General Demographics Bulleted List

---

[ Survey Home ] [ 3rd Survey Home ] [ Graphs ] [ Bulleted Lists ] [ Datasets ]

Age

• Average age across all users was 35.01 years old (median: 35.00 years old), four years older than the last survey in October (See: Graph).

• There are significant differences between European users and US users, with the European population being younger (average age 31.25) than the US (average age 35.68).

• Prodigy users tend to be three years older (38.98 years old) than other Web users (35.01 years old).

• There no significant differences between women and men's age profiles. The average age for women is 35.15 vs 35.05 years old for men. 42% of all surveyed women are between the ages of 30 and 45, while only 37% of the men fall into the same range.

• The youngest users were European (average age 31.25), with the eldest being Prodigy users (average age 39/98). This older average age of Prodigy users concurs with the occupation distributions.

Gender

• Overall, 15.5% of the users were female, 82.0% male and 2.5% "Rather not say!". The actual numbers are 2020 women, 10668 men, and 318 non-responses.

• Compared to the last survey (See: Graphs), women represent a 6% increase and men a 8% decrease. The last survey did not have the "Rather not say!" option.

• In the US, 17.1% of the users were female, 80.3% male and 2.6% "Rather not say!".

• There are were proportionately more males (91.6%) than females (7.2%) in Europe than the USA.

• The percent of users (2.5%) whom chose not to respond to this question via the "Rather not say!" option, is relatively stable across other similar types of sensitive questions, like Disabilities (2.4%).

• For Prodigy, 19.1% of the respondents were female, with 78.8% male. These ratios more so reflect the ratios outside the Web and also suggest that as more major online services join the Web and Internet, more balanced female/male ratios are likely to occur. The US and Prodigy ratios also indicate that the US is integrating women more quickly into the user poplulation than other parts of the world.

• Only 2.0% of Prodigy users opted for the "Rather not say!" option, which is close to a whole percentage point lower than the overall non-response rate.

Specific Job

• Given the complexity of graphing over 25 different jobs, we are planning on generating graphs for each major occupation across samples. This will be done by the end of July or so. In the mean time, the table below looks at all occupations.

Nature of Internet Provider

• This question attempts to try to determine the primary provider of Internet access to the respondents. The question was redesigned to include two new entries: local and major online providers. Major online providers are Prodigy, Compuserve, etc., whereas local providers usually only provide connectivity to a regional area. "Commercial" refers to gaining access through primarily through work rather than one of the other sources.

• The three main areas of primary Internet providers are: local online providers (27.9%), major online providers (27.5%), & educational providers (26.5%).

• Compared to the 2nd survey (See: Graphs), more users are gaining access through non-educational domains. The proportion of users gaining access from educational sources is down 24% to 25.% across all users.

• European users predominantly gain access via educational access points (45.6%) and least gain access via major online providers (6.2%), government (2.0%) and military (0.2%).

• For the Prodigy sample, we see that even though users filled out the survey via Prodigy, 9.8% gain primary access from other sources. This suggests that Prodigy users are "early adopters/seekers" of Web technologies in that access to the Web is not readily provided via work or school, but must be sought out via other means.. The Occupation profiles of Prodigy (See: Occupation Graphs) users supports this hypothesis, as most are employees outside of educational (this category includes students) and computer related fields.

• Women are more likely to gain access via major online providers (33.8%) and educational sources (29.0%) and less likely to use local (20.8%) and commercial (7.8%) providers.

Actual Country/State

• Given the complexity of graphing 80 different countries, we are planning on generating graphs for each major geographical area, e.g. Europe, US, Canada. This will be done by the end of July or so.

• In the mean time, the table below looks at all countries/states weighted equally. From this we see that 13.7% of the respondents came from California, followed by New York (5.5%), Texas (4.8%) and the United Kingdom (3.8%).

• These proportions are quite different from the last survey (See: Graph), with about the same rate of response from California, but increased response rates from New York & Texas.

Number of Dependents

• There are more users with two dependents (16.9%) than users with one dependent (14.5%). Only 7.9% report having three or more dependents.

• Of the users with no dependents (61% over all groups), Europeans have s significantly less number of dependents.

• Prodigy users typically have more dependents, with 20.9% with two dependents, and 9.8% with three or more dependents.

• Overall, female users report having less dependents than males - 63.1% for women vs 60.0% for men with no dependents.

Education

• While the US users are more likely to have a college degree (36.0% vs 25.4%), European users are more likely to have Master's (27.7% vs 18.8%) and Doctoral degrees (13.7% vs 4.06%). While the percentages may differ, there are no significant differences between European and US response profiles.

• Both the US and Europe have the nearly same percent (1.8%) of users who have completed Grammar School. This population of younger users has increased from the last survey.

• Compared to the Second Survey (See: Graphs), these users have fewer advanced degrees, but the same number of college graduates (34%).

• Prodigy users tend to have more representation of users who have completed Grammar School (3.2% vs 1.7%), High School (8.6% vs 6.3%), and some college (33.1% vs 34.8%), with less users with Master's (14.7% vs 19.5%) and Doctoral degrees (2.9% vs 7.4%) than the overall sampled population.

• More women have college (37.4%) and Master's (24.0%) degrees verses the men users (34.3% college and 18.7% master's). Though more men have Doctoral degrees (7.84% men vs 4.4% women), there are no significant differences between response distributions across gender.

How People Heard About the Surveys

• This question attempts to begin to understand the response rates and various populations on the Internet and Web. This is the first time we asked the users to inform us how they found out about the current surveys. We broke the data into groups whom responded via announcements to newsgroups, other WWW pages, and listservs, etc. to see if these groups of users are different from one another. This may very well be one of the most interesting data points for understanding the use of the Web for surveying, as lack of differences between groups would indicate an insensitivity towards sampling methods employed.

• Most important was the lack of substantial differences between gender. This increases the stability of the female/male response distributions we received, as it confirms that no systematic sampling bias seems to be in effect. If any segment of the Web user population were excluded, statistically, we'd expect to find similar gender ratios and response distributions. Note, the below percentages may add up to more than 100% as users we able to select more than one response.

• Overall, 50% of the users found out about 3rd the survey via other WWW pages, with 20.3% finding out via "other" sources, and 17.9% finding out via Usenet newsgroup announcement.

• Memory refers to people who remembered to take the survey, and was the lowest method cited (0.44%). This indicates that reliance on people to remember that it is time to participate in a survey is not a very effective means to accomplish longitude tracking of populations.

• WWW based listserver/mailing lists, e.g. www-announce, etc. accounted for nearly 6% of all respondents finding out about the survey, thus are not tremendously effective means to attract attention or draw samples from.

• The "e-mailing" category refers to users whom received mail from out voluntary survey mailing list. While very few users found out about the 3rd surveys via this method (1.1%) as compared to other methods, we note that the 142 users who did respond accounted for 1/4 of the survey mailing list at the time. Thus, survey exclusive mailing lists are a pretty efficient way to announce the beginning of a survey to people.

• It seems that the European newspapers have circulated the Survey more than US ones, as 11.7% of Europeans reported newspapers as a source, with only 2.4% for US users. European users were also less likely to have found out about the survey via Usenet newsgroups (11.0% European vs 21.9% US).

• Prodigy users were more likely to find out about the survey via Usenet newsgroups than the overall population (28.9% Prodigy vs 17.9% All), but less likely to find out via other Web Pages (42.5% Prodigy vs 50.0% All). Intuitively, this makes sense, as Prodigy users just began to gain access to the Web at the time the survey was conducted, and thus probably had yet to discover as many pages and settle into stable navigation behaviors.

• The three most effective entry points to the survey were: pointers from other Web pages (47.4% women vs 50.5% men), other sources (20.4% women vs 20.2% men), and Usenet news announcements (20.5% women vs 17.4% men).

• There were no significant differences between the response profiles of women and men for the following categories: remembering to take the survey, other Web pages, the newspaper, other sources, and listserve announcements. There were differences found for: finding out via friends, magazines, Usenet news, and the www-surveying mailing list. Given the low effectiveness of all but other Web pages and Usenet news announcements, most of these differences lead to nominal effects.

• Thus, the surveys do not appear to suffer much from sampling biases with respect to gender. If a segment of the Web user population were excluded, statistically, we'd expect to find similar response distribution for women and men.

Hours Per Week of "Fun" Computing

• In the last survey (See: Graph a surprising amount of respondents reported using their computer for work & fun reasons over 50 hours a week. We redesigned the 3rd survey to determine how many of those hours are for work and how many of those hours are for fun.

• Overall, and equal number of users report "for fun computing" as under 5 hours/week (34.5%) and between 6 to 10 hours/week (34.9%), totaling nearly 70% of all users.

• Europeans seem to spend less time doing fun computing than their US counterparts. One possible explanation for this centers around the relatively low proportion of Europeans owning computers (See: Graphs) and the nearly two to 1 ratio of European Internet access being gained through educational sources as compared to US Internet access (See: Graphs). Statistical analysis of the interaction between number of computers owned and hours per week doing fun computing supports this hypothesis.

• Prodigy users spend more time having fun than overall users (38.6% having fun 6 to 10 hours/week for Prodigy users vs 34.9% overall users), though there are no significant differences between response distributions for Prodigy and non-Prodigy responses. That is, the amount of time per week doing fun computing is the same for the traditional Web users as for the early adopters. Thus, we expect to find that as the Web continues to expand its user base, that this distribution will remain stable.

• Women spent less time doing fun computing than males (72.9% of women spent less than 11 hours/week vs 68.9% for men).

Disabilities

• Since the Web does not easily facilitate access by users with disabilities, 91% the users do not report any disabilities.

• Of the other disabilities reported, 3% claim vision impairments with the other disabilities (motor, cognitive, hearing, and multiple disabilities) receiving less than 1% apiece.

• The "Rather not Say!" option accounted for 2.4% of the responses. This percent of users whom chose not to disclose this type of information is stable across other questions like Gender (2.5%), etc.

• Compared to the last survey (See: Graphs), which did not allow for the "Rather not say!" option, the same percent reported vision impairments, with more users (95%) reporting no disabilities.

Income

• This was the most sensitive question asked in the General Survey, with 15.7% of the respondent choosing to select the "Rather not say!" option - nearly seven times higher than other similar questions.

• The below analyzes were performed by the Hermes team. The Hermes team develops and analyzes the consumer attitudes and preferences portions of these surveys.

• The overall median income is between $50,000 and $60,000 US dollars, with an estimated average income of $69,000.

• European respondents continue to lag in income, with an average income of $53,500 US dollars.

• Prodigy users' income is the highest of all sampled groups, with a median income in the range of $60,000 and $75,000 and an estimated average income of $80,000 US dollars.

• Prodigy users were 2.3% more likely to opt for the "Rather not say!" response, totaling 18.0% of all Prodigy responses.

• 15.7% of Prodigy users report an income over $100,000, whereas the entire population reported only 13.2% with an income of $100,000 US dollars.

• 18.3% of the women surveyed would "Rather not say!" their income, whereas 14.5% of the men chose this response. This, nearly one in five, non-response interferes with robust gender comparisons between incomes.

• Statistical examination of the response profiles for women and men do not show significant differences with respect to income.

How Long on Internet

• Overall, 31.0% of the users have been on the Internet less than 6 months, with 29.9% having been on the Internet for 1 to 3 years. 23.1% of the users have been on the Internet for 4 or more years.

• Europeans tend to have been on the Internet for a longer time (30.8% over 4 years and only 19.3% under 6 months).

• As one would intuit, the length of time Prodigy users have been on the Internet is markedly less than overall users. More than half (58.7%) have been on for less than 6 months and only 3.7% having been on the Internet for more than 3 years. Keep in mind that the survey started one month after Prodigy announced access to the Web via it's browser.

• Female users report a similar profile to that of Prodigy, with 39.5% having been on the Net less than 6 months. A higher portion of the sampled female population have been on the Internet for more than 3 years (16.9%).

Major Geographical Location

• For classification of location by major geographical location, 80.6% of the respondents were from the US, 9.8% from Europe, and 5.8% from Canada and Mexico, with all other major geographical locations begin represented, but to a lesser degree. Step towards replicating the survey in other continents and providing some extent of multilingual support might alter these differences.

• 99.0% of all Prodigy users are from the US. This is as would be expected.

• Percentage wise, more women were from the US (88.7%) than men (78.9%), though the proportion of women to men was reversed for all other locations like Europe (11.0% male & 4.6% female). This suggest that as the Web increases its user population, a more balanced women/men ratio is expected.

Marital Status

• Overall, 50.3% of the users are married, with 40.0% being single. The users whom reported being divorced was 5.7%.

• The "Rather not say!" option accounted for 2.7% of the responses, very similar to other like questions, e.g. Gender.

• There are more single European users (58.3%) than married (34.1%).

• The converse occurs in the US, with 37.1% of US users being single and 52.8% being married.

• There are not many widowed or separated respondents (total 1.6% combined).

• The majority of Prodigy users are married (61.4%), with only 27.3% reporting being single. This proportion is significantly higher than the non-Prodigy population.

• The percentage of divorced Prodigy users is slightly higher than others, with 7.3% responding as being divorced.

• Roughly 50% of the women and men users are married, with women being less likely to be single (36.0% vs 41.1% male) and more likely to be divorced (9.4% vs 5.0% male). The response profiles for this question were significantly different.

Number of Computers Owned

• Nearly half the sampled population report owning a single computer (46.1%).

• One in five respondents own three or more computers (20.2%).

• Since only 5.9% of all users report not owning a computer, it seems that those who have online access tend to own a computer. This follows common sense that people who spend a lot of time with computers tend to own at least one computer.

• European users are almost three times more likely (13.1%) not to own a computer than their US counterparts (4.9%).

• Only 1.2% of Prodigy users did not own a computer, with almost one half (46.1%) owning a single computer and 29.1% owning two computers.

• The percent of Prodigy respondents that do not own a computer matches perfectly the number of users that do not subscribe to Prodigy (See: Graphs). One possible read of this is that there were 41 users who filled out the survey using Prodigy that do not own a computer or subscribe to Prodigy (the "over the shoulder respondents"). This provided a good sanity check of the numbers collected.

• Women are twice as likely as men to not own a computer (9.7% female vs 5.2% male), and more likely to own only one computer (52.7% female to 44.5% male). This lends support to the conclusion that for this sampled population, women typically own fewer computers.

Occupation

• Computer (31.4%) and Educational (23.7%) occupations still represent the majority of respondents. Professional (21.9%), Management (12.2%), and "other" occupations (10.8%) fill out the other categories.

• As expected, there is a high degree of dependence among income and occupation.

• Europeans tend to be predominantly in Education (36.4%) and Computers (33.4%) were as the US sample contained mainly Computer (31.1%) and about equally divided amongst the Professional (22.9%) and Educational (21.6%) professions.

• Prodigy users reversed the overall employment profiles, with the dominant occupation being Professional (28.7%), followed by "other" occupation (20.2%), Management (19.5%), Computer (18.2%) and Education (13.4%).

• There are less women in than men in Computers (26.5% women vs 32.4% men) and Management (10.5% women vs 12.7% men), but more women fall into the "other" occupation category (15.7% women vs 9.9% men).

• Education and Professional occupations show marginal differences between gender.

Online Services Subscribed To

• Overall, 39.0% of all users do not subscribe to an online service, with Prodigy (37.0%), America Online (30.0%), and Compuserve (21.1%) being the most dominant subscribe online services.

• European subscribe mainly to either no services (65.6%) or services not listed as part of the question (23.8%).

• For the US respondents, Delphi (2.2%), Genie (1.5%) and Pipeline (0.6%) are the least cited subscribed online services.

• Not surprisingly, nearly all Prodigy respondents subscribe to Prodigy. The fact that this percentage is not 100% indicates that some users may have filled out the survey via Prodigy, but do not subscribe (the "over the shoulder respondents"). See Graphs, for more supporting evidence for this phenomenon.

• Women are more likely than men to subscribe to Prodigy (36.7% female vs 29.0% male) and America Online (27.4% female vs 24.0% male). For the other online services, there were no statistically significant differences in the response profile with respect to gender.

Willingness to Pay Fees for WWW Accesses to Sites

• Overall, 22.6% of the respondents outright stated that they would not pay fees to access material from WWW sites. This is the same ratio as observed in the 2nd survey (See: Graphs).

• Europeans tend to be most concerned about both cost and quality (66.8%) of information. That is, they would only pay if both the cost and quality were right. Much fewer would pay for access solely based upon the quality of information provided (7.21%) than the cost (3.29%).

• US users tend to be most concerned about both the cost and the quality (67.3%) of information. That is, they would only pay if both the cost and quality were right. Much fewer would pay for access solely based upon the cost (4.8%) and less towards the quality (2.7%) of the information.

• The same trend of concern more towards cost (6.4%) than quality (2.4%) is also observed with Prodigy users, with the majority still concerned mainly about both the cost and quality of information (66.4%).

• There were no statistically significant differences found between the Prodigy and non-Prodigy response distributions for this question. This implies that as the Web increases its user base, we'd expect to continue to find a 20% negative response to paying for access to Web sites.

• Women significantly more concerned about both cost and quality of provided information (70.5% women vs 66.4% men) and less were flat-out opposed to paying at all (20.5% women vs 22.7% men) as compared to men.

Primary Computing Platform

• The distribution of primary computing platforms across all sampled populations closely resembles computer marketing reports like the NPD Group: 52.0% Windows, 26.2% Macintosh, & 8.8% Unix. These three platforms account for 87% of all platforms reported.

• These ratios differ greatly to those collected during the 2nd survey (See: Graphs), where the distribution was: 44% Unix, 29% PC, & 19% Macintosh. Clearly, online services providing access to the Web and the Internet have influenced and helped balanced these ratios.

• These number support the hypothesis that the Web is now mid stream in its evolution, with the first stages being dominated primarily by the technology developers (UNIX based) followed by wider acceptance by the "early adopters/seekers" (PC and Mac based) in the later stages.

• Not surprisingly, Prodigy users almost exclusively use Windows as their primary computing platform (86.5%). The fact that this percentage is not 100% indicates that some users may 1) subscribe to Prodigy and 2) have filled out the survey via Prodigy, but primarily use other platforms, e.g. when at work, etc.

• Overall, women are more likely to use Windows (59.1% women vs 50.9% men) than Macintoshes (22.2% women vs 26.92% men) or Unix machines (7.5% women vs 8.9% men) than men.

Race

• 82.3% of the respondents are white, with none of the other groups reporting over 5% of the responses.

• Europeans have a higher ratio of white users (93.5%).

• There were no statistically significant differences observed among the Prodigy/non-Prodgy and Women/Men sub-samples.

• The ratio of "Rather not say!" was about twice (5.3%) that of other questions.

• These numbers for race may reflect certain fundamental problems with the Web and Internet in general: replication of services and multilingual support. Stated differently, Web users who tried to connect from Asia may 1) not have been able to establish and hold a connection due to transmission speed and delays 2) the costs involved in participating and 3) the lack of a survey translated to their native language. We note that timeouts did occur on some connections (will compute ratio in July as part of a set of recommendations for survey design and execution). Additionally, we know of no online multilingual survey.

Share Machine With Others

• This question attempts to begin to understand the ratio of computers on the Web and the number of users behind each computer. This is really very tricky stuff, as there is no registration mechanism in place to accomplish a census of users and randomly sampling computers is essentially spamming (i.e. not good net behavior). So, take these numbers with a big grain of salt along with a nice icy cold margaritta or other preferred chilly beverage.

• Overall, 85.7% of computers were reported as being shared. This is a 15% increase from the 2nd Survey (See: Graphs). This may very well reflect that as major online service providers provide access to the Web and the Internet, the computers that provide the users with access are being used by more than one person. This can not be the only influence, as since Prodigy represent approximately 1/4 the entire sample size, if all Prodigy users reported sharing their machines, it would not raise the percent past 80%, so some other influences are at play. We will be investigating these influences in early July.

• European users are more likely to be the sole user of their machine (22.3%) than their US counterparts (12.9%), even though they tend to own less computers (See: Graphs).

• To further support the effect of other influences since the 2nd survey on sharing computers, the Prodigy sample did indeed report a higher percentage of shared machines, but only by 4.6%.

• Women are 6% less likely to have to share a machine compared to men. There are significant differences between the response distributions for gender.

Hours Per Week of Computing for Work

• In the last survey (See: Graph) a surprising amount of respondents reported using their computer for work & fun reasons over 50 hours a week. We redesigned the 3rd survey to determine how many of those hours are for work and how many of those hours are for fun.

• The European profile (peaks at 31 to 40 hours/week) of hours of computing for work differs notably from the US profile (more even across all ranges of hours/week).

• One out of every four Europeans (24.8%) use a computer for work between 31 and 40 hours/week (US - 16.2% between 31 and 40 hours/week).

• US respondents had a one in three chance (31.8%) of working with computers for less than 11 hours/week (Europeans - 16.8% under 11 hours/week).

• Over half of the Prodigy users (53.9%) reported using a computer less than 11 hours/week, with nearly 1/3 using a computer less than 5 hours/week for work.

• Women differed from men in hours/week working with computers in two ways: 1) women are more likely to use a computer under 5 hours/week (19.4% female vs 14.8% male), but 2) are more likely to use a computer between 31 and 40 hours/week (19.4% female vs 16.7% male). We are unsure about what underlying correlations may be influencing these results.

---

[ Survey Home ] [ 3rd Survey Home ] [ Graphs ] [ Bulleted Lists ] [ Datasets ]

---

For more information or to submit comments:
send e-mail to www-survey@cc.gatech.edu.

GVU's WWW Surveying Team
Graphics, Visualization, & Usability Center
College of Computing
Georgia Institute of Technology
Atlanta, GA 30332-0280