|USFDC Home||| RSS|
This item is only available as the following downloads:
1 of 26 A peer-reviewed scholarly journal Editor: Gene V Glass College of Education Arizona State University Copyright is retained by the first or sole author, who grants right of first publication to the EDUCATION POLICY ANALYSIS ARCHIVES EPAA is a project of the Education Policy Studies Laboratory. Articles appearing in EPAA are abstracted in the Current Index to Journals in Education by the ERIC Clearinghouse on Assessment and Evaluation and are permanently archived in Resources in Education Volume 12 Number 17April 20, 2004ISSN 1068-2341Policy Issues for Australias Education Systems: Evidence from International and Australian Research Gary N. Marks Julie McMillan John Ainley Australian Council for Educational Research and Melbourne Institute for Economic and Social Researc hCitation: Marks, G., McMillan, J., Ainley, J., (200 4, April 20). Policy issues for Australias educati on systems: Evidence from international and Australian research. Education Policy Analysis Archives, 12 (17). Retrieved [Date] from http://epaa.asu.edu/epa a/v12n17/.AbstractOur purpose here is to discuss education policy iss ues in the context of empirical evidence. We note that many co mmonly held beliefs about Australian education such as, the rel ative performance and participation levels of Australian students; the importance of socioeconomic background on education al outcomes both relative to other countries and chang es over-time; gender differences in mathematics and science; and the labour market situation of early school leavers; are not s upported by empirical research. Such findings have implications for
2 of 26 government policies. We also question current polic y directions toward increasing Year 12 participation, expanding both secondary and post-secondary vocational education a nd reducing class sizes. It is hoped that the discussion will p rovide stimulus to evidence-based debates about Australian education.School EducationStudent PerformanceA fundamental point about Australian education is t hat the performance of Australian secondary school students is high by int ernational standards. The 1994 Third International Mathematics and Science Study (TIMSS) found that Australian performance in mathematics in the junior secondary years was lower than only eight (out of 45) countries. The performa nce of Australian students was significantly better than comparable countries such as New Zealand, England and the United States. The performance of A ustralian students was similar to the performance of students in Canada, I reland, Sweden and France. In science, only four countries outperformed Austra lia: Singapore, Korea, Japan and the Czech Republic. Australia recorded science achievement levels similar to that for England and the United States, as well as most of the countries that were similar to it in mathematics (Lokan et al., 19 96:15-16). In the 1999 TIMSS study, Australian students perfor mance in mathematics was again well above the international average by subst antial 0.4 of a standard deviation. Australian performance was significantly lower than six countries: Singapore, Korea, Chinese Taipei, Japan and Flemish Belgium. It was not different from a second group of countries that inc luded the Netherlands, Canada, Finland, and the Czech Republic. It perform ed significantly better than the United States, England and New Zealand (Mullis et al., 2000:32). In science, Australia also performed above the interna tional average by about 0.5 of a standard deviation. Only Chinese Taipei scored significantly higher than Australia. Australia was not different from Singapo re, Japan, Korea, the Czech Republic, England, Canada and Hong Kong. Its studen ts outperformed those in the United States, New Zealand and Italy (Martin et al., 2000:32). In the recent 2000 PISA study of 15 year olds in ov er 30 industrialised countries, Australian students performed well above the OECD average in the three domains of reading, mathematics and science. Students in Finland were the only national group that performed significantl y better in reading literacy than Australian students. Students in Japan were th e only ones who performed significantly better than Australian students in ma thematics. Japanese and Korean students were the only national groups that performed significantly better than Australian students in science (Lokan e t al., 2001:20-33). Therefore, there is consistent evidence that Austra lian students are performing at levels that can be regarded as very good. Their high performance is not limited to a single subject area. This conclusion is no doubt surprising to many who are mindful of the inadequacies of the Australi an education system. The media generally overlooks this good news finding and many involved in Australia education are not aware of how well Austr alian students perform
3 of 26 relative to students in comparable countries. So it is important to find out what Australia is doing right; is it the high quality of teaching or teacher education, the competition between government and non-governme nt schools, the academic environment of schools, the curriculum, or the communitys interest in students education. Although many of these explana tions may be dismissed out-of-hand, the question remains as to why Austral ian students are performing much better than students in comparable countries. Since Australia spends slightly less of its GNP on education than other co mparable countries (Note 1) it could be argued that Australia spends its resour ces more effectively than other countries. So understanding why Australian st udent performance is high by international standards is also important to pol icy makers who need to know where the education dollar is best spent.Although the performance of Australian students is higher than most comparable countries, there is no evidence that the absolute performance of Australian students has improved over time. Rosier (1980) focussed on changes in mathematics achievement between 1964 and 1978 and concluded that there had been a slight decline in the perform ance of 13-year-olds over that period time. Focusing on a longer time span (1964-1 994) Afrassa and Keeves (1999) concluded that there was a decline in the ma thematics performance of 13-year-olds. The magnitude of that decline was app roximately 30 scale points (or 0.3 standard deviations), a non-trivial decline Over the period from 1975 to 1998 there was no change in performance in reading or mathematics (Rothman, 2002). Comparison of science performance between 1994 and 1998 suggests that the relative position of Australia in the country league table of student performance improved (Martin et al., 2000:3 5). However, in absolute terms there was little change in mathematics and on ly a slight improvement in science.Therefore, the performance of Australian students i s high by international standards, but there is no evidence that this high standing is due to improvements in student learning and thus policy in itiatives over the last 30 years. There are a number of strong arguments to fu rther increase the achievement levels of Australian students. For indi vidual students, proficiency in literacy and numeracy is by far the most powerful i nfluence on a range of educational outcomes including early school leaving tertiary entrance scores, and participation in higher education (Marks et al. 2000; Marks et al., 2001; McMillan & Marks, 2003). In addition, literacy and numeracy are important influences on labour market outcomes such as not be coming unemployed, the duration of unemployment, and income (Marks & Flemi ng, 1998b, 1998c; Marks, forthcoming; McMillan & Marks, 2003). The International Adult Literacy Study shows large labour market differences between high and low literacy groups (Kirsh et al., 1993). At the macro-economic level, there is strong case to improve student performance in literacy and numerac y since the economy is likely to be increasingly reliant on industries bas ed on the manipulation of symbols (words and numbers).An important policy question is to how to improve t he performance of students at the bottom end of the distribution. Poor skills in literacy and numeracy are the strongest risk factor for unsuccessful school to wo rk transitionsa stronger risk factor than low socioeconomic background. It is pos sible for a country to
4 of 26 achieve both high average levels of student perform ance and small variation. This involves policies that lift the performance of weaker students without undermining the performance of other students.Educational Participation In Senior Secondary Schoo l One of the most dramatic changes that has occurred in Australian education over past two decades is the rapid increase in Year 12 participation from 35 per cent in 1980 to a peak of 77 per cent in 1992.. Thi s rate has since declined before rising again to around 76 per (ABS, 1984-200 2). However, participation in the final year of school in Australia is lower t han that in many other countries. According to the OECD, 78 per cent of sixteen year olds in Australia are enrolled in upper secondary school. This figure is lower than the OECD average of 84 per cent and is considerably lower than enrol ment rates at the same age in Austria (90 per cent), Belgium (97 per cent), Ca nada (85 per cent) and Sweden (96 per cent) (OECD, 1998:170). However, sch ool completion in university-oriented programs in Australia is higher (66 per cent) than the OECD average (OECD, 2001a:146).The lower level of participation in Australia poses the policy question of whether participation rates should be increased. This invol ves an assessment of how those who do not complete secondary school are fari ng in the labour market. The early labour market experiences of non-complete rs are highly dependent upon the economic climate. Research on non-complete rs who entered the labour market during the early 1990s showed that th is group were experiencing substantially poorer labour market outcomes than an equivalent group who had left school a decade earlier (Lamb et al., 2000). O n the other hand, research on a more recent group of non-completers who entered t he labour market later in the 1990s when the economy was healthier, presents a far more positive picture (Marks & Fleming, 1998a). Subsequent work following the progress of this group until age 19 shows increasing levels of fulltime work, incomes and occupational status (McMillan & Marks, 2003). Among those who did not go to university, there was little difference between ear ly school leavers and school completers in full-time work in the initial years a fter the leaving school. It was non-completers who left school in Years 11 and Year 12 that were having problems securing full-time work (Marks et al., For thcoming). Furthermore, school completion in itself has little influence on labour market outcomes among 21 to 25 year olds (Marks et al., 2003).Those who do not complete secondary school have poo rer labour market outcomes than those with university qualifications. It is well established, both in Australia and overseas, that university qualificati ons are associated with higher incomes, less unemployment, and more steeply rising occupational and income trajectories. When making comparisons between non-c ompleters and those school completers who do not pursue university stud ies, the evidence that completing school is beneficial is equivocal. In so me regards non-completers fare better than school completers who do not enter higher education: they are more likely to be in full-time employment and recei ve higher hourly earnings, at least initially. However, in other regards non-comp leters experience less successful transitions from school: compared with c ompleters who did not enter higher education, male non-completers are more like ly to be unemployed, and
5 of 26 female non-completers are more likely to be outside the labour force and not studying (McMillan & Marks, 2003).During the last decade, one policy response to the labour market outcomes of school non-completers was to increase participation in school. This involved broadening the curriculum by including courses that potential school leavers would find attractive so would remain at school. Th is policy direction was in part a product of research conducted during the 1980s an d early 1990s that argued that non-completers left school because they were a lienated from the academically orientated curriculum. This is undoubt edly true for some students although the degree of student antipathy with schoo l has been over-stated. Longitudinal research on a cohort of young people w ho were in Year 9 in the mid 1990s shows that the majority of non-completers leave school for positive work related reasons. About 50 per cent say the mai n reason they left school was to get a job or an apprenticeship (whether or n ot they actually had a job to go to), and a further 5 per cent say they wanted to earn their own money. Only 13 per cent said their main reason for leaving was that they did not like school, only 6 per cent left because of the subject choice at their school, and only 2 per cent said they left on the advice of teachers. Inte restingly, only 1 per cent cited financial reasons (McMillan & Marks, 2003). Althoug h, subjective evaluations may include post-hoc rationalizations non-completers are most often st udents with lower achievement levels, so are struggling in senior secondary schoolthese data do indicate that schools and the school curriculum are a much smaller influence on school leaving than gener ally believed. The policy implication of these results is that further effort s to make Year 12 more attractive to potential school leavers may not be the most appropriate strategy. Given that many non-completers have positive reason s for leaving school and the majority do obtain full-time work, is there any reason why a student, keen to leave school and has a clear intention of working i n a particular type of job should not do so? Analysis of the labour market out comes on youth cohorts aged between 17 and 25 shows that prior experience of full-time work has considerably larger effects than qualifications on subsequent full-time employment (Marks et al., Forthcoming; McMillan & M arks, 2003). Experience in full-time work provides a strong basis for conti nued full-time work. The strong influence of prior experience in full-time work on subsequent full-time employment appears to becoming stronger (Marks & Fl eming, 1999). So delaying entry to the labour market for one or two years may not beneficial given the importance of labour force experience.Increasing school completion is likely to have othe r undesirable outcomes. Government school students are more likely to leave school than non-government students, so if Year 12 participatio n is increased, students who would have otherwise left school will be enrolled i n government schools. Therefore, the responsibility of catering for this academically weak group of students will fall on government school systems whi le non-government schools concentrate on maximising the university entrance p erformance of their senior students. Thus, the gap in performance between the government and non-government sectors will widen and government sc hools will be increasingly viewed as a residual category. Parents who can affo rd to send their children to non-government schools will do so. The result will be increasing socioeconomic
6 of 26 inequalities in education.Policy options for students likely to leave school early should consider the prevailing and future economic conditions, the ease at which school leavers can later pursue full-time education and training, the cost to potential employers and the assistance available to students who make unsuc cessful school-to-work transitions.Assessments about the current and future state of t he youth labour market provide crucial contextual background for the formu lation of policy options. The substantially more favourable labour market experie nces of school leavers during the late 1990s compared to the early and mid 1990s, is largely due to the improvement in the macro-economy. In an analysis of unemployment in three Australian youth cohorts, a large contextual effect of the annual unemployment rate was found (Marks & Fleming, 1998b). The OECD r eported that, in general, countries with healthier economies and lower unempl oyment show more successful school to work transitions (OECD, 2000b: 37-43). Therefore, low unemployment is a necessary precondition for allowi ng students to leave school before completion of Year 12.A second issue concerns the ease with which school leavers can later return to full-time education. A problem with early school le aving is that it reduces options for further (especially higher) education. Universi ties typically judge prospective students on their performance in Year 12, so non-co mpleters face barriers if they wish to pursue higher education at a later tim e. Therefore, encouraging universities to adopt flexible entrance requirement s for young people who do not complete Year 12, and providing other forms of further education, would represent a policy alternative to increasing school completion. Many universities already have some but limited provisions for later age entry. A third issue is the cost to employers in employing young people who have just left school. Employers need to be encouraged to em ploy those who have not completed secondary school and to provide associate d training to develop their skills. This could include the further extension of formal training provisions to industries that do not traditionally take apprentic es. This has been the thrust of the new apprenticeships and traineeship schemes. An other policy option is to reduce the marginal cost to employers of employing school leavers. Finally, it is important to assist those young peop le who are experiencing unsuccessful transitions to the labour market. Esti mates from studies of recent school leavers suggest that 10 per cent (or less) o f those who do not enter higher education are facing severe difficulties in obtaining work (Marks et al., Forthcoming; McMillan & Marks, 2003). Policies shou ld be targeted at this small group that are actually experience difficulties rat her than assuming that all school non-completers are at risk. However, before specific policies can be implemented closer monitoring the school to work tr ansition is required because many do not come to the attention of government dep artments if they have not applied for social security benefits.Vet-in-Schools
7 of 26 In Australia, a number of vocational education and training (VET) programs are available to students who are still at school and t his has been a substantial area of growth throughout the 1990s. Nationally, approxi mately one quarter of the student cohort from Year 9 in 1995 had participated in some form of VET as part of their studies in Year 11 and 12 (Fullarton, 2001). These data indicate that some 15 per cent of school students had undert aken some VET-in-School subjects at either Year 11 or Year 12, and 7 per ce nt had completed subjects in both Year 11 and Year 12. Only a few (slightly more than one per cent) had participated in a school-based new apprenticeship o r traineeship. There are substantial differences among jurisdictions in part icipation in VET. The highest level of participation is found in Queensland (41 p er cent) and the lowest in Victoria (12 per cent). Participation in VET in sch ools is also higher among students from government schools and with below ave rage achievement levels (Fullarton, 2001). Lamb et al. (1998) noted that VE T-in-schools tends to attract students with manual occupational backgrounds.There is little research on whether VET-in-schools programs benefit their participants. Malley et al. (2001) argued that most of the participants in VET in schools would have stayed at school anyway and that the availability of VET programs did not encourage potential early school l eavers to remain at school. Fullarton (2001) found that after leaving school th e unemployment rate for the VET-in-schools group was similar to that for the co mparison group. Furthermore, VET-in-schools does not facilitate ent ry to a recognised form of post-secondary vocational education or training. Th ese results indicate that the labour market outcomes of VET-in-schools participan ts should be carefully monitored. It may be more beneficial for such stude nts to directly enter the labour market, and have their training needs met by the TAFE system. Schools are arguably less equipped to provide vocational tr aining since they usually have only weak links to employers and have limited financial and human resources to provide suitable training.Participation In Higher EducationAnother important change in Australias education s ystem is an increasing level of participation in higher education. In 1999, tota l higher education enrolments were 686,000, more than twice the 330,000 students enrolled in 1980 (DETYA, 2000:8,15). Estimates from the Longitudinal Surveys of Australian Youth show that approximately 40 per cent of recent youth coho rts participate in higher education. The comparable figure for the early 1980 s was 20 per cent (Marks et al., 2000).Over the last decade, the growth in higher educatio n enrolments has been between 2 and 10 per cent per annum, a figure that is much higher than the population growth rate. The OECD reports that unive rsity enrolments in Australia increased by over 25 per cent between 199 0 and 1996. However, the growth in university enrolments between 1995 and 19 99 was considerably less, with Australia showing the seventh lowest growth of 21 OECD countries (OECD, 2001a:152).Overall, the OECD estimates the proportion of the a ge cohort entering higher
8 of 26 education in Australia at 45 per cent (this figure includes TAFE diplomas). This participation rate is the same as the OECD average and comparable with the United Kingdom and the United States (OECD, 2001a:1 55). Attrition from university courses is a concern. An Australian longitudinal study of the cohorts that commenced university in 1992 and 1 993 estimated ultimate completion rates of 72 and 71 per cent respectively For the 1992 commencing cohort, 60 per cent had completed an award in their original university by 1997 and 64 per cent had completed an award by 1999 (Mar tin et al., 2001; Urban et al., 1999). However, attrition in Australia is not particularly large compared to other countries. The ratio of graduates to enrolled students in any year is around 27 per cent for Australia, which compares fa vourably with the OECD average of 19 per cent but is less than that for th e United Kingdom and the United States (OECD, 2001a:169).The labour market outcomes of graduates are superio r to those of non-graduates in terms of both reduced unemployment and higher incomes. Analysing pathways over a seven-year period (from t he late 1980 to the mid 1990s), only 6 per of graduates (Note 2) experienced extended periods of unemployment, part-time work, and not being in the labour market. This compares with between 20 and 30 per cent of non-gra duates (Lamb, 2001:8; Lamb & McKenzie, 2001:25). In 1998, unemployment am ong 20-24 year old university graduates was substantially lower (aroun d 3 per cent) than that for other educational groups. Similar differences are f ound in most industrialized countries (OECD, 2000a:270).The higher income returns from university qualifica tions are well documented. The OECD reports higher incomes for university grad uates (compared to the mean income) in all 20 countries investigated (OECD 1998:352). For Australia, Borland (2002) estimates the private rate of return for a university education among adults at 14 percent equivalent to a lifetime net monetary gain of nearly $400,000. In the early career, a university qualifi cation is one of the strongest influences on income, increasing hourly earnings by around 20 per cent, net of other influences (Marks & Fleming, 1998c). The incr ease in income inequality observed in several countries (including Australia) is often attributed, at least in part, to increased returns to degrees.The issue of increasing participation in higher edu cation should be considered and debated. There are compelling arguments in favo ur of increasing participation. First, there is strong demand in the labour market for university graduates. The predictions, 20 years ago, of undere mployment and decreasing wages for graduates has not eventuated. If anything the strong demand for graduates is increasing. Second, much of Australia s economic and employment growth in the mediumto long-term is li kely to be in industries that employ graduates. In addition, industries that have traditionally employed students with a vocational education are likely to become more technologically sophisticated and require a different set of skills Finally, there is considerable unmet demand for higher education. Surveys of Year 9 students indicate that approximately 70 per cent intend to go to universit y. Although not all students in this group are suited to higher education, it does indicate a much higher level of demand than supply. The main argument against incre asing participation in
9 of 26 higher education is cost. Although the Higher Educa tion Contribution Scheme (HECS) and other measures have reduced the per capi ta cost of university education, most undergraduate teaching is supported from taxation revenue. Recent reforms to the HECS system include increasin g participation but there is little debate on what the participation levels shou ld be in 5, 10 or 20 years time and how should it be funded. (Note 3) Post-Secondary Vocational Education and TrainingVocational education and training (VET) is an impor tant part of the Australias post-secondary education system. Most (over 95 per cent) vocational education and training is provided in institutes of technical and further education (TAFE). Courses include a range of vocational training from entry-level employment preparation, through to trades, advanced vocational para-professional and professional courses. In addition, many recreation and leisure programs are offered. In 1997, approximately 121 000 TAFE studen ts graduated with a qualification from a vocational course of at least 200 hours or one semester in duration (NCVER, 1998). Overall, there were 1.4 mil lion enrolments in VET programs in 1997. Participation is characterised by part-time attendance and a wide age range (persons aged 15-24 years comprise 3 8 per cent of the clients). Entry to many courses is possible after Year 10, bu t in practice nearly half of the entrants to vocational courses have completed Year 12. In the early 1980s, the corresponding proportion was one fifth.Apprenticeships are an important component of VET. Over three to four years an apprentice works for an employer (or group of em ployers) and attends a training institution (traditionally a TAFE institut e, typically for a total of 800 hours). Recent changes have occurred in response to perceived limitations in the apprenticeship system, such as inflexibility, a limited range of occupations, old technology, lack of access for women and declin ing numbers. In 1985, traineeships were introduced to provide a shorter a nd more flexible approach to entry-level training. Traineeships typically involv ed a one-year program with an employer incorporating on-the-job and off-the-job t raining, mostly in office-based and retail industries. More recently, apprenticeships and traineeships have been integrated as part of a more unified entry-level training system. In the mid 1990s, 18 per cent of males had participated in an apprenticeship by age 19, 5 per cent had participat ed in a traineeship and 25 per cent participated in a non-apprenticeship TAFE course. The comparable figures for women were, 2 per cent for apprenticesh ips, 7 per cent for traineeships and 29 per cent in a TAFE course (Lamb et al., 1998:20). Participation in vocational education is higher amo ng males, students from lower socioeconomic backgrounds, rural students, an d English-speaking (rather than non-English speaking backgrounds) backgrounds. Furthermore, VET participants are more likely to have attended gover nment or Catholic schools (rather than independent schools), have low achieve ment levels in literacy and numeracy, and to be school non-completers (Lamb et al., 1998, pp. 19-29). This is the opposite pattern to participation for higher education. Overall VET participation increased between the mid -1980s and mid-1990s. However, there were declines in the proportion unde rtaking apprenticeships by
10 of 26 age 19 (among males from 26 to 18 per cent) (Note 4) and increases in the proportion participating in TAFE courses by age 19 (among males from 10 to 25 per cent) (Lamb et al., 1998:20). The OECD estimate s the proportion of 18-21 years olds currently enrolled in non-university ter tiary education (VET) in Australia is around 8 per which is slightly higher than the OECD average of 5 per cent (OECD, 1998:184).Apprenticeships are associated with lower rates of unemployment in youth cohorts and substantially higher levels of full-tim e work. Traineeship are also beneficial but to a lesser extent (Marks et al., Fo rthcoming; McMillan & Marks, 2003). However, TAFE certificates and diplomas do n ot have strong beneficial effects on the labour market outcomes (Long et al., 1996; Marks et al., Forthcoming). These findings for vocational educati on that apprenticeships improve employment prospects but that vocational ed ucation, in general, does not substantially improve labour market outcomes is consistent with other work in Australia (Dockery & Norris, 1996; Nevile & Saun ders, 1998). Furthermore, such findings are similar to that found in other co untries (Ryan, 2001). In part, this reflects the industries and occupations, which these programs provide access. Furthermore, vocational education may benef it enterprises and the overall economy. However, the lack of evidence that vocational education provides substantial benefits to its participants i s a concern. One interpretation is the greater benefit of apprenticeships and train eeships compared to TAFE certificates and diplomas is because the former inv olve full-time employment. Since VET is closely aligned with industry, shifts in employment patterns impact on its development. One source of change arises fro m shifts in employment away from industrial sectors traditionally served b y VET (e.g., manufacturing) and the growth in other sectors such as hospitality and service (clerical and office) industries. This results in changes in the institutional organisation of VET (e.g., within TAFE institutions and between TAFE an d other VET providers), the areas in which programs are provided (e.g., the eme rgence of formal training arrangements such as traineeships and new apprentic eships in industries not previously involved in apprenticeships) and the for ms of provision that emphasise skill-specific modules of training rather than structured courses leading to a qualification. Another source of chang e arises from the shifting vocational demands within industries that emphasise higher-order transferable skills that can be adapted to new workplace demands Changing VET programs in response to these pressure s may result in a closer relationship between VET and higher education, so t hat TAFE diploma programs overlap with university degree programs to a greater extent. As these changes emerge there will be a need for greater att ention to issues concerned with accreditation, recognition of prior learning, and coordination of administration. At present responsibility for admin istration and delivery of VET resides with the state and territory governments (w ithin a national strategic plan developed through a ministerial council on the advi ce of the Australian National Training Authority). In terms of student fees, ther e have been arguments that the principles of the Higher Education Contribution s Scheme should be applied to VET so that funding becomes more comparable with that in universities. This may be premature given that the income returns to V ET are considerably lower than that for university degrees and that much of t he participation in VET is
11 of 26 directed to short duration certificates and trainin g modules. EquityIn Australia, gender differences in educational par ticipation have reversed. In 1970, boys were more likely to complete school than girls and had higher levels of participation in higher education. During the ea rly 1980s, Year 12 participation for girls was only 3 percentage point s higher than that for boys. By the late 1980s, the gender gap favouring girls in Y ear 12 participation had increased to around 10 percentage points. Across th e OECD world, young women show higher levels of educational attainment than young men, the reverse of the situation for older cohorts (OECD, 1 997: 35, 320-321). Similarly, the gender gap in higher education has i ncreased over time from no gap in the early 1980s to about 10 percentage point s during the late 1990s (Marks et al., 2000). These changes have also occur red in most OECD countries. In 17 of 21 OECD countries, school gradu ation rates for women exceed those by men. Differences in university-orie ntated school courses are even stronger (OECD, 2001a:140, 146). Over the last two decades there has been a clear and continuing trend of higher female participation in tertiary study, especially in university programs (Bradley & Ramine z, 1996). University graduation rates are higher for women than for men (OECD, 2001a:166). However, gender differences in graduation rates for second degrees are much smaller, and in advanced degrees men still tend to outnumber women. This pattern is also found in Australia, where 58 per ce nt of first-degree graduates, 52 per cent of second degree graduates, and 40 per cent of advanced degree graduates are women (OECD, 2001a:173). An Australia n longitudinal study of university commencing students in the early 1990s f ound that women were almost ten per cent more likely to complete an awar d course than men (Martin et al., 2001; Urban et al., 1999).In international achievement studies of reading lit eracy, females outperform males. Across all OECD countries in the 2000 PISA s tudy of 15-year-olds, females scored higher in reading literacy than male s. The differences ranged from 14 points in Korea to around 50 points in Finl and and New Zealand, with the average difference being 32 points (one third o f a standard deviation). In Australia, the difference between males and females was 34 scale points, about the average for the OECD. Within Australia there ar e indications that gender differences in reading achievement have changed ove r time. Marks and Ainley (1997) reported a decline in the proportion of boys attaining mastery in reading. In the Australian 1994 TIMSS study, differences bet ween males and females in mathematics and science were not statistically sign ificant. Australia was one of only a few countries in which there was not a diffe rence in favour of males (Lokan et al., 1996). The lack of gender difference s in mathematics and science achievement observed in 1994 was replicated in the 1999 repeat of the TIMSS study (Martin et al., 2000; Mullis et al., 2000). T hese results contrasted with those reported from earlier international studies o f mathematics and science achievement (Comber & Keeves, 1973; Rosier, 1980; R osier & Keeves, 1991). The results have been quite reasonably interpreted as evidence of the impact of programs that promoted participation in mathematics and science among girls
12 of 26 at school, and of the impact of more general social changes. Gender differences are also evident for tertiary en trance scores. In New South Wales, females are more frequently found in the top percentiles for university admission (NSW UAC, 1998:10). In the great majority of Year 12 courses in New South Wales, females outperform males and the g ap appears to have increased throughout the 1990s (Collins et al., 200 0:50,57-60; MacCann, 1995). The Victorian Tertiary Admission Centre (VTAC, 1998 -1999:98-99107) reports higher percentages of females in the top percentile bands, with males more common in the lower bands. Females outperform males in the majority of subjects in Victoria and Western Australia (Collins et al., 2000:55). In the Queensland Cores Skills Test (QCS), there were prop ortionally more males in the very top band, more females in the following hi gh and middle achieving bands, but more males in the lower bands. The trend towards females outperforming males is not limited to the Australia n context (Baker & Jones, 1993).Socioeconomic BackgroundAs a result of a number of large-scale studies cond ucted at a national and international level there is now consistent evidenc e of the magnitude of the relationship between socioeconomic background and e ducational outcomes. Typically, correlation coefficients of approximatel y 0.3 are reported between socioeconomic status and educational outcomes. In P ISA, achievement was positively associated with student socioeconomic st atus in all countries but there were differences between countries in the str ength of this association (Note 5) A measure of the strength of this association is provided by the gain in reading literacy associated with a one internationa l standard deviation increase on the index of socioeconomic status. For Australia the size of this measure of the association was 32 scale points, very close to the OECD average of 34 scale points. (The standard deviation was 100 scale points). The countries showing the weakest effect of socioeconomic backgro und was Korea (14 points) and the strongest was for Germany (45 point s). The effects for the United Kingdom, the United States, New Zealand and Canada were 38, 34, 32 and 26 points respectively. In terms of the socioec onomic distribution of achievement, Australia is around the international average and not a leader in terms of equality of outcomes (OECD, 2001b). Simila r results were reported in the TIMSS studies for mathematics and science among middle primary and junior secondary students (Lokan et al., 1996:40; 1 997:44) (Note ) The influence of socioeconomic background on educat ional outcomes is declining in many OECD countries (Rijken, 1999:51-7 8; Sieben, 2001:33-55). In Australia, there is evidence that the influence of socioeconomic background on early school leaving, participation in Year 12 and higher education is declining over-time (Fullarton et al., 2003; Marks & McMillan 2003; McMillan & Marks, 2003).Socioeconomic background is often considered the mo st important influence on educational outcomes and an important element in th e funding of schools. However, its influence on early school leaving, Yea r 12 completion and University entrance performance is considerably sma ller than that of
13 of 26 achievement in literacy and achievement (Marks & Fl eming, 1998a; Marks et al., 2000; Marks et al., 2001). Both the Australian and international PISA reports demonstrate large variation in achievement scores a mong students with the same socioeconomic backgrounds (Lokan et al., 2001: 163-168; OECD, 2001b:185). This variation reflects the lack of a s trong association between socioeconomic background and achievement.From a policy perspective, it is important to furth er reduce the impact of socioeconomic background. There are countries where the impact of socioeconomic background is considerably weaker tha t it is in Australia. A general rather than streamed curriculum is helpful since school systems characterised by tracking or streaming often (but n ot always) show stronger effects of socioeconomic background (OECD, 2001b:19 5-196). However, a more effective policy focus would be to focus on ed ucational performance rather socioeconomic background since poor performa nce is the primary concern and improving the performance of low perfor ming students will necessary reduce socioeconomic inequality. Furtherm ore, such a focus avoids the predictable criticisms of any measure of socioe conomic background used to fund disadvantaged schools.Ethnic and Indigenous MinoritiesAlthough formulas for school funding often include the proportion of non-English speaking students, these students most often exhibi t superior educational outcomes (Marks & Fleming, 1999; Marks et al., 2000 ; Marks et al., 2001). Differences in middle-secondary school achievement are often minimal so it appears that cultural factors are responsible for their higher performance during the last two years of school. However, at th e primary school level students with language backgrounds other than Engli sh tend to show lower mean achievement levels than students with an Engli sh language background (Lokan et al., 1997:173-178).Indigenous students show much poorer educational ou tcomes than non-indigenous students. The difference between Ind igenous and non-Indigenous students in the PISA assessments of reading, mathematics and science was very large at around 0.8 standard devia tions (Note 7) Similar results were found for mathematical and scientific literacy (Lokan et al., 2001:20-33).The educational participation of Indigenous student s is much lower than that of non-indigenous students. In 1997 the Year 8 to Year 12 school retention rate for Indigenous students was 31 per cent compared to 73 per cent for non-Indigenous students (Long et al., 1999b:37). In 1996 approximately 11 per cent of non-Indigenous 20 to 24 year-olds held a un iversity degree compared to only 2 per cent of 20 to 24 year-old Indigenous Aus tralians (Long et al., 1999b:76). Similarly, the more select group of Indi genous students who compete for a tertiary entrance score show scores, on average, 11 points lower than non-Indigenous students (Marks et al., 2001). Indigenous students remain the most educationally disadvantaged group of young Australians. School Sector
14 of 26 Over the past two decades there has been a shift of school enrolments from the government to the non-government sector. In 1984, 7 5 per cent of school students were enrolled in government schools. By 20 00 the percentage of students in government schools was down to 69 per c ent (ABS, 2001:34). The most current data shows that the percentage of stud ents in government schools is smaller in the secondary sector (64 per cent) th an the primary years (73 per cent) and smaller again for the final year of secon dary school (61 per cent). Across all levels of schooling in 2000, 20 per cent of students were in Catholic schools and 11 per cent were in other non-governmen t schools. For Catholic schools there was little difference in the enrolmen t share at primary (19 per cent) and secondary (21 per cent) levels. For other non-government schools the enrolment share for the secondary school years (15 per cent) was almost double that for the primary school years (8 per cen t). For the final year of secondary school 22 per cent of students were in Ca tholic schools and 17 per cent were in other non-government schools.The shift of enrolments from the government to nongovernment schools poses a significant challenge to the organisation of scho oling in Australia. Schooling in Australia has been largely through comprehensive go vernment schools that have a broadly representative intake, with non-gove rnment schools providing for a smaller number of students. If the current trend continues government secondary schools may come to be regarded as provid ing for a little more than half of the student population. The issue is compou nded because the shift of enrolments is probably not uniformly spread across the social distribution in the community. Some organisational responses such as, t he Schools of the Future program in Victoria and Partnerships 21 in South Australia, have attempted to respond to this challenge by devolving more authori ty to individual schools and shortening the lines of authority for operational d ecisions. In these respects government schools would operate like non-governmen t schools. Neither program has operated for a sufficient time for a co nsidered evaluation of their long-term impact.One of the most dramatic changes in Year 12 partici pation is the substantial decline in school sector differences. In the early 1980s only 30 per cent of those who had attended government schools participated in Year 12 compared to 44 per cent of Catholic school students and 88 per cen t of independent school students. By the late 1990s, 71 per cent of governm ent school students participated in Year 12 whereas the participation r ate of independent school students had remained the same. Participation among students from Catholic schools had become almost as high as that of studen ts from independent schools (Fullarton et al., 2003).School sector has a substantial impact on tertiary entrance performance. On average, students attending independent schools hav e higher mean ENTER (Equivalent National Tertiary Entrance Rank) scores than students attending Catholic schools, who in turn have higher ENTER scores than students attending government schools. Differences in ENTER scores between students attending independent and government schools are re duced by nearly 50 per cent after controlling for differences in Year 9 ac hievement and the socioeconomic backgrounds of students. Differences in ENTER scores between
15 of 26 students attending Catholic and government schools are reduced by about 20 per cent after controlling for prior achievement an d the socioeconomic backgrounds of students. Achievement growth in the final years of school is much greater among non-government than government s chool students (Marks et al., 2001). So, on average, students attending n on-government schools perform better than government school students even when taking into account the socioeconomic and academic mix of students.The interpretation of these differences is not clea r. It is possible that many independent schools have a more defined focus on un iversity entrance than many government schools and do not need to spread t heir efforts over such a diverse range of endeavours (including a wider rang e of vocational courses). In general, research on school effectiveness has point ed to the importance of the academic environment of a school for growth in stud ent performance (see below). The difference between government and indep endent schools in tertiary entrance performance could be attributed partly to differences in resource levels but that seems less likely to provide an explanatio n for differences between Catholic and government schools. It is also possibl e that because of greater flexibility in recruitment strategies, coupled with the availability of financial resources, non-government schools are able to attra ct and retain very capable teachers. Rowe (1999) has argued on the basis of da ta from one state that there are important differences between subject are as within schools and between classes within schools. He interprets this as an indication of the importance of individual teachers.School Differences in PerformanceMost studies of educational outcomes identify diffe rences among schools in student performance. Those differences are, at leas t, partly associated with differences in the social and academic mix of the s tudent population in each school. The extent to which there are differences a mong schools indicates the effect of national patterns of school organisation and the effect of differences in the effectiveness of schools. Where school systems are selective, where residential areas are socially stratified, or where schools are differentially effective, between-school differences will be large r. Technically, the extent of these differences can be represented as the percent age of the variation in student achievement that can be explained by the va riation in the average achievement for each school. If all the students in each school achieved the same score but there were differences between schoo ls then 100 per cent of the variation in student achievement could be attri buted to the school attended. If all students achieved different scores but all t he schools had the same average score then none of the variation in student achievement could be attributed to the school attended.One of the issues investigated in the data from PIS A was the extent to which there were variations between schools in student pe rformance (OECD, 2001b:60-67) (Note 8) This was indicated by the percentage of the varia tion in student scores that could be attributed to differen ces between schools and the percentage that could be attributed to differences among students within schools. On average, a little more than one third ( 36 per cent) of the variance in student achievement was attributable to between-sch ool differences across
16 of 26 OECD countries. Belgium, Germany and Austria, each of which have selective school systems, have around 70 per cent of the vari ance in reading achievement attributable to between-school differen ces. In Italy, the Czech Republic and Greece the figure is around 50 per cen t. At the other end of the scale are Finland, Sweden and Iceland where the per centage of variance attributable to between school differences is less than ten per cent. For Australia, approximately 20 per cent of the varianc e in reading literacy is associated with differences among schools. This fig ure is comparable with that for the United Kingdom and New Zealand, lower than the United States (35 per cent) and just a little higher than Canada (OECD, 2 001b). Furthermore, school differences are considerably smaller once differenc es between schools in the academic mix of students are taken into account. In general terms it can be concluded that in Australia efforts to improve stud ent performance need to be directed to less-successful students within schools rather than to improving particular schools.School Influences on OutcomesDifferences between schools are largely the result of differences between schools in the social and academic mix of students. Once such differences are taken into account there is only a minority of scho ols, in which the school itself is a significant independent influence on student p erformance. In Australia, only 17 per cent of schools had an independent influence on Year 12 participation after taking into account state or territory, and p rior student achievement. This figure declined to 12 per cent after adding school sector to the analysis (Marks et al., 2000). Similarly, only 17 per cent of schoo ls had significant effects on tertiary entrance performance after controlling for student intake (prior achievement and socioeconomic background). After ta king into account other student factors this figure declined to 11 per cent (Marks et al., 2001). This means that only in a minority of schools does the i ndividual school increase or decrease student performance to a significant exten t, net of other factors. Although only a minority of schools significantly l ift school performance, there has been much research on the characteristics of e ffective schools. That is, schools that lift student performance above what is expected given the schools social and academic intake. After reviewing the int ernational literature, Kreft (1993) concluded that more effective schools have: a higher level of parental involvement with the school; higher levels of expec tations among students; frequent monitoring of student performance; greater involvement by parents and teachers; an orderly school atmosphere; and strict discipline. In a review of the US research on unusually effective schools, Levine (1992) identified a large number of correlates including mastery of central l earning skills, students having a sense of efficacy, school resources and support f or teachers. A more recent review of the literature concluded that research on effective schools identifies five factors: strong educational leadership; emphas is on acquiring basic skills; an orderly and secure environment; high expectation s of student achievement; and frequent assessment of student progress (Scheer ens & Bosker, 1997:146). After performing meta-analyses on factors often und erstood as important to school effectiveness, (Scheerens & Bosker, 1997, pp 237-238) conclude that the most powerful factors operate at the classroom level. Hill and Rowe (1996)
17 of 26 reached the same conclusion from the analysis of da ta on student progress through Victorian primary schools. Differences amon g classrooms within schools were greater than differences among schools Some of these differences may be partly attributable to the clust ering of students of similar abilities in the same classrooms but it does appear evident that differences between classrooms are important and that it is wha t individual teachers do that is crucial for student learning.Despite the general factors that have been identifi ed as characteristics of effective schools, there is little that is specific It is difficult to conclude which particular factors (and therefore policy initiative s) make for effective schools. Many inter-correlated factors are canvassed as impo rtant influences, which may vary between school systems. It may well be that va riable-focussed modelling is appropriate for establishing the extent to which sc hools vary and for identifying schools that appear to be effective, but case centr ed forms of analysis (both quantitative and qualitative) are needed to elucida te the ways in which factors cluster to influence outcomes.Providing additional resources to schools, and redu cing class size, are two related and much debated ways of improving educatio nal outcomes. One approach to the investigation of these issues has b een through the econometric analyses of education production functions that mak e use of the natural variation of class size across schools and models s tudent achievement in relation to class size, controlling for student cha racteristics and prior achievement. It is crucial to control for prior ach ievement because in many school systems low-achieving students are often all ocated to smaller classes. Greenwald, Hedges and Laine (1996) applied meta-ana lytic techniques to a series of studies and concluded that increased reso urces were associated with improved student outcomes. This analysis was import ant because it differed from the conclusions of Hanushek (1989), who found little or no effects of school resources. However, even though Greenwald et al (1996) concluded that there was an effect of resources, the magnitud e of the effect was not large. A number of experimental studies of class size and achievement have been reported. Some 20 years ago, Glass and Smith (1979) conducted a meta-analysis of laboratory experiments using instr uctional groups of different size. They concluded that reduced class size could be expected to produce increased student achievement but that benefits are only evident when the class size is reduced below 20. In the United State s policy had been strongly influenced by the results of the Tennessee class-si ze experiment (Finn & Achilles, 1999). In 79 schools, students and teache rs in the kindergarten year were randomly assigned to different class sizes fro m kindergarten through to Grade 3. Small classes contained between 13 and 17 students and large classes contained between 22 and 26 students. There has been a consistent finding that students in the smaller classes showed larger gains in reading and mathematics achievement. The magnitude of the effec t in one year has been variously estimated as 0.21 (Word et al., 1994) or 0.15 standard deviations (Goldstein & Blatchford, 1998). As part of a follow up it was concluded that the benefits of the smaller classes lasted through to t he later years of primary school but with an attenuated magnitude (Nye et al. 2001).
18 of 26 Although the results of the Tennessee experiment ha ve provided support for the proposition that reduced class size produces enhanc ed learning outcomes, the conclusions for practice are not unequivocal. Prais (1996) argues that for a given investment alternative actions such as time f or teacher professional development, devoting resources to students with le arning difficulties, developing better curriculum resources, and varying the time students spend in groups of different size should be seen as better u se of resources. The extent to which the results of this study of the early pri mary years can be generalised to later stages of schooling is untested. In additi on, analysis of the costs of class size reduction programs in the United States have i dentified issues associated with the cost of physical resources (such as rooms) and maintenance of teacher quality when there is a rapid expansion of teacher numbers (Brewer et al., 1999). These issues impact on both the cost and eff ectiveness of class size reduction initiatives in school systems. Therefore, the narrow emphasis of class size as a way to improve school performance needs t o put in the context of its small effects and the possibility that there may be more effective ways to improve student performance.DiscussionIn Australia, as in many other countries debates ab out the education system have generally not engaged with the empirical evide nce. Governments have pursued easy policy options, such as increasing the levels of school completion, expanding vocational education, and reducing class sizes, which are politically less contentious, supported by various interests gr oups and simple enough to be understood by the general public. The empirical evidence on the benefits of such policies is, at best, equivocal. Furthermore, they are unlikely to substantially benefit future cohorts of young peopl e. More difficult issues such as reducing socioeconomic inequalities in education improving indigenous education and reducing differences in student perfo rmance between government and non-government schools are put into the too hard basket. There are a variety of contentious issues that are relevant to many education systems. Should educational outcomes only reflect a bility and effort, or are concepts such as ability, merit, or even effor t too contentious to be considered? Should all students complete school or is it more important for school leavers to gain secure full-time employment? What policies should be implemented to reduce socioeconomic inequalities in education? Should indigenous and minority students have similar educa tional outcomes to non-indigenous students or should higher priority b e given to a culturally appropriate education. Should policies be implement ed to improve the educational and labour market outcomes of boys? The se are difficult questions and can only be resolved by constructive evidence-b ased debate. Such debate may lead to formulation of effective policies which improve student outcomes and reduce socially based inequalities in education NotesThe views expressed in this article are not necessa rily those of the Australian Council for Educational Research or the Melbourne I nstitute for Economic and
19 of 26 Social Research. As a proportion of Gross Domestic Product, public a nd private expenditure on education in Australia (at 5.46 per cent) is sli ghtly below the OECD mean (5.75 per cent). Similarly, public expenditure on education (as a proportion of GDP) is lower in Australia (4.34 per cent) than the OECD mean (4.64) (OECD, 2001a:80). Public expenditure on tertiary education as a proportion of GDP in Australia (1.09) is sligh tly above the OECD mean (1.06 per cent) (OECD, 2001a:81). Expenditure per primary and secondary school student in Australia is the same a s the OECD mean. Expenditure per tertiary student in Australia is hi gher than the OECD mean (OECD, 2001a:59). 1. In this study, graduates were comprised of predomin antly university graduates, although a smaller group of TAFE diploma graduates was also included. 2. Reform of university funding is a difficult issue. One argument is that increases in participation should be funded through taxation. Since Australia collects a smaller proportion of GDP in t axation than many other OECD countries, then governments should simply incr ease taxes. However, there are few taxation options for Austral ian governments. The top marginal tax rate of 48 per cent starting at $6 0,000, is a high tax regime compared to many other industrialized countr ies. Australia has just emerged from a difficult debate about indirect taxe s, so it is very unlikely that the GST will be extended or increased. Many of the European countries, which collect larger proportions of tax, do so because of indirect taxes. Many of the options for increasing tax reven ues such as increasing fuel taxes, taxes on the sale of the family home, a nd death duties, have their own economic, social, and political costs. Fu rthermore, they may not attract sufficient revenue. 3. A similar decline in apprenticeships was found in t he Youth in Transition cohorts. The participation rate by age 19 declined from 18 per cent in the early 1980s to 14 per cent in the mid 1990s (Long e t al., 1999a:8). A decline is also evident in a more recent LSAY cohor t (who had been in Year 9 in 1995). By 2000 when the modal age of the age was 19, 13 per cent had participated in an apprenticeship. 4. An international index based on parental occupation s was used to measure socioeconomic status. 5. The relationship was a little stronger at junior se condary than middle primary level. 6. Based on a sample of nearly 500 indigenous students A total of 192 students in the main sample identified themselves a s of Indigenous origin. The study included an additional 300 indigenous stu dents from the same schools as the main sample as an additional sample. 7. Within Australia, there were relatively few signifi cant differences among jurisdictions (Lokan et al, 2001). In reading liter acy, the performance of students from the ACT was significantly better than that of students from Queensland, Victoria, Tasmania and the Northern Ter ritory. In mathematical literacy, there were few differences a mong jurisdictions, but in scientific literacy both the ACT and Western Aus tralia had higher performance levels than several other states. 8.
20 of 26 ReferencesABS. (1984-2001). Schools Australia (Catalogue no 4221.0) Canberra: Australian Bureau of Statistics. ABS. (2002). Schools Australia (Catalogue no 4221.0) Canberra: Australian Bureau of Statistics. Afrassa, T., & Keeves, J. P. (1999). Changes in Stu dent Mathematics Achievement in Lower Secondary School Schools Over Time. International Education Journal 1 (1), 1-21. Baker, D. P., & Jones, D. P. (1993). Creating Gende r Equality: Cross-National Gender Stratification and Mathematical Performance. Sociology of Education 66 91-103. Borland, J. (2002). New Estimates of the Private Rate of Return to Univ ersity Education in Australia (Melbourne Institute Working Paper No. 14/2002): Me lbourne Institute of Applied Economic and Social Research, The University of Melbourne, Victo ria 3010 Australia. Bradley, K., & Raminez, F. O. (1996). World Polity and Gender Parity: Women's Share of Higher Education, 1965-1985. Research in Sociology of Education and Socializatio n 11 63-91. Brewer, D., Krop, C., Gill, B., & Reichardt, R. (19 99). Estimating the Cost of National Class Size Reductions under Different Policy Alternatives. Educational Evaluation and Policy Analysis 21 (2), 179-192. Collins, C., Kenway, J., & McLeod, J. (2000). Factors Influencing the Educational Performance of Males and Females in School and their Initial Desti nations after Leaving School. Canberra: DETYA. Comber, L. C., & Keeves, J. P. (1973). Science Education in Nineteen Countries: Internatio nal Studies in Evaluation I. New York: John Wiley & Sons. DETYA. (2000). Students 1999: Selected Higher Education Statistics DETYA No. 6444. Canberra: Department of Education, Training & Youth Affairs. Dockery, A. M., & Norris, K. (1996). The 'Rewards' for Apprenticeship Training in Australia. Australian Bulletin of Labour 22 (2), 109-125. Finn, J., & Achilles, C. (1999). Tennessee's Class Size Study: Findings, Implications and Misconceptions. Educational Evaluation and Policy Analysis 21 (2), 97-110. Fullarton, S. (2001). Vet In Schools: Participation and Pathways. (LSAY Research Report No. 21). Melbourne: Australian Council for Educational Resea rch. Fullarton, S., Walker, M., Ainley, J., & Hillman, K J. (2003). Patterns of Participation in Year 12 (LSAY Research Report No. 33). Melbourne: ACER. Glass, G., & Smith, M. L. (1979). Meta Analysis of Research on Class Size and Achievement. Educational Evaluation and Policy Analysis 1 2-6. Goldstein, H., & Blatchford, P. (1998). Class Size and Educational Achievement: A Review of Methodology with Reference to the Study Design. British Educational Research Journal 24 (3), 255-269. Greenwald, R., Hedges, L., & Laine, R. (1996). The Effect of School Resources on Student Achievement. Review of Educational Research 66 361-396. Hanushek, E. (1989). The Impact of Differential Sch ool Expenditures. Educational Researcher 18 (4), 45-65. Hill, P., & Rowe, K. J. (1996). Multilevel Modellin g in School Effectiveness Research. School Effectiveness and School Improvement 7 (1), 1-34. Kirsh, I. S., Jungeblut, A., Jenkins, L., & Kolstad A. (1993). Adult Literacy in America. A First Look at the Results of the National Adult Literacy Survey Washington: National Center for Education
21 of 26 Statistics. Kreft, I. G. G. (1993). Using Multilevel Analysis t o Assess School Effectiveness: A Study of Dutch Secondary Schools. Sociology of Education 66 (104-129.). Lamb, S. (2001). The Pathways from School to Further Study and Work for Australian Graduates (LSAY Research Report No. 19). Melbourne: Australia n Council for Educational Research. Lamb, S., Dwyer, P., & Wyn, J. (2000). Non-Completion of School in Australia: The Changing Patterns of Participation and Outcomes (LSAY Research Report No. 16). Melbourne: Australi an Council for Educational Research. Lamb, S., Long, M., & Malley, J. (1998). Access and Equity in Vocational Education and Train ing. (ACER Research Monograph No. 55). Melbourne: Austra lian Council for Educational Research. Lamb, S., & McKenzie, P. (2001). Patterns of Success and Failure in the Transition f rom School to Work in Australia (LSAY Research Report No. 18). Melbourne: Australi an Council for Educational Research. Levine, D. U. (1992). An Interpretative Review of U S Research and Practice Dealing with Unusually Effective Schools. In D. Reynolds & P. Cuttance (Ed s.), School Effectiveness: Research Policy and Practice. London: Cassell. Lokan, J., Ford, P., & Greenwood, L. (1996). Maths and Science on the Line: Australian Junior Secondary Students Performance in the Third Interna tional Mathematics and Science Study. (Vol. 1). Melbourne: Australian Council for Educati onal Research. Lokan, J., Ford, P., & Greenwood, L. (1997). Maths and Science on the Line: Australian Middle Primary Students Performance in the Third Internati onal Mathematics and Science Study. Melbourne: Australian Council for Educational Resea rch. Lokan, J., Greenwood, L., & Cresswell, J. (2001). 15-up and Counting, Reading, Writing, Reasoning : How Literature are Australia's Students? The PISA 2 000 Survey of Student's Reading, Mathematical and Scientific Literacy Skills. Melbourne: Australian Council for Educational Research. Long, M., Carpenter, P., & Hayden, M. (1999a). Participation in Education and Training. 1980-1994. (LSAY Research Report No. 13). Melbourne: Australia n Council for Educational Research. Long, M., Frigo, T., & Batten, M. (1999b). The School to Work Transition of Indigenous Austral ians: A Review of the Literature and Statistical Analysis. Canberra: Department of Education, Training and Youth Affairs. Long, M., McKenzie, P., & Sturman, A. (1996). Labour Market Participation and Income Consequences in TAFE (ACER Research Monograph Series No. 49). Melbourne : Australian Council for Educational Research. MacCann, R. (1995). Sex Differences at the NSW High er School Certificate after Adjustment for the Effects of Differential Selection. Australian Journal of Educational Research 39 (2), 163-188. Malley, J., Frigo, T., Robinson, L., & Hawke, G. (2 001). The Quest for a Working Blueprint: Vocational Education and Training in Australian Sec ondary Schools. Leabrook, SA: National Centre for Vocational Education and Training (NCVER ). Marks, G. N., & Ainley, J. (1997). Reading Comprehension and Numeracy among Junior Sec ondary School Students in Australia (LSAY Research Report No. 3). Melbourne: Australia n Council for Educational Research. Marks, G. N., & Fleming, N. (1998a). Early School Leaving in Australia: Findings from t he 1995 Year 9 LSAY Cohort. (LSAY Research Report No. 11). Melbourne: Australi an Council for Educational Research. Marks, G. N., & Fleming, N. (1998b). Factors Influencing Youth Unemployment in Australia : 1980-1994. (LSAY Research Report No. 7). Melbourne: Australia n Council for Educational Research.
22 of 26 Marks, G. N., & Fleming, N. (1998c). Youth Earnings in Australia: 1980-1994. (LSAY Research Report No. 8). Melbourne: Australian Council for Ed ucational Research. Marks, G. N., & Fleming, N. (1999). Early School Leaving in Australia: Findings from th e 1995 Year 9 LSAY Cohort (LSAY Research Report No. 11). Melbourne: Australi an Council for Educational Research. Marks, G. N., Fleming, N., Long, M., & McMillan, J. (2000). Patterns of Participation in Year 12 and Higher Education in Australia: Trends and Issues (LSAY Research Report No. 17). Melbourne: Australian Council for Educational Research. Marks, G. N., Hillman, K., & Beavis, A. (2003). Dynamics of the Australian Youth Labour Market: The 1975 Cohort, 1996-2000 (LSAY Research Report No. 34). Melbourne: Australi an Council for Educational Research. Marks, G. N., & McMillan, J. (2003). Declining Ineq uality? The Changing Impact of Socioeconomic Background and Ability on Education in Australia. British Journal of Sociology 54 (4), 453-471. Marks, G. N., McMillan, J., & Hillman, K. J. (2001) Tertiary Entrance Performance: The Role of Student Background and School Factors (LSAY Research Report No. 22). Melbourne: Australi an Council for Educational Research. Marks, G. N. (Forthcoming). The Transition of Youth to Full-Time Employment Melbourne: Australian Council for Educational Research. Martin, M. O., Mullis, I. V. S., Gonzalez, E. J., G regory, K. D., Smith, T., A., Chrostowski, S. J., e t al. (2000). TIMSS 1999 International Science Report. Findings f rom the IEA's Repeat of the Third International Mathematics Study at the Eight Grade Boston: International Study Center Lynch School of Education, Boston College. Martin, Y. M., MacLachlan, M., & Karmel, T. (2001). Undergraduate Completion Rates: An Update. Occasional Paper 01/F Canberra: Higher Education Division, Department o f Education, Science and Training. McMillan, J., & Marks, G. N. (2003). School Leavers in Australia. Profiles and Pathways (LSAY Research Report No. 31). Melbourne: Australian Coun cil for Educational Research. Mullis, I. V. S., Martin, M. O., Gonzalez, E. J., G regory, K. D., Garden, R. A., O'Connor, K. A., et a l. (2000). TIMSS 1999 International Mathematics Report. Findin gs from the IEA's Repeat of the Third International Mathematics Study at the Eight Grade Boston: International Study Center Lynch School of Education, Boston College. NCVER. (1998). Australian VET: TAFE Graduate Destination Survey 19 97 National Report. : Leabrook, SA: National Centre for Vocational Educat ion and Training (NCVER). Nevile, J. W., & Saunders, P. (1998). Globalisation and the Return to Education in Australia. The Economic Record 74 (226), 279-285. NSW UAC. (1998). New South Wales Vice Chancellors Conference. Report on the Scaling of the 1998 NSW Higher School Certificate Sydney. Nye, B., Hedges, L., & Konstantopoulos, S. (2001). The Long Term Effects of Small Classes in Early Grades: Lasting Benefits in Mathematics Achievement at Grade 9. Journal of Experimental Education 69 (3), 245-258. OECD. (1997). Education at a Glance: OECD Indicators 1997. Paris: Organisation for Economic Co-Operation and Development. OECD. (1998). Education at a Glance: OECD Indicators 1998. Paris: Organisation for Economic Co-Operation and Development. OECD. (2000a). Education at a Glance: OECD Indicators 2000. Paris: Organisation for Economic Co-Operation and Development. OECD. (2000b). From Initial Education to Working Life. Making Tran sitions Work Paris:
23 of 26 Organisation for Economic Co-Operation and Developm ent. OECD. (2001a). Education at a Glance: OECD Indicators 2001. Paris: Organisation for Economic Co-Operation and Development. OECD. (2001b). Knowledge and Skills for Life. First Results from t he OECD Programme for International Student Assessment. Paris: Organisation for Economic Co-Operation and Development. Prais, S. (1996). Class Size and Learning: The Tenn essee Experiment What Follows? Oxford Review of Education 22 (4), 399-415. Rijken, S. (1999). Educational Expansion and Status Attainment. A Cros s-National and Over-time Comparison Netherlands: Inter-University Center for Social S cience Theory and Methodology. Rosier, M. J. (1980). Changes in Secondary School Mathematics in Australi a, 1964-1978 Hawthorn, Victoria: Australian Council for Educational Resear ch. Rosier, M. J., & Keeves, J. P. (1991). The IEA Study in Science I.:Science Education and C urricula in Twenty-three Countries Oxford: Pergamon. Rothman, S. (2002). Achievement in Literacy and Numeracy by Australian 14 Year-olds, 1975-1998 (LSAY Research Report No. 29). Melbourne: Australia n Council for Educational Research. Rowe, K. (1999). VCE Data Project (1994-1999): Concepts, Issues, Dir ections, and Specifications. A Research and Evaluation Project Conducted for the Victorian Board of Studies. Melbourne: Centre for Applied Educational Research, University of Melbourne. Ryan, P. (2001). From School-to-Work: A Cross-Natio nal Perspective. Journal of Economic Literature 39 (March), 34-92. Scheerens, J., & Bosker, R. (1997). The Foundations of Educational Effectiveness Oxford: Pergamon. Sieben, I. (2001). Sibling Similarities and Social Stratification: The Impact of Family Background across Countries and Cohorts. The Netherlands: Interuniversity Center for Social Science Theory and Methodology. Urban, M., Jones, E., Smith, G., Evans, C., MacLach lan, M., & Karmel, T. (1999). Completions: Undergraduate Academic Outcomes for 1992 Commencing Students (Occasional Paper 99/G). Canberra: Higher Education Division, Department of Education, Training and Youth Affairs. VTAC. (1998-1999). VTAC Annual Statistics Melbourne: Victorian Tertiary Admissions Centre. Word, E., Johnston, J., Bain, H., Fulton, B., Zacha rias, J., Achilles, C., et al. (1994). The State of Tennessee's Student Teacher Achievement Project: Te chnical Report. Nashville: Tennessee State Department of Education.About the AuthorsGary N. MarksGary N. Marks is a Principal Research Fellow at the Australian Council for Educational Research. Since 1996 he has authored a substantial number of reports and articles based on the Longitudinal Stud ies of Australian Youth Project, a study focusing on the transition from sc hool to work. He is also involved in a longitudinal study of adults and is c urrently working on wealth in Australia and its influence on educational outcomes Both longitudinal studies have a substantial policy focus.Julie McMillanJulie McMillan is a Research Fellow at the Australi an Council for Educational Research, where she works on the Longitudinal Surve ys of Australian Youth
24 of 26 project. Her current research focuses on young peop le's educational and labour market pathways and outcomes. Other research intere sts include the development of measures of socioeconomic disadvanta ge among school students, higher education students, and the genera l population. John Ainley +61 3 9277 5507 (Voice)+61 3 9277 5500 (Fax)Email: firstname.lastname@example.orgJohn Ainley is Research Director of National and In ternational Surveys at the Australian Council for Educational Research. Over t he past two decades he has directed a range of policy-oriented research studie s for state and federal education authorities. He directed a five-year long itudinal study of Progress through High School, conducted national surveys of subject choice and has written research reports from the Longitudinal Surv eys of Australian Youth project. Most recently he completed a five-year lon gitudinal study of children's development of reading proficiency for the Catholic Education Commission in Victoria. The World Wide Web address for the Education Policy Analysis Archives is epaa.asu.edu Editor: Gene V Glass, Arizona State UniversityProduction Assistant: Chris Murrell, Arizona State University General questions about appropriateness of topics o r particular articles may be addressed to the Editor, Gene V Glass, email@example.com or reach him at College of Education, Arizona State Un iversity, Tempe, AZ 85287-2411. The Commentary Editor is Casey D. Cobb: firstname.lastname@example.org .EPAA Editorial Board Michael W. Apple University of Wisconsin David C. Berliner Arizona State University Greg Camilli Rutgers University Linda Darling-Hammond Stanford University Sherman Dorn University of South Florida Mark E. Fetler California Commission on TeacherCredentialing Gustavo E. Fischman Arizona State Univeristy Richard Garlikov Birmingham, Alabama Thomas F. Green Syracuse University Aimee Howley Ohio University Craig B. Howley Appalachia Educational Laboratory William Hunter University of Ontario Institute of
25 of 26 Technology Patricia Fey Jarvis Seattle, Washington Daniel Kalls Ume University Benjamin Levin University of Manitoba Thomas Mauhs-Pugh Green Mountain College Les McLean University of Toronto Heinrich Mintrop University of California, Los Angeles Michele Moses Arizona State University Gary Orfield Harvard University Anthony G. Rud Jr. Purdue University Jay Paredes Scribner University of Missouri Michael Scriven University of Auckland Lorrie A. Shepard University of Colorado, Boulder Robert E. Stake University of IllinoisUC Kevin Welner University of Colorado, Boulder Terrence G. Wiley Arizona State University John Willinsky University of British ColumbiaEPAA Spanish and Portuguese Language Editorial BoardAssociate Editors for Spanish & Portuguese Gustavo E. Fischman Arizona State Universityfischman@asu.eduPablo Gentili Laboratrio de Polticas Pblicas Universidade do Estado do Rio de Janeiro email@example.comFounding Associate Editor for Spanish Language (199 8-2003) Roberto Rodrguez Gmez Universidad Nacional Autnoma de Mxico Adrin Acosta (Mxico) Universidad de Guadalajaraadrianacosta@compuserve.com J. Flix Angulo Rasco (Spain) Universidad de Cdizfelix.firstname.lastname@example.org Teresa Bracho (Mxico) Centro de Investigacin y DocenciaEconmica-CIDEbracho dis1.cide.mx Alejandro Canales (Mxico) Universidad Nacional Autnoma deMxicocanalesa@servidor.unam.mx Ursula Casanova (U.S.A.) Arizona State Universitycasanova@asu.edu Jos Contreras Domingo Universitat de Barcelona Jose.Contreras@doe.d5.ub.es
26 of 26 Erwin Epstein (U.S.A.) Loyola University of ChicagoEepstein@luc.edu Josu Gonzlez (U.S.A.) Arizona State Universityjosue@asu.edu Rollin Kent (Mxico) Universidad Autnoma de Puebla email@example.com Mara Beatriz Luce (Brazil) Universidad Federal de Rio Grande do Sul-UFRGSlucemb@orion.ufrgs.br Javier Mendoza Rojas (Mxico)Universidad Nacional Autnoma deMxicojaviermr@servidor.unam.mx Marcela Mollis (Argentina)Universidad de Buenos Airesmmollis@filo.uba.ar Humberto Muoz Garca (Mxico) Universidad Nacional Autnoma deMxicohumberto@servidor.unam.mx Angel Ignacio Prez Gmez (Spain)Universidad de Mlagaaiperez@uma.es DanielSchugurensky (Argentina-Canad) OISE/UT, Canadadschugurensky@oise.utoronto.ca Simon Schwartzman (Brazil) American Institutes forResesarchBrazil (AIRBrasil) firstname.lastname@example.org Jurjo Torres Santom (Spain) Universidad de A Coruajurjo@udc.es Carlos Alberto Torres (U.S.A.) University of California, Los Angelestorres@gseisucla.edu EPAA is published by the Education Policy Studies Laboratory, Arizona State University
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam a22 u 4500
controlfield tag 008 c20049999azu 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E11-00366
Educational policy analysis archives.
n Vol. 12, no. 17 (April 20, 2004).
Tempe, Ariz. :
b Arizona State University ;
Tampa, Fla. :
University of South Florida.
c April 20, 2004
Policy Issues for Australias Education Systems: Evidence from International and Australian Research / Gary N. Marks, Julie McMillan [and] John Ainley.
Arizona State University.
University of South Florida.
t Education Policy Analysis Archives (EPAA)
xml version 1.0 encoding UTF-8 standalone no
mods:mods xmlns:mods http:www.loc.govmodsv3 xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govmodsv3mods-3-1.xsd
mods:relatedItem type host
mods:identifier issn 1068-2341mods:part
mods:detail volume mods:number 12issue 17series Year mods:caption 20042004Month April4Day 2020mods:originInfo mods:dateIssued iso8601 2004-04-20