Women in Computing around the World: an Initial Comparison of

International Statistics
Vashti Galpin
School of Computer Science, University of the Witwatersrand, South Africa
[email protected], http://www.cs.wits.ac.za/vashti
Abstract
This technical report describes the participation of women in computing in more than 30
countries, by focussing on participation at undergraduate level. A brief discussion covers
how societal and cultural factors may affect womens participation. Statistics from many dif-
ferent sources are presented for comparison. Generally, participation is low most countries
fall in the 10-40% range with a few below 10% and a few above above 40%. In the appen-
dices, the approach to data collection is described and statistics for women in computing and
mathematics are presented.
1 Introduction
This report presents a picture of the participation of women in computing around the world. Much has been published
on the low participation of women in computing in the USA, UK and Australia; and there are researchers active in
other countries seeking to describe and understand their own situations. It is important to consider the situation in
different countries to avoid incorrect assumptions and to nd appropriate solutions.
Recently, steps have been taken to obtain a global picture of the situation with the introduction of the ACM-W
Ambassador program by the ACMs Committee on Women in Computing (ACM-W). Each ACM-W Ambassador will
provide information about the status of women in computing in their country via a website links to these sites can
be found at http://www.acm.org/women.
This paper focusses on participation in academic study, specically university undergraduate level, as this is a
relatively available statistic. Other measures will be used to ll out the picture where necessary. Computing has been
described as a discipline that contains aspects of science, engineering and mathematics [16]. As a result of this complex
nature, it is not possible to infer the situation of women in computing from information about women in science, or
information about women in engineering. Hence, published statistical information that aggregates disciplines is not
useful in determining the status of women in computing. This article presents mostly information about computing,
but presents more general information in places to broaden the picture.
The next section briey considers how cultural factors affect the participation of women in computing. The main
section of the paper then presents data from different countries. The data is presented compactly in tables, and details
of how the tables can be interpreted are given. Finally, a brief discussion of the data is given. The rst appendix
describes how data was collected and the second appendix presents the percentage of women studying computer
science and mathematics at tertiary level for 100 countries.
2 The effect of culture and society
The reasons that women choose to study computing will vary from culture to culture, and from country to country,
and it is beyond the scope of this paper to consider this issue in detail. When seeking solutions for womens low
participation in computing, it is important to consider all cultural and societal factors that may affect this participation.
This also allows us to identify when a solution from one country may or may not be suitable to use in another country.
For example, Mukhopadhyay [34] argues that the internal self-selection model used to explain the participation
of women in science in the USA, cannot be applied to India. A model with the family as decision-maker is more
appropriate and explains why there is lower participation in the applied sciences such as engineering and technology
when compared to the pure sciences such as physics, chemistry and mathematics.
1
In 1994, Science published a special issue comparing women in science across a number of cultures and countries.
Factors that were associated to high numbers of women in science are [4]: girls-only schooling (India, predominantly
Catholic countries), compulsory mathematics and science through secondary school (Poland, Italy), family-friendly
societies (Israel, Mediterranean countries), perceptions of science as a low-status occupation when compared to dis-
ciplines such as engineering (Portugal, Turkey, India), class issues (India, Latin American countries) and recently
developed science capabilities (Portugal, Mexico, Argentina).
It is not clear whether these factors apply to computer science as well. Researchers have investigated cross-
cultural gender issues in computing and some of these studies are now briey described. Janssen Reinen and Plomp
[27] considered primary and secondary school students from Austria, Bulgaria, Greece, India, Japan, Latvia, the
Netherlands, Slovenia and the USA collected in 1989 and 1992. All countries showed gender differences in basic
knowledge and skills except Bulgaria, USA and India. The lack of difference in the USA was attributed to high
number of female teachers, parental encouragement and computer use outside school, and in Bulgaria it was attributed
to high number of female teachers.
In their study of university students in 23 countries, Weil and Rosen [69] found that in Thailand, Italy and Kenya,
males were signicantly more anxious about computers, whereas in Israel and Hungary, women were signicantly
more anxious. There were signicantly more male technophobes in Kenya, and signicantly more female techno-
phobes in the USA, Hungary and Australia. In the USA, Singapore, Kenya, Israel, Hungary, Czechoslovakia, Belgium,
Australia and South Africa [11], men had signicantly more positive cognitions with the reverse in Northern Ireland.
In one country, Indonesia, female students had signicantly more experience than males, and the opposite was found
in Yugoslavia-Croatia, Thailand, Mexico, Japan, Italy, India, Hungary, Germany, Czechoslovakia and Australia.
Makrakis [31] considered computer self-efcacy amongst Japanese and Swedish secondary school pupils and
found no gender differences in either country. The research showed the We can, I cant paradox [47, cited in [18]]
appeared amongst Japanese students of both genders but not amongst the Swedish students, and this was attributed to
a greater focus in Japanese society on group identity. A similar study investigating self-efcacy amongst Romanian
and Scottish higher education students [17] found differences between the male and female students but there was no
interaction of gender and nationality. Collis and Williams [12] found that there were fewer gender-based differences
between Chinese students than Canadian students. The main gender-based difference for Chinese students was in the
perceptions of womens abilities where female students were much more positive than male students. This difference
also occurred amongst the Canadian students.
Within a country, students of different backgrounds can have different experiences. Von Hellens and Nielsen [68]
note that amongst IT students at an Australian university, female Asian students feel they are ignored by non-Asian
students and male Asian students, whereas female non-Asian students feel they are the focus of sexual harassment and
unwanted positive discrimination.
Not all countries have low participation by women. In 1987 more than 50% of application/analyst programmers
and system analyst/designers in Singapore were female, and the majority of graduates from computer courses were
female [29]. Uden [64] argues that this occurs because of government promotion of the use of computers, perceptions
of good career prospects in IT, a preference amongst women for computing as opposed to engineering which also pays
well, exposure to computers at schools level in a gender-neutral manner, and assistance with domestic responsibilities
by older family members or employees.
3 Womens participation in computing
The specic measure that will be considered in the tables that follow is women as a percentage of the total number of
undergraduate computing students or computing graduates. Hence, the data collected concentrates on students taking
computing at tertiary education level as a major subject, in the sense that their qualication will focus on computing.
In some cases, the data refers to students in a particular year of study who are taking computing courses at a major
level. Where is was not possible to determine the type of course, the course has been briey described. Most data
is drawn from universities, but as the tertiary education system varies from country to country, information has been
drawn from other types of institutions such as technical colleges. Other data not specic to computing at tertiary level
will be used to give a more complete picture.
Not all countries publish easily available national data recording the percentages, so data has been obtained from
two other sources: data from individual universities and data reported by individuals attending or presenting classes at
university level. There are four tables, Tables 1-4, grouping together countries from similar geographical areas. For a
discussion of the approach taken in collecting data, see Appendix A.
2
Country Data Year Trend Type Source
Botswana 10% (2) 1998 UG fy, CS Ind, U of Botswana
a
[54]
Eritrea < 10% 2001 UG, CS Ind, U of Asmara
b
[33]
Madagascar 11.1% 1997? Other
c
National [67]
Kenya 11.1% (3) 2001 Deg, CS Ind, U of Nairobi [46]
Libya 35.7% (606) 2002 Deg, CT Insts, Alfateh U, 7th of April U,
Altakadum U [2]
Nigeria 31.3% (10) 1997 UG, CS Ind, Ogun State U [51]
20%-30% 1994-6 UG, CS Ind, U of Benin [1]
32.6% 1994 28.7% 1991 CS Insts, Nigerian polytechnics [23]
South Africa 32.1% (337) 1998 none 1991 Deg, CS&IS National [49]
Tanzania 3% (2) 1996 14% 1993 UG, Inf Insts, U of Dar-es-Salam, Sokoine
and Muhimbili [67]
Uganda 27% (9-10) 2000 UG fy, CS Ind, U of Makerere [21]
Zimbabwe 40.7% (341) 1996 CS Insts, Technical colleges [67]
a
The University of Botswana is the only university in Botswana offering a BSc in Computer Science.
b
The University of Asmara is the only university in Eritrea.
c
Computer Science higher education teachers.
Table 1: Africa
3.1 Reading the tables
The rst four tables have the format described below. Information that is unclear is marked with a question mark, and
gures that are approximate are indicated with .
Country: Some countries have more than one entry, when the data is reported from different sources. The data is
presented as discovered, so there is the possibility of contradictory or disparate data.
Data: In this column, a percentage is given calculated from the total number of women and the total number of
people, and if available, the actual number of women is given in brackets.
Year: This is the year or years that the data comes from. Where year information was given as an academic year
split over two calendar years, the most recent year of the two is used in these tables.
Trend: This indicates any trends in the percentage data. If the percentage in the Trend column is less that that in the
Data column, this means that since the year given in the Trend column, there has been an overall trend for the data to
increase. If the percentage in the Trend column is more than that in the Data column, then since the year given in the
Trend column, there has been a trend for the data to decrease. If none appears in the trend column then since the
year indicated in the Trend column, there has been no discernible trend. In some cases, the trends have been inferred
from the data available, and in other cases, it has been described as a trend by the source of the data.
Type: This column covers the level and the discipline to which the data refers. As mentioned above, the data
collected refers to the study of computing at major level at tertiary institutions. Where it was not possible to determine
whether this was the case, the category Other has been used.
The abbreviations for level are: UG undergraduate study, fy nal year, 1y rst year, dist distance
learning, Deg rst/undergraduate degree (this category excludes postgraduate/graduate qualications), PG
postgraduate study, Acc applicants accepted to degree programmes. The abbreviation UG&MSc is used when
gures are given for both undergraduate and MSc study. If the level could not be determined from the source, the level
is not given.
The abbreviations used for discipline are: CS Computer Science, Inf Informatics, IT Information
Technology, IT&T Information Technology and Telecommunications, Cmp Computing, CmpSci Comput-
ing Science, CmpStd Computer Studies, CSS Computer and System Sciences, CSys Computer Systems,
BusCmp Business Computing, Tech Technology, CT Computer Technology, TechInf Technical Infor-
matics, and IS Information Systems.
3
Country Data Year Trend Type Source
India 20.3% (15) 2002 none 1993 UG, CS Inst, Annamalai U [59]
11.3% (11) 1996 CSS Inst, J Nehru U [10]
7.84% (22,857) 1994 none 1992 Other
a
National [10]
28.42% (27) 1993 7.3% 1982 IT Inst, Andhra U [22]
Iran 41% 1999? CS Insts, vocational and training insti-
tutions [50]
Pakistan 18.2% (8) 2000 PG, CS Insts, Q.A.U. Islamabad [5]
4.99% (685) 1998 Other
b
Insts [5]
Malaysia 51.4% (2,167) 1991 Other
c
National [39]
Singapore > 50% 1987? Other
d
Insts [29]
Thailand 55% (158,286) 1998 57.2% 1996 Other
e
Insts [36]
Turkey 20.4% (1,753) 2001 18.3% 1997 UG, CSys National [37, 3]
Australia 19% 1998 22% 1994 IT&T National [38]
49% 1995 35% 1990 UG, BusCmp Inst, Victoria U [13]
New 20% 1992-6 Deg, CS&IS Inst, Massey U [48]
Zealand 26% 1989-96 UG 1y, Cmp Inst, Auckland U [48]
17-23% 1990-6 UG 1y, CS Inst, Victoria U of Wellington [7]
a
Students studying Engineering and Technology. National statistics do not give separate gures for computing [10].
Most computer science departments are located in faculties of Engineering and Technology [22].
b
Enrollment at universities of engineering and technology.
c
Students enrolled for computer related courses at tertiary institutions.
d
Graduates from computer courses from four public institutions.
e
Participation in computer courses at private vocational institutes.
Table 2: Asia and Australasia
Country Data Year Trend Type Source
Czech 9.6% (51) 2001 UG, Inf Ind, Masaryk U [9]
Republic 25% 1989-94 UG&MSc, CS Ind, Charles U [58]
Denmark 6% 1996 CS National [43]
Finland 20% 1997 31% 1985 UG 1y, IT National [43]
Germany 10.5% (610) 2000 16.2% 1995 Deg,Inf National [57]
8.8% (46) 1999 Inf Inst, U Karlsruhe [28]
9% 1999 Inf Inst, RWTH Aachen [28]
9.5% (2,958) 1994 18.8% 1979 UG, Inf National (West Germany) [41]
Iceland 24% (47) 2000 CS Inst, U of Iceland [52]
28% (42) 1999 UG, CS Inst, Reykjavk U [52]
Netherlands 6.6% (7) 1999 TechInf Inst, Technical U Delft [28]
18% 1992 UG dist, Inf Inst, Open U of the Netherlands [15]
12% 1991 Tech Insts, traditional universities [63]
Norway 23.2% (1,691) 1999 20.4% 1996 Tech National [56]
34% (69) 1998 6% 1996 CS Inst, Norwegian U of Science and
Technology (NTNU) [40]
Slovenia 6.7% (94) 2000? CS&Inf Inst, U of Ljubljana [55]
Spain 25.2% (1,101) 1998 Deg, Inf
a
National [25]
Sweden 30% (16,245) 2000 UG, Tech National [60]
10% (24) 1990-4 UG, CS Inst, Uppsala U [6]
Switzerland 11.4% (122) 2001 4,2% 1995 UG, Inf Inst, ETH Z urich [30, 20]
6.5% (2) 1998 Deg, Inf Inst, EPFL [24]
UK 19% (3,444) 1999 Acc, CmpStd National [42]
a
Students graduating with Diplomado Inform atica, Ing. T ec. Inform atica de Sistemas, and Ing. Inform atica.
Table 3: Europe
4
Country Data Year Trend Type Source
Canada 24% (211) 2000 16% 1993 UG, CS Inst, UBC [62]
12% (55), 1997 UG, CmpSci Inst, Simon Fraser U [61]
USA 26.7% (7,166) 1998 37.1% 1984 Deg, CS&IS National [35]
USA &
Canada
20.4% (2,372) 2000 Deg, CS Inst, PhD granting departments [8]
Mexico 39.2% (55,154) 1999 43.1% 1992 UG, Cmp
a
National [26]
Brazil
b
34.8% (5,641) 1993 UG, CS National [32]
20% 1993 UG, CS Insts, U of S ao Paulo, U of Camp-
inas [32]
Bolivia 34.1% (15) 1997 Inf Insts, private universities [45]
Guyana 54.5% (22) 2001 none 1998 Deg, CS Ind, University of Guyana
c
[14]
a
Students registered for a Licenciatura in Computaci on y Sistemas.
b
Other data records 5-10% female students at University of S ao Paolo, in the last ten years and not more than 20% at other institutions in
the same period [53].
c
The University of Guyana is the only university in Guyana.
Table 4: North and South America
Other is used when the data is not about major-level computing at university or a similar tertiary institution, and
a footnote is given describing the data.
Source: If the data is national, National is used. Inst is used for published or ofcial data from a specic univer-
sity or institution, and Insts for similar data from a group of universities or institutions. Ind is used for data given
by an individual. For both institutional and individual data, the names of the institutions are given where known.
3.2 Other data
The 1998 UNESCO Statistical Yearbook [66] gives gures on participation on women in the subject area Mathematics
and Computer Science, both in terms of enrollment and graduates. This data is of interest, but must be considered with
care as a high participation in mathematics may mask a low participation in computer science or vice versa. The 1999
editions of the Statistical Yearbook does not give this breakdown it only gives gures of female participation for the
broad elds Education, Humanities, Social Sciences, Natural Sciences, Medical Sciences and Others and no future
issues are planned of this yearbook [19]. A summary of the statistics from the 1998 UNESCO Statistical Yearbook
can be found in Appendix B.
3.3 Discussion
As can be seen from the tables, there is a wide range in participation in computing by women. As the information
covers different courses and different levels, it is difcult to do a direct comparison between countries. It can be
seen from the data, participation is between 10% and 40% in most countries and courses, with a wide spread in this
range. Hence, there is a strong indication that there is an underrepresentation of women in computing worldwide, at
least in terms of undergraduate participation. There are some countries and courses where womens participation is
below 10%, some with participation above 40%, and a few where women are in the majority. Some countries and
courses show an increasing trend and some decreasing, so it is not possible to predict future changes. To conclude,
women appear to be under represented in the discipline of computing, when we consider the gures for undergraduate
participation, and there is no clear indication that this will improve in the near future. Ongoing research is required to
determine causes and solutions.
Acknowledgements: The author thanks Sibongile Eland, Hlamalani Huhlwane, Yolanda Martins, Ana Rowe, Francesc
Lucio-Gonz alez and P etur P etursson for their assistance in nding and interpreting information.
Request for Information: The research that lead to this article is ongoing and the author appreciates any information
about countries not discussed in this article, or more recent information about countries mentioned here.
5
A Data collection
For the record, this appendix details the process that was used to collect data for this survey. In general, nding
information of this type is fairly difcult because:
it often doesnt exist because it hasnt been collected.
data exists but it is not identiable. For example, it often not clear whether computer science is categorised
under the headings Natural Science, Engineering or Technology.
data exists, but it is not disaggregated. So for example, even if it is possible to determine that Computer Science
is included under Natural Science, it is not possible to obtain the Computer Science statistics.
it is not clear from the data source to what the data refers. For example, it can be difcult to determine whether
the gures cover all tertiary education or university-level education.
Because of these difculties, the approach taken in this report has been to give the reader as much information as
possible about the data and its source. This is done to show that the data is not necessarily comparable and to give
the important details of the data. As can be seen from the tables in the body of the report, these details include the
actual percentage, the actual count (if available), trend information (if available), the year, the level of study and eld
of study, the type of information and the source of the information.
The majority of this data was collected online in the following fashion:
searching with appropriate terms in a search engine such as Google (www.google.com).
working through lists of national statistical websites such as those at
United Nations (www.un.org/Depts/unsd/gs natstat.htm)
Statistics Netherlands (www.cbs.nl/en/services/links/default.asp)
Other data was collected through journal articles and books. There are a number of limitations to this general approach:
Lack of consistency in terms of what is measured because it comes from a variety of sources.
Lack of detailed information it may be the case with a national statistical site that there is a non-electronic
document with the information being sought for. Clearly a web search will not nd this document.
As, can be seen from the data presented in the rst part of the report, there is clearly substantial variation in what is
being measured. However, within the limitation of the approach taken in data collection, it is a comprehensive survey,
and the author is not aware of any similar work.
B Computer Science and Mathematics
This appendix summarises data from the 1998 UNESCO Statistical Yearbook [66] and from an European Union
Report [44]. These appear to be one of the few sources of global statistics that specically mention computer science
as a eld of study. Unfortunately, it is aggregated with mathematics, and hence high participation in the mathematics
may hide low participation in computer science, and vice versa. However, because of its breadth, this data is included
in this report.
B.1 UNESCO data
The UNESCO data covers general programmes in mathematics, statistics, actuarial science, computer science, and
is presented for the following academic levels:
ISCED Level 5: Programmes leading to an award not equivalent to a rst university degree.
ISCED Level 6: Programmes leading to a rst university degree or equivalent qualication
ISCED Level 7: Programmes leading to a post-graduate university degree or equivalent qualication
6
Country % Country % Country % Country %
Albania
a
52 Estonia 25 Latvia 38 Romania 51
Austria 20 Finland 30 Lithuania 36 Slovakia 18
Belgium
a
18 Germany 22 Macedonia 40 Slovenia 14
Bulgaria 53 Hungary 29 Netherlands 14 Spain 29
Czech Republic 15 Ireland 33 Poland 47 Sweden 35
Denmark 22 Italy 45 Portugal 38 United Kingdom
a
25
a
1997
Table 5: Participation in tertiary level mathematics and computer science 1998 [44]
Data is given for number of students as well as number of graduates. Not all countries have data for the number of
women, and those that do not are excluded.
The approach taken in summarising the data is to present data for all tertiary education, namely ISCED Level 5,
6, and 7 (Tables 6 and 7), and for rst university degrees, namely ISCED Level 6 (Tables 8 and 9). For both of these
types of data, the most common data available is number of students. In a few cases, this data was not available, and
the number of graduates is used instead The tables have the following format:
Country: This is the country name. The same list is used for both sets of tables, even though for ISCED Level 6
some countries have no data.
Percentage: This is the number of women expressed as a percentage of the total number of students.
Year: This is the year that the data comes from. In the case of academic years split over two calendar years, the more
recent year is used.
Count: This is the actual number of female students or graduates.
Type: This is the type of data, either students or graduates. Note that when considering actual numbers, care should
be taken in comparing student numbers with graduate numbers.
Change: For some countries, two years of student numbers are given. In this case, if there has been a change of ve
percent in the percentage gure, this is indicated as either an increase or decrease. If the change is smaller than
this, it is indicated by the word Static. given.
The tables show a wide range of participation rates from 1.3% at ISCED Level 6 to 81.0% at ISCED Level 6. No clear
trends can be determined from the change indicators some countries have seen an increase in participation, some a
decrease, but most show little change. Considering the countries with a majority of women participation, a number
come from the Far East, Middle East and Eastern Europe. A sociological analysis of cultural factors may be able to
explain these patterns, but this is beyond the scope of this report.
B.2 European Union data
The European Union (EU) information covers computer science and mathematics and computer science data for a
number of European countries EU member states as well as countries from Central and Eastern Europe (the PHARE
countries). This data is provided here because it is more recent than the UNESCO data, but it is tabulated separately
because of a change in the ISCED classication system [65]. The data is for the new ISCED levels 5 and 6 which
covers all tertiary education. It is not clear whether this can be directly equated with the previous ISCED levels 5, 6
and 7 as described above. Table 5 presents the the percentage of women studying computer science and mathematics.
There is no clear pattern in this data. All countries with a majority of women are in Eastern Europe. Italy, Macedonia,
Latvia, Lithuania and Portugal have more than 35% participation. The lowest gures (less than 20%) are found in
a range of countries Belgium, Czech Republic, the Netherlands, Slovenia and Slovakia. Again, an analysis of the
factors causing these differences are beyond the scope of this report.
7
Country % Year Count Type Change
Albania 56.2% 1997 224 Students Increase (1991)
Algeria 36.0% 1996 7,006 Students
Armenia 50.9% 1996 665 Students
Australia 25.0% 1996 7,001 Students
Austria 20.5% 1996 3,651 Students Static (1991)
Bahrain 67.5% 1991 329 Students
Belgium 22.9% 1994 2,378 Students Static (1991)
Botswana 30.4% 1997 45 Students
Brazil 40.3% 1994 37,413 Students Static (1990)
Brunei Darussalam 48.8% 1996 22 Students
Bulgaria 53.6% 1997 1,974 Students Static (1991)
Burkina Faso 5.4% 1995 5 Students Static (1991)
Burundi 17.6% 1992 9 Students
Canada 27.6% 1996 16,235 Students Static (1991)
Congo 6.8% 1991 13 Students
C ote dIvoire 18.6% 1994 515 Students
Croatia 26.3% 1997 195 Students Decrease (1993)
Cuba 28.7% 1997 202 Students Decrease (1991)
Cyprus 39.4% 1997 290 Students Static (1991)
Czech Republic 15.2% 1996 464 Students Static (1993)
Denmark 27.2% 1996 1,538 Students Static (1991)
Eqypt 43.4% 1996 708 Students Increase (1991)
Eritrea 3.7% 1998 5 Students
Estonia 45.8% 1997 261 Students Decrease (1993)
Ethiopia 17.8% 1996 171 Students Static (1991)
Finland 16.9% 1996 2,425 Students Decrease (1991)
Georgia 52.1% 1996 795 Students
Germany 22.5% 1996 25,783 Students Static (1993)
Ghana 11.4% 1991 26 Students
Greece 32.5% 1993 4,377 Students Decrease (1990)
Guyana 32.6% 1995 49 Students
Honduras 6.6% 1994 177 Students Decrease (1990)
Hungary 40.3% 1995 1,609 Students Increase (1991)
Iceland 12.0% 1995 32 Students
Indonesia 34.2% 1996 43,861 Students Decrease (1993)
Iran 39.1% 1997 11,225 Students Increase (1991)
Ireland 27.3% 1996 816 Students Static (1991)
Israel 34.6% 1993 2,129 Students Static (1991)
Italy 42.2% 1996 21,016 Students Static (1991)
Japan 20.1% 1992 4,218 Students
Jordan 42.1% 1997 4,044 Students Increase (1991)
Kazakstan 38.6% 1995 2,502 Students
Korea 33.2% 1997 49,206 Students Static (1991)
Kuwait 80.9% 1996 507 Students
Kyrgyzstan 72.5% 1994 1,359 Students
Lao Peoples DR 28.5% 1993 62 Students
Latvia 31.5% 1997 1506 Students
Lebanon 37.6% 1996 840 Students
Lesotho 32.5% 1997 13 Students
Lithuania 43.2% 1997 831 Students
Table 6: ISCED Levels 5, 6 and 7
8
Country % Year Count Type Change
Macau 20.5% 1997 15 Graduates
Macedonia 64.2% 1997 560 Students Increase (1993)
Madagascar 26.2% 1997 152 Students Increase (1991)
Malaysia 50.9% 1991 2,322 Students
Mali 4.9% 1991 3 Students
Malta 13.3% 1991 4 Students
Mauritania 13.2% 1994 7 Students
Mauritius 29.9% 1991 38 Students
Mexico 46.6% 1995 29,937 Students Increase (1991)
Mongolia 48.7% 1997 626 Students
Mozambique 27.1% 1993 38 Students
Myanmar 62.2% 1995 9,702 Students
Netherlands 9.9% 1996 794 Students Decrease (1991)
New Zealand 32.0% 1996 438 Students Increase (1990)
Nicaragua 48.1% 1995 1,849 Students Decrease (1990)
Norway 21.4% 1996 337 Students Decrease (1991)
Oman 59.0% 1992 36 Students
Palestine 31.7% 1997 683 Students
Panama 53.1% 1994 830 Students
Papua New Guinea 32.5% 1995 27 Students
Paraguay 46.0% 1993 1,215 Students
Poland 56.2% 1994 7,214 Students Static (1991)
Portugal 46.4% 1995 5,961 Students Static (1991)
Qatar 66.0% 1992 113 Students
Romania 51.5% 1997 6,423 Students
Russian Federation 54.9% 1995 84,717 Students
Rwanda 13.6% 1990 6 Graduates
Saudi Arabia 76.7% 1996 6,197 Students Increase (1991)
Senegal 12.4% 1992 22 Students
Slovakia 21.8% 1997 224 Students Static (1993)
Slovenia 42.9% 1997 89 Students Increase (1992)
South Africa 34.9% 1994 10,653 Students
Spain 30.7% 1996 27,006 Students Static (1991)
Sri Lanka 33.4% 1995 106 Students
St Kitts and Nevis 46.2% 1993 31 Students
Sudan 29.9% 1991 260 Students
Sweden 28.9% 1996 4,651 Students Increase (1991)
Switzerland 14.3% 1996 413 Students Static (1991)
Syrian Arab Republic 39.0% 1995 345 Students Decrease (1991)
Tanzania 2.7% 1996 2 Students
Togo 2.9% 1990 6 Students
Tunisia 27.5% 1997 708 Students Increase (1991)
Turkey 34.8% 1995 10,010 Students Static (1991)
UAE 64.7% 1991 167 Students
Uganda 17.9% 1996 48 Students Increase (1991)
United Kingdom 23.9% 1995 26,423 Students Static (1991)
USA 37.1% 1995 26,749 Graduates
US Virgin Islands 67.4% 1993 60 Students
Yemen 26.3% 1992 19 Students
Zimbabawe 36.6% 1996 343 Students Static (1990)
Table 7: ISCED Levels 5, 6 and 7 (Cont.)
9
Country % Year Count Type Change
Albania 46.4% 1991 58 Students
Algeria 32.9% 1996 4,368 Students
Armenia
Australia 24.7% 1996 5,343 Students
Austria 20.9% 1996 3,192 Students Static (1991)
Bahrain
Belgium 28.4% 1994 664 Students Static (1991)
Botswana 25.4% 1997 16 Students
Brazil 40.4% 1994 37,413 Students Static (1990)
Brunei Darussalam
Bulgaria 53.9% 1997 1,935 Students Static (1991)
Burkina Faso 5.4% 1995 5 Students Static (1991)
Burundi 17.6% 1992 9 Students
Canada 27.8% 1996 7,578 Students Static (1991)
Congo 7.1% 1991 13 Students
C ote dIvoire
Croatia 26.4% 1997 195 Students Decrease (1993)
Cuba 28.8% 1997 202 Students Decrease (1991)
Cyprus 57.0% 1997 142 Students
Czech Republic 15.0% 1996 399 Students
Denmark 27.2% 1996 1,538 Students Static (1991)
Eqypt 62.0% 1996 529 Students Increase (1991)
Eritrea 3.8% 1998 5 Students
Estonia 39.9% 1997 144 Students Decrease (1993)
Ethiopia 8.7% 1996 49 Students Static (1991)
Finland 16.8% 1996 1,890 Students Static (1991)
Georgia 52.0% 1996 714 Students
Germany 22.5% 1996 25,358 Students Static (1993)
Ghana 10.7% 1991 22 Students
Greece 34.1% 1993 4,115 Students Decrease (1990)
Guyana 29.8% 1995 25 Students
Honduras 6.4% 1994 170 Students Decrease (1990)
Hungary 21.5% 1991 94 Students
Iceland 10.3% 1995 14 Students
Indonesia 27.4% 1996 14,272 Students Decrease (1993)
Iran 41.2% 1997 10,025 Students Increase (1991)
Ireland 40.7% 1996 399 Graduates
Israel 35.8% 1993 1,771 Students Static (1991)
Italy 42.0% 1996 20,745 Students Static (1991)
Japan 20.0% 1992 3,852 Students
Jordan 32.4% 1997 2,141 Students Static (1991)
Kazakstan 38.6% 1995 2,502 Students
Korea 35.2% 1997 35,391 Students Increase (1991)
Kuwait 81.0% 1996 506 Students
Kyrgyzstan
Lao Peoples DR 28.6% 1993 62 Students
Latvia 32.6% 1997 1,302 Students
Lebanon 37.6% 1996 840 Students
Lesotho
Lithuania 39.8% 1997 517 Students
Table 8: ISCED Level 6
10
Country % Year Count Type Change
Macau 21.4% 1997 15 Graduates
Macedonia 64.3% 1997 560 Students Increase (1993)
Madagascar 16.1% 1991 71 Students
Malaysia 49.4% 1991 678 Students
Mali 4.9% 1991 3 Students
Malta 13.3% 1991 4 Students
Mauritania 13.2% 1994 7 Students
Mauritius 17.5% 1991 11 Students
Mexico 47.3% 1995 29,311 Students Increase (1991)
Mongolia 47.5% 1997 549 Students
Mozambique 27.1% 1993 38 Students
Myanmar 62.1% 1995 9,423 Students
Netherlands 16.0% 1991 450 Students
New Zealand 24.6% 1996 172 Students Increase (1990)
Nicaragua 48.3% 1995 1,843 Students Decrease (1990)
Norway 14.6% 1996 86 Students Decrease (1991)
Oman 59.0% 1992 36 Students
Palestine 32.3% 1997 676 Students
Panama 53.7% 1994 746 Students
Papua New Guinea 32.5% 1995 27 Students
Paraguay 46.0% 1993 1,215 Students
Poland 57.8% 1994 6,679 Students Static (1991)
Portugal 49.4% 1995 5,102 Students Static (1991)
Qatar 73.9% 1992 68 Students
Romania 52.7% 1997 5,786 Students
Russian Federation 56.4% 1995 77,510 Students
Rwanda
Saudi Arabia 74.0% 1996 5,166 Students Increase (1991)
Senegal 12.4% 1992 22 Students
Slovakia 22.5% 1997 194 Students Static (1993)
Slovenia 43.0% 1997 86 Students Increase (1992)
South Africa 31.3% 1994 4,542 Students
Spain 30.6% 1996 25,618 Students Static (1991)
Sri Lanka 18.8% 1994 9 Graduates
St Kitts and Nevis
Sudan
Sweden 29.8% 1996 4,476 Students Increase (1991)
Switzerland 14.0% 1996 331 Students Static (1991)
Syrian Arab Republic
Tanzania 2.7% 1996 2 Students
Togo 1.3% 1990 1 Students
Tunisia 27.3% 1997 618 Students Increase (1991)
Turkey 37.0% 1995 7,209 Students Static (1991)
UAE 64.7% 1991 167 Students
Uganda 16.7% 1996 36 Students Increase (1991)
United Kingdom 24.9% 1995 16,576 Students Static (1991)
USA 35.1% 1995 13,369 Graduates
US Virgin Islands 67.4% 1993 60 Students
Yemen 35.7% 1992 10 Graduates
Zimbabawe 2.0% 1996 2 Students
Table 9: ISCED Level 6 (Cont.)
11
References
[1] G. Aikhorin. Personal communication, 2001.
[2] M. Arteimi. Personal communication, 2002.
[3] R. Ayfer. Personal communication, 2002.
[4] M. Barinaga. Surprises across the cultural divide. Science, 263:14681472, 11 March 1994.
[5] N.A. Bhatti. Personal communication, 2001.
[6] C. Bjorkman, I. Christoff, F. Palm, and A. Vallin. Exploring the pipeline: towards an understanding of male dominated
computing culture and its inuence on women. ACM SIGCSE Bulletin, 30(2):6469, June 1998.
[7] J. Brown, P. Andreae, R. Biddle, and E. Tempero. Women in introductory Computer Science: experience at Victoria University
of Wellington. ACMSIGCSE Bulletin, 29(1):111115, March 1997. (Proceedings of the 28th SIGCSE Technical Symposium).
[8] R.E. Bryant and M.J. Irwin. 1999-2000 CRA Taulbee Survey. Computing Research News, pages 511, March 2001.
http://www.cra.org/CRN/issues/0102.pdf.
[9] I.

Cern a. Personal communication, 2001.
[10] K. Chanana. Treading the hallowed halls: Women in higher education in India. Economic and Political Weekly, pages
10121022, 18 March 2000.
[11] M.C. Clarke. Technological experience and technophobia in South African university students. South African Computer
Journal, 25:1217, August 2000.
[12] B.A. Collis and R.L. Williams. Cross-cultural comparison of gender differences in adolescents attitudes toward computers
and selected school subjects. Journal of Educational Research, 18(1):183191, 1987.
[13] A. Craig. Women and Business Computing: where to from here? gates, 4(1):1319, 1997.
[14] P. Crawford. Personal communication, 2002.
[15] C.K.M Crutzen. The inuence of Feminist Theory on Informatics course design. In A. Adam, J. Emms, E. Green, and
J. Owens, editors, Women, Work, and Computerization, Breaking Old BoundariesBuilding New Forms, pages 5975. Else-
vier Science, 1994.
[16] P.J. Denning, D.G. Comer, D. Gries, M.C. Mulder, A. Tucker, A.J. Turner, and P.R. Young. Computing as a discipline.
Communications of the ACM, 32(1):923, 1989.
[17] A. Durndell, D. Haag, D. Asenova, and H. Laithwaite. Computer self efcacy and gender: East and West Europe. In E. Balka
and R.K. Smith, editors, Women, Work, and Computerization: Charting a Course to the Future, pages 7885. Elsevier Science,
2000.
[18] A. Durndell and K. Thomson. Gender and computing: a decade of change. Computers and Education, 28(1):19, 1997.
[19] UNESCO Institute for Statistics. Publications, 2002. http://www.uis.unesco.org/en/pub/pub0.htm.
[20] Frauenf orderung am Departement Informatik, ETH Z urich. Statistik der frauenanteile, 2001.
http://www.inf.ethz.ch/frauen/einfuehrung/statistik.html.
[21] P. Gasaatura. Personal communication, 2001.
[22] Padmaja Gogula and Shashi M. Training of women in IT skills: a decade of progress at Andhra University, Visakhapatnam.
In A. Adam and J. Owens, editors, Proceedings of the 5th IFIP Conference on Women, Work, and Computerization, Breaking
Old BoundariesBuilding New Forms, pages 488500, UMIST, Manchester, July 1994. IFIP.
[23] A.M. Hoffman-Barthes, S. Nair, and D. Malpede. Scientic, technical and vocational education of girls in Africa: summary
of 21 national reports, 1999. UNESCO working document,
http://unesdoc.unesco.org/images/0011/001180/118078eo.pdf.
[24] IEEE-WIE Switzerland. Statistical data of the students who graduate at EPFL in 1998, 2000.
http://www.ewh.ieee.org/r8/switzerland/WomenInEngineering/stat.htm.
[25] Instituto Nacional de Estadstica, Espa na. INEbase: Estadstica de ense nanza universitaria, 1999.
http://www.ine.es/inebase/cgi/um.
[26] Instituto Nacional de Estadstica Geografa e Inform atica, Mexico. Sistema de Indicadores para el Seguimiento de la Situaci on
de la Mujer en M exico: Educaci on, 1999. http://www.inegi.gob.mx.
[27] I. Janssen Reinen and T. Plomp. Information technology and gender equality: a contradiction in terminis. Computers and
Education, 28(2):6578, 1997.
[28] H. Kruppa and Zs. Lipt ak. Frauenf orderung am Departement Informatik, ETH Z urich. Personal communication (Internet
Research conducted in October 2000), 2002.
[29] Lim Gek Kheng. WIT: the Singaporean way. The Computer Bulletin, 2(1):1820, 1990.
12
[30] Zs. Lipt ak. Personal communication, 2002.
[31] V. Makrakis. Cross-cultural gender differences in student self-efcacy: the impact of culture. In R. Aiken, editor, Information
Processing 92, volume II, pages 233238. Elsevier Science, 1992.
[32] M.C. Monard and M.C. Nicoletti. A perspective of Brazilian women in Computer Science academia. In A. Adam
and J. Owens, editors, Proceedings of the 5th IFIP Conference on Women, Work, and Computerization, Breaking Old
BoundariesBuilding New Forms, pages 91100, UMIST, Manchester, July 1994. IFIP.
[33] C. Mueller. Personal communication, 2001.
[34] C.C. Mukhopadhyay. The cultural context of gendered science: a theoretical model derived from studies in India. gates,
3(1):1118, 1996.
[35] National Center for Education Statistics. Digest of Education Statistics 2000, 2001. US Department of Education, Washington,
D.C. http://nces.ed.gov/pubs2001/digest.
[36] National Statistical Ofce Thailand. Gender Statistics, 2001. http://www.nso.go.th/gender/eedu.htm.
[37] National Student Selection and Placement Center. Higher education statistics 1996, 1997, 1998, 1999, 2000, 2001. Turkey.
[38] E. Newmarch, S. Taylor-Steele, and A. Cumpston. Women in IT what are the barriers. In Network of Women in Further
Education Conference, March 2000.
http://www.deet.gov.au/iae/analysis/womeninit.htm.
[39] C. Ng Choon Sim and C. Yong. Information technology, gender and employment: a case study
of the telecommunications industry of Malaysia. In S. Mitter and S. Rowbotham, editors, Women
encounter technology: changing patterns of employment in the Third World. UNU/INTECH, 1995.
http://www.unu.edu/unupress/unupbooks/uu37we/uu37we00.htm.
[40] NTNU. Jenter & data, 2002. http://datajenter.ntnu.no/2002/.
[41] V. Oechtering and R. Behnke. Situations and advancement measures in Germany. Communications of the ACM, 38(1):7582,
January 1995.
[42] Ofce of Science and Technology. SET statistics 2001, 2001.
http://www.set4women.gov.uk/ost/setstats/index.htm.
[43] Peer Review Programme. Executive Summary: Reducing the Gender Digital Divide in Skills and Employment, Germany,
2000. http://peerreview.almp.org/en/GERMdec00.htm.
[44] S. Pilos. Educating young Europeans: similarities and differences between the EU member states and the PHARE countries,
2000. Eurostat. http://europa.eu.int./comm/eurostat.
[45] Republica de Bolivia: Instituto Nacional de Estadstica. Estadisticas de educaci on anuario estadstico 2000, 2000.
http://www.ine.gov.bo.
[46] C. Ronge. Personal communication, 2001.
[47] J. Sanders. Closing the computer gap in schools. In J.Z. Daniels and J.B. Kuhle, editors, Proceedings of the 4th GASAT
Conference, Lansing, MI, 1987. National Sciences Foundation.
[48] L. Selby, K. Ryba, and A. Young. Women in computing: what does the data show? ACM SIGCSE Bulletin, 30(4):62a67a,
December 1998.
[49] Y. Shapiro and J. Jacobs. South African Graduate Statistics 1999: Proles and Recent Trends. HSRC, Pretoria, 1999.
[50] L. Shashaani and A. Khalili. Gender and computers: similarities and differences in Iranian college students attitudes towards
computers. Computers and Education, 37:363375, 2001.
[51] A. Shode. Personal communication, 2001.
[52] G. Sigurdard ottir. Women in the Icelandic information society. In Conference about Women and the Information Society,
Reykjavk, Iceland, April 2000.
http://www.simnet.is/konur/erindi/engish gudbjorg erindi.htm.
[53] D.M. Silva. Information technology & women in developing countries:a Brazilian perspective, 2000. Unpublished document.
[54] R. Sinnappan. Personal communication, 2001.
[55] Statistical Ofce of the Republic of Slovenia. Statistical Yearbook 2001, 2001.
http://www.gov.si/zrs/leto01/kazalo01.htm.
[56] Statistics Norway. Current Education Statistics: Universities and Colleges, Key Figures 2000, 2001.
http://www.ssb.no/english/subjects/04/utdanning as en/200103/au200103.pdf.
[57] Statistisches Bundesamt Deutschland. Studienanfaengerinnen und absolventinnen im fach informatik, 2001. Wiesbaden,
Germany.
[58] J. Strbrn a. Personal communication, 2001.
13
[59] M. Suriya. Personal communication, 2002.
[60] Swedish Institute. Fact sheets on Sweden: Higher education in Sweden, 2001.
http://www.si.se/docs/infosweden/engelska/fs83m.pdf.
[61] SWIFT. Simon Fraser University Fall 1997 headcount in engineering and science programs, 1998.
http://taz.cs.ubc.ca/swift/survey stats/scigende.html.
[62] SWIFT. Percentage of male to female undergraduates in Science at UBC, 2000.
http://taz.cs.ubc.ca/swift/survey stats/ubcdepSt.html.
[63] G.J. Teague and V.A. Clarke. Attracting women to tertiary computing courses: two programs directed at secondary level.
ACM SIGCSE Bulletin, 25(1):208212, March 1993. (Proceedings of the 24th SIGCSE Technical Symposium).
[64] L. Uden. Gender bias: the east-west paradox. In G. Lovegrove and B. Segal, editors, Women into Computing: selected papers,
1988-1990, pages 386391. Springer-Verlag, 1991.
[65] UNESCO. International standard classication of education: ISCED 97, 1997.
http://unesdoc.unesco.org/images/0011/001113/111387eo.pdf.
[66] UNESCO. Statistical Yearbook 1998. UNESCO Publishing and Bernan Press, 1998.
[67] UNESCO. Access of girls and women to scientic, technical and vocational education in Africa, 1999.
http://www.chinaacc.edu.cn/wenyuan/unesco/05/.
[68] L. von Hellens and S. Nielsen. Australian women in IT. Communications of the ACM, 44(7):4652, July 2001.
[69] M. Weil and L.D. Rosen. The psychological impact of technology from a global perspective: a study of technological
sophistication and technophobia in university students from twenty-three countries. Computers in Human Behavior, 11(3):95
133, 1995.
14