Abstract
Extant research has examined women’s under-representation (relative to men) in international careers in respect to prejudice towards women in host countries, organizational resistance to selecting women, women’s own disinterest in taking international work and host country societal expectations of women. Although research has suggested the benefits for global organizations of having management diversity, the literature has given limited consideration to barriers and opportunities for international careers for women from developing countries. This study addresses this gap in examining the perceptions of 243 professional women in Sri Lanka about factors they perceive affect opportunities for international careers. The findings suggest the women perceive that positive female role models and family/husband support may facilitate opportunities, whereas prejudice in host countries, organizational gender discrimination and home country societal values emphasizing a primary responsibility of women as wives, mothers and daughters may hinder opportunities. The research has business implications in that, given a global ageing population and increasing numbers of nontraditional families, organizations seeking to achieve gender and ethnic diversity among their international cross-cultural managers need to be cognizant of supporting employees with extended family responsibilities.
Introduction
It has been argued that an increasingly globalized environment means managing widely dissimilar employee populations, markets, cultures and modes of work (Beechler and Woodward, 2009). Since the late 1980s, cross-cultural management researchers have identified that cultural diversity should be used as an advantage in global organizations (see Adler et al., 1986). Soderberg and Holden (2002) proffered that human resource management emerges as the main developer of cross-cultural management competencies across a whole range of international management – functions that plainly call for cross-cultural awareness. Moreover, companies with a higher representation of women in senior management positions are said to outperform companies with fewer women at the top (Catalyst, 2007; cited in Beechler and Woodward, 2009) and Tung (2004) pointed out that female expatriates may be model global managers. Despite such assertions, Adler (2002) asked how prepared are organizations to recognize their success depends on the most talented people in the world among their international managers and suggested the answer would appear to be not very. In the context of a war for talent (see Beechler and Woodward, 2009), women remain under-represented internationally relative to men. In 2015, women still only comprise 19 per cent of international assignees (BGRS, 2015) and women from developing countries are in fewer numbers. Addressing this under-representation of women (and women from developing countries particularly), which limits diversity of the international management cohort, necessitates understanding the factors that prevent women from developing countries participating in international careers.
This article focuses on a developing country, Sri Lanka, and reports the results of 243 responses to a survey of Sri Lankan professional women, which addressed the research question: What factors do Sri Lankan professional women perceive as hindering or assisting their international career opportunities?
In exploring Sri Lankan women’s perceptions of factors that may limit or facilitate their international career opportunities, our study contributes to cross-cultural management research in providing insight into how both perceived host country context and home country contextual factors affect participation in international careers. The research is positioned within the women in international management literature which has explored individual, organizational and national drivers and barriers to women’s international careers.
The organizational analysis has considered whether there is corporate resistance to selecting women for international assignments, individual analysis has examined the extent of personal (dis)interest in undertaking international work, and host country analysis has explored whether organizations or individuals perceive women will not be well received as managers in certain locations and experience foreigner prejudice. Our research also explores how societal expectations of women’s roles may interact with foreigner prejudice, corporate resistance and their own (dis)interest limiting their international careers.
The research is significant for two reasons. First, the study broadens extant literature by examining how home country societal values have bearing on women’s international career opportunities and how this may interact with perceived foreigner prejudice, organizational resistance and individual (dis)interest. Second, it provides better understanding of facilitators and constraints on international career opportunities for women from a developing country, which is important for organizations seeking to maximize gender, ethnic and cross-cultural diversity of their international managerial workforces.
Sri Lanka has been chosen for this study of a developing country because of women’s achievements relative to many of its neighbouring countries (see Malhotra and DeGraff, 1997). The majority of Sri Lankans live in poverty working in low-skilled jobs in agriculture and development (including gender development) is relatively low by world standards (World Economic Forum, 2011). Many women have suffered greatly following 30 years of civil war, leaving many widowed, financially destitute, lacking work skills and facing social stigma and isolation given widow (withawi) has ‘connotations of a deplorable and pitiable condition’ (De Soysa, 2000). Yet, compared to some of its Asian regional neighbours (e.g. Bangladesh and Pakistan), Sri Lanka has performed impressively on human development indicators since 1948, particularly for women. Sri Lanka has high education levels, literacy and maternal health and low infant mortality rates (Thurairajah and Baldry, 2010). High levels of education augur well for women’s managerial prospects and international employability. This does need to be considered though against a backdrop of a society which is patriarchal, has conventional gender practices (Fernando and Cohen, 2013) and could be viewed as conservative in emphasizing women’s caring responsibilities as wives/mothers/daughters as taking precedence over careers. Regardless of Sri Lankan women’s ethnicity (Sinhalese, Tamil and Moors), they are socialized into gendered behavioural norms about how to dress, think, behave and be respectable or good women (Adikaram, 2014). The intersection of gender with race and class further constrains opportunities for some women as women in poorer families and/or from rural areas do not enjoy equal access to education and hence career prospects. Moreover, where families cannot afford to support all children through education, sons are likely to be favoured (see Thurairajah and Baldry, 2010; Wickramagamage, 2012). It has been argued that some Sri Lankan women may suffer a triple disadvantage of gender, class and caste (Wickramagamage, 2012). In particular, workforce participation of women in the Northeast, mainly Tamils (who are primarily Hindu) and Sri Lankan Moors (who are primarily Muslim), has been relatively low. One possible explanation might be ‘cultural norms of Tamil and Muslim [Moors] women engaged in household work and income generation within the home’ (Wanasundera, 2006: 2).
In examining Sri Lankan professional women’s perceptions of international career opportunities, we use our own broad definition as including organizationally assigned or self-initiated, commuter, short-term or long-term assignments of a project-based, fixed or continuing duration (excluding migration). The research did not intend to examine the types of international work that women might undertake but to investigate the perceived issues hindering or assisting them in undertaking international careers in any form.
Literature review
Research on barriers to women in international management
Adler’s (1984a, 1984b, 1994) pioneering research explored three key reasons why women may be under-represented among international assignees: corporate resistance to selecting women; a perception women would suffer foreigner prejudice in the host country, especially in countries markedly culturally different from their own country; and women’s own lack of interest. Research since the 1980s (summarized in a comprehensive review by Altman and Shortland, 2008) has continued to analyse national, organizational and individual reasons why women remain under-represented in international careers. Recent research has also considered impact of home country culture. Key contributions of this research are now presented.
Host country context/foreigner prejudice
The perceived impact of the host country environment has been subject to some debate in the literature since consideration was first given to whether foreigner prejudice in host country locations would make women ineffective in international management roles (see Adler, 1994). Foreigner prejudice has been largely refuted with research having found that women are not necessarily treated differently than their male counterparts in international roles (see Stroh et al., 2000; Taylor and Napier, 1996). Adler (1994) suggested, as international managers, women are viewed as foreigners first and women second and not necessarily subject to the same discrimination which might be experienced by local women – being a woman might even be an advantage for an international posting in that the women (being in smaller numbers than their male colleagues) are visible and thus included. Napier and Taylor (2002) called for a more nuanced assessment of women’s acceptance in certain locations and argued the level of economic development and education within the host country may determine foreign women’s acceptance as managers. Traavik and Richardsen (2010) said there are women who ‘are pushed out of their home countries due to political or economic conditions and are forced to relocate’ (p. 2799) and argued whether organizational transferees or independent women face challenges in being foreign and women.
Organizational policies and practices
Adler’s research (1984a, 1984b) considered that much of the perceived disadvantage for women in international careers reflected corporate resistance towards women, which resulted in men primarily being selected. Research has found many organizations only have informal selection processes that tend to favour men who are the decision-makers (Harris, 2001; Harris and Brewster, 1999; Menzies, 2009). These practices suggest organizations enforce an ‘expatriate glass ceiling’ (see Insch et al., 2008), which limits opportunities for women in a similar manner as the glass ceiling in domestic organizations. Further, a challenge for women even if selected for international positions is that whilst working internationally they do not have the same career development activities, organizational support or mentoring as male colleagues (Linehan et al., 2008). This argument was reinforced by Hutchings et al. (2010) who noted cultural barriers to after-hours socializing can also make it challenging for women from some cultures to actively participate in informal networking overseas (which affects promotion/being headhunted for other positions) or at home which impacts on selection for international assignments. Further, Altman and Shortland (2011) proffered women’s lack of international experience impacts on their ability to reach the highest levels of organizations domestically, which implies a double dose of discrimination in career attainment.
Women’s own motivation
Research also explored whether women had limited international management opportunities due to their own lack of interest (Adler, 1984a) but has found women’s under-representation in international careers may not be due to lack of interest in the work or belief in their own capabilities but rather may reflect concerns for family commitments. Extensive research has focused on obstacles for international relocation of dual-career couples and the importance of both spousal support and organizational support to facilitate relocation (see, for instance, Fischlmayr and Kollinger, 2010; Mäkelä et al., 2011; Välimäki et al., 2009). Moreover, children may restrict women’s ability to translate what might be a general interest in expatriation into practice (see Tharenou, 2008). Resch (2003, cited in Fischlmayr and Kollinger, 2010) claimed work–life balance research has traditionally focused on the challenges of reconciling family and work obligations for married women with children, but Fischlmayr and Kollinger (2010) argued single employees also experience conflict between their work and family roles. Significantly expatriate research is almost absent in considering relocation of other family beyond spouse and children (see Haslberger and Brewster, 2008), even though the Global Mobility Trends survey (BGRS, 2015) now considers whether assignees have accompanying parents whilst working internationally.
Home country societal values
Recent research has explored the role of home country societal values and argued that whilst women may have an interest in international careers, a lack of societal support for them to undertake international careers can affect their willingness to accept international work. The limited research in this field has considered women in Arab Middle Eastern countries with findings having suggested one of the greatest obstacles for the women to take international work is lack of social support (Hutchings et al., 2010). Despite increased emphasis on higher education in the Middle East, gender equality in senior roles in business and government has not eventuated (Madichie, 2013) and even short-term international assignments/work trips can prove difficult, given societal expectations that women should be foremost committed to roles as wives and mothers (Hutchings et al., 2011). In other research, Shortland (2009: 378) has considered both home and host country contexts and revisited the question of host country prejudice against foreign managers in stating that a promising theoretical lens for further research will involve undertaking explorations of the ‘home country cultural prejudice’ angle, namely why women’s reception abroad would be perceived negatively when research suggests the contrary. Insch et al. (2008) also noted that for women to secure international assignments, it is important to recognize the cultural values of their upbringing which act as agents in career development.
Having examined the extant women in international management research, we now consider the specifics of Sri Lankan culture and society in relation to women’s work.
Sri Lankan culture and society
Sri Lankans share some cultural traits with other Asian countries (De Alwis, 2002) and mix this with Western management practices such as individual performance rewards (see Chandrakumara, Glynn, Gunathilake and Senevirathne, 2010). Although Hofstede did not examine Sri Lanka, Mediwake (cited in Hofstede and Hofstede, 2013), in a two-country study of China and Sri Lanka, suggested Sri Lanka could be scored as follows: power distance 72, individualism 28, masculinity 35, uncertainty avoidance 53 and long-term orientation 49. No major studies of cultural values (including Hofstede and the Globe Project) have included Sri Lanka and much written about Sri Lankan values is from Sri Lankan researchers. There is no published analysis of gender differences in relation to cultural dimensions for Sri Lanka, thus this warrants urgent academic attention.
Nanayakkara (1999) revealed that the sociocultural system consists of caste, ethnicity, status, religion and education and is based on different philosophical foundations which exist among the ethnic groups within Sri Lankan society. Sri Lanka is more focused on quality of life, and although it is predominantly a status-oriented society and people ascribe status to jobs and organizations (Chandrakumara, et al., 2010), this may not result in higher level positions for women.
Women and work in Sri Lanka
Sri Lanka is expected to achieve gender equality sooner than most of its neighbours (World Economic Forum, 2011). In respect to a gender gap overall, of 136 countries, Sri Lanka was ranked 55 with women’s economic participation and opportunity ranked 109, and educational attainment ranked 48. This contrasts with Bangladesh 75, 121 and 115, India 101, 124 and 120, and Pakistan 135, 135, and 129 (World Economic Forum, 2011). In Sri Lanka’s patriarchal society, women are viewed primarily as care givers. This expectation presents challenges for women working full-time and a moral code restricts them from being out alone at night or working late hours (Fernando, 2012). Submissiveness and unobtrusiveness are regarded as important attributes of women in South Asia generally, and women are viewed as being more tolerant of gender stereotyping than in the West (Thanacoody et al., 2006, cited in Fernando, 2012). De Alwis (2002) argued education of women, work outside the home, calls for women’s political rights and political office are perceived as potential threats to women’s ‘traditional’ roles within Sri Lanka.
Women were first mentioned in Sri Lankan government planning documents as a target group for development programmes in 1970. Successive governments have emphasized women’s progress as an important aspect of national development, but such efforts do intersect with traditional values and class and caste issues. In 1971, a Standardization Policy was introduced to universities as an affirmative action programme for students from areas which previously had poor educational facilities. Jayawardena and Jayaweera (cited in Hewamanne, 2006) argued that the significant contribution of women in agriculture and home industries, particularly, is perceived as extensions of their domestic role, and rural development programmes and vocational training programmes merely reinforce gender stereotypes. Young women dominate in the export processing zones, and liberalization and privatization of government business has pushed many workers into the informal sector (Samaratunge and Nyland, 2007), which may include both rural and urban workers across a range of industries and occupations but is characterized as being dominated by non-formal, non-standard employment practices (Rasseedin, 2002).
Many women have sought self-employment, particularly those widowed, during the civil war. Ayadurai (2006) suggested that these women make an important contribution to economic growth but need assistance from government and non-government agencies alike in expanding their entrepreneurial capabilities and benefitting other women through expansion of businesses. Although Sri Lankan women have gained participation through government and non-government development organizations, Thurairajah and Baldry (2010) suggested government policies are gender neutral but not gender sensitive. This means that whilst policy suggests equity in treatment, women have not been actively empowered through training/development opportunities or mentoring specifically for women’s needs (see Thurairajah and Baldry, 2010). In 1960, Sri Lanka became the first country to have a female prime minister. However, this has not translated into ongoing high participation rates of women in politics and senior public administration. Whilst women may have more opportunities in family-owned businesses (Kodagoda and Samaratunge, 2015), generally in the private sector, women are unable to enter middle levels of management with ordinary level qualifications (high school), and it is said this leads to a greater degree of disadvantage for women (Thurairajah and Baldry, 2010). Educational opportunities for women are dependent upon family situation and, where there may be a number of children in a family, males will be given preference in education, particularly in rural areas. Whilst this is true for many developing economies, the cost of education affects women disproportionately in relation to class, caste and ethnicity (see Jayewardene, 1995).
Research on Sri Lankan women working internationally has highlighted the ill treatment of women in low-skilled employment in foreign factories and as household help (Thurairajah and Baldry, 2010) in areas such as the Middle East (see Tidball, 2011), yet the literature has not specifically explored Sri Lankan professional women and their opportunities for international careers. This research gap likely results from the limited opportunities women have in middle and senior management roles domestically and thus restrictions to move into positions which may provide access to international careers. The current research is significant in identifying the factors that Sri Lankan women perceive as facilitating or impeding international careers.
Methods
Research design
A quantitative approach is suitable when the research problem calls for ‘(a) the identification of factors that influence an outcome, (b) the utility of an intervention, or (c) understanding the best predictors of outcomes’ (Creswell, 2009: 18). As this study examines the factors that hinder or assist Sri Lankan professional women’s international career opportunities, a quantitative methodology was adopted. A survey was deemed the most appropriate tool for collecting data from a large number of respondents and to provide an initial examination of Sri Lankan professional women’s perceptions of issues that hinder or assist their international career opportunities and followed an approach undertaken in many other studies within the women in international management literature.
The survey was four pages in length including a section on demographic data. It was expected to take approximately 30 min to complete. The questionnaire included items measured on a Likert scale to examine issues of foreigner prejudice, organizational resistance and women’s own (dis)interest. Additionally, two multi-response questions required the respondents to identify factors that they believed were barriers or opportunities for international careers.
Procedure and access
One of the researchers, based in Sri Lanka, contacted senior managers of organizations across a range of industries with a request to distribute the questionnaire to professional women in their organizations. The organizations’ names were accessed via lists publicly available through professional associations and websites. Where management of the organizations agreed to participate, the questionnaire was provided. Self-stamped envelopes addressed to the researcher were also provided to the managers of participating organizations who then made these available to all professional women. Management were asked to request the women complete the questionnaire at their convenience and return it anonymously by post to the locally based researcher. After receiving the survey by email, the women printed the survey, completed the hard copy and posted it in a public mailbox. Respondents were advised not to provide their names or other identifying information. At the completion of the data collection process, the Sri Lankan-based researcher provided the questionnaires to the other researchers to undertake the data analysis.
Sample
Of 320 potential respondents, 243 returned usable questionnaires (all questions were answered). Such a high response rate may suggest potential bias in that the views are those of women who have a particular interest in international work. Alternately, the response rate could be attributed to Sri Lanka’s high power distance culture (see Kailasapathy and Metz, 2012), meaning women responded because of management’s request. The Sri Lankan professional women 1 were primarily employed in the banking/finance/insurance, health and education industries in Colombo (Central Bank of Sri Lanka, 2014). Only nine of the respondents said they had any prior international work experience (mostly of 1 year or less and mostly in other Asian countries). Demographic information is provided in Table 1.
Demographic data of respondents.
Measurements
Perceived foreigner prejudice
This scale used items adapted from those developed by Stroh et al. (2000) and Vance and Paik (2001). The scale consisted of two items with a 5-point Likert-type scale, ranging from 1 (strongly disagree) to 5 (strongly agree) and one item on a 7-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree) to test the Sri Lankan women’s perceptions of foreigner prejudice. An example item was ‘Host country cultures disadvantage Sri Lankan women’. The Cronbach α coefficient was 0.55. Since this scale had a small number of items, it is appropriate to report the mean inter-item correlation, especially because it was common to find low Cronbach α value (Pallant, 2007). In this case, as shown in Table 2, the mean inter-item correlation value was 0.29, with values ranging from 0.19 to 0.49. This is an optimal range according to Briggs and Cheek’s (1986) recommendation of 0.2 to 0.4 (Pallant, 2007: 95).
Summary item statistics: Perceived foreigner prejudice.
Personal disinterest in international work
This scale used items adapted from those developed by Stroh et al. (2000) and Vance and Paik (2001). The scale consisted of seven items with a 5-point Likert-type scale, ranging from 1 (strongly disagree) to 5 (strongly agree) to test the Sri Lankan women’s personal disinterest in international work. An example item was ‘I am less likely to accept an international assignment because I have a husband’. Cronbach α coefficient of this scale was 0.73.
Perceived corporate resistance
This scale used items developed by Stroh et al. (2000). This scale included three items on a 5-point Likert-type scale, ranging from 1 (strongly disagree) to 5 (strongly agree), to test the women’s perceptions of corporate resistance. An example item was ‘Sri Lankan companies are hesitant to send women on international assignments’. Cronbach α coefficient of the scale was 0.60. Since this scale had a small number of items, it is appropriate to report the mean inter-item correlation, especially because it was common to find low Cronbach α value (Pallant, 2007). Table 3 presents the mean inter-item correlation value of this variable. In this case, the mean inter-item correlation (M) was 0.34, with values ranging from 0.26 to 0.46. This is nearly an optimal range according to Briggs and Cheek’s (1986) recommendation of 0.2 to 0.4 (Pallant, 2007: 95).
Summary item statistics: Perceived corporate resistance.
Multi-response items
In order to gain further insight into the women’s views and explore the impact of societal values, we included two multi-response, tick-a-box items. The first item asked respondents to choose from five options, if applicable, in respect to barriers perceived to have limited their international career opportunities. This question was adapted and expanded from a question used in a study by Hutchings et al. (2010). The second item, allowing for a choice of five options, if applicable, was as follows: Are there any cultural factors which you believe may have assisted your international work opportunities? This question was developed based on earlier literature (as discussed in the literature review), which identified factors that may facilitate opportunities. Both questions allowed respondents to list any other factors they viewed as pertinent.
Data analysis
We used one-way between-groups analysis of variance (ANOVA) 2 to test whether age, education level, occupational level and industry were associated with perceived foreigner prejudice, personal disinterest in international work and perceived corporate resistance. We calculated the Pearson product–moment correlation coefficients 3 to test whether the demographic variables of marital status, having children under 18 and whether raised in a country or city and company size were associated with these. We also used the Pearson correlation coefficients to determine whether personal disinterest in international work and perceived corporate resistance were correlated with each other.
The two multi-response questions were examined in terms of number of respondents selecting (multiple) factors hindering or assisting international career opportunities.
Results
Demographic variables and perceived foreigner prejudice
Age and perceived foreigner prejudice
A one-way between-groups ANOVA was conducted with age as the independent variable and perceived foreigner prejudice as the dependent variable. Subjects were divided into three age groups (group 1: 18–34; group 2: 35–49; and group 3: 50–65). Preliminary assumption testing was conducted to check for normality, linearity, univariate and multivariate outliers and multicollinearity, with no serious violations noted. Levene’s test was not significant, F(2, 227) = 1.37, p > 0.05, indicating the assumption of homogeneity of variance was not violated. There was no significant relationship between age and perceived foreigner prejudice, F(2, 227) = 1.77, p > 0.05.
Marital status, whether has children under 18 whether raised in a country or city, company size and perceived foreigner prejudice
The relationships between marital status, children under 18 and whether raised in a country or city area and perceived corporate resistance was conducted using Pearson product–moment correlation coefficient (Table 4). Preliminary analyses were performed to ensure no violation of the assumptions of normality, linearity and homoscedasticity. There was a small, positive correlation between marital status and personal disinterest in international work, r = 0.16, n = 236, p < 0.05.
Correlations of demographic variables, perceived foreigner prejudice, personal disinterest and perceived corporate resistance.
*Correlation is significant at the 0.05 level (two-tailed).
**Correlation is significant at the 0.01 level (two-tailed).
Education level and perceived foreigner prejudice
ANOVA was conducted with education level as the independent variable and perceived foreigner prejudice as the dependent variable. Subjects were divided into three education level groups (group 1: PhD; group 2: masters/MBA/MPhil; group3: MD/MBBS; group 4: bachelor; and group 5: secondary school). Preliminary assumption testing was conducted to check for normality, linearity, univariate and multivariate outliers and multicollinearity, with no serious violations noted. Levene’s test was not significant, F(4, 212) = 1.51, p > 0.05, indicating the assumption of homogeneity of variance was not violated. There was no significant relationship between the two variables, F(4, 212) = 1.09, p > 0.05.
Occupational level and perceived foreigner prejudice
ANOVA was conducted with occupational level as the independent variable and perceived foreigner prejudice as the dependent variable. Subjects were divided into three occupational level groups (group 1: director/owner/professional and senior management; group 2: middle management; and group 3: supervisor). Preliminary assumption testing was conducted to check for normality, linearity, univariate and multivariate outliers and multicollinearity, with no serious violations noted. Levene’s test was not significant, F(2, 200) = 1.36, p > 0.05, indicating the assumption of homogeneity of variance was not violated. A significant relationship between the two variables was found, F(2, 200) = 5.22, p < 0.05.
Post hoc comparisons using Tukey’s honestly significant difference (HSD) test (Table 5) indicated the mean score for group 3, M = 3.31, SD = 0.64, was significantly higher than group 1, M = 2.878, SD = 0.80. Group 2 (M = 3.05, SD = 0.70) did not differ significantly from either group 1 or group 2. This indicated women working as supervisors perceived significantly more foreigner prejudice than women holding positions of director, owner or professional/senior management.
Multiple comparisons: Occupational level as independent variable.
*The mean difference is significant at the 0.05 level.
Industry and perceived foreigner prejudice
ANOVA was conducted with industry as the independent variable and perceived foreigner prejudice as the dependent variable. Subjects were divided into three industry groups (group 1: banking, insurance and finance; group 2: education; and group 3: others). Preliminary assumption testing was conducted to check for normality, linearity, univariate and multivariate outliers and multicollinearity, with no serious violations noted. However, Levene’s test was significant, F(2, 195) = 9.38, p < 0.05, indicating the assumption of homogeneity of variance was violated. Therefore, Welch and Brown–Forsythe statistics should be used in this circumstance. As shown in Table 6, the significance value of these (Welch’s F = 10.64 and Brown–Forsythe’s F = 11.74) were both <0.05. This indicates a significant relationship between industry and perceived foreigner prejudice.
Robust tests of equality of means.
aAsymptotically F distributed.
Post hoc comparisons using Tukey’s HSD test (Table 7) indicated the mean score for group 1, M = 3.25, SD = 0.62, was significantly higher than group 2, M = 2.61, SD = 0.94, and group 3, M = 2.91, SD = 0.62. Group 2 did not differ significantly from group 3. This means that women working in banking, insurance and financial sectors significantly perceived more foreigner prejudice than those working in education and other sectors.
Multiple comparisons: Industry as independent variable.
*The mean difference is significant at the 0.05 level.
Demographic variables and personal disinterest in international work
Age and personal disinterest in international work
ANOVA was conducted with age as the independent variable and personal disinterest in international work as the dependent variable to determine whether there were any significant differences between the means of different age groups. Subjects were divided into three age 4 groups (group 1: 18–34; group 2: 35–49; and group 3: 50–65). Preliminary assumption testing was conducted to check for normality, linearity, univariate and multivariate outliers, and multicollinearity, with no violations noted. Levene’s test was significant, F(2, 236) = 4.71, p < 0.05, indicating the assumption of homogeneity of variance 5 was violated. Therefore, Welch and Brown–Forsythe statistics were used in this circumstance to determine whether there was a significant relationship between age and personal disinterest. As shown in Table 8, the significance value of these (Welch’s F = 3.71 and Brown–Forsythe’s F = 3.35) were both <0.05. This indicates a significant relationship between age and personal disinterest in international work.
Multiple comparisons: age as independent variable.
*The mean difference is significant at the 0.05 level.
Post hoc comparisons using Tukey’s HSD 6 test indicated the mean score for group 3, M = 3.10, SD = 0.76, was significantly higher than group 1, M = 2.79, SD = 0.58. Group 2 (M = 2.91, SD = 0.81) did not differ significantly from either group 1 or group 3. This means women aged from 50–65 significantly showed less interest in international work compared with those aged from 18–34.
Marital status, whether has children under 18, whether raised in a country or city, company size and personal disinterest in international work
The relationships between marital status 7 , children under 18 8 , whether raised in a country or city area 9 and personal disinterest in international work was conducted using Pearson product–moment correlation coefficient (Table 4). There was a small, positive correlation between marital status and personal disinterest, r = .16, n = 236, p < 0.05. This indicated married women were less interested in international work than single women.
There was also a small, positive correlation between the size of organization and personal disinterest, r = .14, n = 222, p < 0.05l, indicating women employed in organizations with 500+ employees were less interested in international work than women in an organization with fewer than 500 employees.
Education level and personal disinterest in international work
ANOVA was conducted with education level as the independent variable and personal disinterest in international work as the dependent variable to determine whether there were any significant differences between the means of different education level groups. Subjects were divided into five education level groups (group 1: PhD; group 2: masters/MBA/MPhil; group3: MD/MBBS; group 4: bachelor; and group 5: secondary school). Levene’s test was not significant, F(4, 220) = 0.57, p > 0.05, indicating the assumption of homogeneity of variance was not violated. There was no significant relationship between the two variables, F(4, 220) = 2.22, p > 0.05.
Occupational level and personal disinterest in international work
ANOVA was conducted with occupational level 10 as the independent variable and personal disinterest as the dependent variable to determine whether there were any significant differences between the means of different occupational level groups. Subjects were divided into three occupational level groups (group 1: director/owner/professional and senior management; group 2: middle management; and group 3: supervisor). Levene’s test was not significant, F(2, 207) = 0.29, p > 0.05, indicating the assumption of homogeneity of variance was not violated. There was a significant relationship between the two variables, F(2, 207) = 7.16, p < 0.05.
Post hoc comparisons using Tukey’s HSD test indicated the mean score for group 3, M = 3.19, SD = 0.70, was significantly higher than group 1, M = 2.78, SD = 0.74 and group 2 (M = 2.78, SD = 0.70). Group 2 did not differ significantly from group 1. This indicated women working as supervisors were less interested in international work than women holding a managerial position.
Industry and personal disinterest in international work
ANOVA was conducted with industry 11 as the independent variable and personal disinterest in international work as the dependent variable to determine whether there were any significant differences between the means of different industry groups. Subjects were divided into three industry groups (group 1: banking, insurance and finance; group 2: education; and group 3: others). Levene’s test was not significant, F(2, 202) = 0.10, p > 0.05, indicating the assumption of homogeneity of variance was not violated. There was a significant relationship between the two variables, F(2, 202) = 11.88, p < 0.05.
Post hoc comparisons using Tukey’s HSD test indicated the mean score for group 1, M = 3.08, SD = 0.67, was significantly higher than group 2, M = 2.65, SD = 0.73 and group 3 (M = 2.53, SD = 0.71). Group 2 did not differ significantly from group 3. This indicated women in banking/insurance/finance were less interested in international work than women in education, health and other sectors.
Demographic variables and perceived corporate resistance
Age and perceived corporate resistance
ANOVA was conducted with age as the independent variable and perceived corporate resistance as the dependent variable to determine whether there were any significant differences between the means of different age groups. Subjects were divided into three age groups (group 1: 18–34; group 2: 35–49; group 3: 50–65). Levene’s test was not significant, F(2, 233) = 2.32, p > 0.05, indicating the assumption of homogeneity of variance was not violated. There was no significant relationship between age and perceived corporate resistance, F(2, 233) = 0.42, p > 0.05.
Marital status, whether has children under 18, whether raised in a country or city, company size and perceived corporate resistance
The relationships between marital status, children under 18, whether raised in a country or city area and perceived corporate resistance was conducted using Pearson product–moment correlation coefficient (Table 4). Preliminary analyses were performed to ensure no violation of the assumptions of normality, linearity and homoscedasticity. No correlations were found between these demographic variables and perceived corporate resistance.
Education level and perceived corporate resistance
ANOVA was conducted with education level as the independent variable and perceived corporate resistance as the dependent variable to determine whether there were any significant differences between the means of different education level groups. Subjects were divided into three education level groups (group 1: PhD; group 2: masters/MBA/MPhil; group3: MD/MBBS; group 4: bachelor; and group 5: secondary school). Levene’s test was not significant, F(4, 217) = 0.87, p > 0.05, indicating the assumption of homogeneity of variance was not violated. There was no significant relationship between the two variables, F(4, 217) = 1.64, p > 0.05.
Occupational level and perceived corporate resistance
ANOVA was conducted with occupational level as the independent variable and perceived corporate resistance as the dependent variable to determine whether there were any significant differences between the means of different occupational level groups. Subjects were divided into three occupational level groups (group 1: director/owner/professional and senior management; group 2: middle management; and group 3: supervisor). Levene’s test was not significant, F(2, 204) = 0.66, p > 0.05, indicating the assumption of homogeneity of variance was not violated. No significant relationship between the two variables was found, F(2, 204) = 0.47, p > 0.05.
Industry and perceived corporate resistance
ANOVA was conducted with industry as the independent variable and perceived corporate resistance as the dependent variable to determine whether there were any significant differences between the means of different industry groups. Subjects were divided into three industry groups (group 1: banking, insurance and finance; group 2: education; and group 3: others). Levene’s test was not significant, F(2, 199) = 0.26, p > 0.05, indicating the assumption of homogeneity of variance was not violated. No significant relationship between the two variables was found, F(2, 199) = 0.38, p > 0.05.
Correlation between perceived foreigner prejudice, personal disinterest and perceived corporate resistance
The relationship between perceived foreigner prejudice, personal disinterest in international work and perceived corporate resistance was conducted using Pearson product–moment correlation coefficient (Table 4). Preliminary analyses were performed to ensure no violation of the assumptions of normality, linearity and homoscedasticity. There was a small, positive correlation between personal disinterest in international work and perceived corporate resistance, r = .14, n = 237, p < 0.05, with high levels of perceived corporate resistance associated with higher level of personal disinterest in international work. There was a medium, positive correlation between personal disinterest in international work and perceived foreigner prejudice, r = 0.35, n = 231, p < 0.05, with high levels of perceived foreigner prejudice associated with higher level of personal disinterest in international work.
Multi-response questions
Respondents were able to select multiple (provided) items in respect to perceived barriers and opportunities. Respondents also had the option to mention other issues besides the selectable items, but of 243 respondents, only 69 did so. In descending order of the number of times mentioned, the perceived barriers included family commitments (children/other relatives), lack of family/husband support, lack of organizational selection for roles, lack of female mentors and stereotypes about female managers. Other factors that were mentioned by some respondents can be summarized, in descending order of number of times mentioned, as societal expectations regarding women’s primary responsibility for children/relatives/husbands, women’s own commitment as being firstly to children or parents, lack of opportunities in the organization or for women at low levels, political influence/favouritism or top management selecting only friends and family, perceived societal unsuitability for unmarried women to move abroad, preferring to raise children in Sri Lanka and their own lack of interest in working internationally.
In respect to perceived support, in descending order the factors included family support, social attitudes/values, husband’s support, role models among family/friends and other. Other responses, in descending order of the number of times mentioned were changing social attitudes, being motivated by friends’ international experiences, their husband first taking an international assignment, having a father already working abroad, support from local government, education, language and having an interest in working internationally.
Discussion
Host country environment/foreigner prejudice
The results found Sri Lankan women in supervisory positions were more likely to expect to experience foreigner prejudice than were those in higher level positions, and women working in banking/insurance/finance perceived more foreigner prejudice than those in other sectors. There were no other significant relationships between demographic factors and perceptions of the host country environment. Although some earlier research suggested female expatriates face the challenge of being foreign and a woman (Traavik and Richardsen, 2010), other research has said women do not suffer disadvantage in host countries, and this included a study of expatriate women from India, another South Asian country (see Varma et al., 2006). Moreover, visibility in being a woman may actually be an advantage in an international position (Adler, 1994). It might be expected that being from a developing country would give the women additional visibility among the female international managers who are mostly from the developed world. It is feasible though that this added visibility could have been perceived negatively by the respondents and reflect double disadvantage of gender and ethnicity. The women in lower level positions may have felt they would be stereotyped when working in expatriate locations where many people from Sri Lanka are employed in low-skilled and service roles. Given that some respondents mentioned that it would not be seen as socially suitable for single women to move overseas on their own, this likely reinforces a view that women may feel insecure outside their home country, particularly in contexts that are very cultural distant from Sri Lanka and have differing views about women’s roles and positions in work and nonwork environments. As the banking/insurance/finance industry is highly internationalized, it can be surmised the women expected prejudice because of male domination.
Organizational policies
In respect to corporate resistance of organizations in selecting women, or alternately, facilitating international career opportunities, the data showed that overall women perceived that organizations in Sri Lanka generally provided few international opportunities for women. This reflects the political influence/favouritism/gender discrimination which women referred to and limitations in women’s domestic opportunities as having a flow-on effect to international opportunities. This finding supports research on international glass ceilings (Insch et al., 2008) and discrimination in international assignment selection (Harris, 2001; Menzies, 2009). Importantly, the results found a strong relationship between corporate resistance and women’s own interest, which could mean that women lacked interest if they thought organizations would not consider them for international careers. The findings indicate that organizations provide insufficient peer support and mentoring for women. Consistent with Linehan et al. (2008), we suggest that for Sri Lankan women to have knowledge of managing cross-culturally relies on there being more female role models in senior positions and with international experience, particularly for more junior women and those in the banking/finance sector who also perceived the likelihood of foreigner prejudice when working internationally.
Women’s own interest
The results provided considerable evidence of women’s own interest as affecting whether they would actually undertake international careers with several important relationships between demographic data and women’s personal lack of interest in international work. These included that women aged 50–65 indicated less interest in international work than those 18–35, married women showed less interest than single women, those with secondary school education showed less interest than those with higher education, supervisors showed less interest than those in higher level positions, those in large organizations showed less interest than those in smaller and medium-sized organizations and those in banking/finance/insurance were less interested than those in other industries.
Some of these results reinforce earlier literature suggesting married women (and particularly those with children) would be less likely to undertake international work (Tharenou, 2008), and it might be expected that women in lower level positions and with lower levels of education might have less awareness of international career opportunities. Older women may be coming towards the end of their work life and thus considering retiring rather than relocating for work. It is interesting that those in larger organizations expressed less interest, given these organizations are likely to have more international opportunities and a greater probability of providing support.
The responses reflect that women may view their primary commitment to focus on children/parents/family – consistent with societal values (see Fernando, 2012). Given the importance of extended family in the context of a collectivist culture, women may be reticent to work internationally, preferring their children are raised close to grandparents and other family members. Whilst earlier research has suggested the importance of spousal support for successful international assignments (Mäkelä et al., 2011, Välimäki et al., 2009), in this cultural context, lack of support and their husband not already being chosen for an international assignment are factors in women not even considering international careers.
Interestingly, there was a small positive correlation between personal disinterest and perceived corporate resistance and a stronger positive correlation between personal disinterest and perceived foreigner prejudice, both suggesting women’s disinterest was related to whether they thought they would experience foreigner prejudice (small perception) and whether they thought organizations would even provide them with opportunities (strong perception).
Home country societal values
Sri Lankan values, including being highly collectivist and masculine (Mediwake cited in Hofstede and Hofstede, 2013) and having gender stereotypes (Adikaram, 2014; Fernando, 2012; Hewamanne, 2006) means women are expected to follow societal roles. Responsibilities for children/parents/husbands were seen to limit international careers and arguably may also impact on opportunities that might be provided by an organization. Thus, this research underpins other research suggesting women’s international opportunities are impeded by societal values emphasizing family commitments as a woman’s primary focus (Hutchings et al., 2010; Hutchings et al., 2011). Whilst research has indicated women’s child-caring responsibilities may impact on disposition to undertake international careers (Tharenou, 2008), our research extends this in arguing that home country societal values with an expectation to care for not only children but other relatives limits women’s international careers. The respondents also mentioned home country issues that might assist women to work internationally, namely increasing support from families for women’s education and greater family involvement in caring responsibilities, being motivated by friends’ positive international experiences and a broader social agenda of greater economic and political representation of women.
Conclusions, contributions, implications and future research
Conclusions
The findings disagree with some earlier research refuting foreigner prejudice in suggesting that women in lower positions/some industries in a developing economy do perceive they will experience prejudice working in a host country. The findings reinforce earlier research that women perceive disadvantage in organizational selection for international positions and do not have strong role models for such work. The findings support earlier research in suggesting that whilst women may be interested in international work, their willingness to do so is affected by child-rearing. However, the research is significant in extending this literature to consider that national societal expectations of their responsibilities to other family members also hinders them working in international positions.
Theoretical contributions
In extending the extant literature from a focus largely on women from developed countries to a study of women in a developing country, our research reinforces earlier literature suggesting organizational resistance (and lack of supporting policies) as well as women’s own commitment to family may limit women’s international careers, but our research refutes earlier research in suggesting that women in this developing economy do expect to experience foreigner prejudice. Our research is most significant though in finding that, whilst women may indeed be interested in international careers, home country societal expectations to care for elderly parents and other relatives constrain their opportunities. Thus, our research indicates that perceived barriers to women’s international career opportunities need to be viewed as interrelated, not siloed. That is, it is critical to move beyond organizational and individual issues to also consider home country societal values and how these may impact on organizational resistance/support and women’s own (dis)interest.
Practitioner implications
Enhancing women’s global mobility and increasing the international pool of educated and cross-culturally ready talent depends upon organizations providing policies that redress bias in selection and offering training and development to better position women for global roles. This research highlights that women may not consider international careers until they perceive that there is more home country support to do so, and thus we argue that changes in societal and organizational views about women’s roles in Sri Lanka need to occur in concert in order to open up international career opportunities.
The findings indicate that there is need for global organizations to support a wider range of family situations when relocating women internationally. Although Haslberger and Brewster (2008) noted that few people on international assignments would have accompanying family other than spouse or children, given an ageing population worldwide and growing numbers of family households comprising aged parents and other family members, single parent and blended families, organizations need to consider how to facilitate relocation of such families. Namely, organizations could provide extra resources for physical relocation of extended families as well as ongoing household carer assistance for various family members whilst expatriates are working/travelling. Doing so would assist organizations to maximize gender, ethnic and national diversity of cross-cultural managers.
Limitations and issues for future research
The research extends theory on women in international management by not only analysing previously examined barriers but also providing more insight into how home country societal values might affect women’s international career opportunities. We recommend future interview-based research to allow for more in-depth consideration of factors that may affect women’s perceptions of undertaking international careers such as perceived foreigner prejudice, language skills, educational opportunities and extended family commitments.
The research was limited in being conducted in one region of Sri Lanka in which the respondents may have been predominantly Sinhalese, and thus our research did not allow for an analysis of class and ethnic differences. Many of the women identified as being raised in an urban area, and though some noted challenges for rural women, the findings may not adequately reflect rural women’s views and their employment opportunities, education or foreign language ability. Thus, future research might directly consider the influence of geographic and ethnic variations in women’s perceptions of factors influencing international careers.
Footnotes
Notes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.