Community, Work & Family,

Vol. 12, No. 3, August 2009, 299
312

RESEARCH ARTICLE
Work and family divided across borders: the impact of parental
migration on Mexican children in transnational families
Claudia Lahaie*, Jeffrey A. Hayes, Tinka Markham Piper, and Jody Heymann

Institute for Health and Social Policy, McGill University, 1130 Pine Ave. West, Montreal, QC
H3A 1A3, Canada
(Received 30 June 2008; final version received 19 March 2009)

Using new data collected in high emigration communities within Mexico, we

explore the impact of partial family migration on children left behind in Mexico.
Multivariate results suggest that households where respondents have a spouse
who was a caregiver and who migrated to the USA are more likely to have at least
one child with academic, behavioral, and emotional problems than non-migrant
households. This finding supports efforts to decrease the need for families to cross
borders either by decreasing the economic necessities for migration or by
designing immigration policies aimed at decreasing the separation of families
across borders and increasing family support after a caregiver’s departure to the
USA. The end goal of these efforts and policies is to improve children’s health and
well-being in communities with high levels of migration.
Keywords: migration; transnational families; family well-being; children’s well-
being; children’s health; caregiving

À partir de nouvelles données collectées au Mexique auprès de communautés

résidant dans des zones où la migration est très forte, nous explorons l’impact de
l’émigration partielle d’une famille aux Etats-Unis sur le bien-être des enfants
demeurés au Mexique. Les résultats de modèles multivariés suggèrent que les
foyers ayant des répondants avec un époux qui avait plusieurs tâches parentales
avant sa migration aux Etats-Unis ont plus de chance d’avoir au moins un enfant
avec des problèmes au niveau académique, comportemental et émotionnel que les
foyers qui ne sont pas touchés pas la migration. Ces résultats appuient les mesures
visant à diminuer les besoins des familles à traverser les frontières telles la
diminution des mauvaises conditions économiques qui poussent à émigrer ainsi
que l’augmentation du support apporté aux familles suite au départ pour les
Etats-Unis d’un parent ou de toute autre personne qui assumait beaucoup de
responsabilités parentales. L’objectif ultime pour tous ces efforts et ces politiques
est l’amélioration de la santé et du bien-être des enfants vivant dans des
communautés à forts courants migratoires.
Mots-clés: migration; familles transnationales; bien-être des familles; bien-être des
enfants; santé des enfants; charge parentale

Mexicans represent by far the largest immigrant group in the USA (Massey, Alarcon,
Durand, & Gonzalez, 1987). Between 1997 and 2006, the legal inflow of Mexicans to

*Corresponding author. Email: [email protected]

ISSN 1366-8803 print/ISSN 1469-3615
# 2009 Taylor & Francis
DOI: 10.1080/13668800902966315
http://www.informaworld.com
300 C. Lahaie et al.

the USA totaled more than 1.6 million people (US Department of Homeland Security,
2007). In addition, millions more have immigrated to the USA without documentation
in order to better meet their family’s basic needs. As the border has tightened,
undocumented immigrants are increasingly remaining in the USA and not returning to
their home country. Before the 1986 Immigration Reform and Control Act (IRCA), the
average probability of return migration was estimated at 45%; currently, this rate has
decreased to approximately 25% (Massey, 2005, p. 9). The result of this migration trend
is the creation of transnational families, families who are divided between countries,
one of origin and one of employment. Research shows that transnational families are
predominantly families with low education and low income (Frank & Hummer, 2002)
who seek a better future by sending members to work in the USA.
Despite the presence of this large transnational movement, few large representative
studies have examined child outcomes in Mexican sending communities, defined as
those families with children remaining in Mexico who have at least one family member
who migrated to the USA. The main findings in this emerging literature describe the
financial effects of migration on children’s health (Frank & Hummer, 2002;
Kanaiaupuni & Donato, 1999) and educational outcomes (Kandel & Kao, 2001;
Massey, Goldring, & Durand, 1994; Mier, Rocha, & Romero, 2003). Very few studies
have examined the consequences of parental availabilities on children’s outcomes.
Also absent from this literature are studies related to the emotional and behavioral
outcomes of Mexican children living in transnational families.
The driving question of this study is: when families experience separation from
adult members due to migration, what impact does it have on children living in these
changing family environments? To answer this question, this study examines whether
migration significantly impacts Mexican children’s academic, emotional, and
behavioral outcomes. Using data we collected from a nationally representative
sample of Mexican sending families, the objectives of this paper are twofold. First,
the paper seeks to determine the impact of a caregiver’s migration to the USA on
(their) children’s academic and emotional experiences. Second, the paper examines
the influence of care settings and working conditions of remaining adult family
members on emotional health and educational outcomes of children whose parents
migrated to the USA. To our knowledge, this is the first study that specifically
examines the role of the caregiver in Mexican transnational families, those who
remain with their children in Mexico, as well as the ones who migrated to the USA.
This paper proceeds as follows. First, there is a review of the relevant research on
transnational families and on parental involvement. Second, both the data and
methodology used are explained. Third, descriptive tabulations and regression
models that examine the impact of migration on children’s academic/behavioral
outcomes, emotional problems, and drop-out rates are presented.

Influence of parental involvement and availability on child outcomes

Extensive research on parental involvement (Coleman et al., 1966; Epstein, 1983;
Gordon, 1979; Henderson, 1987) suggests the importance of parental availability on
the health and educational well-being of children. Parental engagement in their
children’s education improves children’s achievement, even after the students’ ability
and family socioeconomic status are taken into account (Epstein, 2001). For children
who are at risk academically or behaviorally, the availability of parents to meet with
 Community, Work & Family 301

teachers and specialists and to address their children’s development and emotional
problems is particularly important (Heymann & Earle, 2000). One study showed that
parental involvement reduces by a third of a standard deviation the gap in
mathematic scores between children of immigrants and children of American-born
parents when entering kindergarten (Lahaie, 2008). In addition to educational
achievement, children who reside in nurturing and responsive environments also
benefit in social and cognitive growth (Garmezy, 1993). Research on Latino families
also identifies the important role of family involvement on the social and cognitive
performance of youth and children (Belksy & MacKinnon, 1994; Solis, 1995).
Children who live in transnational families may experience the absence of at least one
of their parents. Since parental presence is critical for child well-being, the absence of
the parent(s) combined with the decrease in the time available to the other parent or
family members remaining in Mexico may have negative consequences on children
living in transnational families.

Mexican family member migration and child outcomes

Despite the increase in the phenomenon of temporary migration from Mexico to the
USA, few studies have examined its impact on the well-being of children remaining
in their home country. Most research on temporary migration/transnational families
focuses on the impact of remittances on child outcomes (Frank & Hummer, 2002;
Kanaiaupuni & Donato, 1999; Kandel & Kao, 2001). Kandel and Kao (2001)
examined the impact of a family member’s labor migration on children’s educational
outcomes. They found that while remittances may allow children to continue
attending school and help them perform better, they may also reduce their
motivation to attend further years of schooling beyond the average number of
years. Along with Mier et al. (2003), they also found that the impact of remittances is
only significant in primary school, as these facilitate school completion by removing
the need for children to drop out and engage in work and contribute to the
household income. However, a separate study, on children between 10 and 13 years
old who resided in small cities with mothers with low education levels, found that
remittances were not significant on schooling completion (Borraz, 2005). Apart from
remittances, other factors influence the educational achievement of children living in
transnational families. In fact, Kandel and Kao (2000) found that the opportunity
for migration appears to be a stronger influence on children’s career aspirations than
the additional economic resources provided by remittances that would allow the
achievement of educational goals. The greater the migration of family members or
community members to the USA, the more migration becomes instituted as a norm
(Kanaiaupuni & Donato, 1999; Massey et al., 1994), and the less children are
motivated to seek additional schooling (Kandel & Massey, 2002).

Current study
The emerging research on transnational families primarily focuses on the impact of
remittances on the well-being of the family members who remain in Mexico. Studies
that extend beyond remittances have primarily focused on one community or state.
Our investigation furthers the research on transnational families in Mexico and in
the USA. It is the first study to gather data to examine the effects of a caregiver’s
302 C. Lahaie et al.

migration to the USA, and care and working conditions in Mexico on the well-being
of Mexican children. Second, in addition to expanding our understanding of the
impact of migration on the educational outcomes of Mexican children, this study
will also address a gap in the literature by looking at the impact of migration on the
emotional health of these children. Third, existing nationally representative samples,
such as the Mexican Migration Project (MMP) and the Mexican Census Bureau
(INEGI) provide information on migration but do not focus on family well-being.
Other samples, which include data on health and education, are either not based on
representative samples or limited to one state. We have designed the Transnational
Working Families Survey in order to collect information regarding migration and
households’ ability to meet family needs.

Data
In 2004, after 12 years of in-depth studies on related issues in the USA, Mexico, and
other nations, the Project for Global Working Families designed a survey of 1509
Mexican households regarding migration and family well-being. With field admin-
istration by Berumen y Asociados, we surveyed all municipalities in Mexico that had
50,000 or more people with at least 20% of the households reporting the migration of
a family member from Mexico to the USA. The municipalities fitting these criteria
were Comonfort, Guanajuato (22% migrant households); Romita, Guanajuato (23%
migrant households); Ixmiquilpan, Hidalgo (21% migrant households); La Barca,
Jalisco (22% migrant households); and Rio Grande, Zacatecas (24% migrant
households). Multistage cluster disproportionate sampling was used to obtain
equivalent numbers of respondents from each municipality including migrant and
non-migrant households.
Households with children under the age of 15 and a primary caregiver who
regularly received an income from a salaried or informal job were eligible for
enrollment. Migrant households were defined as households with a member who had
gone to live in the USA in the last five years. Through in-person interviews information
was collected on basic demographic characteristics of all members of the household, as
well as on health and educational outcomes, working conditions, caregiving, and
experiences of international migration. Questionnaire items were piloted in a range of
international settings, including Mexico, prior to the implementation of the survey.
The average length of the interviews was one hour. In each household, primary
caregivers (94% female) were interviewed because of their knowledge of the children in
the household and of caregiving patterns. To verify accuracy in data collection, a
supervisor revisited 10% of the households. A double-blind data entering process was
used to minimize data entry errors. Households were revisited four times before being
classified as non-responding. During the study period, 1509 of 1861 selected
respondents were successfully interviewed, with a response rate of 81%. As a result
of our sampling frame, 755 of the 1509 households surveyed had at least one family
member who had migrated to the USA in the last five years.1

Methods
We use multivariate logistic regression models to address our three research
questions. We look at both the effect of migration and the availabilities of the
 Community, Work & Family 303

caregiver who remains in Mexico on children’s educational and emotional outcomes.

Because municipalities were the primary sampling unit in the clustered sample
design, in all analyses we correct confidence intervals for the non-independence of
observations within municipalities.
The basic logistic regression equation is
Oi IMi b1 Carei b2 WCi b3 HHi go i
where i represents the respondent in each household, Oi refers to the outcomes for
children of respondent i (indicating whether at least one of the respondent’s children
has: (a) academic and behavioral problems, (b) emotional problems, or (c) dropped
out), IMi represents the immigration variables, Carei represents care settings
variables, WCi represents the working condition variables, while HHi stands for
household characteristics which we are controlling for, and oi the error term.

Variables used in model

To address the research question posed above, we focus on three outcomes:
behavioral and academic problems, early school exits or dropping out, and emotional
problems. The first two outcomes refer to children’s educational achievement and the
third is used as an indicator for emotional health. These outcomes were obtained
from the respondents’ answers to the following three questions:
1. Have any of the children you care for ever had any trouble at school

behavioral or academic?
2. Have any of the children you care for dropped out of school?
3. Do any of the children you care for have emotional problems?
We examine four migration variables. The first migration variable indicates
whether or not a caregiver migrated to the USA in the last five years, the second
variable identifies whether a spouse has migrated to the USA in the last five years,
the third variable was created to include migrant household members who were both
a spouse and a caregiver, and the final migration variable in the model is monthly
remittances. In this paper, remittances are defined as the amount of money a migrant
living in the USA at the time of interview is sending to family members in Mexico on
a monthly basis.
Our analyses considered a set of five care setting variables. These five variables
represent the different kinds of care arrangements: (a) respondent care, (b) informal
adult care (relative and non-relative care), (c) formal care (such as preschool), (d)
sibling care, and (e) self-care. Three variables measuring the primary caregiver’s
working conditions were examined in the context of their relationship with the
children’s outcomes: (a) number of hours worked per week, (b) whether or not it is
very hard to take time off from work, and (c) whether or not the respondent is self-
employed. Since 72% of the sample is self-employed, we decided to use the variable
‘very hard to take time off’ as an alternative for ‘leave of absence’, to better represent
the working reality of this sample.
Based on the findings from the literature, for each of our models, we control for
a set of household characteristics that is expected to affect our outcome measures
and may also be related to migration status. Household characteristics include
primary caregiver’s gender, age, marital status (married or partner, never married,
304 C. Lahaie et al.

separated/divorced/widowed), years of education of both respondent and spouse,

household income, number of adults in the household, three variables for the
number of children aged 0
5, 6
14, and 15
17, and one variable for whether or not
the respondent is taking care of an adult (at least 18 years old). We use one
missing data variable in our analyses for the spouse’s education to ensure that cases
missing on this covariate are not excluded from the sample (Allison, 2001). The
rate of missing data for spouse’s education is 10%.

Descriptive characteristics of the sample

Table 1 presents descriptive characteristics for the entire sample by immigration
status. It provides the mean difference between migrant and non-migrant households
and summarizes significance tests for these differences. As mentioned previously, half

Table 1. Sample description. Means and migrant
non-migrant differences in Mexican

household characteristics.

Full sample Migrant Non-migrant Difference

Migration
One family member migrated to 50.03 100.00 0.00 n/a
the USA
Care settings
Respondent care 57.46 57.35 57.56 0.08
Formal 2.32 1.85 2.79 1.20
Informal 25.65 26.62 24.67 0.87
Sibling care 7.69 0.07 0.08 0.59
Self-care 6.43 6.75 6.10 0.52
Working conditions
Respondent is self-employed 71.50 72.98 70.03 1.27
Days worked during a week 5.44 5.51 5.33 2.30
Hard to take time off to take care 74.95 74.17 75.73 0.70
Demographics
Marital status
Married/partnered 66.47 55.44 77.59 9.37*
Never married 9.01 7.55 10.48 1.99*
Separated/divorced/widowed 24.45 36.95 11.94 11.81*
Respondent’s number of years of 7.76 7.22 8.30 4.86*
schooling
Spouse’s number of years of schooling 7.93 7.20 8.44 4.27*
Number of adults 1.74 1.79 1.70 1.41
Number of children 0
5 years old 0.77 0.79 0.76 0.66
Number of children 6
14 years old 1.43 1.40 1.46 1.07
Number of children 15
17 years old 0.32 0.29 0.35 1.80
Family income per month in pesos 1895.7 1995.0 1796.3 0.78
Respondent cares for at least one adult 7.02 7.15 6.90 0.19
Notes: Data are given in percentages of the total sample of 1503 respondents. The significance test
(Difference) is a t-test with *pB0.05. Formerly married category includes respondents who have a spouse
living in the USA.
 Community, Work & Family 305

of the sample has at least one household member who migrated to the USA in the
last five years.
There were no statistically significant differences in working conditions of the
caregivers in Mexico or in childcare settings. In the overall sample, respondents work
five days a week and 75% of them find it very difficult to take time off to address
personal or family issues. In addition, 57% of the respondents are taking care of their
children themselves without any additional childcare/support. The second most
common care arrangement is informal care, defined as care provided by relatives and
non-relatives excluding childcare center and preschool, representing 26% of the
sample.
There were also no significant differences in the number of children or adults’
caregiving responsibilities. Overall, there are 7% of the respondents from the sample
who take care of at least one adult. The average family income is 1896 pesos per
month.
In the sample, migrant families differed from non-migrant families in their
number of years of schooling. On average, non-migrant respondents have 8.30 years
of schooling while migrant respondents have 7.22 years. The same case holds for the
spouse, migrant spouses (7.20 years) averaging fewer years of schooling than non-
migrant spouses (8.44 years).
Table 2 describes the migrant members from sample households. The 755
migrant households report sending 908 household members to the USA in the last
five years.
The majority of migrants were neither a caregiver nor a spouse before leaving for
the USA. However, 34% of migrants were spouses who did not provide care before
leaving the household. Only 3% provided care without being a spouse or partner, and
11% of the migrants were both a caregiver and a spouse before leaving for the USA.
Finally, the monthly remittances sent by the migrant household member amount to
309 pesos, representing one-sixth of the average family income for the whole sample.

Results
Among the migrant households, half had at least a member who left for the USA
who was neither a caregiver nor a spouse. The remaining half of the migrant sample
was composed of more than a quarter of migrant family members who were a
caregiver as well as a spouse. In the overall sample, 10% of respondents report that
one or more of their children have behavioral and academic problems at school. Five
percent believe that at least one of their children has emotional problems. Finally, 9%
have had a child leave school early or ‘dropped out’.

Role of the family member prior to migration

Results are shown in Table 3. Our findings demonstrate that the strongest predictor
of educational and emotional health outcomes of Mexican children is a caregiver
who left the house. A household with a caregiver-spouse (a caregiver married to a
respondent) who has migrated to the USA is 2.6 times as likely to have at least one
child with behavioral and academic problems as a household with a caregiver-spouse
who has not migrated. We also found that a household with a caregiver-spouse who
has migrated to the USA is 3.6 times as likely to have a child with emotional
306 C. Lahaie et al.
Table 2. Migrant Mexican household characteristics.

N %

Relationship of household member who left to the USA to respondent

Spouse or partner 337 37
Father 24 3
Mother 19 2
Child 197 22
Sibling 236 26
Total number of household members who left to the USA 908 100
Role of migrant in the household previous to his departure to the USA
Caregiver and spouse 84 11
Spouse only 253 34
Caregiver only 45 3
Neither caregiver nor spouse 526 52
Remittances per month in pesos 309

problems as a household where the caregiver-spouse has not migrated. Conversely,

households where a spouse who was not a caregiver has migrated (mostly males),
have a 61% decrease in the odds of having a child who is going to drop out of school.
However, remittances were not the driving force, since they were not found
significant for any outcome.

Role of care settings and working conditions

Two aspects of working conditions are significantly associated with children’s
outcomes. First, a household with a respondent having a very hard time taking time
off from work for personal or family issues is 2.6 times as likely to have at least one
child with emotional problems as a household which does not have a respondent
having this lack of flexibility. Second, each additional hour per day a parent works is
associated with 8% increase in the odds of having at least one child in the household
with behavioral and academic problems.
The impact of care settings and working conditions on children’s academic and
behavioral outcomes, emotional problems, and drop-out rates is also documented in
Table 3. A household with children taking care of themselves at home in ‘self-care’ is
3 times as likely to have at least one child with behavioral and academic problems as
a household with children who are cared for by a respondent. A household having
children who take care of themselves is also more likely to have at least one child
dropping out and having emotional problems than a household with children who
are cared for by a respondent.

Role of household characteristics’ variables

Table 3 also presents findings which include household characteristics’ variables that
play a similar role to those from other studies on the educational outcomes of
Mexican children. First, respondents who are separated/divorced/widowed2 are 2.7
times as likely to have at least one child with behavioral and academic problems as
respondents who are either married or with a partner. This finding indicates that the
 Community, Work & Family 307
Table 3. Logistic regression on the effects of migration, caregiving, and work conditions on
children’s educational and mental health outcomes (odds ratio).

Academic/ Emotional Drop out

behavior problem
problem

Migration
One caregiver in USA (who is a spouse) 2.64** 3.59** 0.70
(1.27
5.48) (1.48
8.72) (0.28
1.77)
One caregiver in USA (who is not a spouse) 2.09 1.08 0.82
(0.67
6.54) (0.23
5.05) (0.26
2.62)
Spouse in USA (who was not a caregiver) 1.33 0.81 0.39*
(0.75
2.33) (0.38
1.75) (0.16
0.96)
Remittances per month (1000 pesos) 0.93 0.88 1.00
(0.78
1.10) (0.73
1.06) (0.86
1.17)
Care settings
Informal care (omitted: 1.51 1.57 0.87
respondent’s care)
(0.79
2.87) (0.73
3.40) (0.44
1.70)
Formal care 0.69 0.89 1.15
(0.10
4.69) (0.16
5.01) (0.11
12.39)
Sibling care 1.37 0.89 1.12
(0.64
2.93) (0.30
2.65) (0.49
2.59)
Self-care 3.08** 1.48 1.53
(1.42
6.71) (0.37
5.87) (0.58
4.02)
Working conditions
Hours worked per day 1.08* 1.07 0.99
(1.01
1.15) (0.98
1.18) (0.91
1.08)
Very hard to take time off 1.24 2.56* 1.35
(0.59
2.61) (1.14
5.75) (0.53
3.43)
Self-employed 1.29 1.10 0.88
(0.74
2.23) (0.50
2.41) (0.43
1.78)
Demographics
Age 1.00 1.00 0.97*
(0.98
1.02) (0.97
1.04) (0.94
1.00)
Never married (omitted: spouse/partner) 1.56 0.92 3.55
(0.32
7.56) (0.10
8.07) (0.60
20.89)
Separated/divorced/widowed 2.67* 2.22 2.84
(1.02
6.97) (0.46
10.74) (0.53
15.22)
Respondent’s education 0.95 0.99 0.91*
(0.89
1.01) (0.93
1.05) (0.84
0.99)
Spouse’s education 1.01 1.04 0.90*
(0.93
1.09) (0.94
1.15) (0.83
0.98)
Household income (1000 pesos) 1.00 1.01 1.00
(1.00
1.01) (1.00
1.01) (0.99
1.01)
No. of adults in household 0.79* 0.69 1.05
(0.66
0.96) (0.44
1.08) (0.84
1.31)
Care of adults 0.85 2.11 1.58
(0.39
1.88) (0.91
4.91) (0.61
4.07)
N 1503 1503 1503
**pB0.01; *pB0.05.
Notes: For each model, we also controlled for the respondent’s gender as well as the number of children
in the household aged 0
5, 6
14, and 15
17 years old.
308 C. Lahaie et al.

absence of a parent, whether due to legal separation or migration, is associated with

behavioral and academic problems for at least one child in the household.
On the other hand, the presence of each additional adult in the household is
associated with a 21% decrease in the odds of having at least one child in the
household with academic or behavioral problems.

Discussion
This investigation on transnational families studied the association between
migration and caregiver’s availabilities and the educational and emotional outcomes
of Mexican children. Our focus on the well-being of children in particular is designed
to address a gap in the emerging literature on transnational families.
These data strongly suggest that when a caregiver-spouse migrates to the USA,
children who remain behind in these migrant households are the ones doing worse in
terms of academic, behavioral, and emotional outcomes. The negative effect of the
loss goes beyond the quality and availability of care arrangements, but the significant
association found between respondents having children taking care of themselves
(self-care) and having at least one child with academic and behavioral problems may,
in fact, exacerbate the effect of this loss.
These findings add to the existing literature on transnational families which finds
that the migration of a family member incurs substantial social and psychological
changes among the family who remains behind due to the increased care
responsibilities (Chaney, 1985; De Snyder & Neily, 1993). They also support the
extensive literature on parental involvement emphasizing the importance of parental
availability on the health and educational well-being of children (Coleman et al.,
1966; Epstein, 1983; Gordon, 1979; Henderson, 1987) and the literature on non-
migrant working families which finds that working conditions impact children’s
educational and health outcomes (Heymann, 2000, 2006). However, the specific
impact on the well-being of children who remain in transnational families had
previously not been studied. Results from these data provide important policy
implications specific to the reality of children living in transnational families. To help
families try to stay together, there is a need to increase job opportunities in local
communities, decrease the need for caregivers to migrate, or increase opportunities
for families to migrate as a whole.
Second, our results also suggest that, overall, adequate support for children and
good working conditions make a difference on the well-being of children who live in
transnational families. Children taking care of themselves (self-care) are three times
as likely to experience behavioral and academic problems as children in other care
arrangements. In addition, a household with a caregiver who has difficulties taking
time off from work for personal or family matters is far more likely to have a child
with emotional problems. Our hypothesis is that children’s lack of access to their
parents when they need them may negatively impact their mental health. Lastly,
households characterized by caregivers who work a large number of hours are
associated with having a child with behavioral and academic problems.
Specific policy implications derived from the findings are threefold. First, border
and economic policies which increase the likelihood of children being raised in
transnational families are damaging. Immigration policies which facilitate family
unification and economic policies that decrease the pressure to migrate for adequate
 Community, Work & Family 309

wages would both benefit children in the long term. Second, Mexican families would
benefit from an expanded number of public day care centers to support poor working
families and transnational families to alleviate the additional burden caused by a
caregiver’s absence and decrease the damaging educational and behavioral effects of
children taking care of themselves. Educational programs such as early childhood
centers and after-school programs have been shown to improve school readiness,
increase academic achievement, and decrease the likelihood of drop out (Crosnoe,
2007; Magnuson, Lahaie, & Waldfogel, 2006; Waldfogel & Lahaie, 2007). Third,
different approaches to motivate children to continue with their schooling when
faced with immediate migratory opportunities to the USA may be considered, such
as the expansion of the Oportunidades program. These conditional cash transfers
were found to benefit Mexican children’s educational achievement (De Janvry, Finan,
Sadoulet, & Vakis, 2006). Finally, for those who have an employer, policies that may
alleviate certain burdens in balancing work and caregiving responsibilities may be
considered: work flexibility (such as the possibility to take time off or to work fewer
days/hours, as well as the flexibility as to when starting and ending workdays) and
family leave (such as parental and sick leaves). For those who are self-employed it is
crucial to organize their working conditions. In that sense, there are global models
that have been successfully developed by collective organizations.3
The last significant contribution from this study is the association between a non-
caregiving spouse who migrated to the USA with a decrease in drop out for Mexican
children living in these migrant families. Why does the absence of migrant fathers
(who are not caregivers) help their children remain in school? Our data show that
remittances do not play a role since they were found not to be significant. It is
noteworthy that in this data set, migrant families are less educated than the non-
migrant ones. While our data set does not allow this to be examined, it seems plausible
that children may see firsthand the great costs to migrating and may be aware that a
higher education results in a greater probability of staying in Mexico. This
is consistent with results in the USA that link educational achievement with
household income (Brooks-Gunn, Duncan, & Maritato, 1997; Hanson, McLanahan,
& Thomson, 1997).
Our definition of the marital status category ‘separated’, which includes
respondents both separated legally or because of distance, may underestimate the
strength of the association between having one caregiver who moved to the USA and
having at least one child with academic and behavioral problems. In addition, our
definition of migrant household (five years) may underestimate the impact of
migration on these households, both because we do not know the impact of house-
hold members migrating before this five-year window and because in some of the
households in the migrant group, the migrating member had returned to Mexico by
the time of interview. Moreover, households in the non-migrant group may have had
members migrate to the USA outside of the five-year window used in the interview.
Each of these logically suggests a conservative bias to the migration effects we have
estimated which intimates that the effects of migration are larger than our estimates in
this study.
There are some limitations in this study. First, our use of cross-sectional data
raises the concern that the estimated associations may not reflect causal relation-
ships, but may be the result of omitted variable biases. While this is possible, our
regression models control for key household and economic characteristics. Second,
310 C. Lahaie et al.

surveys can contain self-report or recall bias. The likelihood of this is mitigated by
the type of outcomes respondents are asked to recall and the limited recall period.
Further studies could be enriched by the availability of testing to identify subtle
differences.
Studying transnational families is more relevant than ever as the trend of families
divided across borders continues to expand and may have serious negative
consequences on children and family outcomes as shown in this study and previous
ones (Kandel & Kao, 2001; Kandel & Massey, 2002). We need to better understand
the needs and problems of this growing population. If we can increase our
understanding of how to best support Mexican transnational families, we will be
better situated to develop effective policies and targeted programs to meet their needs.
Additional research is required to examine the adaptive processes transnational
families undergo, especially when it is a caregiver-spouse who moves to the USA.

Acknowledgements
The research presented in this paper was made possible due to a grant from the Rockefeller
Foundation. The authors would like to thank Edmundo Berumen for his valuable advice
regarding the fielding of the survey. We are indebted to Andrea Diaz Varela for her research
assistance.

Notes
1. To be considered a migrant household, at least one member had to migrate to the USA in
the last five years (between 1999 and 2004). Each sampled household that had a recent
migrant was selected for the full interview while one in three of the sampled households
meeting the other requirements was asked to complete an interview.
2. It is important to note that most respondents who have a spouse living in the USA reported
themselves as ‘separated’. The separated/divorced/widowed category includes both
respondents separated legally or because of distance.
3. For more information, refer to examples such as programs from the Women in Informal
Employment: Globalizing and Organizing (WIEGO).

Notes on contributors
Claudia Lahaie received her PhD in Social Work from Columbia University. She is currently a
Postdoctoral Fellow at the Institute for Health and Social Policy where she studies the impact
of family environment on the educational and health outcomes of children of immigrants.

Jeffrey Hayes received his PhD in Sociology from the University of Wisconsin
Madison. His
research focuses on inequality and well-being across the life course. He is currently a Research
Associate at the Institute for Health and Social Policy, McGill University.

Tinka Markham Piper received her MPH and MSW degrees from Columbia University.
Currently, she is the Student Programs Manager at the Institute for Health and Social Policy
at McGill University and directs the Institute’s research internship and policy fellowship
programs.

Jody Heymann is founding director of the Institute for Health and Social Policy and founding
director of the Project on Global Working Families. Heymann holds a Canada Research Chair
and is a Professor in the faculties of Medicine and Arts at McGill University.
 Community, Work & Family 311

References
Allison, P. (2001). Missing data. Thousand Oaks, CA: Sage.
Belksy, J., & MacKinnon, C. (1994). Transition to school: Developmental trajectories and
school experiences. Early Education and Development, 5(2), 106
119.
Borraz, F. (2005). Assessing the impact of remittances on schooling: The Mexican experience.
Global Economy Journal, 5(1), 1
30.
Brooks-Gunn, J., Duncan, G.J., & Maritato, N. (1997). Poor families, poor outcomes: The
well-being of children and youth. In G.J. Duncan & J. Brooks-Gunn (Eds.), Consequences of
growing up poor (pp. 1
17). New York: Russell Sage.
Chaney, E.M. (1985). Women who go and women who stay behind. Migration Today, 10, 7
13.
Coleman, J.S., Campbell, E.Q., Hobson, C.J., McPartland, J., Mood, A.M., Weinfeld,
F.D., et al. (1966). Equality of educational opportunity. Washington, DC: US Department
of Health and Human Services.
Crosnoe, R. (2007). Early child care and the school readiness of children from Mexican
immigrant families. International Migration Review, 41(1), 152
181.
De Janvry, A., Finan, F., Sadoulet, E., & Vakis, R. (2006). Can conditional cash transfer
programs serve as safety nets in keeping children at school and from working when exposed
to shocks? Journal of Development Economics, 79(2), 349
373.
De Snyder, S., & Neily, V. (1993). Family life across the border: Mexican wives left behind.
Hispanic Journal of Behavioral Sciences, 15, 391
400.
Epstein, J.L. (1983). Longitudinal effects of family
school
person interactions on student
outcomes. Research in Sociology of Education and Socialization, 4, 101
128.
Epstein, J.L. (2001). School family, and community partnerships: Preparing educators and
improving schools. Boulder, CO and Oxford: Westview Press.
Frank, R., & Hummer, R. (2002). The other side of the paradox: The risk of low birth weight
among infants of migrant and nonmigrant households within Mexico. International
Migration Review, 36(3), 746
765.
Garmezy, N. (1993). Children in poverty: Resilience despite risk. Psychiatry, 56(1), 127
136.
Gordon, I.J. (1979). The effects of parent involvement in schooling. In R.S. Brandt (Ed.),
Partners: Parents and schools (pp. 4
25). Alexandria, VA: Association for Supervision and
Curriculum Development.
Hanson, T.L., McLanahan, S., & Thomson, E. (1997). Economic resources, parental practices,
and children’s well-being. In G.J. Duncan & J. Brooks-Gunn (Eds.), Consequences of
growing up poor (pp. 190
238). New York: Russell Sage.
Henderson, A. (1987). The evidence continues to grow: Parent involvement improves achieve-
ment. Columbia, MD: National Committee for Citizens in Education.
Heymann, S.J. (2000). The widening gap: Why working families are in jeopardy and what can be
done about it. New York: Basic Books.
Heymann, S.J. (2006). Forgotten families: Ending the growing crisis confronting children and
working parents in the global economy. New York: Oxford University Press.
Heymann, S.J., & Earle, A. (2000). Low-income parents: How do working conditions affect
their opportunity to help school-age children at risk? American Educational Research
Journal, 37(4), 833
848.
Kanaiaupuni, S.M., & Donato, K.M. (1999). Migradollars and mortality: The effects of
migration on infant survival in Mexico. Demography, 36(3), 339
353.
Kandel, W., & Kao, G. (2000). Shifting orientations: How U.S. labor migration affects
children’s aspirations in Mexican migrant communities. Social Science Quarterly, 81(1),
16
32.
Kandel, W., & Kao, G. (2001). The impact of temporary labor migration on Mexican
children’s educational aspirations and performance. International Migration Review, 35(4),
1205
1231.
Kandel, W., & Massey, D. (2002). The culture of Mexican migrant: A theoretical and empirical
analysis. Social Forces, 80(3), 981
1004.
Lahaie, C. (2008). School readiness of children of immigrants: Does parental involvement play
a role? Social Science Quarterly, 89(3), 684
705.
Magnuson, K., Lahaie, C., & Waldfogel, J. (2006). Preschool and school readiness of children
of immigrants. Social Science Quarterly, 87(5), 1241
1262.
312 C. Lahaie et al.

Massey, D., Alarcon, R., Durand, J., & Gonzalez, H. (1987). Return to Aztlan. Berkeley:
University of California.
Massey, D.S. (2005, June 13). Backfire at the border. Why enforcement without legalization
cannot stop illegal immigration. Executive Summary. Trade policy analysis. Washington,
DC: CATO Institute.
Massey, D.S., Goldring, L., & Durand, J. (1994). Continuities in transnational migration: An
analysis of nineteen Mexican communities. American Journal of Sociology, 99(6),
1492
1533.
Mier, M., Rocha, T., & Romero, C.R. (2003). Inequalities in Mexican children’s schooling.
Journal of Comparative Family Studies, 34(3), 415
454.
Solis, J. (1995). Diversity of Latino families. In R. Zambrana (Ed.), Understanding Latino
families: Scholarship, policy, and practices (pp. 62
80). London: Sage.
US Department of Homeland Security. (2007). Yearbook of immigration statistics, 2006.
Washington, DC: US Government Printing Office.
Waldfogel, J., & Lahaie, C. (2007). The role of preschool and after-school policies in improving
the school achievement of children of immigrants. In J.E. Lansford, K. Deater-Deckard &
M.H. Bornstein (Eds.), Immigrant families in contemporary society (pp. 177
193). New
York: Guilford.