• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Demography. Author manuscript; available in PMC Mar 27, 2012.
Published in final edited form as:
PMCID: PMC3313449
NIHMSID: NIHMS358139

Children’s Experiences after the Unintended Birth of a Sibling

Abstract

This study examines whether children with a younger sibling whose birth was unintended experience larger declines in the quality of their home environment and larger increases in behavioral problems than children whose younger sibling’s birth was intended. We use data from the NLSY79 to estimate crosslag regression models that assess changes in the home environment and children’s behavioral problems after the birth of a sibling (intended or unintended). Results are consistent with our hypotheses, finding that, indeed, unintended births have negative spillover effects. Compared to children whose sibling’s birth was intended, both boys and girls whose sibling’s birth was unintended experienced larger declines in the quality of their home environment, and boys had larger increases in behavioral problems. We also find some unexpected evidence that mistimed births may have larger negative effects. This deserves further research, and we offer some possible explanations that could guide those investigations.

Unintended childbearing has been consistently associated with a host of negative health statuses for children and their mothers (Brown and Eisenberg 1995; Baydar et al. 1997a, 1997b). These have ranged from infant health problems and infant mortality, to maternal depression, low self-esteem, problematic parent-child relationships, and even child abuse (Axinn et al. 1998; Barber et al. 1999; Baydar et al. 1997a, 1997b; Brown and Eisenberg 1995; Kost et al. 1998; Najman et al. 1991; Salmon and Drew 1992; Zuravin 1991). Although the negative associations between unintended childbearing and child well-being are strong and long-lasting, there are questions about the causal nature of this association (Brown and Eisenberg 1995). Some argue that unobserved heterogeneity and selection factors produce the association between unintended pregnancy and unfavorable child outcomes (Joyce et al. 2000, 2002; Korenman et al. 2002). For example, using fixed-effects sibling models, Joyce and colleagues (2000) found that the birth weight and cognitive ability of children born from unintended pregnancies did not differ discernibly from their siblings born from intended pregnancies. They concluded that unintended pregnancy has little association with child well-being measures such as birth weight and cognitive ability. However, Joyce and colleagues also suggest that their estimates could be biased downward if there are negative “spill over” effects from the unintended pregnancy to the existing children.

The current paper specifically tests a spill-over hypothesis by examining whether the unintended birth of a sibling is more problematic for existing children than an intended sibling birth. Crosslag change models are used to analyze the overall quality of the home environment before and after the birth of a sibling. Specifically, we assess whether an unintended birth is associated with a disproportionately sharper drop in the quality of the home environment than an intended birth. This is similar to the approach adopted by Guo and VanWey (1999), who used change models to test whether spill-over effects altered older children’s cognitive test scores after the birth of a sibling. They proposed that if family size is causally related to an individual’s success, that individual’s test scores should change after the birth of a sibling. Similarly, we propose that if unintended childbearing is related to existing children’s environment, then the unintended birth should produce a change in the home environment. If such spill-over effects exist, it would suggest that there are unique effects for the family, including existing older siblings, resulting from an unintended birth (East 1998; East and Jacobson 2000). That is, the mechanisms linking unintended childbearing to child well-being are likely to apply to all children within a family, not just to the child born from the unintended pregnancy.

A second and related goal of this paper is to determine whether existing children experience greater increases in behavioral problems following the unintended birth of a sibling as opposed to an intended sibling birth. On average, children exhibit increases in acting-out and immature behaviors following a sibling’s birth (Stewart et al. 1987; Stewart 1990). However, it is unknown whether existing children experience disproportionately more behavioral problems following the birth of an unintended sibling.

There are multiple reasons we might expect children to experience a greater decline in the quality of the home environment and a greater increase in behavioral problems following an unintended birth rather than an intended birth. Mothers who have unintended births experience more depression compared to other mothers (Barber et al. 1999; Beck 2001; Ispa et al. 2007; Najman et al. 1991; Salmon and Drew, 1992). A substantial literature documents the reductions in parenting quality associated with maternal depression (e.g. see Gerdes et al. 2007).

In addition, parents experience stress and hardship from an unintended birth because they have not accumulated sufficient (or what they perceive as sufficient) social or financial resources for that child (Barber et al. 2003). High stress levels interfere with mothers’ patient and nurturing parenting (Anhalt et al. 2007; McLoyd 1990; McLoyd et al. 1994; Winraub and Wolf 1983). Children whose parents are experiencing stressful events exhibit more behavioral problems such as temper tantrums, quarrelsome, negativistic, and explosive behavior, relative to children of non-stressed parents (Elder et al. 1985; Pettit et al. 2001). Certainly, an unintended birth may impede a mother’s ability to be nurturing and responsive to her older children (Barber et al. 1999), which in turn could exacerbate existing children’s behavior problems. The current study examines whether children who experience the birth of an unintended sibling exhibit a greater decrease in the quality of the home environment and a greater increase in problematic behaviors than children whose sibling’s birth was intended.

Family Context and Sibling Constellation Factors

Intention status of a sibling’s birth may also be a mediator or a moderator in the relationship between other characteristics and quality of the home environment (or behavioral problems). For example, the consequences of the unintended birth of a sibling may depend on the family context in which the birth occurs. Or, one reason that families with fewer resources have lower home environment quality may be because they tend to have unintended births. In the current study, we consider various family contextual factors to determine whether the impact of the birth of an unintended sibling has more negative effects on existing children under specific family conditions, or whether intention status explains those factors’ relationship to the home environment and/or behavioral problems. Specifically, we consider birth spacing, sex composition and number of existing children, the family’s resources, and the quality of the home environment and children’s behavioral problems before the birth.

Closely spaced births have been linked to poor academic performance (Powell and Steelman 1990, 1993), and are more likely to be considered “mistimed” by parents. The birth of a closely spaced sibling may exacerbate the difficult adjustment to an unintended birth, and thus closely spaced unintended sibling births may be particularly detrimental to the availability of home resources and children’s behavioral adjustment. In addition, one reason why closely spaced births may be related to home resources and behavior may be because those births are less likely to be intended.

Sex composition and number of existing children are strong determinants of childbearing intentions, with most Americans preferring at least one boy and one girl (Pollard and Morgan 2002). However, the research on sex composition and birth order is mixed – some scholars posit a difficulty for existing children if the sibling is a girl (Butcher and Case 1994), others if the sibling is a boy (Powell and Steelman 1989, 1990), and still others if the sibling is a different sex (Conley 2000). Building on the demographic finding that families prefer one child of each sex, we hypothesize that having a sibling who is the same sex will exacerbate parents’ difficulty adjusting to an unintended birth. We further expect that same-sex siblings may compete more fiercely for the same parental resources, which may have a particularly important effect on behavioral problems (Behrman, Pollak and Taubman, 1986; Conley, 2004). The specific gender of the child also may be directly related to whether that child is considered unintended, as well.

Birth order has been linked to the extent to which children can successfully marshal family-level social resources, such as parental time, energy, and attention (Powell and Steelman 1990, 1993) as well as financial resources (Powell and Steelman 1995; Steelman and Powell 1991). Birth order has also been linked to personality characteristics that may be relevant to adjustment after the birth of a sibling (e.g., Sulloway 1996). Higher order births are particularly likely to be evaluated as unwanted by mothers.

A variety of types of family resources may be important buffers that facilitate its adaptation to an unintended birth. For example, a family with high income may be better able to hire extra help during the period of adjustment to the unintended birth. Or, a mother with strong cognitive ability may be better able to conjure solutions to the problems presented by an unintended birth. In addition, families with fewer resources are less able to plan their births, are consequently more likely to have unintended births, and this may partially explain the relationship between resources and their consequences for children.

The quality of the home environment and existing behavioral problems may also be related to the change in these factors simply because of the starting level – for example, families with a very low quality home environment may be less likely to experience a decline in home quality simply because there is not much room below the starting level. We do not have strong directional hypotheses about these factors, but investigate them to determine whether a pattern emerges.

Summary of Hypotheses

In sum, we have two major hypotheses:

  1. The home environment for an existing child will deteriorate more after the birth of an unintended sibling than after the birth of an intended sibling.
  2. Existing children will experience a larger increase in behavioral problems after the birth of an unintended sibling relative to the birth of an intended sibling.

We also hypothesize that these main effects may vary depending on pre-birth levels of the dependent variable, characteristics of the existing child, or of the birth. Finally, we hypothesize that intention status may mediate the effects of family characteristics on the quality of the home environment and behavioral problems.

Data

We use data from multiple waves of the 1979 National Longitudinal Survey of Youth. The NLSY79 is an ongoing survey initiated in 1979 for a nationally representative sample of 12,686 young men and women who were then 14 to 22 years of age.2 Data have been collected annually on these youth through 1994 and biennially since. In addition, beginning in 1986, information was collected bienially from the female sample about all children living in the household (NLSY-C). Thus, the NLSY79 combined with the NLSY-C is a rich source of information on the experiences of young women as they became mothers, and on the birth, childhood, and early adulthood of each of their children. We use data through the 2004 survey wave at which point over 80% of the women still eligible to be in the sample were interviewed.3 Births up to and including the year 2004 represent a cross-section of children born to the cohort of women who were between 39 and 47 years of age in 2004. The NLSY website reports that these children represent approximately 90% of possible births that will occur to the cohorts of women born between 1957 and 1964. This results in 2,976 sibling births to 2,317 focal children with valid measures of the quality of the home environment before and after the sibling birth, and 1,144 sibling births to 968 focal children with valid measures of behavioral problems before and after the sibling birth. Note that these sibling births include the next sibling (in terms of birth order) as well as higher order siblings.

Measures

Intention Status of the Birth

The key independent variable is the intention status of the younger sibling’s birth; we compare change among focal children who experience the unintended birth of a younger sibling to change among focal children who experience the intended birth of a younger sibling. In addition, we include a measure of the intention status of the focal child’s own birth, to account for the potentially different home quality and behavioral trajectories of focal children born from unintended births themselves.

About all of their recent births, women were asked, “Just before you became pregnant the (first, second, third, etc.) time, did you want to become pregnant when you did?” If they answered yes, the birth is classified as intended. If they answered “no,” they were asked, “Did you want a (nother) baby but not at that time, or did you want (none/no more) at all?” If they answered that they wanted another baby, their birth is classified as mistimed. If they answered that they wanted none/no more babies, their birth is classified as unwanted.4

For each model, we operationalize the intention status of the birth in two different ways – first as a dichotomy: unintended (either mistimed or unwanted) versus intended, and second as a trichotomy: intended, mistimed, or unwanted.5

Because women answered questions about their pregnancies and births in reference to the period since last interview, and not all women are interviewed at every survey, it is not a straightforward task to code the intention status of each birth. However, when comparing our coding to that of Joyce et al. (2000), we had identical coding for 95 percent of cases. For an additional 3 percent, we have a code where they had missing data. Of the remaining 2 percent of cases, either our data are coded ‘mistimed’ and theirs is coded ‘unwanted’ or vice versa, but we were still in agreement that the pregnancy was unintended.

We use these measures to include the intention status of the older child’s birth, as well. Descriptive statistics for this and all other variables we use in these analyses are presented in Table 1.

Table 1
Means and Standard Deviations of Measures Used in Analyses (N = 2,976)

Dependent Variables

We focus on two dependent variables: an indicator of the quality of the home environment, and mother’s report of a focal child’s behavioral problems. These measures are used as both dependent and independent variables – the measure from the interview before the birth is an independent variable, and the measure from the interview after the child’s birth is the dependent variable.

To represent an observation in each model, an existing child must have experienced the birth of a younger sibling and must have valid measures of the dependent variable from both the interview before and the interview after that birth. If a measure of the dependent variable was not available from the interview directly preceding the birth, it was taken from the closest interview prior to the birth up to and including three surveys prior. This is a stringent data requirement, resulting in relatively few cases for many of the child development indicators in the dataset. For example, the behavioral problems index is only asked of children age 4 and older, so a child who experienced a younger sibling birth at age 4 might be age 3 at the assessment prior to the birth and age 4 or 5 at the assessment after the birth. Thus, s/he would not be included in the behavioral problems analysis because s/he would not have a valid observation before the birth. Many of the cognitive development scales are administered only to older children, and thus do not result in enough cases for this analysis. The HOME measure (described below), however, was scored for all children of all ages, and thus results in a relatively large number of observations for this model. Although this stringent data requirement results in fewer cases for analysis, controlling for prior levels of the exact dependent variable produces the most careful estimates of whether experiencing a younger sibling’s birth depends on the intention status of that birth.

HOME Score

The Home Observation for Measurement of the Environment (HOME) inventory provides a scale indicating the quality of the home environment for a specific child of any age (Caldwell and Bradley 1984). It is based on a semi-structured interview and observations. It focuses on the quality and quantity of stimulation. Each HOME item was coded as absent (“0”) or present (“1”). Because the number of items at each developmental period varied, scores were normalized such that the standard score mean was 100 and the standard deviation was 15. Although the items in the assessment vary slightly with the child’s age, because the scores are normalized to mean 100 and standard deviation 15, we treat the scale as equivalent across age groups.

Generally, the HOME measure is the most widely used measure of the home environment, and has been shown to be reliable over time and across cultures, and is predictive of children’s cognitive development at different age points (Bradley, Corwyn, Burchinal et al., 2001a, 2001b; Bradley & Corwyn, 2005). Baker and Mott (1989) also reported that the items on the HOME (as assessed in the NLSY) have high reliability, high validity, good discrimination indices, and strong and cohesive factor loadings across age. All of the age-specific versions of the HOME have undergone extensive norming and standardization and each has acquired considerable validation as applied in a wide array of studies throughout the world (for reviews, see Bradley, 1982, 1994). Recently, five biennial assessments of the HOME data within the NLSY data set were analyzed (for data collected from 1986 through 1994; Bradley et al., 2001a) and the across-age validity of the HOME within the NLSY data was confirmed. There is also no evidence that measurement error of the HOME is linked with historical era (as specifically discussed in Bradley & Corwyn, 2004, p. 21).

Behavior Problems Index (BPI)

A questionnaire assessing behavioral problems was completed by the NLSY mothers of children age 4 and above. The Behavioral problems Index (BPI) index rates children on six problem areas, including antisocial, anxious-depressed, hyperactive, headstrong, dependent, and peer-conflicting behaviors (Baydar and Brooks-Gunn 1991; Chase-Lansdale and Gordon 1996; Peterson and Zill 1986; Zill 1990). A raw total score for each child is obtained by summing affirmative answers to each statement. The NLSY data also include nationally normed scores by gender, constructed from the nationally representative sample in the 1981 National Health Interview Survey. The NLSY data also include six BPI sub-scales derived from factor analysis – antisocial, anxious/depressed, headstrong, hyperactive, immature dependency, and peer conflict/social withdrawal. Each of these scales is nationally normed, as well. The mean for each normed score is approximately 100, with a standard deviation of approximately 15.

The Behavior Problems Index (BPI) was developed to assess behavior problems in children ages 4 to 17 by Peterson and Zill (1986), with items drawn primarily from the very widely-used Child Behavior Checklist (Achenbach & Edelbrock, 1981). The BPI is considered the standard in the field and consists of 28 items that yield six subscales, each of which has adequate internal consistency (αs range from .54 to .69; Chase-Lansdale, Mott, Brooks-Gunn & Phillips, 1991). Good reliability is apparent for the total scale (.86; Baker & Mott, 1989), and the two scales of internalizing (α=.62) and externalizing behavior problems (α=.77; Chase-Lansdale & Gordon, 1996). The BPI was developed and standardized on over 15,000 children in the National Health Interview Survey (Zill, 1988) and underwent extensive norming and shows good across-age validity (Chase-Lansdale, Mott et al., 1991). The BPI has consistently shown good discriminant validity in discriminating between clinic and non clinic samples (Zill & Snyder, 1981), between children from high-conflict and low-conflict marriages (Peterson & Zill, 1986), between children of divorced and remarried parents and those of nondivorced families (Zill, 1988), and to correlate highly with harsh parenting and mother-child conflict (Corwyn & Bradley, 2005). Moreover, Achenbach and colleagues have demonstrated very good test-retest reliability, interparent agreement, cross-cultural similarity, and concurrent validity of the original Child Behavior Checklist items with items from the Quay-Peterson Revised Behavior Checklist and Conners Behavior Problem Checklist (Achenbach, Howell, Quay & Conners, 1991). Measurement error of the CBC items was specifically tested by comparing clinical intake scores to home interview survey scores of clinically referred children, with no difference in problem scores between the two assessment methods (Achenbach et al., 1991, pp. 40-41).

Control Variables

We also include a number of control variables, chosen because they are associated with unintended childbearing as well as the quality of the home environment and behavior of children. These measures refer to the mothers (family level), the focal children, and the younger siblings.

Mother/Family-Level Measures

Family income is a measure of total family income from all sources, ascertained from the mother for the year prior to the interview, and presented in $1,000 units. In other words, a family with an income of $20,000 is coded 20, and a variation of one-unit is equivalent to a $1,000 difference. All income information is converted to 1979 dollars using the consumer price index. For example, a family that reported $50,000 of income from all sources in 1982 is coded 37.62 (50 ÷ 1.33). In other words, $50,000 in 1982 has the same buying power as $37,616.58 in 1979. Home ownership is a dichotomous measure coded 1 if the mother owned her home.6 If the mother was coresiding with a spouse or partner, mother living with a spouse or partner is coded 1. This person may not be the father of the sample child, however.7 By searching the relationship codes in the mother’s household listing, we are also able to determine if the mother was living with her parents (coded 1). Next is a simple count of the number of mother’s children who live in her household, where 0 means that she does not reside with any of her children. Mother’s education is coded as her highest grade of school completed, which ranges from 0 to 20 years. As a measure of mothers’ cognitive skills we use her percentile score on the Armed Forces Qualifying Test (AFQT) which was completed by all NLSY79 women in 1980. These percentile scores are divided by ten so that a mother who scored 50 is coded 5, and a one-unit difference between mothers for this variable is equivalent to 10 percentile points. We also include the mother’s age at the focal child’s birth, which ranges from 15 to 40; her race/ethnicity coded as non-Hispanic white (reference category), Latina, or African-American; and her region of residence in the U.S. at the time of the assessment: south (reference category), northeast, north central, or west.

Focal Child/Sibling Birth-Level Measures

We additionally include multiple measures referring to the focal child. A dichotomous variable indicates whether the focal child is female. Focal child’s birth order is coded 1 through 10, with 1 indicating first born. Existing research has suggested that later-born children may have different developmental trajectories, and thus it may be related to the focal child’s reaction to the younger sibling’s birth. We use a spline function to estimate the effect of the focal child’s age at assessment. We divide the age range into groups corresponding to the ages when the age-specific measures included in each assessment change (ages 0-3, 3-6, 6-10, and 10+ for the HOME, and ages 4-10 and 10+ for the BPI). These spline functions measure the linear effect of each additional month of age on the change in the dependent variable. The spline function also controls for the likelihood that a closely spaced sibling may be more likely to be unintended, and may cause particular declines in the home environment as well as behavioral problems in existing children (e.g., Powell and Steelman 1990). Characteristics of the focal child’s assessment (after the younger sibling’s birth) include temporal order, which refers to the sequence in which an individual observation took place among all observations of the focal child, and controls for effects of repeated measurement of the home environment/behavior as well as continued participation in the longitudinal study. Total number of assessments indicates the total number of observations the focal child received, and also controls for repeated measurement and longitudinal participation. Finally, wave is an index of the year the focal child’s assessment took place – with codes ranging from 1 (1986) to 8 (2002) – and controls for changes over time in home environment and child behavior experienced by all participating families.

Finally, we include a dichotomous indicator of whether the younger sibling is the same sex as the older sibling, which may also be related to behavioral problems for the existing child, as well as intention status (Powell and Steelman 1990).

Analytic Strategy

We use crosslag regression models to assess the change in the focal child’s HOME and BPI after a younger sibling’s birth. In this method, the score of the dependent variable after the birth is regressed on the score of that same variable before the birth plus the intention status of the younger sibling’s birth as well as the other independent variables. Because our central focus is on the process of change in the home environment and behavior, we use crosslag (or lagged dependent variable) regression models. These models allow us to assess change in the focal child’s dependent variables after the birth of a younger sibling, while also including time-invariant indicators of the situation before the younger sibling birth as controls.8

To be included in each model, an existing child must have a younger sibling and valid HOME or BPI measure must be available from both before and after the sibling’s birth. If a measure of any of the independent variables was not available from the interview directly preceding the sibling’s birth, it was taken from the interview prior to the pre-birth interview. In addition, because births are clustered within mothers in the data, we estimate two-level models.9

Note that these are not representative samples of births. For example, none of the focal children is an only child. For the BPI analysis, our sample is restrictive – because behavioral problems of young children are difficult to reliably assess, mothers were not asked about the behavioral problems of their children under age 4. Thus, the focal child must be at least age 4 at the younger sibling’s birth in order to have had a BPI assessment at age 4 and another after the younger sibling’s birth. (The HOME analyses are less restrictive because, barring missing data, all children would have a HOME assessment before the birth of a younger sibling, except for those who are less than one year older than their sibling, and even then they will often have a HOME assessment unless the interview prior to the younger sibling’s birth occurred very soon before the birth).

The model is as follows for the ith focal child-younger sibling pair of mother j:

Level-1 Model (focal child/sibling birth level)

yij=πoj+π1j(focalchildsprebirthscore)+π2j(siblingbirthunintended)+π3j(childbirthlevelcontrols)+eij

where yij is focal child i’s score after the younger sibling birth to mother j, π0j is the focal child’s expected baseline score after the sibling’s birth, π1j is the slope for the focal child’s score before the sibling’s birth, π2j is the difference in the expected baseline score change between intended and unintended sibling births (effect of unintended), Σπ3j is a vector of slopes for child/birth-level controls, and eij is the random error for focal child-sibling birth pair, ij.

Level-2 model (mother/family level)

π0j=β00+β01(motherfamilylevelcontrols)+r0j

π1j=β10

π2j=β20

π3j=β30

β00 is the overall average focal child score after the sibling birth, Σβ01 is a vector of differences in the overall average focal child score by mother/family-level controls, r01 is the random effect for the expected baseline score (the correlation between focal child scores within the same mother/family), β10 is the slope for pre-birth score, β20 is the difference in the overall average focal child score change between intended and unintended sibling births (effect of unintended), and Σβ30 is a vector of child/birth-level controls slope differences. Note that there are many other possible random effects terms, which represent the extent to which each variable or set of variables influences the correlation between births within a family. We only include the random intercept term because that most closely matches our theoretical framework and specific hypotheses. We also test models that include interactions between intentionality and the initial level of the dependent variable; we do not present those specifications here, but briefly describe those results below.

Results

Table 2 presents estimates of the relationship between intention status of the younger sibling’s birth and change in the focal child’s HOME score (hypotheses 1). Next, table 3 presents the relationships between younger sibling’s intention status and change in the focal child’s total behavioral problems index (BPI), separately for boys and girls (hypothesis 2). Finally, Tables 4 and and55 present the relationships between intention status of the younger sibling’s birth and the focal child’s BPI subscales – Table 4 for boys, and Table 5 for girls.

Table 2
Two-Level Estimates of the Change in HOME Score after the Birth of a Younger Sibling, according to Intention Status of Younger Sibling’s Birth (standard errors in parentheses), n=2,317 subjects, 2,976 observations
Table 3
Two-Level Estimates of the Change in Total BPI Score after the Birth of a Younger Sibling, according to Intention Status of Younger Sibling’s Birth (standard errors in parentheses)
Table 4
Two-Level Estimates of the Change in BPI Subscales for Boys after the Birth of a Younger Sibling, according to Intention Status of Younger Sibling’s Birth (standard errors in parentheses), n=492 subjects, 575 observations
Table 5
Two-Level Estimates of the Change in BPI Subscales for Girls after the Birth of a Younger Sibling, according to Intention Status of Younger Sibling’s Birth (standard errors in parentheses), n=476 subjects, 569 observations

HOME Score

Table 2 presents estimates for two-level crosslag models of the change in the focal child’s home environment after the birth of a younger sibling. Model 1 includes control variables only. Because the average overall change in the home environment is negative after the birth of a younger sibling, positive coefficients indicate less negative change, and negative coefficients indicate stronger negative change . First note that the strongest predictor of HOME score after the birth is HOME score before the birth. This is typical for crosslag models. In addition, many of the control variables measuring characteristics of the families, the focal children, and the assessment itself are strong predictors of the quality of the home environment. Focal children experience less negative change if they are from families with higher income, families who own their own home, families with two parents, as well as families with more highly educated and cognitively able mothers. Female children, and families who live in the northeast and north central parts of the country (relative to the south) also experience less negative change. On the other hand, children from African American and Latina families, families with many children, and higher birth order children experience more negative change than otherwise similar children. Indicators of continued participation in the study are related to more negative change in the home environment. This may be because mothers answer “yes” to fewer questions after completing many assessments, because the quality of the home environment for children is generally declining, or some other reason. Finally, the spline functions indicate the linear effect of each additional month of age on the change in the dependent variable, allowing the slope of the effect to differ within the four groups. Among the focal children ages 0-3, each additional month of age at the younger sibling’s birth results in a larger decrease in the quality of the home environment. For focal children ages 3-10, each additional month of age at the younger sibling’s birth results in a smaller decrease in the quality of the home environment. For focal children ages 10 or older, age is not associated with the magnitude of change in the home environment. This pattern may be because focal children who are much older at the birth of their younger sibling engage with their parents in ways that are quite different from the baby. On the other hand, very young children’s routines may be more easily accommodated in the presence of a new baby. Although the models do not directly assess the ideal birth spacing in terms of the quality of the home environment, the spline functions are consistent with existing literature suggesting that larger birth intervals may be better (e.g., Zajonc and Mullally 1997).

Model 2 illustrates that, relative to children who experience the intended birth of a younger sibling, focal children who experience the unintended birth of a younger sibling have more negative change in their home environment. Note that this is net of the age difference between siblings (equivalent to the age of the focal child at the younger sibling’s birth), as well as the focal child’s own birth intention status.

Model 3 decomposes unintended births into two types: mistimed and unwanted births. This model indicates that only mistimed sibling births are associated with significant negative change in the home environment. Both models 1 and 2 are consistent with hypothesis 1, described above: the home environment for an existing child will deteriorate more after the unintended birth of a younger sibling, relative to the intended birth of a younger sibling.

In addition, because model 1 is nested within both models 2 and 3, we can compare the magnitude of the coefficients in model 1 to their magnitude in models 2 and 3 to get a sense of the extent to which intention status of the younger sibling’s birth explains the influence of family characteristics on the change in the home environment after a birth. Many of the coefficients change little between model 1 and model 2 or 3, indicating that these characteristics influence change in the home environment regardless of the intention status of the birth. Other coefficients change more substantially – e.g., a 9.5% decrease in the effect of home ownership, a 7.3% decrease in the effect of living with a spouse/partner, and a 7.7% decrease in the effect of birth order. These decreases indicate that part of the reason these characteristics affect change in the home environment is because they are related to families’ risk of an unintended birth. Families who do not own their own home are at higher risk of an unintended birth, and this is one reason that home ownership affects the change in the home environment after a birth. Similarly, women who do not live with a spouse or partner are more likely to have unintended births, and this is one reason their children experience declines in the home environment after a younger sibling’s birth. Finally, children who are higher in the family’s birth order experience larger declines in their home environment after a younger sibling’s birth, partly because those births are more likely to be unwanted.

Behavioral Problems Index (BPI)

Table 3 presents estimates for two-level crosslag models of change in mother’s reports of the focal child’s behavioral problems after the birth of a younger sibling. Models 1 through 3 focus on boys only; models 4 through 6 focus on girls only. Models 1 and 4 include control variables only. Models 2 and 5 use the dichotomous measure of intention status (intended versus unintended), and models 3 and 6 use the trichotomous measure (intended, unwanted, or mistimed). Recall that these models focus on sibling pairs with relatively large birth intervals – because the focal child must be at least age 4 to receive the behavioral problems assessment, the BPI is only constructed for focal children ages 4+. This means that our analyses focus on sibling births to children who are at least age 4, and do not examine changes in behavioral problems among, for example, 2-year-olds who experience the birth of a younger sibling.

Models 1 and 4 show the relationship between the family background measures and focal child’s characteristics on the one hand, and behavioral problems on the other. Children from families with more resources tend to experience smaller increases in behavioral problems (income for boys, home ownership, mother’s education and AFQT scores for girls). Girls whose mothers live with a spouse or partner experience larger increases in behavioral problems. Boys and girls from large families experience smaller increases in behavioral problems. In terms of age, among this sample of relatively widely spaced sibling pairs, focal boys between ages 4 and 10 appear to experience more behavioral problems when they are closer to age 10 rather than 4. There is no difference in behavioral change by age among boys who are at least 10. Finally, the coefficient on the indicator of temporal order demonstrates that mothers report fewer behavioral problems during subsequent assessments of the same child. Note that although these models demonstrate slightly different patterns in terms of gender, none of these gender differences is statistically significant.

Model 2 illustrates that mothers report significantly larger increases in behavioral problems for boys after the birth of an unintended younger sibling, relative to the birth of an intended younger sibling. This is consistent with hypothesis 2, described above. Model 3 indicates that this relationship is limited to the experience of a mistimed younger sibling birth, not an unwanted younger sibling birth.

For girls, the intention status of the younger sibling’s birth is not related to the change in behavioral problems reported by mothers. This is not consistent with hypothesis 2 above, but is consistent with the idea that the consequences of an unintended birth will differ according to the context in which it occurs. It appears that these consequences may depend somewhat on the gender of the older child, but this gender difference is not statistically significant in a pooled gender model.

Also note that intention status of the younger sibling’s birth does not at all mediate the relationship between the family background measures or the focal child’s characteristics. In other words, there is very little change in those coefficients between model 1 and either model 2 or 3.

Table 4 presents parallel models for the seven BPI subscales, among boys. Panel A uses the dichotomous measure of intention status, and Panel B uses the trichotomous measure. For all subscales, we see a larger increase in mothers’ reports of behavioral problems for a focal child after the birth of an unintended younger sibling, relative to the birth of an intended younger sibling. These relationships are largely limited to mistimed births, and do not apply to unwanted births. In addition, for most subscales, mothers of unwanted births on average reported a slightly greater increase in behavioral problems than mothers of intended births, but these coefficients are not statistically different from zero. Because most of these coefficients are in the same direction as the coefficients for mistimed births, the dichotomous measure of intention status, which combines unwanted and mistimed births, is statistically significant for all of the subscales except dependent behavior, but the differences between these coefficients are likely not statistically significant.

Table 5 presents parallel models for girls. In striking contrast, for the most part mothers do not report larger increases in behavioral problems after a mistimed, unwanted, or unintended younger sibling birth compared to an intended sibling birth. The exceptions are conflict and hyperactive behavior – girls who experience the birth of an unwanted or mistimed younger sibling experience a larger increase in conflict, and there is a strong relationship between having an unwanted younger sibling and larger increases in hyperactive behavior.

Of course, these results do not necessarily mean that boys truly experience more behavioral problems than girls after a mistimed younger sibling’s birth. Instead, this difference may be an artifact of mothers’ reporting differences. For example, mothers may be more likely to report behavioral problems for their sons than for their daughters, perhaps due to gendered expectations of sons’ and daughters’ behavior in the presence of a baby.

Interaction Effects

In addition, we tested 80 interaction models; we regressed eight dependent variables (the total BPI scale and the seven subscales) on ten interaction terms (income, education, HOME score, mother living with spouse/partner, mother living with parents, child’s prior behavioral problems, birth order, armed forces qualifying test score, age of focal child {spline}, and younger sibling same sex as older sibling); each interaction term was included in a separate model. These models did not provide evidence for any of the interaction hypotheses – out of 80 interaction coefficients, only 5 were statistically significant at the .05 level, and they did not constitute a strong pattern. Of course, given that we are using a .05 significance level, we would expect that, on average, eight of those interaction terms would be statistically significant by random chance, even in the absence of any true interaction effect.

Summary and Conclusions

Overall, results suggest that children experienced a greater decline in the quality of their home environment after the birth of a younger sibling if that birth was unintended rather than intended. This is mainly limited to mistimed births, and largely does not apply to unwanted births.

Perhaps as a result of these negative changes in the environment, mothers report that boys exhibit larger increases in behavioral problems after the birth of a younger sibling when that birth is unintended. Again, this is mainly limited to mistimed births, not unwanted births. Girls do not exhibit such a strong pattern of increases in behavioral problems. This is consistent with the earliest research on unintended childbearing, the “Prague Study”, by David and colleagues (David and Matejcek 1981; Dytrych et al. 1975; Kubicka et al. 2002). They found that the negative experiences of children born from pregnancies twice denied abortion were stronger for boys. Other studies have also found that boys are more vulnerable to problematic adjustment than are girls under stressful family conditions such as poverty or parental divorce (Elder et al. 1984; Hetherington et al. 1989). And, there is evidence that mothers are less supportive and nurturing toward sons than daughters under stressful family conditions, which partially accounts for boys’ greater tendency toward acting-out behaviors (Galambos and Silbereisen, 1987; Nix et al. 1999).

Overall, these results are consistent with our hypotheses, that the home environment for an existing child declines more after the unintended birth of a younger sibling, relative to the intended birth of a younger sibling, and that existing children experience greater increases in behavioral problems after the unintended birth of a younger sibling, relative to the intended birth of a younger sibling. We found no evidence, however, that the effect of having an unintended younger sibling depends on the social context of that birth. Further research should explore, in greater detail, whether certain types of families or families with more resources are better able to weather an unintended event as monumental as a birth.

Study results are also consistent with the notion that unintended births affect the overall quality of the home environment because of the strain it places on social and financial resources, and because mothers of unintended births on average lack the resources to successfully cope with the unintended birth. Existing research has documented the importance of the home environment for children’s behavioral problems (Baydar et al. 1997a, 1997b). However, further research will be required to uncover the complex relationship between parental resources, the home environment, and its consequences for children. These analyses suggest that unintended births influence the quality of the home environment, and also have serious consequences for children’s behavioral problems, and thus the intention status of births may explain some of the relationship between the quality of the home environment and children’s development.

Geronimus and Korenman’s (1992) important and controversial assertion that much of the disadvantage associated with teenage childbearing is causally prior to the teen birth sparked a debate and a series of critiques of the sibling comparison approach. A key critique of that approach is that family-level consequences of teen childbearing impact the sisters of those teen mothers, and those “spill over” effects must be considered when interpreting sibling comparison models (East and Jacobson 2000). Taking an approach similar to Guo and Vanwey (1999), our models of change after the unintended birth of a sibling demonstrate exactly these “spill over” effects, and we argue that our analyses suggest caution when interpreting sibling comparison models of unintended childbearing as evidence that intention status of births has no causal consequences for children.

Guo and Vanwey’s (1999) approach has also sparked controversy, and our analyses are susceptible to some of the same critiques as theirs. Our models focus only on relatively short-term consequences of unintended births. Analyses of change in the home environment focus on the change between the wave before the unintended younger sibling’s birth and the wave after the birth – a relatively short time period. Although some existing research suggests long-term deficits in mother-child relationships among families with unintended births (Barber et al. 1999), further research will need to explore these consequences in greater detail. Similarly, analyses of behavioral problems after the birth of an unintended younger sibling concentrate on change between two specific time points – approximately one year before the younger sibling’s birth and approximately one year after the younger sibling’s birth. Furthermore, these measures of behavioral problems are only assessed for children age 4 and older. Thus, these analyses focus only on children who turned age 4 before the birth of their younger sibling.

In light of previously published research suggesting that unwanted births may have more severe consequences than mistimed births, one part of our results is unexpected and difficult to interpret – that the negative consequences of an unintended birth for existing children (in terms of behavioral problems and the home environment) may be limited to mistimed (but not unwanted) births. The substantive rationale for this difference is quite weak. We believe it deserves further research. We offer two possible explanations that could be examined in the future. The evaluation of a second or higher birth as mistimed may be formed in direct response to the recentness of the previous child’s birth. Thus, frustration or impatience resulting from a mistimed birth may be more easily directed toward existing children (older siblings), while the emotional consequences of an unwanted birth may be more easily directed at an unwanted child him/herself. In other words, the concept of mistimed is more relational than the concept of unwanted. Or, it could be that abortion is more likely for unwanted rather than mistimed births, and thus there is a greater selection in the births from unwanted pregnancies.

Furthermore, nearly all existing research on the consequences of unintended childbearing relies on mothers’ retrospective reports of the intention status of her births. Future data collection and analysis efforts should focus on prospective evaluations, to test the hypothesis that mothers who are experiencing problems with their newborns or young children are more likely to retrospectively view those births as unintended, which in turn may produce an artificial correlation between intention status and a wide variety of post-birth problems.

While the debate continues about the true consequences of unintended pregnancy, there is no doubt that harmful consequences are associated with it. Yet researchers have paid less attention to a key question: why are unintended pregnancies so prevalent, and what, if anything, can social scientists do to better understand and prevent their occurrence?

Acknowledgments

This research was supported by a center grant to the University of Michigan (R24 HD041028), R01-HD39285 from NICHD (Barber), R01-HD39285 from NICHD (East), and APR-006013 from the Office of Population Affairs (East).

Footnotes

2High levels of immigration since 1979 mean that our results should not be interpreted as representative of all children in the U.S. in the corresponding age ranges. Rather, our sample is representative of children born to this particular cohort of women who resided in the U.S. in 1979. This is a limitation of the NLSY79 data, as well as of any other long-term, longitudinal study based on a cohort.

3The original 1979 cohort included oversamples of military personnel and economically disadvantaged whites that were later dropped.

4An abundance of existing research has suggested that childbearing preferences are complex and that our current measures of intention status are too narrowly defined (e.g., Bachrach and Newcomer 1999; Barber 2001; Bruce 1990; Casterline et al. 1997). Unfortunately, large, longitudinal, intergenerational datasets like the NLSY have not yet begun to include more complex measurement of this concept in their design. Although dichotomous measures of women’s feelings about their pregnancies are much simpler than the true feelings, analyses of more complicated measures in the National Survey of Family Growth indicate that these categorical measures correspond to the more complicated measures in predictable ways, and are a useful simplification (Abma and Mosher 2008).

5Which model is more appropriate is ambiguous. If families with unwanted and mistimed births are theoretically similar in terms of the mechanisms that might produce change, then it is best to combine them into a category called “unintended births.” However, if the theoretical reasons for expecting change are different for mistimed and unwanted births, then it is best to keep them separate. In some instances, the unintended nature of both types of births makes them similar in terms of many of the mechanisms producing family change. However, we present both models to investigate the extent of similarity.

6Mothers who reside with their parents are coded 0.

7We also re-ran all models with a measure of whether the mother lived with the child’s father. The results were quite similar; thus, we only present models including whether the mother lived with her spouse/partner. Also note that parents’ marital status at the birth was not a statistically significant predictor of the change in any of our dependent variables. We do not include marital status in our analyses, instead focusing on the correlated but distinct concept of parental living arrangements.

8Note that Johnson (2005) advocates for a change score model, particularly when there is measurement error in the independent variables, or when important exogenous predictors of the outcome are unobserved. The crosslag model more closely fits our hypotheses, because a time-invariant variable (intention status) is our key predictor of interest, and because we hypothesize about several pre-birth family and child characteristics. In addition, as Johnson’s (2005) simulation demonstrates, the null crosslag model is estimated correctly, even in the face of substantial measurement error in the independent variables, as long as exogenous predictors (selection factors and stable individual characteristics) are well controlled. Because the dataset used here was specifically designed to investigate child development indicators, including the quality of the home environment and behavioral problems, an extensive set of variables to account for the selection into unintended childbearing is included in the dataset and in our models. If we instead implemented our independent variables as change scores (between the pre- and post-birth waves), this would include some change occurring after the birth of the sibling, which would be mediators in our model rather than exogenous controls.

9Note that we also ran the models selecting one focal child-younger sibling pair from each family. The results were similar, so we present the models using all of the child pairs in the tables.

References

  • Abma Joyce C., Chandra Anjani, Mosher William D., Peterson Linda S., Piccinino Linda J. Fertility, Family Planning and Women’s Health: New Data from the 1995 National Survey of Family Growth. National Center for Health Statistics, Vital Health Statistics; Hyattsville, MD: 1997. Series 23, No. 19. [PubMed]
  • Anastasi Anne. Intelligence and Family Size. Psychological Bulletin. 1956;53:187–209. [PubMed]
  • Anhalt K, Telzrow CF, Brown CL. Maternal Stress and Emotional Status During the Perinatal Period and Childhood Adjustment. School Psychology Quarterly. 2007;22:74–90.
  • Axinn William G., Barber Jennifer S., Thornton Arland. The Long-term Impact of Parents’ Childbearing Decisions on Children’s Self-esteem. Demography. 1998;35(4):435–444. [PubMed]
  • Bachrach Christine A., Susan Newcomer. Intended Pregnancies and Unintended Pregnancies: Distinct Categories or Opposite Ends of a Continuum? Family Planning Perspectives. 1999;31(5):251–252. [PubMed]
  • Baker PC, Mott FL. Center for Human Resources Research. The Ohio State University; Columbus, OH: 1989. NLSY Child Handbook 1989: A guide and resource for the National Longitudinal Survey of Youth 1986 Child Data.
  • Barber Jennifer S. Ideational Influences on the Transition to Parenthood: Attitudes Toward Childbearing and Competing Alternatives. Social Psychology Quarterly. 2001;64(2):101–127.
  • Barber Jennifer S., Axinn William G., Thornton Arland. The Influence of Unwanted Childbearing on Mother-child Relationships. Journal of Health and Social Behavior. 1999;40:231–257. [PubMed]
  • Barber Jennifer, Elizabeth Cooksey, Linda Young-DeMarco, Dawn Hurst. Unintended childbearing and children’s cognitive ability and achievement; Paper presented at the annual meetings of the Population Association of America; Minneapolis, MN. 2003.
  • Baydar Nazli, Jeanne Brooks-Gunn. Effects of Maternal Employment and Childcare Arrangements on Preschoolers’ Cognitive and Behavioral Outcomes: Evidence from the Children of the National Longitudinal Survey of Youth. Developmental Psychology. 1991;27:932–945.
  • Baydar Nazli, April Greek, Jeanne Brooks-Gunn. A Longitudinal Study of the Effects of Birth of a Sibling During the First 6 Years of Life. Journal of Marriage and the Family. 1997a;59:939–956.
  • Baydar Nazli, Patricia Hyle, Jeanne Brooks-Gunn. A Longitudinal Study of the Effects of the Birth of a Sibling During Preschool and Early Grade School Years. Journal of Marriage and the Family. 1997b;59:957–965.
  • Beck CT. Predictors of postpartum depression: An update. Nursing Research. 2001;50:275–285. [PubMed]
  • Behrman Jere R., Pollak Robert A., Taubman Paul. Do Parents Favor Boys? International Economic Review. 1986;27(1):33–54.
  • Blake Judith. Family Size and Achievement. University of California; Los Angeles: 1989.
  • Bradley R. The HOME inventory: A review of the first fifteen years. In: Anastasiow NJ, Frankenberg W, Fandall A, editors. Identifying the developmentally delayed child. University Park Press; Baltimore, MD: 1982.
  • Bradley R. The HOME Inventory: Review and reflections. In: Reese HW, editor. Advances in child development and behavior. Vol. 25. Academic; San Diego, CA: 1994. pp. 241–288. [PubMed]
  • Bradley RH, Corwyn RF. ‘Family process’ investments that matter for child well-being. In: Kalil A, DeLeire T, editors. Family investments in children’s potential: Resources and parenting behaviors that promote success. Lawrence Erlbaum Associates Publishers; Mahwah, NJ: 2004. pp. 1–32.
  • Bradley R, Corwyn R. Caring for children around the world: A view from HOME. International Journal of Behavioral Development. 2005;29:468–478.
  • Bradley R, Corwyn R, Burchinal M, Pipes McAdoo H, Garcia Coll C. The home environments of children in the Unites States Part I: Variations by age, ethnicity, and poverty status. Child Development. 2001a;72:1844–1867. [PubMed]
  • Bradley R, Corwyn R, Burchinal M, Pipes McAdoo H, Garcia Coll C. The home environments of children in the Unites States Part II: Relations with behavioral development through age 13. Child Development. 2001b;72:1868–1886. [PubMed]
  • Brown SS, Eisenberg L. The Best Intentions: Unintended Childbearing and the Well-being of Children and Families. National Academy Press; Washington, DC: 1995. Demography of Unintended Childbearing; pp. 21–49.
  • Bruce J. Fundamental Elements of the Quality of Care: A Simple Framework. Studies in Family Planning. 1990;21(2):61–91. [PubMed]
  • Butcher Kristin F., Anne Case. The Effect of Sibling Sex Composition on Women’s Education and Earnings. Quarterly Journal of Economics. 1994;109(3):531–563.
  • Caldwell B, Bradley R. Home Observation for Measurement of the Environment. University of Arkansas; Little Rock: 1984.
  • Casterline John, Perez Aurora E., Biddlecom Ann E. Factors Underlying Unmet Need for Family Planning in the Philippines. Studies in Family Planning. 1997;28:173–191. [PubMed]
  • Chandra Anjani, Martinez Gladys M., Mosher William D., Abma Joyce C., Jones J. Fertility, Family Planning, and Reproductive Health of U.D. Women: Data from the 2002 National Survey of Family Growth. National Center for Health Statistics, Vital Health Statistics; Hyattsville, MD: 2005. Series 23, No. 25.
  • Chase-Lansdale Lindsay, Gordon Rachel A. Economic Hardship and the Development of Five- and Six-year Olds: Neighborhood and Regional Perspectives. Child Development. 1996;67:3338–3367.
  • Conley Dalton. Sibship Sex Composition: Effects on Educational Attainment. Social Science Research. 2000;29:441–457.
  • Conley Dalton. The Pecking Order: Which Siblings Succeed and Why. Pantheon; New York: 2004.
  • David Henry P., Zdenek Matejcek. Children Born to Women Denied Abortion: An Update. Family Planning Perspectives. 1981;13(1):32–34. [PubMed]
  • Dytrych Z, Zdenek Matejcek, Schüller V, David Henry P., Friedman HL. Children Born to Women Denied Abortion. Family Planning Perspectives. 1975;7:165. [PubMed]
  • East Patricia L. Impact of Adolescent Childbearing on Families and Younger Siblings: Effects That Increase Younger Siblings’ Risk for Early Pregnancy. Applied Developmental Science. 1998;2(2):62–74.
  • East Patricia L., Jacobson Leanne J. Adolescent Childbearing, Poverty, and Siblings: Taking New Direction from the New Literature. Family Relations. 2000;49:287–292.
  • Elder Glenn, Liker TJ, Cross C. Parent-child behavior in the Great Depression: Life Course and Intergenerational Influences. In: Baltes P, Brim O, editors. Life-span development and behavior. Vol. 6. Academic Press; Orlando, FL: 1984. pp. 109–158.
  • Elder Glenn, Nguyen T, Caspi A. Linking Family Hardship to Children’s Lives. Child Development. 1985;56(1):361–375. [PubMed]
  • Frost JJ. Family Planning Clinic Services in the United States, 1994. Family Planning Perspectives. 1996;28:92–100. [PubMed]
  • Frost JJ. Public or Private Providers? U.S. Women’s Use of Reproductive Health Services. Family Planning Perspectives. 2001;33:4–12. [PubMed]
  • Galambos Nancy, Rainer Silbereisen. Influences of income change and parental acceptance on adolescent transgression proneness and peer relations. European Journal of Psychology of Education. 1987;1:17–28.
  • Gerdes AC, Hoza B, Arnold LE, Pelham WE, Swanson JN, Wigal T, Jensen PS. Maternal Depressive Symptomatology and Parenting Behavior. Journal of Abnormal Child Psychology. 2007;35:705–714. [PubMed]
  • Geronimus Arline T., Sanders Korenman. The Socio-economic Consequences of Teen Childbearing Reconsidered. Quarterly Journal of Economics. 1992;107:1187–1214.
  • Gipson Jessica D., Koenig Michael A., Hindin Michelle J. The Effects of Unintended Pregnancy on Infant, Child, and Parental Health: A Review of the Literature. Studies in Family Planning. 2008;39(1):18–38. [PubMed]
  • Guo Guang, VanWey Leah K. Sibship Size and Intellectual Development: Is the Relationship Causal? American Sociological Review. 1999;64:169–187.
  • Henshaw Stanely K. Unintended Pregnancy in the United States. Family Planning Perspectives. 1998;30:19–23. [PubMed]
  • Hetherington EM, Stanley-Hagan M, Anderson E. Marital Transitions: A Child’s Perspective. American Psychologist. 1989;44(1):303–312. [PubMed]
  • Ispa JM, Sabe MR, Porter N, Csizmadia A. Pregnancy Acceptance, Parenting Stress, and Toddler Attachment in Low-Income Black Families. Journal of Marriage and Family. 2007;69:1–13.
  • Johnson David. Two-Wave Panel Analysis: Comparing Statistical Methods for Studying the Effects of Transitions. Journal of Marriage and Family. 2005;67:1061–1075.
  • Joyce Theodore, Robert Kaestner, Sanders Korenman. The Effects of Pregnanacy Intention on Child Development. Demography. 2000;37:83–94. [PubMed]
  • Joyce Theodore, Robert Kaestner, Sanders Korenman. On the Validity of Retrospective Assessments of Pregnancy Intention. Demography. 2002;39:199–213. [PubMed]
  • Korenman Sanders, Robert Kaestner, Theodore Joyce. Consequences for Infants of Parental Disagreement in Pregnancy Intention. Perspectives on Sexual and Reproductive Health. 2002;34:198–205. [PubMed]
  • Kost K, Landry David J., Darroch Jacqueline E. Predicting Maternal Behaviors During Pregnancy: Does Intention Status Matter? Family Planning Perspectives. 1998;30:79–88. [PubMed]
  • Kubicka Ludek, Zdenek Roth, Zdenek Dytrych, Zdenek Matejcek, Henry David. The Mental Health of Adults Born of Unwanted Pregnancies, Their Siblings, and Matched Controls: A 35-Year Follow-up Study from Prague, Czech Republic. Journal of Nervous & Mental Disease. 2002;190(10):653–662. [PubMed]
  • McLoyd Vonnie C. The Impact of Economic Hardship on Black Families and Children: Psychological Distress, Parenting, and Socioemotional Development. Child Development. 1990;61(1):311–346. [PubMed]
  • McLoyd Vonnie C, Jayaratne Toby E., Rosario Ceballo, Julio Borquez. Unemployment and Work Interruption Among African American Single Mothers: Effects on Parenting and Adolescent Socioemotional Functioning. Child Development. 1994;65(2):562–589. [PubMed]
  • Mosher William D. Advance Data from Vital and Health Statistics. National Center for Health Statistics, Vital Health Statistics; Hyattsville, MD: 1990. Use of Family Planning Services in the United States: 1982 and 1988. No. 184.
  • Najman Jake M., John Morrison, Williams Gail M., Andersen Margaret J., Douglas Keeping J. The Mental Health of Women Six Months after They Give Birth to an Unwanted Baby: A Longitudinal Study. Social Science and Medicine. 1991;32:241–247. [PubMed]
  • Nix Robert L., Pinderhughes Ellen E., Dodge Kenneth A., Bates John E., Pettit Gregory S., McFadyen-Ketchum Steven A. The Relation Between Mothers’ Hostile Attribution Tendencies and Children’s Externalizing Behavior Problems: The Mediating Role of Mothers’ Harsh Discipline Practices. Child Development. 1999;70(4):896–909. [PubMed]
  • Peterson JL, Zill N. Marital Disruption and Behavior Problems in Children. Journal of Marriage and the Family. 1986;48:295–307.
  • Pettit Gregory S., Laird Robert D., Dodge Kenneth A., Bates John E., Criss Michael M. Antecedents and Behavior-Problem Outcomes of Parental Monitoring and Psychological Control in Early Adolescence. Child Development. 2001;72(2):583–598. [PMC free article] [PubMed]
  • Pollard Michael S., Philip Morgan S. Emerging Parental Gender Indifference? Sex Composition of Children and the Third Birth. American Sociological Review. 2002;67:600–613. [PMC free article] [PubMed]
  • Powell Brian, Steelman Lala C. The Liability of Having Brothers: Paying for College and the Sex Composition of the Family. Sociology of Education. 1989;62:134–147.
  • Powell Brian, Steelman Lala C. Beyond Sibship Size: Sibling Density, Sex Composition, and Educational Outcomes. Social Forces. 1990;69:181–206.
  • Powell Brian, Steelman Lala C. The Educational Benefits of Being Spaced Out: Sibship Density and Educational Progress. American Sociological Review. 1993;58:367–381.
  • Powell Brian, Steelman Lala C. Feeling the Pinch: Age Spacing and Economic Investments in Children. Social Forces. 1995;73(4):1465–1486.
  • Salmon P, Drew NC. Multidimensional Assessment of Women’s Experience of Childbirth: Relationship to Obstetric Procedure, Antenatal Preparation, and Obstetric History. Journal of Psychosomatic Research. 1992;36:317–327. [PubMed]
  • Steelman Lala C., Brian Powell. Sponsoring the Next Generation: Parental Support for Higher Education. American Journal of Sociology. 1991;96:1505–1521.
  • Stewart RB. The Second Child: Family Transition and Adjustment. Sage; Newbury Park, California: 1990.
  • Stewart RB, Mobley LA, Van Tuyl SS, Salvador MA. The First-born’s Adjustment to Birth of a Sibling: A Longitudinal Assessment. Child Development. 1987;58:341–355. [PubMed]
  • Sulloway Frank J. Born to Rebel: Birth Order, Family Dynamics, and Revolutionary Genius. Pantheon; New York: 1996.
  • Winraub M, Wolf B. Effects of Stress and Social Supports on Mother-Child Interactions in Single- and Two-Parent Families. Child Development. 1983;54(4):1297–1311. [PubMed]
  • Zajonc RB, Mullally PR. Birth order: Reconciling conflicting effects. American Psychologist. 1997;52(7):685–699.
  • Zill N. Behavioral Problems Index Based on Parent Report. Child Trends; Washington, D.C.: 1990. Publication 9103.
  • Zuravin Susan J. Unplanned Childbearing and Family Size: Their Relationship to Child Neglect and Abuse. Family Planning Perspectives. 1991;23:155–61. [PubMed]

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...