Transnationalization of Educational Aspirations: Evidence from China

Nuhoglu Soysal, Yasemin; Cebolla Boado, Héctor Article — Published Version Transnationalization of Educational Aspirations: Evidence from China Sociological Research Online Provided in Cooperation with: WZB Berlin Social Science Center Suggested Citation: Nuhoglu Soysal, Yasemin; Cebolla Boado, Héctor (2024) : Transnationalization of Educational Aspirations: Evidence from China, Sociological Research Online, ISSN 1360-7804, SAGE Publications, Thousand Oaks, CA, Vol. 29, Iss. 3, pp. 632-649, https://doi.org/10.1177/13607804231189626 This Version is available at: https://hdl.handle.net/10419/307959 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ https://doi.org/10.1177/13607804231189626 Sociological Research Online 2024, Vol. 29(3) 632 –649 © The Author(s) 2023 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/13607804231189626 journals.sagepub.com/home/sro Transnationalization of Educational Aspirations: Evidence from China Yasemin Nuhoglu Soysal WZB, Germany Héctor Cebolla Boado CSIC, Spain Abstract The international migration of students has garnered a lot of attention from researchers due to its growing popularity and significance. However, the current state of research in this field is limited. On one hand, there is a scarcity of high-quality, large-scale data, and existing studies primarily focus on students who are already abroad. On the other hand, the field predominantly revolves around Bourdieusian-inspired arguments that narrowly view international education as a strategic investment by parents and a means of perpetuating social advantage. This article addresses these limitations by utilizing nationally representative survey data from China, the largest single source of international students globally. Our findings challenge the existing literature by revealing that parental aspirations to send their children abroad are more widespread across diverse social backgrounds than previously suggested. Furthermore, we observe that exposure to transnational environments amplifies aspirations for international education across various parental backgrounds and mitigates differences in aspirations based on parental education levels. We argue that these empirical patterns reflect the global standardization and diffusion of models and ideals of self, of which international education has increasingly become a part within the context of the transnationalization of higher education itself. Keywords China, cultural capital and social reproduction, educational aspirations, global standardization of models and ideals of self, international education transnationalization of higher education The increase in the number of internationally circulating higher education students in the last 50 years has been striking: from approximately 250,000 in 1965 to 5 million in 2020. Since the 1990s, there has been more than a fourfold increase. Asian countries source Corresponding author: Yasemin Nuhoglu Soysal, Berlin Social Science Center, WZB Reichpietschufer 50, 10785 Berlin, Germany. Email: [email protected] 1189626SRO0010.1177/13607804231189626Sociological Research OnlineSoysal and Boado research-article2023 Article Soysal and Boado 633 more than half of today’s international students, with China, India, and South Korea leading as the primary sending countries. Chinese students constitute the single largest and fastest-growing body of international students. The total number of Chinese nationals studying overseas in 2019 was 703,500, marking an increase of 41,400, or 6.3%, over the previous year (Ministry of Education, PRC, 2022a). According to official figures, the Chinese represent 17.5% of the total stock of students abroad (UNESCO, 2022). These figures explain the growing interest in studying the dynamics surrounding the international migration of students and the privileged focus China has in this literature. Although rapidly growing, research in this field has significant limitations. On one hand, there is a lack of high-quality, large-scale representative data, and the existing research tends to focus on students who have already moved across borders, neglecting the drivers of international migration for education that are entangled with student origins (CebollaBoado et al., 2018). On the other hand, the literature does not pay sufficient attention to the processes that have transformed higher education worldwide. In this article, we focus our analysis on student origins and study parental aspirations and intentions to send their children abroad for their higher education, using the China Family Panel Study (CFPS), a large-scale representative sample of Chinese households produced by Peking University since 2010 (Xie and Hu, 2014). Going beyond the conventional understanding of international education (IE) as a mechanism of elite reproduction, we focus on the link between education and migration as an aspiration and self-expression, which are increasingly standardized in the context of the transnational transformations of higher education itself. In addition to its prominent contribution to the global flow of international students, the choice of China is also relevant because of its highly dynamic migration and educational context. Since 1978, migration and educational expansion have been connected with the ‘modernization’ of China (i.e. transition to a market economy and broad social reforms), which significantly increased the possibility of upward social mobility for the younger generations (Liu, 2007). This expectation of mobility created unprecedented levels of internal migration, which have been captured in recent national survey data sets such as the CFPS. For generations born in the 1980s and the 1990s, rural to urban migration has been the principal means of improving family well-being (Xu and Xie, 2015). At the same time, China has experienced a rapid expansion of its education system (Treiman, 2013). The reintroduction of the standardized National Higher Education Entrance Examination (Gaokao) in 1978 inaugurated the modern system of higher education in China and furthered the idea of ‘meritocracy’ in education (Liu, 2013). In addition, the ‘one-child policy’, along with increasing middle-class incomes, has fed into this educational expansion by making it possible to pool resources that multigenerational families can invest in children’s education (Cai et al., 2010). According to the Ministry of Education, the number of registered students in all higher education institutions in 2021 totaled 44.3 million across the country (Ministry of Education, PRC, 2022b). The drastic expansion of tertiary education enrolments over the last two decades in China transformed the system into a social space of fierce competition (Samir et al., 2010). Combined with vast regional differences in access to and quality of education, this has boosted internal and international migration for educational reasons. 634 Sociological Research Online 29(3) Explaining IE aspirations Youth, including large segments of the disadvantaged population, have been documented to be highly aspirational and optimistic in their educational expectations (Alexander et al., 2008; Khattab, 2003; Soysal and Cebolla-Boado, 2020). Scholars of social stratification often depict youth aspirations and expectations as rational calculations pursuing either social influence (Haller, 1982; Sewell and Hauser, 1972) or maximizing economic gains (Breen, 2001; Morgan et al., 2013), or both (Zimmermann, 2020). Scarcer views in the field offer cultural accounts of educational aspirations, arguing that aspirations should be understood as assertions of identity shaped by cultural schemas and standards of the virtuous individual (Frye, 2012). Research on IE has been mainly inspired by the social stratification scholarship but reinterprets it from a Bourdesian perspective in line with a growing interest in habitus and cultural capital in broader sociology to account for inequalities by social background (Jaeger, 2022). Two strands of arguments are manifest (Lipura and Collins, 2020). The first emphasizes positional gains, such as a degree from a prestigious university and language competence, and the competitive advantage that IE enables (Brooks and Waters, 2011; Findlay et al., 2012; Findlay and King, 2010; Gerhards and Hans, 2013; Waters, 2008). The second highlights the non-material benefits of IE, such as self-growth and cosmopolitan experiences and outlook (Collins, 2013; Marginson, 2014; Tran, 2015). Both strands follow the Bourdesian logic of capital conversion and conceptualize IE as cultural capital, which can ultimately be commodified over the life course to ensure intergenerational transmission of social advantage. Thus, the literature often discusses IE in the context of middleand upper-class families strategizing about international schools and sending their children abroad to maximize cultural capital in the family while at the same time participating in the reproduction of the symbolic power of international higher education in a hierarchically structured global field (Yu, 2021). In the context of China, scholars emphasized the utilization of educational migration by the newly rich to convert economic capital to cultural capital, which in turn commodified IE as a status symbol accessible only to the socially and politically advantaged (Xiang and Shen, 2009). Others saw studying abroad as a strategy for well-to-do Chinese families to compensate for potential failure in a highly competitive domestic educational system (Zhou et al., 2019). In sum, IE as an elite social reproduction strategy is highly prevalent in the current research agenda. Despite its predominance, we find studies starting from such a view narrow and restrictive in their explanatory accounts. Notably, they do not consider broader environmental factors such as the diffusion of transnationally standardized models of individual and life course, which precipitate educational and mobility aspirations across different social groups and contexts. Our article innovates on this aspect and invites rethinking individual and familial level explanations of IE and how aspirations for IE are formed by privileging transnational cultural models and standards in our analytical accounts. In the broader macro-sociological literature, world society scholarship offers arguments and empirical models that elaborate global cultural models and their diffusion Soysal and Boado 635 (Jepperson and Meyer, 2021; Meyer, 2010). As their research has shown, a significant aspect of the global education field is its emphasis on the increasingly standardized models of the agentic, self-motivated, and aspirational individual (Lerch et al., 2022; Meyer and Jepperson, 2000). Individuals with such traits are predicted to succeed in labor markets and overall life goals and especially so in knowledge society contexts (Hasse and Krucken, 2013). Since the 1990s, the agentic and aspirational individual model has been transmitted through scientific ideologies and policies of education, espoused by international organizations, government, and market agencies of various sorts, equipping individuals on a global level with uniform narratives of the self (Frank and Meyer, 2002; Lerch et al., 2017; Soysal and Wong, 2015). Particularly in higher education, a highly transnationalized field, we observe a standardized conception of the student that centers around individual agency, ambition, competitiveness, and openness to new experiences (Soysal and Cebolla-Boado, 2020). This conception cuts across higher education sectors worldwide, affecting self-orientations and societal expectations. From this perspective, IE emerges as a transnationally standardized aspiration, an assertion and projection of self-transformation in line with an idealized future (Frye, 2012). As such, educational and mobility aspirations become taken-for-granted goals in themselves, in contrast to the narrower understandings of IE as a strategic investment and elite reproduction. Conceptualizing IE as a transnationally standardized aspiration suggests a different distribution of aspirations than the elite reproduction perspective would predict. We formulate the following specific hypotheses: 1. Bourdieusian accounts of IE inspired by elite reproduction arguments suggest a positive selection by relevant background factors among parents choosing to send their children abroad. Thus we expect parents with higher social backgrounds (income, education), who are more involved with their children’s education, and whose children perform better in school to be more likely to aspire to and invest in IE for their children. 2. As a transnationally standardized aspiration, IE implies convergence of aspirations across background variables. Thus we expect parents’ aspirations and intentions for IE to be independent of their social background, their involvement in children’s education, and children’s school performance. To address the transnational sources of IE aspirations more directly, we further consider the effect of being embedded in less or more transnational environments. Following the world society theory (Meyer, 2010), if we understand the diffusion of cultural models as a function of transnational linkages, we should expect higher likelihoods of IE aspirations with higher transnational exposure. Moreover, transnational embeddedness may not only increase the average level of aspirations for IE but may also mitigate the social background differences in aspirations. Exposure to the transnational could thus be a booster of aspiration and an equalizer, making IE an object of desire for many and more diverse people. We formulate the following additional hypotheses: 636 Sociological Research Online 29(3) 3. Families embedded in transnationally exposed environments are more likely to consider IE an option for their offspring. 4. Social background differences in aspirations for IE are less evident in transnationally exposed environments, where models and standards to emulate are likely to be more widespread. Data, variables, and methods In this article, we use data from the CFPS, a large-scale representative sample of Chinese households produced by Peking University since 2010 (Xie and Hu, 2014). We use the 2010 wave, the only wave with questions about aspirations for IE. CFPS collects household and individual information through computer-assisted, person-to-person interviews of all family members beyond the age of nine. The sample is drawn from a multi-stage probability sampling with implicit stratification in three stages: county (or equivalent), village (or equivalent), and household. Consequently, the sampling covers 25 geographical units in China–Hong Kong, Macao, and the provinces of Xinjiang, Qinghai, Inner Mongolia, and Hainan being explicitly excluded. Half of the households come from the following provinces with large urban concentrations: Shanghai, Liaoning, Henan, Gansu, and Guangdong. The questionnaires collect essential information on families, households, and their broader social context. The CFPS is the best-available empirical representative tool for household dynamics and the involvement of parents in their children’s education. The children’s questionnaire, completed by parents, contains complete information on school education and includes researcher-conducted assessments of children’s cognitive abilities. In the 2010 wave, parents in the sample were asked whether they had ever thought of sending their child abroad for study and, if so, when they thought it would be an appropriate age. Table 1 shows the distribution of responses to these questions for the entire CFPS sample (ages 0 to 15) and three separate definitions of our analytic samples for analysis. The first refers to the number of respondents for whom basic controls are available. The second refers to those for whom the index of parental involvement has valid values (the variables contributing to this index are scant in valid responses). Finally, the last column shows the observations for whom valid test scores (numeracy and literacy in the Chinese language) are available. The CFPS provides this information only for students aged 10 to 15. For the sake of simplicity, and in light of the small number of respondents who considered IE before tertiary education, we collapsed this trichotomous variable into a dummy (1: considered IE; 0: did not consider IE) to be used as the dependent variable in our subsequent analyses. Our independent variables consist of the following. For social background, we use the highest level of parental education as the relevant stratification variable (1: primary, 2: secondary; 3: bachelor’s degree and above). We also use household income, measured as a continuous variable, which we recodified into deciles in the analyses. For the prior performance of the child, we include performance in mathematics. This is the result of a cognitive test, which was standardized across ages (mean 0; standard Soysal and Boado 637 deviation 1). Parental involvement is a synthetic index built by using regression scoring after factor analyses of information regarding whether the parents talk to their child about school; if they supervise homework; and if they control time spent watching TV (response categories: very often, often, sometimes, rarely, never). Table 2 in the Appendix includes the result of the corresponding factor analysis. Note that these two variables, performance and involvement, are only available for a limited subset of cases (students aged 10–15). Finally, we use provincial data to build a synthetic index of exposure to the transnational at the province level from two different sources.1 First, we include information from the National Statistical Bureau on the number of patent applications for inventions of industrial enterprises/province population; the number of foreign visitors/province population; the number of registered websites/province population; total value of foreign investments/gross domestic product (GDP); total value of imports and exports of foreign-funded enterprises (1000 US dollars)/GDP. Second, we utilize data from Borsi et al. (2022) on higher education density in the 31 Chinese provinces since we regard universities as active agents in the transnationalization of aspirations and life course ideals. Universities are globally oriented institutions that actively contribute to exposing their environments to the transnational. Using factor analysis and then regression scoring (see Table 3 in the Appendix), all these measures are merged into a single synthetic index of exposure to the transnational. We use the continuous version of the index for the descriptive analyses and a recodified version of it in our treatment effect modeling strategy in which we differentiate between being above (treatment) and below (control) the mean exposure to the transnational.2 As additional controls, we include gender (1: female, 0: male) and whether the child has siblings (0) or is a single child (1). Table 4 in the Appendix summarizes the distribution of all variables. Throughout our empirical analyses, we use linear regression models rather than the conventional logistic regression models, which are often affected by omitted variables and allow a less straightforward interpretation (Mood, 2010). The final set of analyses is conducted using IPWRA (inverse probability weighting with regression adjustment) treatment effects to explain differences in the slope of the effect of our independent variables on the inclination to consider IE. IPWRA helps us consider the non-random distribution of people across provinces that are differentially exposed to transnational settings. Table 1. Sample size and responses to the question: ‘Have you ever considered sending your child abroad to study?’. CFPS sample (ages 0–15) Basic analytic sample Sample including parental involvement Sample including children’s test scores (ages 10–15) Did not consider 3327 3289 1514 1273 Before tertiary education 231 230 89 67 For tertiary education 795 787 336 239 Total N 4353 4306 1939 1639 CFPS: China Family Panel Study. Legend: Percentages (and the number of cases in parentheses). 638 Sociological Research Online 29(3) This helps to control for selection biases across levels of exposure to the transnational before any interpretation of the correlation between exposure and aspirations is made. Findings Figure 1 shows the descriptive distribution of the aspirations for IE as directly obtained from the CFPS. It shows that almost one in four respondents (23.5%) has considered sending their offspring to study abroad. Tertiary education is by far the most preferred educational stage; only around 5% of the parents considered secondary education outside China. Given the distribution of this variable, in the following analyses, we used a dichotomous version of it in which respondents having ever considered IE at any level were codified as ‘1’ and the rest as ‘0’. Social background effect Our first set of models examines parental socio-economic backgrounds, specifically parental education and income, the most common proxies of status used in the literature on IE. In Figure 2, the panel on the left shows the predicted probabilities for the deciles of household income. The model predicts that one in four Chinese parents report considering IE, regardless of household income level. Only the wealthiest 10% are more likely to consider sending their children abroad (the predicted probability is above 0.32 for this group). This only partially supports the elite reproduction hypothesis, as even among the poorest Chinese households in our sample parents that report considering IE are more than 20%. The panel on the right (Figure 2) shows similar results for educational levels. For parents whose educational attainment is primary or less, the predicted probability of Figure 1. Unconditional preferences for international education by level. Source: CFPS 2010. Soysal and Boado 639 considering sending their children abroad is slightly below 0.2; for parents with an educational level at least secondary it is 0.24, and for parents with an educational level tertiary and above it is close to 0.35. These figures show some social gradient (see also Zhou et al., 2019); however, they also clearly indicate that aspirations for IE are more homogenously distributed by social background than the elite reproduction literature suggests. Figure 3, considering parental involvement in children’s education and children’s academic performance, points to the same conclusion. Although there is a positive slope suggesting parents who are involved and whose children are academically successful are more inclined toward IE, it would be misleading to ignore that a sizable proportion of even the least involved parents and parents with the least successful children broadly have IE aspirations for their children. In sum, our results point to an interesting regularity suggesting that while there is some positive effect of social background, aspirations for international higher education are fairly widespread among a significant proportion of Chinese parents – a finding that deviates from the existing literature. Transnational effect Our next set of analyses assesses whether transnational exposure at the provincial level increases the likelihood that parents express IE aspirations and attenuates the effects of social background. We employ two empirical strategies. First, we estimate the effect of exposure to the transnational on the likelihood of considering IE using linear probability models, as we did with the parental socioeconomic variables. The results from our basic model (Figure 4) suggest that higher levels of exposure to the Figure 2. Probabilities of considering IE by household income and parental education. Source: Our elaboration from CFPS 2010. Estimated from linear probability models. Model 1 in Table 5 in the Appendix. 646 Sociological Research Online 29(3) Yu J (2021) Consuming UK transnational Higher Education in China: A Bourdieusian approach to Chinese students’ perceptions and experiences. Sociological Research Online 26(1): 222– 239. Zhou X, Li J and Jordan LP (2019) Parental intent for children to study abroad: The aole of educational aspiration and children’s characteristics. Cambridge Journal of Education 49(6): 789–807. Zimmermann T (2020) Social influence or rational choice? Two models and their contribution to explaining class differentials in student educational aspirations. European Sociological Review 36(1): 65–81. Author biographies Yasemin Nuhoglu Soysal is Research Professor of Global Sociology at WZB Berlin (https://www. wzb.eu/en/persons/yasemin-soysal), Professor of Sociology at Free University Berlin, and leading research member of the SCRIPTS Cluster of Excellence (https://www.scripts-berlin.eu). Prior to it, she was Professor of Sociology and member of the Center for Migration Studies and Human Rights Center at University of Essex, and John L. Loeb Associate Professor of Sociology and Faculty Associate of the Center for European Studies and the Center for International Affairs at Harvard University. Her research brings transnational and global perspectives to the study of the historical development and contemporary reconfigurations of the nation-state, citizenship, and human rights, with specific interest in the transnational transformations of the national. Héctor Cebolla Boado is Senior Scientist at the National Research Council, Institute of Economics, Demography and Geography (Madrid, Spain). He obtained his PhD, from the University of Oxford. He has been an associate professor of Sociology at UNED (Madrid, Spain), and a visiting professor at Pompeu Fabra University (Barcelona, Spain) and the University of Bielefeld (Germany). His research deals with the explanation of ethnic differentials in education and health among migrants and natives, both in countries of origin and destination. Date submitted 25 January 2023 Date accepted 7 June 2023 Appendix Table 2. Parental involvement; Factor analysis, Factor loadings. Variable Loadings Uniqueness Speak about child’s problems 0.4562 0.7773 Ask about their problems 0.5420 0.6998 Supervise their developments 0.5581 0.6824 Impose restrictions on their behavior 0.4172 0.8085 Eigenvalue: 1.00; Difference: 0.84. Soysal and Boado 647 Table 4. Distribution of variables involved in the empirical analyses. N Mean SD Min Max Considers IE 4306 0.24 0.42 0.00 1.00 Female 4306 0.48 0.50 0.00 1.00 Primary education 4306 0.31 0.46 0.00 1.00 Secondary education 4306 0.59 0.49 0.00 1.00 Tertiary education 4306 0.10 0.30 0.00 1.00 Age 4306 7.35 4.37 0.00 14.00 Household income/1000 4306 2.23 2.89 0.00 50.00 One child 4306 0.16 0.37 0.00 1.00 Mean test scores 1639.00 −0.00 0.87 −3.43 2.98 Transnational exposure index 4300 −0.00 0.98 −0.56 3.59 Inventions 4300 5.09 10.11 0.04 28.26 Foreign investments 4300 0.04 0.05 0.00 1.36 Websites 4300 0.01 0.02 0.00 0.21 Foreign visitors 4300 0.00 0.00 0.00 0.02 Exportsimports 4300 2.30 3.71 0.03 12.41 Index of density of higher education 4300 0.17 0.16 0.00 1.00 GDP per capita 4300 24,942.55 13,152.41 2292.00 69,165.00 Speak about child’s problems 1941 2.85 1.18 1.00 5.00 Ask about their problems 1940 2.01 0.90 1.00 5.00 Supervise their developments 1940 2.84 1.35 1.00 5.00 Impose restrictions on their behavior 1941 2.54 1.12 1.00 5.00 Parental involvement 1939 0.01 0.59 −1.74 1.17 IE: international education; GDP: Gross domestic product. Table 3. Exposure to the transnational; Factor analysis, Factor loadings. Variable Loadings Uniqueness Chinese higher education density index 0.1610 0.4842 Number of Patent Applications for Inventions of Industrial Enterprises above Designated Size (piece) 0.9225 0.0205 Number of Foreign Visitors/Population 0.9538 0.0355 Websites/Population 0.9204 0.0417 Foreign investment/GDP 0.8559 0.1615 Exports and imports/GDP 0.9520 0.0135 Eigenvalue: 4.27; Difference: 3.53. 648 Sociological Research Online 29(3) Table 5. Linear probability models with province fixed effects. (1) (2) (3) (4) Female 0.00085 (0.013) −0.0066 (0.021) −0.0045 (0.019) 0.00079 (0.013) One child 0.034 (0.018) 0.021 (0.027) 0.0013 (0.026) 0.035* (0.018) Parental education Secondary 0.051*** (0.015) 0.035 (0.023) 0.017 (0.021) 0.050*** (0.015) Tertiary 0.15*** (0.026) 0.058 (0.046) 0.092* (0.041) 0.14*** (0.027) Household income 2nd decile 0.00046 (0.032) −0.036 (0.050) 0.0053 (0.047) −0.0020 (0.032) 3rd decile 0.0095 (0.027) 0.0019 (0.042) 0.013 (0.038) 0.0058 (0.027) 4th decile 0.030 (0.025) −0.0050 (0.039) 0.0088 (0.036) 0.029 (0.025) 5th decile 0.013 (0.025) −0.030 (0.039) 0.016 (0.037) 0.015 (0.025) 6th decile 0.0052 (0.025) −0.055 (0.039) −0.017 (0.037) 0.0081 (0.025) 7th decile 0.023 (0.029) −0.020 (0.046) −0.0019 (0.042) 0.030 (0.029) 8th decile 0.013 (0.027) −0.029 (0.045) 0.032 (0.040) 0.019 (0.027) 9th decile −0.0045 (0.028) −0.031 (0.047) −0.028 (0.041) −0.0042 (0.028) 10th decile 0.096** (0.030) 0.081 (0.052) 0.080 (0.048) 0.098** (0.031) Test scores 0.051*** (0.013) Parental involvement 0.055*** (0.016) Transnational Exposure −0.064*** (0.014) Exposure*exposure 0.029*** (0.0056) Constant 0.17*** (0.020) 0.21*** (0.032) 0.19*** (0.030) 0.14*** (0.021) N 4306 1639 1939 4300 F 6.31 2.91 2.33 7.37 r20.019 0.024 0.017 0.025 *p < 0.05; **p < 0.01; ***p < 0.001. Soysal and Boado 649 Table 6. Inverse probability weighting with regression adjustment treatment effect models. Treatment: the effect of exposure to transnational; treatment status predictor: per capita GDP. (1) ATE 0.11*** (0.035) P0 mean 0.24*** (0.0096) Control group Female 0.017 (0.018) One child 0.021 (0.020) Parental education Secondary 0.085** (0.041) Tertiary 0.048* (0.029) Household income 2nd decile −0.035 (0.042) 3rd decile 0.018 (0.035) 4th decile 0.035 (0.035) 5th decile 0.049 (0.036) 6th decile 0.0095 (0.036) 7th decile 0.015 (0.043) 8th decile −0.027 (0.045) 9th decile 0.019 (0.043) 10th decile 0.066 (0.051) Constant 0.18*** (0.029) Treatment group Female Treatment group 0.071 (0.068) One child 0.037 (0.075) Parental education Secondary 0.16 (0.14) Tertiary 0.055 (0.094) Household income 2nd decile 0.077 (0.19) 3rd decile −0.18 (0.092) 4th decile 0.16 (0.15) 5th decile −0.031 (0.13) 6th decile 0.066 (0.13) 7th decile 0.27 (0.15) 8th decile −0.076 (0.10) 9th decile −0.063 (0.15) 10th decile 0.24 (0.13) Constant 0.23** (0.094) Treatment equation GDP per capita 0.00021*** (0.0000076) Constant −4.89*** (0.18) N 4300 GDP: Gross domestic product; ATE: Average treatment effect *p < 0.05; **p < 0.01; ***p < 0.001.