Another valuable national information source for describing children and their living arrangements is the decennial United States Census. The census provides the most comprehensive enumeration of the American population available, although this information is collected far less frequently than the Current Population Survey and other large sample surveys. The last national census was conducted in April, 1990, so the information it provides is now dated by over 6 years. In using data from the census to describe current patterns it is important to assess the chance that the characteristics and relations being examined have changed since the census was taken. As a rule, small-area data and information related to sectors that can fluctuate rapidly -- like the economy -- are the least likely to maintain short-term stability.
By law, the Census Bureau cannot distribute detailed household-based data as it is collected, as a protection of the people's privacy rights. Most of the publicly available data from the census is produced and distributed in the aggregate form of Summary Tape Files (STF). These extracts contain a broad range of fields, arrays, and cross-tabulations that give counts of population units (persons, families, households, etc.) across pre-defined arrangements of characteristic traits. The STF records are reproduced, in the same format, for many geographic levels and places -- nation, region, state, county, place, minor civil division, tract, block, Metropolitan Statistical Area, etc. The analyst can refer either to relationships between various data tables for one single geographic unit, or obtain similar data from many geographic units to compare variation across and within places.
The STF structure provides a relatively rich data structure concerning children, families, households, and the living arrangements of persons. However, for describing a population as specific as "children living with relatives -- parent not present", the topic of this project, only counts and rudimentary age characteristics are directly available for analysis from these tabulations. To define and describe the context of kinship caregiving, the following distribution of children was developed from the STF tables:
| Own-Child with Two Parents | Related Child, No Parent Present -- (Kinship Care) |
| Own-Child with Mother, Father not Present | Unrelated Child |
| Own-Child, with Father, Mother not Present |
Care must be taken in interpreting these categories, as they compress a complexity of possible arrangements into a short and mutually exclusive list. For example, the "Own Child, with Mother, Father not Present" category is usually referred to with shorthand terms like "Mother Only," and is usually presumed to contain children in simple single-mother families. What defines this category is that the child's mother is present, and neither the natural father of the child nor a different husband of the mother is present. The household may contain grandparents, other adult relatives, a mother's "partner" or boyfriend, and other persons, but it is defined strictly by child-parent and(step-parent) relations.
Each of these categories is further classified into two age groups, for children ages 0-5 and children ages 6-17. The method for obtaining this categorization from the STF tables is described fully in Appendix 2. Details could be added to extend this classification. Each of the own-child categories can be divided into children in primary nuclear families and children living in parent-child subfamilies within extended households. Unrelated children can also be further divided -- into those living in household settings, those in institutions, and those in noninstitutional group quarters. Unfortunately, no information is available in the STF tables that will allow for more specific classification of the "related children--no parent present" group, which is of greatest interest here.
Two major qualifications regarding the accuracy of counts in these census living arrangement categories should be addressed here. The first is that all census counts are subject to some bias due to the under-enumeration of certain hard-to-locate population groups. Groups known to be systematically undercounted in the census include young adult minority males, homeless persons, resident aliens, and African American infants. Census undercount would affect the living arrangement data most seriously if the unenumerated children have systematically different living patterns than other groups.
The second accuracy issue has more direct substantive bearing. Recognizing the diversity of family and household structures employed by the American public in caring for our children, the census recently developed a very sensitive methodology for finding and properly classifying parent/child subfamilies that live within larger household units. Children in a nuclear family that live in a grandparent's home, for example, would once have been classed only as "relatives" of the household head, but they can now be identified in the more meaningful category of own-children in a "related subfamily.(8) Parent-child subfamilies are also recognized when they are "unrelated subfamilies," i.e. when the subfamily bears no direct kinship relation to the head of the household. However, any children now living in an "adult-relative and child" equivalent of a subfamily are not tracked as carefully. If this type of "kinship subfamily" has no marriage or kinship-based relationship to the defined head of the household where they reside, the children are probably identified by the census as "unrelated" members of those households.
Table 2.1 presents the national distribution of children in these living arrangement categories as reported by the 1990 census. At the national level:
Each of the three own-parent categories shows similar age composition, with just over one-third of the children being under 6 years of age. A slightly higher proportion of children in mother-only arrangement tend to be in the older (6-17) age category than children in the other two own-parent groups. In contrast, the related child and unrelated child groups contain a noticeably higher percentage of older children than the own-child groups. Only around one-fourth of these children are under 6 years of age, with the kinship group (23.6 percent ages 0-5) having slightly fewer young children than the unrelated group (26.1 percent ages 0-5). National living arrangement distributions for each age group are presented graphically in Figure 2.1.
By examining the full distribution of living arrangements instead of looking just at children living in kinship care settings, we can shift our frame of comparative reference. The conditional percentage of living in any of these arrangements, given that the child does not live with two parents, is shown in Panel D of Table 2.1. In this necessarily higher "risk" group, the percentage of children living with relatives approaches 8 percent. Similarly, children living in kinship settings comprise just over one-half (51.3 percent) of all children living in arrangements in which no parent is present (Panel E).
These national census data provide very little direct description of the kinship population other than counts and age groupings. These numbers are useful for identifying the size and level of kinship caregiving, but they do very little to help us better understand which children are involved in kinship care settings and how they differ from other children. It is particularly unfortunate that this census-based information is not classified by race/ethnicity, because the national CPS data have shown this to have an important influence on living arrangement types, including kinship caregiving.
Although the census STF data provide little direct descriptive information about kinship care at the national level, they allow extension of this work through examination of variations observed within and across places. This section addresses the distribution of American child living arrangements at the state level.
The full distribution of child living arrangements for each state is presented in Table 2.2a-c both as counts and percentages. (Table 2.2a includes all children under 18 years of age, and the same information is shown in Table 2.2.b for children ages 0-5 and in Table 2.2.c for children ages 6-17). A brief inspection of the percent distributions will show a wide variation in the pattern of child caretaking across the states. The District of Columbia is a clear outlier, with just over one-third of its children living with two parents and almost one-half living in settings with a mother present and father absent. Across the fifty states, the percentage of children living with two parents still varies significantly, from 61.6 percent in Mississippi to 83.9 percent in Utah.
Looking at kinship care directly, the percentage of children who lived in the care of relatives in 1990 ranges from a low of 0.8 percent in both Minnesota and North Dakota to highs of 3.7 percent in Mississippi and 6.0 percent in Washington, D.C.. Although the differences between these kinship percentages is rather small, the proportional differences can be quite large. For example, a child in Mississippi is over four and one-half times more likely to live in a kinship care arrangement than is a child from Minnesota.
The same basic patterns are replicated in Tables 2.2b and 2.2c for the separate age groups. As was seen in the national percentage, the 0-5 age group is significantly less likely to live in a non-parental (relative or unrelated) setting than are children in the 6-17 year age group. Overall, the across-state variations for both age groups seem to mirror what was observed for all children in Table 2.2a.
Two features of the data in Table 2.2 stand out. First, the state-by-state distributions of child living arrangements show an apparent tendency to vary regionally, or at least, many geographically proximate states seem to have similar child living patterns. Second, each state's distribution appears to be dominated by the first living arrangement category, the number of children living with two parents. The values of all of the other categories are very much bounded, or restricted, by the percentage of children in two-parent homes. For example, because only 61.6 percent of Mississippi children live with two parents, 38.4 percent of the child population remains to be divided across the Mother Only, Father Only, Relative and Unrelated categories. In Utah, by contrast, only 16.1 percent of the child population fit into these categories. Because this "pool" of children who do not live with two parents varies so greatly across states, we need to be careful in interpreting direct numerical differences in the population percentages for the various categories across states. Although these raw population percentages accurately represent the final net impact of children living in a certain care setting, it is not as clear that they can usefully represent the processes and tendencies by which children come into these arrangements.
Table 2.3 presents the child living arrangement percentage data for states, now ordered within census regions instead of alphabetically. To aid in interpreting these patterns, this table also adds a new series of "conditional percentages" of children living in given arrangements. Columns (3)-(7) replicate the population percentages that were presented above in Table 2.2a, with the sole change that column (3) is modified to show the percentage of children not living with two parents instead of the percentage that do live with two parents. Columns (8)-(11) present the percentage of children in each of the other living arrangements, given that these children are not living in a two-parent family setting. These conditional percentages reflect our understanding that the original population percentages can be separated into two parts: the likelihood that a child lives without both parents, and the likelihood that a child living with less than two parents lives in the particular type of care arrangement. The formal mathematical relationship is represented by a simple equation:
| Proportion(kinship care) = | Prop (not/2 parents) * | Prop (kinship | not2 parents) |
| Population proportion of children in kinship care | Population proportion of children not living living with 2 parents | Conditional proportion of kinship: i.e., proportion of those children not living with two parents who are in kinship care. |
Thus, two states can have similar proportions of children living with relatives, yet have very different underlying relationships. For example, Arkansas and Louisiana are neighboring states that have similar (2.9 percent and 3.1 percent) population levels of children living in kinship care. However, in Arkansas, 28.6 percent of all children do not live with two parents, and 10.3 percent of these live in kinship settings. In Louisiana, 36.3 percent of all children do not live with two parents, but only 8.5 percent of these live in kinship settings. By the formula above:
Arkansas .29 = .286 * .103 and Louisiana .31 = .363 * .085
This formally expresses the relationships observed -- that although children in Arkansas are more likely than children in Louisiana to live with both of their parents, because a higher proportion of those not living with both parents are in kinship care settings in Arkansas, the two states have similar proportions of children in kinship living arrangements.
Although technically correct, this last description of this decomposition and the conditional relationship is fairly sterile and free of interpretive power. Introducing some inferences about the meaning of these components can help to bring more meaning to their relationship. In this vein, it shall be (provisionally) assumed that maintenance of two-parent care situations for children is both "preferable" and "preferred" in American society, and that the nuclear family is the primary care arrangement. The likelihood of a child leaving a two-parent family is very low, and the processes by which children do enter mother-only, father-only, relative, and unrelated care settings tend to occur only in the absence of (or disbanding of) the nuclear family unit. These other four categories appear then to result from processes that sort out children from a residual group that cannot be cared for in a two-parent living setting.(9)
Following this line of argument, the component presented in column (3), the proportion of children not living with two parents, will be loosely interpreted as representing the "level of family disruption," or the extent to which caring for children differs from that of the nuclear family. The "disruption" can result from parental breakup through separation, divorce, or death, or it could be the result of family "non-formation." But in the aggregate, this indicator will be held as a proxy for family disruption and disorganization. The conditional percentages in columns (8) through (11), then, can be interpreted as "tendencies" of children of disrupted family situations to locate, or to be located, in a given care arrangement. Returning to the two-state example discussed above, the relationship might now be described as Louisiana showing higher statewide levels of initial family disruption than Arkansas, but with Arkansas demonstrating a greater tendency to place the children from these disrupted families in kinship settings.
Column (12) presents a second type of conditional percentage, the percent of children living in kinship settings, given that they live with neither of their parents. This indicator directly measures the relative share of children in kinship versus unrelated living arrangements, assuming that they will live in either of these two types of arrangement.
The conditional percentages in columns (8) through (12) are necessarily higher than the comparable population percentages in columns (3) through (7) because they are computed from a smaller and more restricted population base. Because the numbers are larger, the absolute differences between these conditional percentages viewed across states tend to be larger than those observed with population percentages. At the same time, one effect of controlling for the variation due to "family disruption" has been to reduce the degree of proportional variation within these columns. A clear example is the District of Columbia, which for the most part shows conditional living arrangement percentages similar to those of its neighboring states. The reason D.C. is such an extreme outlier in its distribution of population percentages is almost entirely explained by the extremely high numbers of children not living with two parents, and not by its consequent placement tendencies and patterns.
Nationally, the percentage of children without two parents who live with their mothers-only varies across states from of 61.9 percent in Alaska to 75.7 percent in Louisiana. Similarly, the conditional percentage living with their fathers-only ranges from 9.8 percent in Washington, D.C. to 21.1 percent in Alaska; the conditional percentage living with relatives varies from 4.0 percent in Vermont to 10.3 percent in Arkansas; and the conditional percent with unrelated persons ranges from 4.6 percent in Alabama to 11.5 percent in Utah.
Clearly, regional patterns and regularities do exist in these data. Table 2.4 presents the same indicators, summed across states, for each of the four census-defined regions of the nation. The South has the highest "family disruption" level, with 30.5 percent of its children living without two parents, while the Midwest has the lowest levels at 25.5 percent.
The regional distribution of these children between alternative living arrangements, given that they are not with two parents, also shows clear patterns. The West has the lowest level of Mother-Only arrangements (64.9 percent), and the highest level of Father-Only (17.4 percent) and unrelated (9.7 percent) arrangements. The Midwest and Northeast have patterns similar to each other, with the highest levels of Mother-Only placement (73.6 percent and 73.2 percent, respectively), and low levels of Father-Only and kinship arrangements. The South shows the highest level of kinship arrangements (9.2 percent), the lowest level of unrelated placements (6.3 percent), and moderate levels for both single-parent only arrangements. The South is the only region where the level of kinship arrangements exceeds the level of unrelated arrangements.
Looking back to the individual state information in Table 2.3, we can see substantial variation remaining between the states within each region, but that the overall regional patterns remain evident. For example, West Virginia has the lowest conditional kinship percentage of any southern state, but at 8.4 percent it is larger than the conditional kinship percentage in all but two of the non-southern states (California and Hawaii).
It is not possible to explain these regional, or state, patterns from the data at hand. A variety of cultural, racial, economic and social influences vary by place -- and any or all of them might affect living arrangement patterns. The South has a higher concentration of African Americans and is more rural than the rest of the nation. The populations of the Midwest and Northeast are heavily urbanized. The West contains a higher proportion of recent immigrants and the largest Hispanic population. All of these factors and many others could contribute to these differences, and cannot really be pursued without better individual-level data. Even though explanation is elusive, systematic patterns of child living arrangements across these places provides evidence that the choices involved in how we care for our children are clearly linked to other social, cultural, and economic influences in our society.
It has been suggested in the above discussion that living arrangement indicators tend to vary across the states in systematic ways. This was particularly evident in the way that regional "patterns' seemed to be identifiable and persistent in these data. To examine the relationship between these living arrangements, and to draw some fundamental insight into their distribution, we performed a correlation analysis on statewide living arrangement indicators.(10)
This analysis uses three levels of indicators. Indicators 1 - 5 are population percentages of 1) children not living with two parents (family disruption), 2) own children living with mother, 3) own children living with father, 4) children relative (kinship) care, and 5), children living in unrelated care situations. Indicators 6 - 9 are percentages conditioned on less than two parents present for 6) own child living with mother, 7) own child living with father, 8) child living in relative (kinship) care and 9) child living in unrelated care situations. Indicator 10 is the percentage living in relative arrangements conditioned on no parents being present.
Correlation coefficients between each of these indicators are presented in Table 2.5, arranged in the format of a series of blocks (identified by letters) highlighting the three different levels of indicators being correlated--population percentages and two types of conditional percentages. The correlation is a measure of their mutual relationship between two variables, or the degree to which a change in the value of one variable can lead us to expect a change in the other. A correlation coefficient (r) can vary between 1.00 (a perfect positive relationship, where an increase in one implies an increase in the other) and -1.00 (a perfect negative relationship, where an increase in one implies a decrease in the other). When a correlation coefficient is 0.00, we say that the two variables are independent, that is, that information about value of one of the variables gives us no information about the expected value of the other. The square of the correlation coefficient (r2)is a statistical indicator of the actual amount of variation either of the variables can explain in the other variable.
Block A presents correlations of the "family disruption" indicator (percent of children not living with two parents) with the population-level percentages of the remaining four living arrangement types. The absolute percentages of children for both Mother-Only care (r=.97) and Relative Care (r=.89) are very strongly and positively correlated to the relative size of the available "pool" of children not living in two-parent families. Conversely, the percentage of children in Unrelated Care arrangements does not appear to co-vary significantly with the level of family disruption (r=.10).
Block B presents correlations between the population-level percentages for each of the alternative living categories. Strong positive relationships are evident in the correlation of Mother-Only and Relative care (.84), and in the correlation between Father-Only and Unrelated care (.63). Mother-Father and Relative-Father showed moderate positive relationships, while Mother-Unrelated and Relative-Unrelated showed no significant relationships.
The correlations in Block C represent relationships between the "family disruption" indicator and the conditional percentages of children in each living arrangement, given family disruption. This is an important set of relationships because, as we have seen, the final population percentages for these living arrangements are the product of the two percentages being correlated. The primary observed relationships here are a strong positive correlation (.71) between Relative Care and Family Disruption and a strong negative correlation (-.64) between Unrelated Care and Family Disruption. This can be interpreted as follows: as the presumed "pool" of children available for alternative care arrangements becomes larger (due to fewer children living in two-parent families), the likelihood that any child is in Relative Care becomes greater and the likelihood that any child is in Unrelated Care becomes smaller. This suggests that the increased "risk" of Relative Care caused by family disruption is further reinforced by an increasing likelihood of being in a relative arrangement. In the Unrelated Care case, increased "risk" is counteracted by a decreasing likelihood of being in an unrelated care arrangement. The conditional correlation between Mother-Only care and Family Disruption is positive but not statistically significant. Thus, while we saw in Block A that the percentage of children in mother-only care is clearly dependent on the level of family disruption, this finding suggests that the rate at which mother-only arrangements occur does not change significantly with the number of children at "risk." The correlation between Father-only and Family Disruption is mildly negative.
Skipping to Block E, we see correlations among the conditional percentages of moving to each arrangement given less than two parents. Here we see a very strong negative relationship between Mother-Only and Father-Only (-.93), strong negative relations between Mother-Only and Unrelated (-.73) and between Relative and Unrelated (-.59), and a strong positive relationship between Father-Only and Unrelated (.73). At this level of conditional likelihood, kinship care is independent of Mother-Only care, has a very weak negative relationship to Father-Only care, and a strong negative relation to Unrelated Care.
Block D, in the center, represents the correlations between the population percentages and the conditional percentages of each arrangement. All of the coefficients along the diagonal are positive, as a higher conditional percentage contributes to a higher population percentage. For Relative Care, this joint coefficient is very high (.94) because, as we have seen, these two percentages tend to increase together. The remaining coefficients in the table are mostly rather strong and follow a distinct pattern. Overall, the Mother-Only and Relative Care percentages vary together positively, the Father-Only and Unrelated Care percentages vary together positively, and the Mother-Only and Relative Care percents both vary negatively with the Father-Only and Unrelated percents(11).
Blocks F, G, and H introduce the second type of conditional percentage, that of being in a relative care arrangement given the condition that no parent is present. It is most readily interpreted as representing the Relative-Care versus Unrelated-Care dimension. This number is necessarily positively related to each of the other Relative Care percentages and negatively related to the Unrelated Care percentages. What is of interest is its strong positive correlation to the population percents of family disruption (.78) and Mother-Only (.80), and moderate positive correlation to the conditional percentage for Mother-Only care (.39).
The clear implication of this correlation analysis is that, at the state level, kinship care arrangements appear to be a response attached to what we have termed "family disruption," measured by the percentage of children not living with two parents. This "disruption" can be a product of either the nonformation or the breakup of families. Kinship care levels also are seen to co-vary closely with levels of Mother-Only care, while the relationships between these living arrangements and the percentage of children in Unrelated and Father-Only care arrangements tend to be weak or negative. It appears from these findings that the processes or conditions that lead children into the Unrelated and Father-Only care arrangements are different, and often in opposition to, the processes and conditions that lead children to Mother-Only and Relative care arrangements.
(8)" This is an important substantive distinction. Child living arrangements are often reported by relationship to the household head. The 1994 Current Population Survey estimated that over 5.4 million American children lived in households headed by grandparents or other nonparent relatives. Of these children, only 43 percent (or 2.1 million) did not have a parent present in the household. Therefore, what we here call "kinship care" represents significantly fewer than half of the population of children living in households headed by relative adults.
(9) The interpretation that the two-parent nuclear family is "primary" and other arrangements "residual" is not empirically justified, and these data cannot support such a causal inference. Rather, this is an inductively grounded organizing principle, which is subject to future empirical examination and revision. The working hypothesis is that children are most likely to remain in nuclear families unless those families are disrupted. An ancillary hypothesis would imply that children are more likely to remain with a single parent than either relatives or strangers if the parent-child living arrangement is not disrupted. A single-parent home is considered more likely to be "at-risk" of disruption than a two-parent home, other things being equal, so we would expect a significant amount of adaptive caretaking to occur around children living with one parent only. Clearly some children live outside of a parental unit for reasons other than disruption of the household (e.g. protective removal, institutionalization, school choice, etc.), but these factors should not have a disproportionate effect of the overall pattern of living arrangements.
(10) It is important to notice that this is an "ecological" analysis that examines the relation between properties of state distributions of child living arrangements. Conclusions cannot be casually assumed to apply at the individual level. Also, it should be noted that Washington, D.C. has been excluded from this correlation analysis to remove the extremely skewed influence of its population-level indicators.
(11) The sole exception to this pattern is the conditional percent in relative care having a weak positive relationship to the population percentage in Father-Only Care.
