‘Because it’s boring, irrelevant and I don’t like computers’: Why high school girls avoid professionally-oriented ICT subjects

Available online at www.sciencedirect.com

Computers & Education 50 (2008) 1304–1318

www.elsevier.com/locate/compedu

‘Because it’s boring, irrelevant and I don’t like computers’:Why high school girls avoid professionally-oriented ICT subjects

Neil Anderson *, Colin Lankshear, Carolyn Timms, Lyn Courtney

James Cook University, School of Education, P.O. Box 6811, Cairns, QLD 4870, Australia

Received 13 August 2006; accepted 7 December 2006

Abstract

The current paper details results from the Girls and ICT survey phase of a three year study investigating factors asso-ciated with low participation rates by females in education pathways leading to professional level information and com-munications technology (ICT) professions. The study is funded through the Australian Research Council’s (ARC) LinkageGrants Scheme. It involves a research partnership between Education Queensland (EQ), industry partner Technology Oneand academic researchers at (affiliation removed for review purposes). Respondents to the survey were 1453 senior highschool girls. Comparisons were drawn between Takers (n = 131) and Non Takers (n = 1322) of advanced level computingsubjects. Significant differences between the groups were found on four questions: ‘‘The subjects are interesting’’; ‘‘I amvery interested in computers’’; ‘‘The subject will be helpful to me in my chosen career path after school’’; and ‘‘It suitedmy timetable’’. The research has demonstrated that senior high school girls tend to perceive advanced computing subjectsas boring and they express a strong aversion to computers.� 2007 Elsevier Ltd. All rights reserved.

Keywords: Gender studies; Secondary education; Human–computer interface; Pedagogical issues; Teaching/learning strategies

1. Introduction

This paper begins with an overview of the trend, which is international, of low and falling female partici-pation rates in ICT professional careers and pathways. By way of contextual background, we offer a shortreview of factors associated with this trend based on research conducted in numerous countries around theworld since the 1980s. Particular attention was paid to factors associated with subject choices within formaleducation, since our specific focus is on the situation with respect to girls’ choice of subjects in their final twoyears of secondary schooling in Queensland, Australia as advanced computing subjects are recognized as lead-ing into ICT career pathways (Anderson, Klein, & Lankshear, 2005; Anderson, Lankshear, Courtney, &Timms, 2006). Forthcoming is a brief account of the scope and design of the current project, and the roleof the survey component within the larger study, before describing the survey instrument and reporting thedata collection and results. The paper concludes with an account of the findings from the survey with respect

0360-1315/$ - see front matter � 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.compedu.2006.12.003

* Corresponding author. Tel.: +61 74042 1722; fax: +61 740421312.E-mail address: [email protected] (N. Anderson).

N. Anderson et al. / Computers & Education 50 (2008) 1304–1318 1305

to female students who opted not to take advanced level ICT subjects in their final two years and their impli-cations for the second data collection phase involving group interviews.

1.1. A downward spiral: Girls and professional ICT careers and pathways

When we speak of professional ICT careers, we refer to roles like ‘designing and developing software andhardware systems; providing technical support for computer and peripheral systems; and creating and man-aging network systems and databases’ (Sandy & Burger, 1999, p. 5). These roles subsume the work of com-puter analysts, programmers, software engineers, computer managers, internet architects, webmasters,learning resources managers, and the like (Millar & Jagger, 2001). Accordingly, when we speak of pathwayswithin formal education to ICT careers we are referring to subjects from the Queensland senior school curric-ulum such as Information Processing Technology (IPT), Information Technology Systems (ITS) and BusinessCommunication Technology (BCT). Of these, IPT and ITS are widely seen as more ‘prestigious’ subjectsbecause of the ways they are weighted, vis-a-vis subjects like BCT in calculating students’ Overall Position(OP) score (Queensland’s tertiary entrance score system) at the completion of Year 12. Consequently, theyare more closely associated contingently with transition to University computing degree programs thanBCT, which tends to be taken by larger numbers of students.

At a moment, when female participation in other science and engineering areas has been growing consis-tently across industrialized countries, the trend has been the opposite in ICT subjects and careers, and thetrend is widespread. Reporting data from a comparative study of the United Kingdom (UK), United Statesof America (USA), Canada, Taiwan, Spain and Ireland, Millar and Jagger (2001) noted that female comput-ing graduates are in decline in all of these countries, with figures in the UK lower than in Ireland and the USA.According to Cohoon (2003, p. 669)

In most industrialized countries, women appear to be a minority in computer science. Data comparingmen’s and women’s CS [Computer Science] education internationally showed that in 1995, men earned twiceas many math and CS degrees as women earned in Canada, Germany, Ireland, Japan, Spain and the US . . .Three times as many math and CS degrees went to men in Austria, New Zealand, Norway and the UK.Only in Italy did men and women earn close to the same number of math and CS degrees. [NB. Interna-tional data were only available with math and CS numbers combined]

In Australia, national figures indicated the same kind of decline apparent in other countries. A study con-ducted across tertiary institutions in 2000 found that female participation in tertiary computer science courseshad fallen from 26.2% in 1989 to 19% in 2000 (Newmarch, Taylor-Steele, & Cumpston, 2000). Recent datafrom the Department of Education, Science and Technology (DEST) indicated that the proportion of womenrelative to men in ICT has fallen steadily from 26.65% in 2001 to 20% in 2005. In March 2005, Australia’sMinister for Communications, Information Technology and the Arts announced members of an AdvisoryGroup to help plan a forthcoming Women in ICT Summit, and stated that women currently comprise ‘onlyabout one fifth of the ICT workforce’ (Coonan, 2005, March 8).

A Queensland government report, Smart Classrooms: Report on the Girls and ICT Framework for

Action(Queensland Government, 2003) claimed that in 2001 only 5% of girls in their final year of secondaryschooling chose IPT, the ICT subject that carries the most weight in calculating student overall performance.The proportion of girls relative to boys studying IPT fell from 26% to 21% between 1998 and 2001. In 19925.9% of boys and 3.4% of girls took IPT. By 2001 the figure was 21% for boys. For girls it was 5%. In the 5year period to 2001 the percentage of women in university ICT courses in Queensland increased from 15% to24%. In 2002, however, the first year intake dropped to 18.2%. Meanwhile, the proportion of womenemployed in computing professions in Queensland fell from 26% to 22% between 1996 and 2001.

1.2. The usual suspects: Research-based factors associated with low female student participation rates in

professionally-oriented ICT courses and subjects

During the past 10–15 years the international trend toward low and declining rates of participation byfemales in formal educational pathways to professional ICT careers has been widely researched (Anderson

1306 N. Anderson et al. / Computers & Education 50 (2008) 1304–1318

et al., 2005; Anderson et al., 2006; Camp, 1997; Frieze, 2005; Margolis & Fisher, 2003; Millar & Jagger, 2001;Zweben & Aspray, 2004). This research has generated a range of factors widely recognised as being associatedwith aspects of this trend. Since our concern in this paper is with factors associated with subject choice in seniorsecondary school, we will only review those factors generally believed to dissuade female students from takingICT subjects and programs. It is important to note that a large proportion of studies that have addressed sub-ject choice have done so with respect to university programs and not with school subject choices. The presentstudy stands out as one of the most substantial to date to address subject choice at the school level.

Five factors are particularly widely cited with the decisions of female students not to take professionalICT career-oriented subjects at secondary and tertiary levels. First, research indicates that many femalestudents associate ICT subjects with a bad image, as being either a ‘nerd’ domain or as a focal point forundesirable elements of ‘male culture’. Other factors are also linked to perceptions of the area being‘male-gendered’ (Margolis & Fisher, 2003). Hence, a second factor identified by numerous researchershas to do with perception of a male bias in software, which is seen as turning female learners away fromICT subjects (see also Gurer & Camp, 2002; Millar & Jagger, 2001). This second factor has an interestingpotential link to a third, which is that female students often feel inferior to and/or dominated or intimidatedby male students within learning settings (Cisco Systems, 2002; Gurer & Camp, 2002). Some research sug-gests that this can be a function of girls viewing themselves as being low in confidence and awareness withrespect to ICT. The potential link to male-biased software comes from the fact that confidence and feelingsof being ‘technologically savvy’ (Association of American University Women (AAUW), 2000) often comethrough gaming, and it is arguable that until recently commercial games were mostly geared to a malemarket (Beckwith et al., 2005).

The two remaining factors have to do with support for learning and awareness of the subjects per se.Research suggests that a relative lack of role models and mentors for females among teachers and peers is feltkeenly in the later years of school and in higher education. With respect to undergraduate courses, Cohoon(2003) argued that students regarded peer support as vital for meeting the demands of being a Computer Sci-ence major. While peer support is important to male and female students alike,

Women don’t have the level of access to same-sex classmates that men have . . . In CS, not all women werecomfortable with relying on male classmates for support . . . [Some] felt that when they didn’t know anyoneit was easier to approach another woman (Cohoon, 2003, p. 671).

The fifth commonly recognized factor is poor knowledge about ICT as a subject. According to Kahle andSchmidt (2004), it seems as though ‘not being informed is the most important reason why women are notenrolling in computer science’ (p. 82).

2. The current project: ‘Developing informed and integrated strategies to address low female participation rates in

professional ICT careers and pathways’

2.1. Background to the research

The current project emerged as a response to observations relating to initiatives being undertaken with aview to encouraging greater participation by girls in ICT at the senior high school level. One of these concernedthe fact that many female participants in the ‘I-Star Program’ (Anderson, 2003), which ran between 2001 and2004 as a Queensland Government funded initiative designed to provide young people with rich and intensiveopportunities for access to ICT, found school computing subjects aversive and described them as ‘boring’. Therural and regional ‘ICT Stars’ component of I-Star Program operated as a series of three-day weekend campsfor 50 female secondary school students each year. Students from urban, rural and remote areas of the stateattended the camps held in schools in far north Queensland (different models emerged in other parts of thestate). Workshops on aspects like multimedia animation, web design and video editing were presented by teach-ers and people from the industry selected on the basis of being ‘switched on’. Students from out of town wereaccompanied by teachers and members of their communities, who also participated in the workshops.

Interestingly, experience of school ICT programs as boring and unattractive had not been identified in themainstream research literature as among the major factors associated with low female participation rates in

N. Anderson et al. / Computers & Education 50 (2008) 1304–1318 1307

ICT subjects. Moreover, initiatives described in the research literature as having been designed to enhancefemale participation rates had not targeted the possibility that girls found ICT subjects tedious – with a viewto trying to make them more interesting and attractive. Formal evaluation of the regional and remote I-StarProgram at the end of each camp, as well as verbal comments volunteered during the workshops, were verypositive. This was consistent with the findings of research literature at the time (Gee, 2004) that reported ado-lescent girls engaging enthusiastically in a range of ICT pursuits in their out of school lives, but that thesetended to be very different kinds of pursuits (e.g., contributing to electronic ‘zines’, maintaining a web page,participating in online role playing games, ‘doing chat’, etc.) (Lankshear & Knobel, 2003) from what wasincluded in school programs. At that time, school ICT subjects in Queensland included little in the way ofmultimedia components. While some subject changes had occurred in the interim, there were no grounds atthe time the present study was proposed for thinking that the perceived boringness of ICT subjects wouldnot still be a potentially significant factor in subject choice decisions.

A second observation was that initiatives being undertaken with a view to addressing the issue of lowfemale participation had tended to be piecemeal and random rather than seeking to address multiple factorssimultaneously from a more holistic and comprehensive perspective. For example, there were self-containedmentoring projects, where successful women from the ICT sector visited schools and raised student awarenessabout opportunities in ICT careers, day trips to ICT establishments and the I-Star Program mentioned above.These initiatives were self-standing, piecemeal, small scale and localized, although another I-Star initiative,‘Gidgits’ (Girls into doing great IT), later developed a more widespread presence. These sorts of programs,along with anecdotal evidence of outcomes, have been reported at an international level by Denner, Werner,Bean, and Campe (2005); Frieze (2005) and Graham and Latulipe (2003).

Third, there had been considerable support for local initiatives from sectors of the ICT industry and, inparticular, from Women in Technology (WIT). The Queensland WIT offered annual scholarships for girlsto pursue ICT university careers and gave an annual award for excellence in ICT study by girls. Individualmembers also participated in mentoring initiatives, and WIT operated a website providing information rele-vant to encouraging female students into ICT career tracks (http://www.wit.org.au/).

2.2. Project conception

Against this background the present project was conceived with a view to build on existing research-basedknowledge about the problem area by means of a collaborative and strategic approach that simultaneouslybuilt on the impetus to address the issue that was already apparent in initiatives on the part of EQ and indi-viduals, groups and companies associated with the ICT industry sector. The ARC ‘Linkage Projects’ schemeoffered an ideal option for a funded research project. The program aims to promote collaborative researchprojects between researchers in higher education and industry. Projects must involve interaction betweenthe researchers and actual and potential users of the research outcomes, therefore, industry partners mustmake a contribution towards the cost and conduct of the research. The scheme seeks to encourage long-termstrategic alliances between industry and higher education institutions in order to apply advanced knowledge toproblems or to provide opportunities to obtain national economic or social advantage.

In accordance with this intent, the authors entered a collaborative venture with EQ and a Queensland-basedcompany, Technology One, as partners and as active research collaborators. Integral to the research design isthe establishment of reference groups representing EQ and the ICT industries sector respectively. These willfunction as working groups during the data analyses and applied interpretation phases, and as reference groupsproviding consultative advice and feedback throughout the project. Initially the reference groups will facilitatedata collection. In their capacity as working groups, they will receive pre-circulated material from the chiefinvestigators at key points during the analyses and interpretation phases and meet for half and/or full dayworking sessions on the data with researchers. The non-formal training component will involve the chief inves-tigators actively modelling research procedures to the working groups as appropriate for the work at hand andthe members of the respective industry groups modelling procedures, in which they are experts, to each otherand the chief investigators. These will include aspects such as decision-making, policy development, prioritisingthat are part of their respective cultures, relevant to achieving the project outcomes, yet outside the experiencesand perspectives of the chief investigators and, in some cases, of the other industry partners.

1308 N. Anderson et al. / Computers & Education 50 (2008) 1304–1318

2.3. The research aim and objectives

The current project aims to identify and interpret factors associated with low female participation rates inICT professional level occupations and education pathways and, in light of these findings, to work with theresearch partners to identify priority strategic response options for EQ and ICT industries. It is anticipatedthat subsequent research and development activities will be designed to enhance female participation rates.

This aim entails five research objectives:

� To obtain a rich and extensive Queensland data set for female students at years 11 and 12, and at the pointof entry to university study pertaining to their views and decisions about ICT programs and careers.� To develop a comprehensive Queensland data set for female employees within professional level career in

the ICT industry pertaining to their retrospective and current views and decisions about ICT programs andcareers.� To analyze this data in ways that incorporate participant perspectives (e.g., by means of member checks

and focused group discussions of preliminary analyses).� To undertake ‘applied interpretation’ of results by means that involve and reflect perspectives of EQ and

ICT industry personnel (e.g., by asking what the results of data analysis might imply at the level of practicalresponse within their respective domains).� To identify possibilities and priorities for co-operative, informed cross-sector strategic responses to the

issue of low rates of female participation in ICT.

This paper reports on the first of these objectives.

3. Methodology

The overall project contains multiple data collection phases involving survey and interview methods withdifferent populations. Data was collected from female high school students in Years 11 and 12 and survey datais currently being collected from women in professional career positions within the ICT industry. In this paperwe are concerned only with the student survey data.

Two populations of female students, approximating to ‘typicality’ by socio-economic status, location, andschool type (single/mixed sex; government schools (GS)/non-government schools (NGS) were identified forsurvey purposes. The first group, ‘Takers’ was comprised of girls taking Board Level ICT subjects (IPT orITS). The second group, ‘Non Takers’ was comprised of girls who did not take these subjects. A small Year12 ‘cohort sample’ was surveyed near at the end of the 2004 school year in order to trial the survey instrumentprior to preparing the final instrument for the survey during 2005.

The survey was designed to serve two research purposes. The first was to provide information about factorsassociated with student decisions about whether or not to include Board Level ICT subjects in their Year 11and 12 programs. The second was to provide an informed basis from which to select subjects to participate inindividual and group interviews during 2006, as well as to identify potentially useful themes to address in theinterviews.

3.1. Pilot survey phase – November 2004

Shortly before the end of the 2004 school year, a short survey was tested with Year 12 female students atfour schools (two urban and two rural) in far north Queensland. An excellent response rate (approximately90%) was obtained, yielding 171 completed surveys. The survey contained three sections. Most items wereforced choice, although short answers responses were sought to further clarify respondents’ reasons for takingor not taking ICT subjects. All participants completed Section A, which asked whether they were taking eitheror both of the Board ICT subjects (and, if so, which).

Respondents not taking Board subjects then responded to three brief items in Section B. One asked the girlsto identify those factors, from a set of 15, that influenced their decision not to take the subjects (e.g., ‘It didn’tsuit my timetable’, ‘The subjects are boring’). An ‘other’ option asked participants to list any other factors was

N. Anderson et al. / Computers & Education 50 (2008) 1304–1318 1309

included to pick up additional factors to the 15 forced choices. This option invited participants to list anyother factors. The remaining questions asked if the girls intended to go on to further study after leaving schooland, if so, did they intend to study any Information Technology courses.

Students who were studying one or both Board subjects completed five further questions. These included aquestion asking about factors influencing their decision to study the Board subjects (which used the same 15factors but with positive rather than negative expressions of the factors – e.g., ‘It suited my timetable’ ‘Thesubjects are interesting’). The other items asked if the girls intended to go on to further study and, if so,whether they would study IT subjects. There were also items asking those who intended to further studyIT subjects to identify reasons or factors influencing their decision, and to identify whether they intendedto study at university or a Technical and Further Education (TAFE) college.

Across the four schools, 87% of female students were not studying a Board ICT subject – the rates of NonTakers ranged from 91% to 79%. Of the 148 Non Takers, 52% (n = 77) said they intended full time study thefollowing year – the rates here ranging from 70% to 46%. Of the 77 students intending to go on to furtherstudy, 12 (16%) intended including ICT courses in their further studies. The three factors most frequently citedfor not taking Board ICT subjects were: ‘The subjects were boring’ (45%), ‘I’m not interested in computers’(50%) and ‘The subject would not be helpful for me in my chosen career path after school’ (49%). Of the 23students (13%) taking either or both Board subjects, 18 (78%) intended to go on to further study; 15 at uni-versity and 3 at TAFE, and only three among these 18 students planned to take ICT subjects in their tertiarystudies.

While nothing substantial can be concluded from such a small scale inquiry, a number of potentially inter-esting results emerged as possible foci for further investigation, particularly, when compared with those fac-tors most frequently identified in the research literature as being associated with low female participation inICT subjects, such as female student being intimidated by male student (Cisco Systems, 2002), being put off bymale based computer games (Millar & Jagger, 2001) and lack of computer confidence and competence (Adya& Kaiser, 2005). For example, respondents did not nominate gender-related issues, the question of the kind ofimages associated with ICT, or the matter of confidence around computers or ‘computer savvy’ as key factorsassociated with their not taking ICT subjects.

Two of the three factors most commonly associated with taking Board subjects in Year 12 overlapped(although positively rather than negatively) with the factors most commonly associated with not taking thesubjects. Eighteen Takers (78%) cited ‘The subjects are interesting’, and 14 (61%) cited ‘I am very interestedin computers’. The other most commonly associated factor was ‘I have a computer at home and often use it’.Eleven Takers (49%) negated the third factor most commonly cited by Non Takers ‘The subject would behelpful in my chosen career after school’.

The pilot study provided valuable information for helping shape the final instrument to be used in the pro-ject. Beyond this, however, it suggested that the reason most girls do not study ICT subjects concerns the samefactors that influence some girls to study them. Different populations of girls have opposite perceptions of thesame phenomena. Moreover, the results tenuously affirm findings of other studies that posit multiple factorrather than single factor explanations for low female participation in ICT careers and pathways (Adya & Kai-ser, 2005; Margolis & Fisher, 2003; Millar & Jagger, 2001). Ascertaining influences of these contradictory per-ceptions suggests a potentially fruitful role for the interview component of the research.

3.2. The survey phase – August–November 2005

3.2.1. Data collectionThe formal survey was conducted between August and November 2005 with Years 11 and 12 girls through-

out Queensland. Thirty-one GS and NGS were selected to participate in the survey, 26 of which subsequentlytook part. Selection was based on an attempt to obtain a study population as typical as possible of the state asa whole, but without any claims to producing any kind of representative sample. Emphasis was on pursuingscale of responses and ensuring that participants ranged over key variables such as socio-economic status, eth-nicity, language background, rural/urban location, GS or NGS systems.

The survey was refined in light of the pilot research. Some new questions were added to determine whethergirls had participated in any out-of-school initiatives intended to enhance interest in using new technologies

1310 N. Anderson et al. / Computers & Education 50 (2008) 1304–1318

and ICT careers and whether they planned to include ICT subjects in their higher education studies. The sur-vey contained two main formatting changes from the pilot version. It used 5-point Likert scales ranging from‘strongly disagree’ to ‘strongly agree’ for each of the factors presented as closed options, rather than the simple‘tick the box’ format used in the pilot survey. In addition, it presented the ‘other factors associated with deci-sions’ option as a separate question, rather than as the final option at the end of a long list within a question.This was intended to serve as a more active approach to eliciting additional factors or reasons in decisions totake or not take ICT subjects.

A total of 5863 surveys were mailed to the original selection of 31 schools. Twenty-six schools administeredthe survey with 1453 girls completing it (a 25% return rate). Of the 1,453 responses, 673 were from Year 12 and764 were from Year 11 students (16 respondents did not indicate their year). A total of 1322 respondents wereNon Takers and 131 were Takers. The survey was voluntary and responses were only accepted from partic-ipants who completed consent forms in accordance with ethical clearance requirements. In some schoolsresponse rates approached 70%, while in others the response rate was below 10%. A more complete break-down of the respondent demographic details appears in Table 1.

The research proposal had sought up to 350 Takers and 1250 Non Takers at each of Years 11 and 12 and tokeep drawing from a reserve pool of schools until targets were met. It became apparent at an early stage ofnegotiations with schools that such targets would be difficult to achieve. Schools were reporting research fati-gue, and this made it difficult to ask staff members to keep reminding students to return completed consentforms. In view of this, the research team began organizing and reviewing data as soon as the first returns camein. Before even half of the eventual returns had come in the responses were no longer producing new infor-mation. With saturation reached it was unnecessary to extend the data collection timeframe and study pop-ulation. Moreover, a trend in the proportion of Takers to Non Takers in the vicinity of 10% was discernedearly in the returns and did not vary significantly as further responses were received.

3.2.2. Preparation of data for analysis

Statistical comparison of responses of Takers with those of Non Takers required some manipulation of thedata set. In line with previous experience in the pilot study, questions for Takers had been framed positively(e.g., ‘‘The subjects are interesting’’) and those for Non Takers had been framed negatively (e.g., ‘‘The subjectsare boring’’). This meant that a statistical comparison of the two groups was only possible after scores for NonTakers had been reversed on those questions which demonstrated polarity with those of the Takers. Table 2indicates the questions which were reverse scored during this process.

4. Results

The data were analyzed using two procedures: initial plotting of means and non-parametric statistics;Mann–Whitney U test comparison.

4.1. Data analysis: Initial plotting

First, initial plotting of Non Takers’ (n = 1322) mean data established the pattern displayed in Fig. 1. It isnoted that for simplicity only those variables relevant to the current discussion are included in Fig. 1. Morecomplete figures displaying all responses of Non Takers are to be found in Appendix A. Fig. 1 demonstratesthat of all variables only ‘‘the subject is boring’’, ‘‘I am not interested in computers’’ and ‘‘The subjects would

Table 1Participant details of Takers and Non Takers

Takers of IPT/ITS (or both) % Non Takers of IPT/ITS % Total

Year 11 81 5.6 677 46.6 758Year 12 47 3.2 632 43.5 679Year unknown 3 .2 13 .9 16

Totals 131 9 1322 91 1453

The subjectwould not be

helpful tome in mychosen

career path

I don't havemuch

experience in using

computers

I am notinterested

incomputers

There aretoo many

boys in thissubject

Thesubjects

are boring

I don't havea computer

at home

I did notthink it

would helpmy OPscore

I don't thinkI would like

the waythese

subjectsare taught

There is notmuch

content forgirls

I think youneed tobe

good atmaths

Did not suitmy

timetable

SA

A

N A or D

D

SD

Fig. 1. Responses of Non Takers (n = 1,322) to selected survey questions (SD = strongly disagree, D = disagree, NA or D = neither agreeor disagree, A = agree, SA = strongly agree).

Table 2Question sets which were reverse scored for statistical comparison between Takers and Non Takers

Takers Non Takers (reverse scored)

The subjects are interesting The subjects are boringI am very interested in computers I am not interested in computersI have a lot of experience using computers I don’t have much experience in using computersI thought it would help my OP score I did not think it would help my OP scoreThe subject will be helpful to me in my chosen career path after

schoolThe subject would not be helpful to me in my chosen career path afterschool

It suited my timetable It did not suit my timetable

N. Anderson et al. / Computers & Education 50 (2008) 1304–1318 1311

be helpful in my chosen career’’ appeared above the ‘‘Neither Agree or Disagree’’ line indicating a slight trendamong participants to agree with these statements.

4.2. Data analysis – Non-parametric statistics

Second, because standard assumptions of analysis of variance (ANOVA) were violated (Levene’s test ofhomogeneity of variance <.05), non-parametric statistics were undertaken which are considered ideal for datameasured on nominal (categorical) and ordinal (ranked) scales (Pallant, 2005). Table 3 reports the findings ofMann–Whitney U test comparisons between Takers and Non Takers on questions adumbrated in Table 2.Statistically significant differences were found between the two groups on four items: ‘The subjects are inter-esting’, ‘I am very interested in computers’, ‘The subject will be helpful to me in my chosen career path after

1312 N. Anderson et al. / Computers & Education 50 (2008) 1304–1318

school’ and ‘It suited my timetable’. Two items showed no significant difference between the groups; ‘I have alot of experience using computers’ and ‘I thought it would help my OP score’. In other words, both Takers andNon Takers believed the subjects would help their OP score and likewise both groups felt they had experiencein using computers. Whereas, the Takers chose the subjects, the Non Takers did not. Conversely, the reasonsfor the differences in the choices seemingly have to do with interest in computers, the subjects, timetable issuesand beliefs about the subject in relationship to future careers.

5. Discussion

The force of the rejection of ICT subjects by Non Takers may be intensified by the fact that timetabling wassignificantly less of an impediment to their choice than it was for Takers. The statistical difference betweenTakers and Non Takers, as measured using Mann–Whitney U test comparisons, went the opposite way tothe other variables. Non Takers were actually more disposed toward taking ICT subjects than were Takersso far as timetabling was concerned, yet they chose subjects that were less boring/more interesting and/or thatthey saw as being more appropriate to their career path. An obvious question that arises here is: Why?

This question is of particular interest because 84 of the Non Takers stated their intention to subsequentlytake ICT subjects in tertiary study. That is, they did not see the Board ICT subjects as relevant for their futurecareers, yet this noteworthy proportion of Non Takers (6.3%) reported tertiary ICT subjects among theirintended future career pathways. This information cuts different ways. On one hand it might point towardthe intensity of their rejection of school ICT subjects. On the other hand, however, it may prompt questionsabout assumptions concerning career pathways that may underpin research and curriculum developmentalike. If the Board subjects are intended to constitute career pathways then they are obviously not as effectivein the minds of many school students as they might be.

At the same time, as research by the AAUW (2000) highlights, there are multiple pathways to ICT careersbesides school subjects. This indicates the value in subsequent research of including a focus on how femalestudents at school levels perceive career pathways so far as ICT professional careers are concerned. In the caseof the current research cohort, this question will be explored in the focus group phase. It will also be worthfocusing interview items on what participants Non Takers consider might make senior school ICT subjectsmore attractive from the standpoint of career pathways.

An interesting outcome of the girls and ICT survey phase is the fact that responses provided little supportfor those factors reported in extant research literature most often associated with low female participationrates in ICT career pathways and professions. It is important to note that the survey did not attempt inany direct or straightforward way to tap these factors. Rather, the strategy was to employ a combinationof potential ‘prompts’ in the forced choice items and opportunities to follow up in the open ended questionoptions. Hence, for example, the forced choice item ‘There are too many boys in this subject’’ provided anopportunity to broadly tap into aspects of male domination or intimidation. Similarly, the image of the subjectwas broadly ‘tapped’ by a forced choice item asking participants whether their decision was influenced by whatparticipants felt their friends might think if they took the subjects. Other items broadly tapped into comfortlevel/experience with computers, and into male bias in the subjects.

n = Takers n = NonTakers

Mean SD U value Significance

The subjects are interesting 130 1314 2.72 1.29 37402.00 .000***

I am very interested in computers 130 1314 2.93 1.32 51,609.00 .000***

I have a lot of experience using computers 129 1315 3.52 1.16 79,032.00 .186 (ns)I thought it would help my OP score. 130 1308 3.53 1.10 78,626.50 .142 (ns)The subject will be helpful to me in my chosen career path after

school.130 1314 2.63 1.28 39,046.50 .000***

It suited my timetable. 130 1306 3.46 1.18 72,152.00 .003**

Note: Discrepancies in participant numbers reflect missing responses for these items.** p < .01.

*** p < .001.

N. Anderson et al. / Computers & Education 50 (2008) 1304–1318 1313

None of these factors emerged as statistically significant when reverse scored responses for Non Takerswere compared with the responses of Takers. Neither did the responses to open choice questions, whichshowed evidence of considerable reflection on the part of respondents, give any indication that image of sub-ject or industry, male bias in computing generally and computing subjects specifically, male domination inclass, or lack of computing ‘savvy’ provide any indication that these factors featured particularly in subjectchoice. Some participants alluded to lack of knowledge about the subjects and their availability, but not ona scale remotely comparable to such clear statements that featured very frequently in open ended responsesas: ‘These subjects are boring’; ‘I am not interested in computers’; ‘Not interested’, ‘Into more hands on, prac-tical things’; ‘I’m not interested in what happens with my computer, all I want to do is use it’.

It is noted that the study design was limited by the need to reverse score responses for Non Takers (seeTable 2), and that the survey was designed to elicit the current participants’ own grounds for decision makingrather than to assess the pertinence of commonly identified factors for the Queensland study population. How-ever, the fact remains that our survey results point in very different directions to those pointed by previousresearch. For example, Non Takers did not report intimidation by the over-representation of males in thesesubjects (see Figs. 1 and 2) (Cisco Systems, 2002).

In this study, the survey instrument was used primarily with a view to sharpening focal themes that couldbe taken up in greater depth by means of qualitative interviewing as well as to provide directions for sub-sequent studies. The survey instrument was an eclectic device. It allowed us to tap into broad trends inresponses to forced choice items. Such items are often problematic when it comes to finer points of meaning.For example, an item like ‘I am (not) interested in computers’ is almost certain to identify people who donot find computers and computing interesting in and of themselves and as phenomena they would like toknow more about. But it cannot tell us anything about whether a person is interested or not in computersas programmable devices, as tools used in design, as access points to online affinity spaces (Gee, 2004) asso-ciated with practices such as gaming, ‘fanfiction’ writing and Anime–Music–Video remixing. Yet this deeperqualitative level might be crucially important to better understanding of what might be involved in makingICT subjects more attractive to learners as potential pathways to ICT careers. Alternatively, it may provideus with valuable information about alternative pathways into ICT careers, such as where programmers have

Non TakersTakers

SA

A

D

SD

It suited timetable

Subject helpful forcareer

Interest in computers

Subjects interesting

N A or D

Fig. 2. Comparison of Takers and Non Takers on statistically different study variables (SD = strongly disagree, D = disagree, NA orD = neither agree or disagree, A = agree, SA = strongly agree).

1314 N. Anderson et al. / Computers & Education 50 (2008) 1304–1318

developed an interest in programming through gaming and gone on to work as games designers, computerprogrammers, and so on.

The eclectic nature of the survey instrument provided some opportunities to obtain clues that could lead tothe kind of qualitative knowledge sought. This was enabled through the open-ended questions and one of theYes–No items. The former provided participants with space to elaborate on their forced choice responses. Thisoption proved especially useful with respect to the trend among Non Takers to freely identify their perceptionof the ICT subjects as boring and to elaborate on what they did not like about computers, and so on. The Yes–No items in question asked respondents if they had participated in any of the special events around computingthat had been organized within Queensland, such as the I-Star camps. It was confirmed that special ICT eventshad a significant relationship with girls’ plans to study ICT in the future (Anderson et al., 2006).

These facets of the survey instrument provided valuable information for designing items for the semi struc-tured interviews in Phase 2 of the data collection, designed to obtain the additional qualitative depth in infor-mation sought for the larger study, and which will be reported elsewhere.

5.1. Conclusion and next steps

In summary, the survey found little support within the current study population for the factors typicallyassociated with low rates of female participation in ICT careers and traditionally recognized career pathwaysin the form of high status school subjects with ‘face validity’ for professional computing. At the same time,it identified two important factors seemingly associated with aversion toward formal subject pathways: theperception that these subjects were boring, and a strongly expressed aversion to computers. These leads arebeing followed in qualitative interviewing of girls in four schools (GS and NGS, urban and rural schools)and the results of the second phase will be reported in subsequent publications.

The third phase of this research project has just been launched, the Women in ICT Industry Survey, whichis a national online survey. Major IT companies (e.g., Technology One, IBM, Mincom), government depart-ments (e.g., EQ, DEST) and women’s ICT organizations (e.g., Australian Women in IT and Science Entity[AWISE]), Females in Information Technology and Telecommunication [FITT], WIT) were contacted toassist in announcing the Women in Industry to their employees or membership lists. Furthermore severalof these organisations have agreed to put a link on their homepage announcing that this research is underwayand providing a link to the survey website. During the two weeks 158 responses were received from all Aus-tralian states. Data analysis is expected to be completed in November 2006, which is expected to add furtherinsight into the low participation rates of females in ICT pathways.

Appendix A

See Table 4.

Table 4Responses of Non Takers to survey questions

Frequency Percent Cumulative percent Mean SD

ICT did not suit my timetable

Valid Not applicable 16 1.2 1.2Strongly disagree 350 26.5 27.7Disagree 262 19.8 47.5Neither agree or disagree 448 33.9 81.4Agree 173 13.1 94.5Strongly Agree 73 5.5 100.0

Mean and SD 2.48 1.20

Total 1322 100.0

Table 4 (continued)

Frequency Percent Cumulative percent Mean SD

I would be worried about what my friends would think

Valid Not applicable 7 .5 .5Strongly disagree 973 73.6 74.1Disagree 226 17.1 91.2Neither agree or disagree 96 7.3 98.5Agree 7 .5 99.0Strongly agree 13 1.0 100.0

Mean and SD 1.37 .73

Total 1322 100.0

I think you need to be good at maths

Valid Not applicable 7 .5 .5Strongly disagree 457 34.6 35.1Disagree 436 33.0 68.1Neither agree or disagree 285 21.6 89.6Agree 116 8.8 98.4Strongly agree 21 1.6 100.0

Mean and SD 2.08 1.04

Total 1322 100.0

There is not much content for girls

Valid Not applicable 7 .5 .5Strongly disagree 497 37.6 38.1Disagree 407 30.8 68.9Neither agree or disagree 310 23.4 92.4Agree 81 6.1 98.5Strongly agree 20 1.5 100.0

Mean and SD 2.02 1.01

Total 1322 100.0

I don’t think I would like the way these subjects are taught

Valid Not applicable 7 .5 .5Strongly disagree 289 21.9 22.4Disagree 361 27.3 49.7Neither agree or disagree 455 34.4 84.1Agree 173 13.1 97.2Strongly agree 37 2.8 100.0

Mean and SD 2.46 1.07

Total 1322 100.0

I did not think it would help my OP score

Valid Not applicable 14 1.1 1.1Strongly disagree 322 24.4 25.4Disagree 336 25.4 50.8Neither agree or disagree 435 32.9 83.7Agree 167 12.6 96.4Strongly agree 48 3.6 100.0

(continued on next page)

N. Anderson et al. / Computers & Education 50 (2008) 1304–1318 1315

Table 4 (continued)

Frequency Percent Cumulative percent Mean SD

Mean and SD 2.43 1.13

Total 1322 100.0

I was discouraged by a teacher or careers counselor

Valid Not applicable 14 1.1 1.1Strongly disagree 635 48.0 49.1Disagree 392 29.7 78.7Neither agree or disagree 234 17.7 96.4Agree 36 2.7 99.2Strongly agree 11 .8 100.0

Mean and SD 1.75 .91Total 1322 100.0

I was discouraged by a family member

Valid Not applicable 6 .5 .5Strongly disagree 669 50.6 51.1Disagree 387 29.3 80.3Neither agree or disagree 194 14.7 95.0Agree 56 4.2 99.2Strongly agree 10 .8 100.0

Mean and SD 1.74 .92

Total 1322 100.0

I was discouraged by friends

Valid Not applicable 12 .9 .9Strongly disagree 711 53.8 54.7Disagree 383 29.0 83.7Neither agree or disagree 183 13.8 97.5Agree 26 2.0 99.5Strongly agree 7 .5 100.0

Mean and SD 1.64 .84

Total 1322 100.0

I don’t have a computer at home

Valid Not applicable 6 .5 .5Strongly disagree 807 61.0 61.5Disagree 292 22.1 83.6Neither agree or disagree 121 9.2 92.7Agree 63 4.8 97.5Strongly agree 33 2.5 100.0

Mean and SD 1.64 .84Total 1322 100.0

The subjects are boring

Valid Not applicable 8 .6 .6Strongly disagree 135 10.2 10.8Disagree 163 12.3 23.1Neither agree or disagree 373 28.2 51.4Agree 329 24.9 76.2Strongly agree 314 23.8 100.0

1316 N. Anderson et al. / Computers & Education 50 (2008) 1304–1318

Table 4 (continued)

Frequency Percent Cumulative percent Mean SD

Mean and SD 3.38 1.28Total 1322 100.0

There are too many boys in this subject

Valid Not applicable 45 3.4 3.4Strongly disagree 479 36.2 39.6Disagree 310 23.4 63.1Neither agree or disagree 387 29.3 92.4Agree 65 4.9 97.3Strongly agree 36 2.7 100.0

Mean and SD 2.04 1.11Total 1322 100.0

I am not interested in computers

Valid Not applicable 8 .6 .6Strongly disagree 181 13.7 14.3Disagree 258 19.5 33.8Neither agree or disagree 311 23.5 57.3Agree 313 23.7 81.0Strongly agree 251 19.0 100.0

Mean and SD 3.13 1.33

Total 1322 100.0

I don’t have much experience in using computers

Valid Not applicable 7 .5 .5Strongly disagree 313 23.7 24.2Disagree 404 30.6 54.8Neither agree or disagree 298 22.5 77.3Agree 236 17.9 95.2Strongly agree 64 4.8 100.0

Mean and SD 2.48 1.19

Total 1322 100.0

The subject would not be helpful to me in my chosen career pathValid Not applicable 8 .6 .6

Strongly disagree 115 8.7 9.3Disagree 167 12.6 21.9Neither agree or disagree 341 25.8 47.7Agree 353 26.7 74.4Strongly agree 338 25.6 100.0

Mean and SD 3.46 1.27

Total 1322 100.0

N. Anderson et al. / Computers & Education 50 (2008) 1304–1318 1317

References

Adya, M., & Kaiser, K. M. (2005). Early determinants of women in the IT workforce: A model of girls’ career choices. Information

Technology and People, 18(3), 230–259.American Association of University Women Educators (AAUW). (2000). Tech-savvy: Educating girls in the new computer age.

Washington: American Association of University Women Educational Foundation.

1318 N. Anderson et al. / Computers & Education 50 (2008) 1304–1318

Anderson, N., Klein, M., & Lankshear, C. (2005). Redressing the gender imbalance in ICT professions: Towards state level strategicapproaches. Australian Educational Computing, 20(2), 3–9.

Anderson, N., Lankshear, C., Courtney, L., & Timms, C. (2006). Girls in ICT Survey: Initial findings. Curriculum Leadership: Electronic

Journal for Leaders in Education, 4(16). Available fromhttp://cmslive.curriculum.edu.au/leader/default.asp?id=13812.Anderson, N. (2003). Making the link: Three successful university, industry and community collaborations around ICT development. In J.

Renner & J. Cross (Eds.), Higher Education without Borders, Sustainable Development in Higher Education. Perth: Edith CowanUniversity.

Beckwith, L., Burnett, M., Wiedenbeck, S., Cook, C., Sorte, S., & Hastings, M. (2005). Effectiveness of end-user debugging softwarefeatures: Are there gender issues? In: Paper presented at the Conference on Human Factors in Computing Systems, April 2–7, 2005,Portland, OR, USA.

Camp, T. (1997). The incredible shrinking pipeline. Communications of the ACM, 40(10), 104–110.Cisco Systems (2002). Gender Initiative. <http://gender.ciscolearning.org> Retrieved July 6, 2005.Cohoon, J. M. (2003). Must there be so few? Including women in CS: Paper presented at the 25th International Conference on Software

Engineering, Portland, OR, USA.Coonan, H., (2005). Women in ICT summary advisory group established. <http://www.minister.dcita.gov.au/media/media_releases/

women_in_ict_summit_advis> Retrieved March 29, 2006.Denner, J., Werner, L., Bean, S., & Campe, S. (2005). The Girls Creating Games Program: Strategies for engaging middle school girls in

information technology. Frontiers: A Journal of Women’s Studies, 26(1), 90–98.Frieze, C. (2005). Diversifying the images of computer science: Undergraduate women take on the challenge! Paper presented at the

Special Interest Group of Computer Science Education ’05 (SIGCSE), February 23–27, St. Louis, MI, USA. <http://delivery.acm.org/10.1145/1050000/1047476/p397-frieze.pdf?> Retrieved June 29, 2005.

Gee, J. P. (2004). Situated language and learning: A critique of traditional school. London and New York: Routledge.Graham, S., & Latulipe, C. (2003). CS girls rock: Sparking interest in computer science and debunking the stereotypes. In Proceedings of

the Special Interest Group on Computer Science Education, 2003 (pp. 322–326).Gurer, D., & Camp, T. (2002). An AMC-W literature review on women in computing. In Road, Special Interest Group on Computer

Science Education (SIGCSE) Bulletin, 34(2), 121–127.Kahle, J., & Schmidt, G. (2004). Reasons women pursue a computer science career: Perspectives from women from a mid-sized institution.

Journal of Computer Science in Colleges, 19(4), 78–89.Lankshear, C., & Knobel, M. (2003). New illiteracies: changing knowledge in the classroom. Buckingham, England and Philadelphia, PA:

Open University Press.Margolis, J., & Fisher, A. (2003). Unlocking the clubhouse: women in computing. Cambridge, MA: Massachusetts Institute of Technology

Press.Millar, J., & Jagger, N. (2001). Women in ITEC courses and careers. London, Department of Education and Skills, Department for

Employment. The Women’s Unit: 156.Newmarch, E., Taylor-Steele, S., & Cumpston, A. (2000). Women in IT – What are the barriers? Conference paper presented to the

Network of Women in Further Education Conference, Net gains: Women, Information Technology and emerging issues. <http://www.dest.gov.au/research/pubs/womeninit.htm> Retrieved March 22, 2006.

Pallant, J. (2005). SPSS survival manual: A step by step guide to data analysis using SPSS (2nd ed.). Crows Nest, NSW: Allen & Unwin.Queensland Government (2003). Smart classrooms: Report on the girls and ICT framework for action 2003–2004. Department of

Education and the Arts, Queensland Government. <http://education.qld.gov.au/smartclassrooms/pdf/girls-ict-frameworkreport.pdf>Retrieved March 6, 2006.

Sandy, M., & Burger, C. (1999). Women and minorities in information technology forum: Causes and solutions for increasing the numbersin the information technology pipeline. Report presented at The National Science Foundation, ‘‘Transitions from Childhood to theWorkforce’’ Forum, November 4–5, Hampton, VA, USA.

Zweben, S., & Aspray, W. (2004). 2001–2003 Traulbee Survey: Undergraduate enrollements drop: Department growth expectationsmoderate. Computing Research News, 16(3), 5–19.