
Product Name
Least restrictive
Intermediate
Most restrictive
Smartfilter:
87.2
88.7
89.0
8e6:
89.1
90.9
92.1
Websense:
83.9
91.3
93.8
Cyberpatrol:
85.7
85.7
87.2
Symantec:
87.8
89.3
90.5
N2H2:
89.5
92.8
94.0
Average:
87.2
89.8
91.1
AOL PC:
---
92.1
94.8
The percent of recommended health-information URLs blocked:
Product Name
Least restrictive
Intermediate
Most restrictive
Smartfilter:
0.0
1.0
16.8
8e6:
0.5
1.2
10.9
Websense:
0.5
1.4
39.4
Cyberpatrol:
0.5
0.9
26.5
Symantec:
1.4
8.4
28.5
N2H2:
0.3
3.8
23.2
Average:
0.5
2.8
24.2
AOL PC:
---
1.2
15.0
The percent of non-pornographic health sites on "condom," "gay," and "safe sex" that were blocked at the least restrictive level, by product:
Product Name
Least restrictive
Smartfilter:
16
8e6:
8
Websense:
4
Cyberpatrol:
15
Symantec:
10
N2H2:
5
Authors’ Conclusion:
“The extent to which Internet filters may adversely impact young people’s access to online health information depends in large measure on how the filters are configured by the schools or libraries installing them. … The major filtering products on the market allow administrators a great deal of leeway in determining how much material will be blocked. At their least restrictive settings, overblocking by filters has a negligible impact on access to general health information – especially when compared to other factors that can affect search results, such as spelling errors, limited search skills, and uneven quality of search engines. However, even at their least restrictive settings, filters could have a modest impact on those seeking information on sexual health issues; on average, filters incorrectly blocked about one in ten sites on safe sex, condoms, or health issues pertaining to gays. A determined searcher would likely still find the information he or she was searching for, particularly by altering search terms slightly: far fewer health sites were blocked on terms such as ‘birth control,’ ‘STD,’ or ‘herpes.’”
This study makes the case for the Children’s Internet Protection Act (CIPA), which requires public schools and libraries that accept federal subsidies for Internet access to block access to illegal pornography. When software is configured to the “least restrictive level,” in compliance with the CIPA, there is no unacceptable “overblocking” and “underblocking” as the opponents of filtering and the CIPA claim.
It means that filtered Internet access provides a voluminous amount of appropriate health and sex-ed information while blocking most of the smut. Hopefully, we’ve heard the last of the great myth about filters blocking breast cancer information.
|
|
Concerned Women for America 1015 Fifteenth St. N.W., Suite 1100 Washington, D.C. 20005 Phone: (202) 488-7000 Fax: (202) 488-0806 |
Feedback / Questions? || Problem with this page? || Archives
