Access the full text.
Sign up today, get DeepDyve free for 14 days.
This article reports an analysis of the relationship between women’s increased participation in higher education and other recent social changes over the last four decades. To date, women’s increased involvement in higher education has been studied as either a force for or a consequence of other sociocultural changes. Drawing on data from key international data sets and with a focus on a range of Organisation for Economic Co-Operation and Development (OECD) countries, this article details an exploratory factor analysis of women’s higher education participation and a range of other variables identified as indicators of or mediators for social change. This analysis reveals the existence of four underlying factors showing the structural interrelationship between the variables. Keywords globalization, higher education participation, gender, sociocultural variables, factor analysis Wotipka, 2001). In other words, women undertaking higher Introduction education comprise an increasingly larger proportion of their Over the past four decades, the number of higher education female peers as well as being a higher proportion of the stu- students has increased globally, and universities have moved dent body. Just what this development means in terms of its from elite to mass and then to global higher education in social effects is as yet not clear. This study presents one most developed countries (Taylor, 2003, p. 813). A particular attempt to identify factors relating to women’s participation feature of this development has been the increasing rate of in higher education and other social changes and to investi- women’s participation in higher education. In the vast major- gate any potential interrelationships. ity of developed countries as well as those in transition, A useful indicator for examining the participation levels women now comprise the majority of tertiary students of women in higher education is defined by United Nations (Leathwood & Read, 2008). For example, based on data Development Program (UNDP) as the Gross Tertiary derived from United Nations Educational, Scientific and Enrollment Ratio (GTER) (UNDP, 2008). GTER reflects Cultural Organization (UNESCO; 2016), over the last 40 changes in enrollment numbers and population size by divid- years, the proportion of women in higher education has ing the total tertiary school enrollment, regardless of age, by increased from 39% to 56.4% in North America and Western tertiary school-aged population. Europe, 34% to 51.3% in East Asia and Pacific, 35% to The current study looked at the case of a range of OECD 56.3% in Latin America and the Caribbean, 21% to 32.8% in countries as an example of the increased participation of South and West Asia, and 22% to 45.6% in Sub-Saharan women in higher education as these countries report the Africa. However, in Sub-Saharan Africa, and South and West highest level of GTER a feature which may have clear Asia, men remain in the majority, although there are some implications for developing countries. The statistics for notable differences between countries and within regions. each indicator have been collected over the past four decades Importantly for this study, the proportion of women’s partici- based on the initial needs of the larger project so that pation in higher education has also risen from 33% in 1970 to 54% in 2012 in Organisation for Economic Co-Operation and Development (OECD) countries (UNESCO, 2016). University of South Australia, Adelaide, Australia These increases are observable whether one thinks in Corresponding Author: terms of women’s enrollments as a share of total enrollments, Somayeh Parvazian, University of South Australia, Adelaide, SA 5000, or as a percentage of the appropriate age cohort of women Australia. eligible to attend institutions of higher education (Ramirez & Email: email@example.com Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License (http://www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 SAGE Open significant comparisons can be made between different of a simple linear relationship between variables should be countries and times, relying on the available data. (Note treated with great caution. that not all OECD countries collect data in similar ways and Although the studies mentioned above used regression nor do they collect data at the same times or time spans.) analysis, to demonstrate the complexity of the relationship On average, OECD countries report a relatively high between variables, none of them examined the bidirectional GTER of 74% in 2012 compared to the average GTER of relationship between higher education and the full range of 21% in 1980. Of these OECD countries in 1980, GTER was variables at any one time. In contrast, this study investigates highest for women in United States and Canada and lowest any consequential relationship between these variables and in Luxembourg and Turkey. Although the growth in the par- the underlying factors behind them, outlining the structure of ticipation rate of women in higher education is observable complexity and relationship between variables of interest. for all the OECD countries in the time period 1980-2012, in Overall, this research seeks to provide a systematic and countries such as Turkey and Luxembourg, a tenfold increase detailed analysis of the relationship between the expansion has led to a remarkable growth of GTER from 3% in 1980 to of women’s participation in higher education, and the 55% in 2011 in Turkey, and from 1% in 1980 to 19% in 2010 changes made in women’s lives and the broader society as a in Luxembourg. Of course, the low base makes the percent- basis for future structural modeling to present a more nuanced age increase look very dramatic but the interesting point here and complex picture of the interactions between variables. is that within the OECD, there was a wide variation of Although lack of information about current opportunities women enrolling in higher education in recent times. and possibilities may lead to inappropriate policies and Studies show several implications for the global position- plans, an understanding of the effects of these changes for ing of women in terms of status and power that appear to women and the way in which they can influence women’s follow from their increased access to higher education. lives and general social attitudes, has the potential to improve Researchers have identified the sociocultural variables asso- national policies in various fields such as employment. ciated with these changes as changes in family structure and This study uses standard indicators defined by interna- fertility behavior (Abbasi-Shavazi, Lutz, Hosseini-Chavoshi, tional organizations such as UNDP. Selected indicators & Samir, 2008; Abbasi-Shavazi & McDonald, 2006; Fisher include Women’s Labor Force Participation, Percentage of & Charnock, 2003; Gottard, Mattei, & Vignoli, 2015; Yu, Women in Managerial and Professional Occupations and 2006), changes in social norms and values (Bertrand, Cortés, Percentage of Women in Parliament, Women’s Wages as a Olivetti, & Pan, 2016; Gilbert, 2014; Salehi-Isfahani, 2001; Percentage of Men’s Wages, Women’s Age at First Marriage, Shaditalab, 2005), and women’s improved access to job Total Fertility Rate, Divorce Rate, Marriage Rate, and GTER opportunities, income, and security (Frenette & Coulombe, of Women. 2007; Mok, 2016; Vedadhir, 2002; Watts, 2003; Woodd, The next section reviews the literature that describes a 2013; Yu, 2006). Although the direction of all such variables causal relationship between women’s participation in higher seems positive, the question remains about the degree to education and other sociocultural variables through an explo- which they are seen as most desirable and/or important for ration in theory and previous empirical research. This is fol- women in their different societies. lowed by an explanation on the methods used and an These variables impact on both personal lives and society exploratory factor analysis (EFA) of women’s socioeco- as a whole, although the quality and intensity of their impact nomic indicators in selected countries. Finally, the results are may be different across people and places. The literature sug- compared to previous knowledge on the topic and a hypoth- gests that the relationship between these variables is very esis is produced for examining the model of complexity and complex, and it is impossible to draw a firm conclusion relationship between the variables. about which is the cause and which is the consequence. It is quite likely the variables are driven by each other that is they Literature Review are causes and consequences of one another. For example, with regard to family structure, the increasing divorce rate Empirical research tends to proceed from an idea of higher has been at least partly attributable to women’s increased education as benefiting the individual woman through (a) the economic independence over the past few decades (Kalmijn, potential earning power and greater labor force participation 2013; Poortman, 2005). It could be argued that the causal of women (Ahituv & Lerman, 2007; Benavot, 1989; Carnoy, relation works in both directions: With divorce more preva- 2006; Crompton, Lewis, & Lyonette, 2007; Mok, 2016; lent, women might reenter the labor market in larger num- Woodd, 2013), (b) changes in fertility behavior and family bers or become the primary wage earner instead of the arrangements (Carnoy, 2006; Gilbert, 2014; Gottard et al., secondary wage earner (Fernández & Wong, 2014). In addi- 2015; Stevenson & Wolfers, 2007; Weston, Qu, & Parker, tion, women might be more cautious about giving up paid 2004), and (c) shifts in individual beliefs and values employment in favor of full-time homemaking when divorce (Crompton et al., 2007; Gilbert, 2014; Moghadam, 2015). rates are high (Johnson & Kalb, 2002; Raymo, 2015; Van The current study is designed to investigate the social Damme, Kalmijn, & Uunk, 2009). Consequently, any claim effects of women’s increasing participation but its starting Parvazian et al. 3 point is an examination of the individual effects as seen in determination of life chances in most developed countries existing research in the OECD context. (Lauder, Brown, Dillabough, & Hasley, 2006; Lauder, Young, Daniels, Balarin, & Lowe, 2012). However, empiri- cal evidence from most European countries shows that the Potential Earning Power and Greater Labor Force increasing participation in higher education has not achieved Participation of Women equitable social access. In general, some studies show that the increase in higher education enrollments has occurred Two global approaches are identified for the basis of wom- mainly within the middle class, while other people, males en’s greater employability due to higher educational attain- and females without educational credentials, continue to face ment. The first counts women’s greater employability as a disadvantages in access to life chances in most developed function of the knowledge and skills transmitted by higher countries (Brennan, 2008; Brown, 2003, 2013; Quaye & education and its contribution to workforce development. Harper, 2014; Tomlinson, 2003). So it would appear that The second approach investigates the use of higher education women’s increasing access to higher education may be lim- credentials by employers to identify the potential social and ited to one particular group of women, suggesting an inter- cultural capital of the individuals (Brennan, 2008; Brown, relationship of factors relating to class and educational level. 2016). Whichever approach is used, higher education is seen globally to affect women’s willingness and ability to enter the labor market as it raises their potential earning power, Changes in Fertility Behavior and Family provides them with necessary credentials for employment, Arrangements and changes their attitudes toward women’s traditional roles in the household and in the workplace (Ahituv & Lerman, Another highly discussed issue in empirical research in 2007; Benavot, 1989; Bianchi, 2011; Carnoy, 2006; Carnoy this field is fertility behavior and family structure. Studies et al.2012; Crompton et al., 2007). show that there have been remarkable worldwide changes Cross-national studies expect women’s increased educa- in family life in the past 40 years (Stevenson & Wolfers, tional attainment to reduce the wage differentials between 2007; Thornton, 2013). Divorce rates, cohabitation outside men and women because women will increase their commit- of marriage and single parenthood have increased. Fertility ment to wage employment, be less inclined to work in part- rates have significantly decreased in most developed and developing countries. More women are choosing not to time jobs, and have a more continuous employment history. Thus, women’s increased educational attainment is seen as have any children at all and greater percentages of women likely to cause sex differentials in the rate of return on educa- are willing to participate in paid work as well as being responsible for domestic work and child care (Baxter, tion to decrease (Benavot, 1989; Oreopoulos & Petronijevic, Haynes, Western, & Hewitt, 2013; Carnoy, 2006; Gilbert, 2013). Several empirical studies support the idea that in 2014; Weston et al., 2004). It appears that these changes in recent decades, the increasing participation of women in higher education has helped to decrease the gender gap in social relations and sexual mores have yielded both costs and benefits. earnings (Gill & Leigh, 2000; Loury, 1997). However a lon- Many developed countries have experienced women’s gitudinal study by Frenette and Coulombe (2007) in Canada increasing educational achievement and, at the same time, showed different factors driving the gender gap decline in decreasing fertility, a connection shown to be significantly each decade. For example, changes in family characteristics linked (Abbasi-Shavazi et al., 2008; Hazan & Zoabi, 2015; were the main driving force for the decline in earning gaps in Yu, 2006). Other factors cited as contributing to the decline the 1980s, while the rise in educational attainment is shown in fertility rates include the following: the progress in con- to be the main driving force of gender gap decrease in the traceptive technology, delayed age of marriage, increasing 1990s. The convergence in earnings has slowed down during lifetime childlessness and the falling average number of the 1990s, a feature which is explained by Blau and Kahn children (Goldscheider, Bernhardt, & Lappegård, 2015; (2006, p.5) as due to “changes in labour force selectivity,” Oppenheimer, 2013; Qu, Weston, & Kilmartin, 2000), and “gender differences in unmeasured abilities and labour mar- increasing participation of women in paid employment. ket discrimination,” and “changes in the relative advantage of supply and demand shifts.” The main educational expla- Consequently, researchers have stressed the increasing nation for the gender earning gap is that women mainly study opportunity costs for women to bear and rear children in areas that lead to less well-rewarded jobs (Bobbitt-Zeher, (Castles, 2002; Del Bono, Weber, & Winter-Ebmer, 2012; 2007; Frenette & Coulombe, 2007; Gerber & Schaefer, Johnson & Kalb, 2002; McDonald, 2001; Qu et al., 2000; 2004), a finding that shows the strong connection between Shreffler & Johnson, 2013; Yu, 2006). Yu (2006) argues, gender roles and economic reward. not only is the opportunity cost of raising children affected Of course, changes in women’s employability and earn- by the education of parents, but education also affects fer- ings affect society as well as individuals at a time when edu- tility through various other channels such as income, fam- cational credentials are becoming increasingly central to the ily structure, and fertility preference. 4 SAGE Open between the variables. Although the literature studied in this Shifts in Individual Beliefs and Values article captures a number of variables affecting women’s Education has different effects on measurable characteristics higher education participation or considers higher education such as employment rate, marriage, childbearing, and participation of women as a cause to various changes in income; however, these factors do not fully explain the women’s life, the current study aims to fill in the literature effects of increase in women’s higher education participa- gap by studying the consequential interrelationship between tion. A great part of the change in women’s higher education these variables. participation appears to be associated with unobservable norms and values which are harder to document and quan- Method and Materials tify, yet have a genuine influence on human affairs, which becomes most evident in periods of change. The studies discussed above have used regression analysis Modernization theory explains higher education as a which structures the explanation for the key variable in terms modernizing institution and focuses on the associated of another. Thus, women’s participation in higher education changes in norms and values. From this perspective, higher is inevitably seen as produced by one of the other variables education alters the individual’s traditional norms and values being examined. Instead of this approach, we suggest the use by developing modern values such as “openness to new of structural equation modeling (SEM) for studying the com- ideas,” “independence from traditional authority,” a “will- plex interrelationship between variables under study. In this ingness to plan and calculate future exigencies,” and “a approach, the causal relationship is not structured in but strong sense of personal and social efficacy” (Benavot, 1989, rather links between variables are free to emerge during the p. 16). Others have argued that when a significant amount of investigation. SEM is a collection of statistical techniques the population is exposed to modernizing institutions such as that allow a set of relationships between one or more inde- universities, the level of individual modernity increases and pendent variables and one or more dependent variables, and the educational expansion affects the economic development unlike regression models, the response variable in one equa- through its effects on individual beliefs and values (Brown & tion in SEM may appear as a predictor in another equation Lauder, 2006; Eggins, 2003). (Ullman & Bentler, 2003). Variables in SEM may influence Changes in women’s rights along with educational each other reciprocally, either directly or through other vari- achievements have increased gender equality through soci- ables that act as intermediaries (Fox, 2006). A SEM study ety, including in family life. These changes have been her- involves two steps: (a) producing a hypothetical structural alded as transforming women’s opportunities to control and model of the relationship between a set of observed depen- shape their personal lives (Goldin & Katz, 2008). In family dent variables and a set of continuous latent variables and (b) relations, based on a social partnership of interdependence model evaluation and modification that describes the rela- and mutual adjustment, couples can decide how to divide tionships produced in the first step (Schumacker & Lomax, their labor most effectively to satisfy personal needs and 2004). family responsibilities, which does not require that all duties In this article, we produce the hypothetical structural be split evenly down the middle. Today, women have more model by investigating the consequential relationship control over the course and rhythm of their lives than ever between variables under study and the underlying factors before. They also struggle with more choices about how to behind them, outlining the structure of complexity and rela- achieve self-fulfillment (Gilbert, 2014). tionship between variables of interest. In recent decades, the increasing education and employ- EFA is used to determine the factor structure of the ment of women has challenged the traditional family struc- observed variables, on the basis of factor loadings. EFA is ture which was based on gender differences (Gilbert, 2014; recommended as a first step when the researcher does not Goldscheider et al., 2015; Johnson & Kalb, 2002; McDonald, have a substantive theoretical model (Schumacker & Lomax, 2001). The rise in educational attainment has weakened the 2010), and it is a method used for theory building, which is traditional gender roles, on one hand, and resulted in new different from other factor analysis methods such as confir- norms and values, on the other. Although traditional and matory factor analysis (CFA)—used for testing an existing modern roles are hard to harmonize, this disjunction can lead theory—or principal components analysis (PCA)—used for to a social identity crisis especially in those sections of the reducing a set of variables into a smaller number of compo- society which are mostly affected by these institutions and nents (Matsunaga, 2015). cultural changes (Crompton et al., 2007; Gilbert, 2014). Factor analysis enables the consideration of the effects of Table 1 summarizes the literature discussed in this sec- participation in higher education on selected sociocultural tion, introducing the variables shown to affect higher educa- variables as well as the impact these variables have on higher tion participation and/or to be affected by it thereby depicting education itself. Factor analysis differs from multiple regres- a relationship between these variables. As can be seen in sion analysis in that it considers all variables simultaneously Table 1, the same factors often appear as both provocation with no distinction as to dependent or independent variables and consequence, indicating the complex interactions (Hair, Anderson, Tatham, & Black, 1998, p. 97). Thus, factor Parvazian et al. 5 Table 1. Commonly Referenced Variables and the Relationship Between Them. Variable To be affected by To affect Age at first marriage •• Tertiary enrollment (Abbasi-Shavazi, Lutz, Hosseini- •• Total fertility rate (Bumpass et al., 2009; Qu, Chavoshi, & Samir, 2008; Bumpass et al., 2009; Weston, & Kilmartin, 2000) Raymo & Iwasawa, 2005) •• Tertiary enrollment (Carnoy et al., 2012; •• Labor force participation (Abbasi-Shavazi et al., Shaditalab, 2005) 2008; Yu, 2006) Marriage rate •• Tertiary enrollment (Bumpass et al., 2009; Raymo & •• Total fertility rate (Atoh, 2001; Bumpass et al., Iwasawa, 2005) 2009) Divorce rate •• Labor force participation rate (Kalmijn, 2013; •• Labor force participation rate (Poortman, 2005; Poortman, 2005) Raymo, 2015) Gross tertiary enrollment •• Labor force participation (Lauder, Brown, •• Wages as a percentage of men’s wages in ratio Dillabough, & Hasley, 2006; Ramirez & Wotipka, nonagricultural occupations (Bobbitt-Zeher, 2001) 2007; Frenette & Coulombe, 2007; Johnson & Kalb, 2002; Mok, 2016; Woodd, 2013) •• Labor force participation rate (Brown, 2016; Johnson & Kalb, 2002; Van Damme, Kalmijn, & Uunk, 2009) Labor force participation •• Divorce rate (Poortman, 2005) •• Divorce rate rate •• Tertiary enrollment (Johnson & Kalb, 2002; Lauder •• Tertiary enrollment (Lauder et al., 2006) et al., 2006; Watts, 2003b) Wages as a percentage •• Tertiary enrollment (Bobbitt-Zeher, 2007; Frenette of men’s wages & Coulombe, 2007; Johnson & Kalb, 2002) in nonagricultural occupations Percentage of managerial Wages (Miyoshi, 2008) and professional positions Share of seats in Parliament •• Tertiary enrollment (Norris & Inglehart, 2001) •• Percentage of managerial and professional positions (Norris & Inglehart, 2001) Total fertility rate •• Age at first marriage (Qu et al., 2000; Upadhyay & Gupta, 2013) •• Marriage rate (Bumpass et al., 2009; Qu et al., 2000) •• Divorce rate (Qu et al., 2000) •• Tertiary enrollment (Gottard, Mattei, & Vignoli, 2015; Raymo & Iwasawa, 2005) analysis will allow for the inclusion of higher education par- The sudden growth in higher education participation is ticipation in the model without anticipating a one-way rela- seen as a global phenomenon (Marginson & Van der Wende, tionship between variables. In this case, an exploratory study 2007) with all countries affected, regardless of the position is used to reveal the structure of relationships between of governments (Marginson, 2004). Since the process of selected variables. change occurs on a global scale, the results and incentives of To conduct the EFA, principal axis factoring with a this process should similarly be studied at a global level. As Promax rotation was used to extract factors from the data, a consequence, a cross-national study is able to reveal resulting in four factors that captured 63% of variability in dynamic interaction between the variables listed and the the original data set. wider society. This type of cross-national study is beneficial as it produces a large-scale and representative sample, per- mitting macro-level analysis through aggregated data and Data Selection enabling inferential statistics to be used. The decision to Based on the literature and availability of the data, nine choose OECD countries was made to obtain as much data as sociocultural indicators of women defined by international possible from a group of countries that have already achieved organizations were selected for further study, each of which a high participation rate of women in higher education. These has been established as connected with women’s participa- countries provide cultural and geographic diversity through tion in higher education (see Table 1). A complete list of indi- which to explore cross-national differences over the course cators, definitions, and sources are provided in Table 2. of time, depending on the available data. Data were collected 6 SAGE Open Table 2. Indicators, Definitions and Sources. Indicator Definition Source Women’s age at first marriage Mean age of women at first marriage (below age 50 years) 1 Marriage rate The crude marriage rate is the number of marriages formed each year as a ratio 1 to 1,000 people. This measure disregards other formal cohabitation contracts and informal partnerships. Divorce rate The crude divorce rate is the number of divorces granted per 1,000 people. It 1 does not account for separations where partners remain married officially or the breakdown of informal partnerships. Total fertility rate The average number of children that would be born to a woman by the time 1 she ended childbearing if she were to pass through all her childbearing years conforming to the age-specific fertility rates of a given year. Gross tertiary enrollment ratio of Tertiary school GER is the total tertiary school enrollment, regardless of age, 2 women expressed as a percentage of the tertiary school-aged population. 3 Women’s labor force participation Labor force participation rate is the proportion of the population ages 15 and 5 older which are economically active: all people who supply labor for the production of goods and services during a specified period. Women’s wages as a percentage This variable shows average wages per worker in manufacturing as a whole and 6 of men’s wages in nonagricultural has been arranged according to the ISIC Revision 3, or its former version, occupations ISIC Revision 2. Percentage of women in managerial Shows the distribution of the economically active female population in 6 positions managerial occupations as defined by ILO. Major Group 0/1 Professional, technical and related workers and the Major Group 2 Administrative and managerial workers from the ISCO-1968 and Major Group 1 Legislators, senior officials and managers from the ISCO-1988 Percentage of women in professional Shows the distribution of the economically active female population in 6 jobs professional occupations as defined by ILO. Major Group 2 Administrative and managerial workers from the ISCO-1968 and Major Group 2 Professionals from the ISCO-1988 Percentage of women in Parliaments Women in parliament are the percentage of parliamentary seats in a single or 7 lower chamber held by women. Source. (1) Society at a Glance: OECD Social Indicators 2006 Edition—OECD© 2007—ISBN 9264028188 (English). (2) UNESCO Institute for Statistics. 2009. (3) Education Indicators. Paris: UNESCO. Available online at: http://www.uis.unesco.org. (4) The World Bank. Available online at http://data. worldbank.org/indicator. (5) The World Bank. Available online at http://data.worldbank.org/indicator/SL.TLF.CACT.FE.ZS. (6) LABOURSTA–ILO database of labor statistics. Available online at United Nations, Women’s Indicators and Statistics database (www.ipu.org). Note. GER = gross enrollment ratio; ILO= International Labour Organization; ISIC = International Standard Industrial Classification of all economic activities; ISCO = International Standard Classifications of Occupations; OECD = Organisation for Economic Co-Operation and Development; UNESCO = United Nations Educational, Scientific and Cultural Organization. at an aggregated level for each particular indicator being with different data collection schedules. Considering the recorded across 30 different OECD countries. The main assumptions of EFA, missing values were treated by listwise sources for the data were the OECD statistical data bank deletion reducing the data set into a final data set of 69 (Organisation for Economic Co-operation and Development, records. Each record includes simultaneous data on all the 2010), UNDP data bank (UNDP, 2010), and the World Bank nine variables at a same point of time for each country. In (World Bank, 2010). In some cases, national statistical screening the data, a number of countries were removed bureau websites were used to collect additional data for the from the data analysis as these countries had only joined missing values. OECD since 1995, and data were not adequately collected on Overall, this article reports on a large-scale study built them by OECD (especially for the period 1970-1995). These from individual country statistics. Although each country has countries included Korea, Slovak Republic (2000), Hungary its own survey and census schedule, the data were not avail- (1996), Mexico (1994), Poland (1996), and Slovenia (2010). able for all countries at the exact same time. For example, A selection of other countries was also removed because one country might collect data at 5-year intervals from 1955, they provided no data on one or more variables under study while another country collects national data in 2-year inter- (e.g., Germany had not reported gross tertiary enrollment vals from 1960. Thus, to compare the data for these two ratios and therefore was screened out in the listwise dele- countries, the only available data are for 1960, 1970, 1980, tion). Finally, after removing countries with a large number 1990, 2000, and 2010. This pattern of data becomes scarcer of missing values, the data set was reduced to just 15 OECD when all 30 countries are considered simultaneously, each countries including Australia, Austria, Belgium, Denmark, Parvazian et al. 7 Table 3. Pattern Matrix. Factor Variables Marrying later Stronger public presence Child bearing Gaining independence Women’s age at first marriage .970 Women’s labor force participation .474 .541 Gross tertiary enrollment ratio of women .415 Women’s wages as percentage of men’s wages .454 Total fertility rate −.832 Divorce rate −.642 Marriage rate −.629 .408 Percentage of women in professional and .684 managerial occupations Percentage of women in Parliaments .738 Variance explained by each factor 44.79 16.06 10.5 9.1 Extraction method: Principal axis factoring. Rotation method: Promax with Kaiser normalization. Rotation converged in five iterations. Finland, Iceland, Japan, Netherlands, New Zealand, Norway, between subsets of the items. The data indicated no multi- Portugal, Sweden, Switzerland, Turkey, and United collinearity, and we were able to regard the factors as Kingdom. Considering the reduced country list and avail- indicative of separate constructs without substantial over- ability of data for each country, the listwise deletion left us lap. Because the assumptions of factorability were com- with 4 or 5 points in time (spread over the 40-year time span) fortably observed, we retained all items in the ensuing for each country under study. analysis (Field, 2009). Data Screening and Assumption Testing Results and Discussion To conduct statistical procedures with the sample, the data The analysis revealed four factors which were structurally had to be prepared to make it amenable to the particular pro- involved with women’s higher education participation. They cedures to be undertaken. The data were screened for multi- emerged as follows. A factor analysis of the nine items using variate Normality by using a Kolmogorov–Smirnov test of Promax rotation and Kaiser Normalization was conducted. Normality. With the exception of women’s Gross Tertiary Using a scree plot, Jolliffe’s criterion (eigenvalues above Enrollment Ratio, Labor Force Participation, and Age at 0.7), and the interpretability of the results, our study resulted First Marriage, the remaining variables did not pass the mul- in a four-factor solution. tivariate Normality tests at this stage. A minimum cutoff of 0.40 was used to identify items To achieve Normality, we transformed the original data, loading on each factor, based on the sample size of n = 69 using the skewness and kurtosis statistics as indicator of (Hair et al., 1998). The factors, item loadings, and eigenval- Normality. After transformation, these statistics were all ues are given in Table 3. within a tolerable range of ±2, making the data acceptable as Based on the factor loadings in Table 3, three items multivariate Normal. loaded onto the first factor, namely, Women’s Age at First We further conducted a number of tests to determine Marriage, Women’s Labor Force Participation, and Gross whether the data were appropriate for factor analysis. The Tertiary Enrollment Ratio of women. All these variables sample size was n = 69, after using listwise deletion for have the same sign, suggesting that the changes in these missing values, satisfying the recommendation of at least variables complement one another and as such, are all five observations per item (Field, 2009). We also checked important together and should not be considered the Kaiser–Meyer–Olkin (KMO) Measure of Sampling separately Adequacy which was 0.78, above the recommended value This suggestion is supported by the studies indicating that of 0.6, indicating a good sample size to support our analy- women delay marriage in favor of building their career and sis. Each item had a minimum correlation of .3 with at attaining higher education (Abbasi-Shavazi et al., 2008; Yu, least two other items, suggesting reasonable factorability 2006). Since women’s Age at First Marriage has the highest (Hair et al., 1998). The Bartlett Test of Sphericity yielded loading on this factor—twice as large as the other two vari- significant results, approximately χ (36) = 270.214, p < ables—we name this factor as marrying later (Hair et al., .05, supporting the existence of significant correlations 1998). 8 SAGE Open Table 4. Structure Matrix. Factor Variables Marrying later Stronger public presence Child bearing Gaining independence Women’s age at first marriage .871 .427 .430 Women’s labor force participation .760 .484 .796 Gross tertiary enrollment ratio of women .628 .663 .457 .496 Women’s wages as percentage of men’s wages .499 .728 .614 .614 Total fertility rate −.706 Divorce rate −.660 Marriage rate −.441 Percentage of women in professional and .670 .406 managerial occupations Percentage of women in Parliaments .715 .913 .724 Extraction method: Principal axis factoring. Rotation method: Promax with Kaiser normalization. Three items load onto a second factor highly related to The pattern matrix only reflects the direct path from a fac- women’s empowerment. This factor related to Percentage of tor to a variable, it contains the unique correlations between Women in Managerial and Professional Occupations and variables and factors and the variance shared among factors Percentage of Women in Parliament. This factor also relates has been removed. In contrast, the structure matrix reflects to Marriage Rate; however, the effect of marriage is nega- all the possible paths from a factor to a variable including the tively related to the percentage of women in managerial and factor to factor paths (Field, 2009).Women’s Age at First professional occupations and percentage of women in Marriage, Women’s Labor Force Participation, Gross Parliaments. The negative relation of marriage to the other Tertiary Enrollment Ratio of women, Percentage of Women two variables appears to suggest that while marriage may in Parliaments, and Women’s Wages as a Percentage of have an important relation to this factor, it needs to be con- Men’s Wages have high loadings in the structure matrix for sidered separately. This factor was labeled stronger public all three factors. However, Women’s Marriage Rate only has presence as indicated by the higher loading of the two first high loadings on Factor 2 while Total Fertility Rate only variables on this factor. loads high on Factor 3 in the structure matrix. Interestingly, The two items that load onto Factor 3 relate to the Total Factor 4 stands alone on Divorce Rate; however, it shares Fertility Rate and Women’s Labor Force Participation. Gross Tertiary Enrollment Ratio of women with all the other Because Women’s Labor Force Participation has a positive factors. sign and Total Fertility Rate has a negative sign, it seems the In summary, the results from the pattern and structure two variables are in contrast with each other. The relation- matrix show that Gross Tertiary Enrollment Ratio of women ship between these variables is supported by the studies indi- loads high on all four factors, which supports the conclusion cating the decrease in fertility rate followed by increasing in the literature on the effect of higher education participa- participation of women in paid employment and conse- tion on different aspects of women’s lives (Abbasi-Shavazi quently the increasing opportunity costs for women to bear et al., 2008; Abbasi-Shavazi & McDonald, 2006; Fisher & and rear children (Castles, 2002; Johnson & Kalb, 2002; Charnock, 2003; Frenette & Coulombe, 2007; Shaditalab, McDonald, 2001; Qu et al., 2000; Yu, 2006). This factor was 2005; Vedadhir, 2002; Watts, 2003; Yu, 2006). Once again labeled childbearing due to the higher loading of Total the inference to be drawn from this study is that there is no Fertility Rate on this factor compared to Women’s Labor simple linear relationship between women’s involvement in Force Participation. higher education and other aspects of their lives. However, Items for Factor 4 represented Women’s Wages as a the impression remains that higher education has affected a Percentage of Men’s Wages and Divorce Rate. The two vari- broad range of women’s lives as indicated by the steadily ables are negatively related which is supported by those stud- increasing uptake of higher education across different coun- ies indicating that higher family income decreases the tries and cultures. divorce rate (Sayer & Bianchi, 2000). This factor was labeled The information presented in Tables 3 and 4 pictures the gaining independence. structure underlying the variables under study and can be The factors are also interrelated. Factors 1 and 3 are used as a hypothesis in further research. highly interrelated by virtue of Women’s Labor Force The overall picture produced by the EFA is in some sense Participation impacting on both. Factors 2 and 3 are also in line with expectations from the literature. Women with interrelated, both sharing Marriage Rate. university degrees are likely to marry later and have fewer Parvazian et al. 9 children than their less educated peers. However, the analy- The study has shown the inadequacy of explanations sis also shows several interesting and perhaps less antici- describing simple linear relationships between women’s pated outcomes such as the lower divorce rate coincident access to higher education and social change. By detailing with a higher public presence, the latter being also working some of the components of this complexity, the analysis pre- in opposite direction to marriage. In general, these results sented here provides a robust basis for future investigations. open up many questions for further exploration. In this study, we are not making assumptions about individu- However, the main outcome of this work lies in the way in als, but we are using aggregated country-level statistics to which it shows the intricate interrelationships between the find out generalizable trends from group data, not to claim factors involved and the lack of any clear causal explanation that any one woman from any of the groups would behave in in terms of any one factor. The study presented here reveals the same typical way. In this case, further qualitative studies a dynamic and fluid situation, much less controlled and con- were conducted by the authors (Parvazian & Gill, 2012) to trollable than has been seen in other work on women and demonstrate how some of the trends played out in the lives of higher education. individual women—who took different strategies to forge new ways of creating spaces for themselves. This study fur- ther indicated that cultural differences play an important role Research Limitations and Conclusion in women’s lives in various countries. A comparative study The database used in this study is limited to data collected by of the OECD countries, shows that Japan is noted as an international organizations in the last four decades and there exception to other OECD countries given that in this country was a lack of information for some indicators in different the employment rates of tertiary graduate women are similar years. We acknowledge the possible influence of data collec- to, or lower than, the rates of low-educated women. tion times on the results but the trends still stand regardless Explanations for this difference were sought within Japan of the data collection times. In studying cultural changes, a through further quantitative and qualitative analysis to iden- statistical analysis will be able to show the relationships tify the interrelationships between changes in women’s between the selected variables; however, we may only spec- higher education participation, labor force participation, and ulate about any effects of women’s participation in higher earning in terms of potential connections to other aspects of education on social norms and values. Existing statistics on the national social settings. Results show that women’s posi- family structure and fertility behavior can indicate some tion and life choices are so culturally embedded in Japan that changes in norms and values but there is a need to explore such issues cannot be completely changed in a short period some of the outcomes in greater depth in terms of the reali- of time. ties of women’s lives. Hence, qualitative, interpretive meth- odology is suggested for further research at ground level to Implications for Further Study understand the changes in social norms and values experi- enced by female university graduates in depth and detail. The factors emerging from this study can supply the basis for By analyzing the factors involved in women’s increased hypothesis testing in further research. In this case, SEM is participation in higher education this study has identified suggested for further study both as a measurement model and some key interrelationships. Although we acknowledge a structural model. The measurement model is a multivariate the limitation of our findings in terms of a finely nuanced regression model that describes the relationships between a account of the experience of individual women, we argue set of observed dependent variables and a set of continuous that the issue of increasing gender equity in terms of uptake latent variables. The structural model describes three types of higher education has been revealed as much more com- of relationships in a set of multivariate regression equations: plex than previously envisaged. Prior research on this topic the relationships among factors, the relationships among has been generally concerned with the effects of women’s observed variables, and the relationships between factors and higher education participation on their economic and fer- observed variables that are not factor indicators. However, tility behavior, which has resulted in a steady state picture, the SEM analysis on the variables under study in this article wherein one dimension—higher education—is seen as the requires at least 20 observations per variable, a number not change agent. In contrast, this research has shown the available from the current data sets provided by international interrelationship between women’s participation in higher organizations. Although OECD is continually collecting data education and sociocultural aspects of their lives, treating on the sociocultural variables of its member countries, in a all the variables in dynamic interaction. This article argues few years time, the basic data for the SEM study might be that the changes in women’s lives are not simply a result of available and further research can be done using SEM women’s increased involvement in higher education. analysis. Rather, the argument put forward here is that the changes On the basis of the findings detailed here, we urge further in higher education are taking place at the same time as a investigation of the interrelationship between women’s range of other changes, all of which interact and affect access to and participation in higher education to further each other. establish women’s position as essential players in the 10 SAGE Open knowledge economies of both developed and developing Ahituv, A., & Lerman, R. I. (2007). How do marital status, work effort, and wage rates interact? Demography, 44, 623-647. countries. Atoh, M., (2001). Why are cohabitation and extra-marital births so few in Japan? Paper presented at the EURESCO confer- Declaration of Conflicting Interests ence. In The second Demographic Transition in Europe, Bad The author(s) declared no potential conflicts of interest with respect Herrenalb, Germany, 23–28 June 2001. to the research, authorship, and/or publication of this article. Baxter, J., Haynes, M., Western, M., & Hewitt, B. (2013). Gender, justice and domestic work: Life course transitions and per- Funding ceptions of fairness. Longitudinal and Life Course Studies, 4, 78-85. The author(s) received no financial support for the research, author- Benavot, A. (1989). Education, gender, and economic development: ship, and/or publication of this article. A cross-national study. Sociology of Education, 62, 14-32. Bertrand, M., Cortés, P., Olivetti, C., & Pan, J. (2016). Social Notes norms, labor market opportunities, and the marriage gap 1. The Organisation for Economic Co-Operation and for skilled women (No. w22015). Cambridge, MA: National Development (OECD) is a group of 30-member countries that Bureau of Economic Research. cooperatively discuss, develop, and refine a broad range of Bianchi, S. M. (2011). Changing families, changing workplaces. economic and social policies including macroeconomics, edu- The Future of Children, 21(2), 15-36. cation, and science. Blau, F. D., & Kahn, L. M. (2006). The US gender pay gap in the 2. This article reports parts of a larger project which included five 1990s: Slowing convergence. ILR Review, 60(1), 45-66. steps: (a) documentary analysis on the global reforms in higher Bobbitt-Zeher, D. (2007). The gender income gap and the role of education sector over the last 40 years; (b) descriptive analy- education. Sociology of Education, 80(1), 1-22. sis of the changes in women’s higher education participation Brennan, J. (2008). Higher education and social change. Higher and the other eight variables under study for each country over Education, 58, 381-393. the past 40 years; (c) studying the interrelationship between Brown, P. (2003). The opportunity trap: Education and employ- variables under study(which is reported in this article); (d) ment in a global economy. European Educational Research studying the differences between countries based on their fac- Journal, 2, 141-179. tor scores and structure model (based on factor scores, Japan Brown, P. (2013). Education, opportunity and the prospects for showed the lowest score in all factors emerged from the study social mobility. British Journal of Sociology of Education, 34, and Sweden showed the highest score on all factor). Therefore, 678-700. similar exploratory factor analysis (EFA) methods were used Brown, P. (2016). The transformation of higher education, creden- on these individual countries based on additional country-level tial competition, and the graduate labour market. In J. E. Cote data collected from these countries’ bureau of statistics, and the & A. Furlong. (Eds.), Routledge handbook of the sociology of structure of model was compared to see whether there are dif- higher education (pp. 197-207). New York: Routledge. ferences between countries. Interestingly Sweden showed the Brown, P., & Lauder, H. (2006). Globalisation, knowledge and the same structure model as the whole OECD data set while Japan myth of the magnet economy. Globalisation, Societies and showed a different structure model). (e) The case of Japanese Education, 4, 25-57. data was followed up with qualitative methods to allow the Bumpass, L. L., Rindfuss, R. R., Choe, M. K., & Tsuya, N. O. inclusion of some immeasurable factors such as culture in the (2009). The institutional context of low fertility: The case of study. This case study of Japan was completed with studies on Japan. Asian Population Studies, 5(3), 215-235. Japan’s national policies and cultural standards compared to Carnoy, M. (2006). The new family and flexible work. In H. other OECD countries. Lauder, P. Brown, J. A. Dillabough, & A. H. Halsey. (Eds.), 3. Numbers in parentheses refer to the date these countries joined Education globalization and social change (pp. 371-380). New OECD. York, NY: Oxford University Press. 4. In statistics, the Kolmogorov–Smirnov test (K-S test) is a non- Carnoy, M., Loyalka, P. K., Androushchak, G. V., & Proudnikova, parametric test for the equality of continuous, one-dimensional A. (2012). The economic returns to higher education in the probability distributions that can be used to compare a sample BRIC countries and their implications for higher education with a reference probability distribution. The two-sample K-S expansion. Rochester: Stanford University. test is one of the most useful and general methods for testing Castles, F. G. (2002). Three facts about fertility: Cross-national les- for Normality of the distribution. In this case, samples are stan- sons for the current debate. Family Matters, 63, 22-27. dardized and compared with a standard Normal distribution. Crompton, R., Lewis, S., & Lyonette, C. (2007). Introduction: The unravelling of the “male breadwinner” model—And some of References its consequences. In R. Crompton, S. Lewis, & C. Lyonette. Abbasi-Shavazi, M. J., Lutz, W., Hosseini-Chavoshi, M., & Samir, (Eds.), Women, men, work and family in Europe (pp. 1-16). K. C. (2008). Education and the world’s most rapid fertility UK: Palgrave Macmillan. decline in Iran. IIASA Interim Report. Laxenburg, Austria: Del Bono, E., Weber, A., & Winter-Ebmer, R. (2012). Clash of International Institute for Applied Systems Analysis (IIASA) career and family: Fertility decisions after job displacement. Abbasi-Shavazi, M. J., & McDonald, P. (2006). Fertility decline Journal of the European Economic Association, 10, 659-683. in the Islamic Republic of Iran: 1972-2000. Asian Population Eggins, H. (2003). Globalization and reform: Necessary conjunc- Studies, 2, 217-237. tions in higher education. In H. Eggins(Ed.), Globalization and Parvazian et al. 11 reform in higher education (pp. 1-8). Berkshire, UK: Open uni- Marginson, S., & Van der Wende, M. (2007). To rank or to be versity press. ranked: The impact of global rankings in higher education. Fernández, R., & Wong, J. C. (2014). Divorce risk, wages and work- Journal of Studies in International Education, 11(3-4), ing wives: A quantitative life-cycle analysis of female labour 306-329. force participation. The Economic Journal, 124, 319-358. Matsunaga, M. (2015). How to factor-analyze your data right: Do’s, Field, A. (2009). Discovering statistics using SPSS. London, UK: don’ts, and how-to’s. International Journal of Psychological Sage. Research, 3, 97-110. Fisher, K., & Charnock, D. (2003, March). Partnering and fertility McDonald, P. (2001). Family support policy in Australia: The need patterns: Analysis of the HILDA Survey, Wave 1 (Vol. 13). Paper for a paradigm shift. People and Place, 9(2), 14-20. presented as the HILDA Conference, Melbourne, Australia. Moghadam, V. (2015). Gender and globalization: Female labor and Fox, J., (2006). Teacher’s corner: Structural equation modeling women’s mobilization. Journal of World-Systems Research, 5, with the sem package in R. Structural Equation Modeling, 366-389. 13(3), 465-486. Mok, K. H. (2016). Massification of higher education, graduate Frenette, M., & Coulombe, S. (2007). Has higher education employment and social mobility in the Greater China region. among young women substantially reduced the gender gap in British Journal of Sociology of Education, 37, 51-71. employment and earnings? Ottawa, Ontario: Statistics Canada, Miyoshi, K. (2008). Male–female wage differentials in Japan. Business and Labour Market Analysis. Japan and the World Economy, 20(4), 479-496. Gerber, T. P., & Schaefer, D. R. (2004). Horizontal stratification Norris, P., & Inglehart, R. (2001). Cultural obstacles to equal repre- of higher education in Russia: Trends, gender differences, and sentation. Journal of Democracy, 12(3), 126-140. labor market outcomes. Sociology of Education, 77, 32-59. Oppenheimer, V. K. (2013). Work and the family: A study in social Gilbert, N. (2014). A mother’s work: How feminism, the market, and demography. New York: Academic Press. policy shape family life. New Haven, CT: Yale University Press. Oreopoulos, P., & Petronijevic, U. (2013). Making college worth it: Gill, A. M., & Leigh, D. E. (2000). Community college enrollment, A review of research on the returns to higher education (No. college major, and the gender wage gap. Industrial & Labor w19053). National Bureau of Economic Research. Relations Review, 54, 163-181. Organisation for Economic Co-Operation and Development. (2010). Goldin, C., & Katz, L. (2008). Gender differences in careers, edu- OECD.Stat: The Organisation for Economic Cooperation and cation and games. American Economic Review, 98, 363-369. Development. Retrieved from http://dx.doi.org/10.1787/data- Goldscheider, F., Bernhardt, E., & Lappegård, T. (2015). The gen- 00285-en der revolution: A framework for understanding changing fam- Parvazian, S., & Gill, J. (2012). Women and cultural change in ily and demographic behavior. Population and Development Japan: The implications of increasing access to higher edu- Review, 41, 207-239. cation. TASA 2012 Conference, University of Queensland, Gottard, A., Mattei, A., & Vignoli, D. (2015). The relationship Brisbane, 26 November 2012. between education and fertility in the presence of a time vary- Poortman, A. R. (2005). Women’s work and divorce: A matter of ing frailty component. Journal of the Royal Statistical Society, anticipation? A research note. European Sociological Review, Series A: Statistics in Society, 178, 863-881. 21, 301-309. Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Qu, L., Weston, R., & Kilmartin, C. (2000). Effects of changing Multivariate data analysis (5th ed.). Upper Saddle River, NJ: personal relationships on decisions about having children. Prentice Hall. Family Matters, 57, 14-19. Hazan, M., & Zoabi, H. (2015). Do highly educated women choose Quaye, S. J., & Harper, S. R. (2014). Student engagement in higher smaller families? The Economic Journal, 125, 1191-1226. education: Theoretical perspectives and practical approaches Johnson, D., & Kalb, G. R. (2002). Economic analyses of families: for diverse populations. New York, NY: Routledge. Existing research findings. Melbourne Institute of Applied Ramirez, F. O., & Wotipka, C. M. (2001). Slowly but surely? The Economic and Social Research, Melbourne: The University of global expansion of women’s participation in science and engi- Melbourne. neering fields of study, 1972-92. Sociology of Education, 74, Kalmijn, M. (2013). The educational gradient in marriage: A com- 231-251. parison of 25 European countries. Demography, 50, 1499-1520. Raymo, J. M., & Iwasawa, M. (2005). Marriage market mismatches Lauder, H., Brown, P., Dillabough, J. A., & Hasley, A. H. (Eds.). in Japan: An alternative view of the relationship between wom- (2006). Education, globalization and social change. New en’s education and marriage. American Sociological Review, York, NY: Oxford University Press. 70(5), 801-822. Lauder, H., Young, M., Daniels, H., Balarin, M., & Lowe, J. (Eds.). Raymo, J. M. (2015). Living alone in Japan: Relationships (2012). Educating for the knowledge economy? Critical per- with happiness and health. Demographic Research, 32, spectives. London, England: Routledge. 1267-1298. Leathwood, C., & Read, B. (2008). Gender and the changing face Salehi-Isfahani, D. (2001). The gender gap in education in Iran: of higher education: A feminized future? A feminized future? Evidence for the role of household characteristics. In D. Salehi- McGraw-Hill Education. Isfahani (Ed.), Labor and human capital in the Middle East: Loury, L. D. (1997). The gender earnings gap among college-edu- Studies of markets and household behavior (pp. 235-255). cated workers. ILR Review, 50, 580-593. Cairo and Reading: Ithaca Press. Marginson, S. (2004). Competition and markets in higher educa- Sayer, L. C., & Bianchi, S. M. (2000). Women’s economic indepen- tion: A ‘glonacal’analysis. Policy Futures in Education, 2(2), dence and the probability of divorce: A review and re-examina- 175-244. tion. Journal of Family Issues, 21, 906-943. 12 SAGE Open Schumacker, R. E., & Lomax, R. G. (2004). A beginner’s guide institutional determinants. European Sociological Review, 25, to structural equation modelling. Mahwah, NJ: Lawrence 183-197. Erlbaum. Vedadhir, A. A. (2002). Women and higher education, a new iden- Schumacker, R. E., & Lomax, R. G. (2010). A beginner’s guide to tity for women in higher education. Ketab E Zanan, 15, 28-45. structural equation modeling (3rd ed.), New York: Taylor & Watts, M. (2003). The evolution of occupational gender segrega- Francis Group. tion in Australia: Measurement and interpretation. Australian Shaditalab, J. (2005). Iranian women: Rising expectations. Critique: Journal of Labour Economics, 6, 631-655. Critical Middle Eastern Studies, 14, 35-55. Weston, R., Qu, L., & Parker, R. (2004). “It’s not for lack of want- Shreffler, K. M., & Johnson, D. R. (2013). Fertility intentions, ing kids”: A report on the fertility decision making project. career considerations and subsequent births: The moderat- Melbourne: Australian Institute of Family Studies. ing effects of women’s work hours. Journal of Family and Woodd, M. (2013). The move towards a different career pattern: Economic Issues, 34, 285-295. Are women better prepared than men for a modern career? Stevenson, B., & Wolfers, J. (2007). Marriage and divorce: Career Development International, 5, 99-105. Changes and their driving forces. The Journal of Economic World Bank. (2010). Data: World Bank. Available from http://data. Perspectives, 21(2), 27-52. worldbank.org Taylor, W. (2003). Steering change in tertiary education. In H. Yu, P. (2006). Higher education, the bane of fertility? An investi- Eggins. (Ed.), Globalization and reform in higher education gation with the HILDA Survey. Retrieved from https://ideas. (pp. 11-31). Berkshire, UK: Open University Press. repec.org/p/auu/dpaper/512.html Thornton, A. (2013). Reading history sideways: The fallacy and enduring impact of the developmental paradigm on family life. Author Biographies Chicago, IL: University of Chicago Press. Tomlinson, S. (2003). Globalization, race and education: Continuity Somayeh Parvazian is a data analyst at the National Centre for and change. Journal of Educational Change, 4, 213-230. Vocational Education Research (NCVER) in South Australia. She Ullman, J. B., & Bentler, P. M. (2003). Structural equation model- has several years of experience in statistical analysis in socio-eco- ing. In Handbook of Psychology. Four (vol. 24, pp. 607-634). nomic areas, in particular the analysis of gender equality in fields of John Wiley & Sons, Inc.,Wiley Online Library. Retrieved education and employment. from http://onlinelibrary.wiley.com/doi/10.1002/0471264385. Judith Gill is an adjunct associate professor at the University of wei0224/abstract South Australia. Her research interests reflected in her many pub- United Nations Development Program. (2008). Statistics of the human development report. Author. Retrieved from http://hdr. lications concern gender and education and women and work. undp.org/en/statistics/. Recent books include A Girl’s Education? Schooling and the for- United Nations Development Program. (2010). Statistics: United mation of gender, identities and future visions. Gill Judith with Nations Development Program. Retrieved from http://hdr. Kathy Esson & Ros Yuen (2016): London, Palgrave Macmillan., undp.org/en/statistics/ and Challenging Knowledge , Sex and Power: Gender, Work and United Nations Educational, Scientific and Cultural Organization. Engineering. Gill J with Mills J., Franzway S., & Sharp R (2013): (2016). UNESCO data centre: UNESCO Institute for Statistics. New York, Routledge. Retrieved from http://stats.uis.unesco.org/unesco/TableViewer/ Belinda Chiera is a program director and senior statistics lecturer document.aspx?ReportId=143&IF_Language=eng at the School of Information Technology and Mathematical Upadhyay, S., & Gupta, P. (2013). Marriage, process and pre- Sciences at the University of South Australia. She has many years paredness among youth: Insight from youth in India. In XXVII of experience in statistical methods across a vast quantity of appli- International Population Conference, Busan, Republic of cations focused on uncovering insights hidden in large databases, in Korea, 26–31 August 2013. particular in Big Data analytics centered on social network analysis Van Damme, M., Kalmijn, M., & Uunk, W. (2009). The employ- and causal system modelling. ment of separated women in Europe: Individual and
SAGE Open – SAGE
Published: May 16, 2017
Keywords: globalization; higher education participation; gender; sociocultural variables; factor analysis
Access the full text.
Sign up today, get DeepDyve free for 14 days.