|
Get AfricaFocus Bulletin by e-mail!
Print this page
|
Note: This document is from the archive of the Africa Policy E-Journal, published
by the Africa Policy Information Center (APIC) from 1995 to 2001 and by Africa Action
from 2001 to 2003. APIC was merged into Africa Action in 2001. Please note that many outdated links in this archived
document may not work.
|
APIC: Web Visitor Survey
APIC: Web Visitor Survey
Date distributed (ymd): 991121
APIC Document
+++++++++++++++++++++Document Profile+++++++++++++++++++++
Region: Continent-Wide
Summary Contents:
This posting contains an analysis of responses to a survey on
the Africa Policy Web Site, from more than 9,000 visitors who
filled in the survey between June 1997 and mid-August 1999.
The survey shows that the web site reaches a very different
audience than the e-mail electronic distribution list --
including many new to African issues. In particular, web
visitors filling out the survey are on average much younger
and have far fewer already existing connections with Africa
than subscribers to the distribution list. The largest
institutional sector represented is the educational sector,
including a significant number of secondary school teachers
and students.
We invite those of you among our e-mail subscribers who never
or rarely visit the site to check it out, if you have not done
so recently. We predict you will find it a useful resource
when you need to locate information on African issues on the
web (start with our own search page at
http://www.africapolicy.org/search.htm; or use the regional or
thematic pages to find a host of additional sources). Perhaps
even more important, you can refer the site to others who are
less informed and less familiar with African issues than you
are but who are searching for resources for themselves or
their students.
A more detailed version of this analysis is available at:
http://www.africapolicy.org/survey/webrep99.htm
Previous reports from annual surveys of e-mail distribution
list subscribers can be found at:
http://www.africafocus.org/docs99/srep99.php
http://www.africapolicy.org/survey/report97.htm
http://www.africapolicy.org/survey/report96.html
+++++++++++++++++end profile++++++++++++++++++++++++++++++
Africa Policy Web Site
Reader Survey Analysis
September 1999
A reader survey form first placed on the web site in June 1997
(
http://www.africapolicy.org/survey/websurv.htm) has proved to
be
(1) a highly effective tool in eliciting reader feedback
and,
(2) through offer of a free Africa's Regions poster, also
effective in attracting new visitors to the site and prospects
for other APIC information services.
The number of visitors filling in the survey grew modestly in
1997 and the first half of 1998, averaging a little more than
three a day in 1997 and almost five a day in the first six
months of 1998. In August 1998, however, the site's offer of
a "free poster" was discovered by the "freebies" community on
the internet. With links to our site appearing on many of
these sites (sometimes as "freebie of the day" or "best poster
offer"), the demand for posters -- and the influx of new
visitors to the site -- grew in rapid and sometimes
unpredictable spurts. The number filling in the survey
averaged almost 15 a day in the last six months of 1998 and
more than 20 a day in 1999. By fall 1999 the average was
fluctuating between 15 a day and 20 a day.
This report analyzes survey returns both in terms of visitor
demographics and evaluation of the web site and in terms of
the survey's free poster offer as a means of outreach to new
constituencies. With the rise of the links to "freebie"
sites, a higher proportion of survey respondents were
attracted simply by the offer and were first-time visitors.
At times, the demand grew so rapidly as to indicate abuse,
which was identified and protected against by new software
checks on the visitors to ensure that they did indeed visit
pages on the site before filling in the survey. The fact that
the site was widely publicized on "freebie" sites in Russia
and other former Soviet bloc countries introduced a further
distortion.
For the purpose of analysis, therefore, the respondents were
divided into three groups. First, Eastern Europe (including
Russia), clearly not representative of visitors to the site as
identified by web hit logs, was excluded for separate
analysis on a later occasion. The remaining responses were
divided into first-time visitors and repeat visitors.
Unlike the annual surveys of the e-mail distribution list
(1996-1999), the web survey is unlikely to be a representative
subset of visitors to the site. Results from the web survey
represent only the respondents themselves -- a substantial
fraction of the visitors, and arguably the most interested
rather than purely casual visitors -- but not a majority of
those seeing the site.
For analysis of the core constituency to which the site
appeals, the relevant subgroup is repeat visitors. From June
1997 through mid-August 1999, there were 2,596 survey
responses in this group.
Africa Policy Web Site Core Constituency
1. Overview
It is notable that there is relatively small overlap
between the audience identified by respondents to the annual
survey of the Africa Policy Electronic Distribution List and
that identified by respondents to the web survey. Only 5.5%
of these repeat visitors to the web site said they subscribed
to the Africa Policy Electronic Distribution list. Even if
some of these subscribe to the list after filling out their
web survey, this would hardly change the conclusion that those
reached by the web and those reached by e-mail are quite
distinct groups of people. Similarly, the e-mail survey
revealed that a significant fraction of subscribers
(approximately 75%) never or very rarely visit the web site,
although more than 90% of them do have access to the web.
This reinforces APIC's strategic decision to present the same
information through both e-mail and the web, in order to reach
larger audiences than is possible using either vehicle alone.
Please note that the data discussed below is for the full
period June 1997 through mid-August 1999, and that the
percentages therefore indicate averages for that period.
2. Countries and Continents of Residence
Three-quarters of the core constituency of the web site is in
the USA and Canada. Among African countries, only South
Africa, in sixth place with a bit over 1% of the survey
responses, ranks among the top 25 in repeat visitors filling
in the survey. The European country most widely represented
is the Netherlands with more than 4%, followed by the UK with
3%.
Nevertheless, there is a wide range of countries represented,
including 16 African countries in addition to South Africa, 13
other countries in the Americas, 20 countries in Europe (not
including 19 Eastern European countries excluded from this
section of the analysis), and 23 countries in the Middle East,
Asia and the Pacific. Altogether, this core constituency
includes respondents from 75 countries (again not including
the 19 Eastern European countries).
Table 1: Regions of Residence of survey respondents (excluding
Eastern Europe)
Region Number Percent
USA/Canada 1955 75%
W. Europe 353 14%
Asia/Pacific 163 6%
Africa 66 2.5%
Other America 59 2.3%
This shows a significant contrast with the pattern for the
e-mail Africa Policy Electronic Distribution List, for which
in 1999 the proportion of respondents in Africa reached 12.5%
and the proportion in the USA and Canada was 72%.
Table 2: Countries of Residence, in order by number of survey
respondents
COUNTRY Number Percent
USA 1824 70.26
CANADA 131 5.05
NETHERLANDS 114 4.39
UK 69 2.66
AUSTRALIA 42 1.62
SOUTH AFRICA 34 1.31
BELGIUM 29 1.12
GERMANY 25 0.96
INDONESIA 23 0.89
SWEDEN 17 0.65
ITALY 15 0.58
BRAZIL 15 0.58
and under 15 each for New Zealand, Mexico, France, Sri Lanka,
Turkey, Philippines, Denmark, Scotland, Israel, India,
Malaysia, Argentina, Portugal, Kenya, Ireland, Switzerland,
Singapore, Peru, Malta, Uzbekistan, Trinidad and Tobago,
Mauritius, Kazakhstan, China, Austria, Uganda, Pakistan,
Greece, Finland, Ethiopia, Colombia, Chile, Zambia, Venezuela,
South Korea, Hong Kong, Egypt, Cyprus, Cote D'Ivoire, Tunisia,
Thailand, Tanzania, Syria, St. Croix, Spain, Senegal, Namibia,
N. Ireland, Mozambique, Liberia, Lebanon, Jordan, Japan, Iran,
Ghana, Ecuador, Costa Rica, Cambodia, Burundi, Bolivia,
Bermuda,Bahrain, Angola
3. Connections with Africa
As shown in Table 3, even these repeat visitors to the web
site were a group with much less Africa experience and
connections than the subscribers to the e-mail distribution
list. Fourteen percent were born in Africa, as compared to
21% of the list subscribers. Fifteen percent had worked in
Africa, as compared to 49% of the list subscribers. Twenty
percent had visited Africa, as compared to 43% of the list
subscribers. Only 46% said they had activist concerns, as
compared to 60% of the list subscribers. The proportions
saying they had business or academic interests, however, was
in each case slightly greater than those for the list
subscribers.
Table 3: Connections with Africa
Connection Web Survey E-mail Survey
Repeat Visitors 1999 (percent)
(percent)
Born in Africa 14.2 21.1
Worked in Africa 14.8 48.8
Visited Africa 20.3 42.6
Born in, worked in or
visited Africa 37.2 80.6
African Diaspora 8.7 12.3
African Diaspora or born
in Africa 19.6 31.2
Business interests 15.1 14.3
Academic interests 55.2 52.8
Other professional
interests 25.2 35.4
Activist concerns 46.0 59.8
4. Other Demographic Characteristics
Repeat visitors to the web site were strikingly younger than
subscribers to the e-mail list. List subscribers were roughly
divided into thirds in the age ranges 20 to 35, 36 to 50, and
over 50, with almost none under 20. Web visitors, however,
included 15% under 20 and only 6% over 50 (see table 4).
Table 4: Age
AGE Percent Percent
(repeat web visitors) (1999 list survey)
Under 20 15% less than 0.5%
20 to 35 56% 29%
36 to 50 23% 41%
Over 50 6% 30%
In terms of sex, repeat visitors to the web site were 49%
female, as compared to only 44% of the subscribers to the
distribution list.
As would be expected from the age profile, the educational
level completed among the web visitors was also very distinct
from list subscribers. 75% of the list subscriber survey
respondents held advanced degrees, 24% more held college
degrees, and only 1% reported secondary level schooling.
Among repeat web visitors, however, 33% held advanced degrees,
42% college degrees, and 25% secondary school level only.
The range of institutional affiliations of web visitors and
list subscribers also differed, although not so dramatically
as the age and education profiles.
Table 5: Institutional Affiliation
Institution Percent Percent
(repeat web visitors) (1999 list survey)
Education 50% 43%
NGO 13% 21%
Business 13% 7%
Government 8% 8%
Media 3% 3%
Religious 5% 12%
Other 7% 6%
5. Evaluation of Web Site
Among the repeat visitors responding to the survey, the
average ratings for both quality and accessibility of the web
site was in each case 4.2 on a five-point scale in which
"good" is counted as 4 and "very good" is counted as 5. This
is almost identical with the ratings given documents by
subscribers to the electronic distribution list.
Respondents were also close to list subscribers in saying that
in the case of documents calling for action they contacted
policymakers (34% doing so at least "sometimes", as compared
with 38% for list subscribers) or passed on the documents to
others (49% doing so at least sometimes, as compared to 52%
for list subscribers).
These numerical ratings are probably less revealing than the
comments by approximately one-fifth of those who filled in the
survey as repeat visitors (see full report cited above for
citations).
B. First-Time Visitors
The number of survey responses from first-time visitors, from
June 1997 through mid-August 1999, totals 6,411. The utility
of the responses as a evaluation of the site or for analysis
of the demographic profile of even of first-time visitors is
very doubtful, given that a large number are coming to the
site specifically in order to get a free poster. It is better
conceived as an analysis of prospects.
In order to protect against abuse, however, a new feature
using Javascript was introduced at the end of June, to ensure
that visitors take a tour of the site before being allowed to
fill out the survey. Earlier, in 1998, a software feature
(also using Javascript) was introduced to block submissions
from other sites (including several in Russia) that had copied
the survey already filled out in order to allow their users to
submit the survey without ever visiting our site at all.
1. Regions of Residence
Table 6: Regions of Residence of survey respondents (excluding
Eastern Europe)
Region Repeat Visitors First-time Visitors
USA/Canada 75% 85%
W. Europe 14% 8%
Asia/Pacific 6% 4%
Africa 2.5% 1.4%
Other America 2.3% 0.7%
First-time visitors filling out the web survey are even more
likely than repeat visitors to come from the USA or Canada.
This is probably a result of the fact that the "freebie" site
phenomenon is still primarily limited to North America,
Western Europe and Eastern Europe.
2. Countries of Residence
Table 7: Top Countries of Residence, in order by number
COUNTRY % of Repeat Visitors % of 1st-time Visitors
USA 70.26 80.1
CANADA 5.05 4.9
NETHERLANDS 4.39 2.5
UK 2.66 2.3
AUSTRALIA 1.62 1.7
SOUTH AFRICA 1.31 0.7
BELGIUM 1.12 0.7
GERMANY 0.96 0.5
INDONESIA 0.89 0.5
Countries represented in the first-time visitor survey
respondents, with less than 0.5% of the total respondents,
also included 19 additional African countries* and 47
additional non-African countries.
3. Connections with Africa
Table 8 shows that as expected, the first-time visitors among
the survey respondents have on average fewer connections with
Africa than do the repeat visitors, and much less than the
subscribers to the e-mail list. This means that the poster
offer to the survey is reaching out to constituencies which
include a significant number with new or limited interest in
the continent. The relatively high proportion indicating an
academic interest, together with other evidence from the
addresses and comments, indicates that many of these
respondents are teachers and secondary-school students seeking
information for their classrooms.
Table 8: Connections with Africa,
Comparison of web survey respondents and e-mail survey
respondents
Connection First-time (%) Repeat (%) E-mail 1999 (%)
Born in Africa 6.4 14.2 21.1
Worked in Africa 6.0 14.8 48.8
Visited Africa 11.2 20.3 42.6
Born in, worked in
or visited Africa 19.5 37.2 80.6
African Diaspora 4.6 8.7 12.3
African Diaspora
or born in Africa 10.2 19.6 31.2
Business interests 8.4 15.1 14.3
Academic interests 47.2 55.2 52.8
Other professional
interests 16.6 25.2 35.4
Activist concerns 36.9 46.0 59.8
4. Other Demographic Characteristics
The age distribution of first-time visitors was roughly
similar to that of repeat visitors, showing a similar
dominance of the age ranges 35 and under. In terms of sex, the
first-time visitors were 59% female, perhaps indicating the
prominence of teachers (including home-school teachers) among
those visiting the site specifically in order to get the
poster. The educational level completed among first-time
visitors was also somewhat lower than among repeat visitors,
with 24% having advanced degrees and 40% having college
degrees.
First-time visitors were also more likely than repeat visitors
to come from the educational sector (see Table 9).
Table 9: Institutional Affiliation
Institution Percent Percent
(repeat visitors) (first-time visitors)
Education 50% 55%
NGO 13% 11%
Business 13% 11%
Government 8% 6%
Media 3% 3%
Religious 5% 6%
Other 7% 9%
Selected comments from first-time visitors are available in
the full report cited above.
This material is being reposted for wider distribution by the
Africa Policy Information Center (APIC). APIC's primary
objective is to widen international policy debates around
African issues, by concentrating on providing accessible
policy-relevant information and analysis usable by a wide
range of groups and individuals.
|
