Table of Contents | Previous | Next |
5.0 Functioning of Head Start Families
5.1 Overview
Chapter 5 presents information gathered from the parent interviews about the functioning of the Head Start families, including their home safety practices, use of discipline and household rules, their psychological well-being, and their exposure to violence.
5.2 Home Safety Practices
Head Start families were asked to indicate whether they engaged in various home safety practices, such as using a child safety seat or seat belt for their children, keeping medicines in childproof bottles, having an operating smoke detector, and having a first aid kit. During the fall 1997 interview, all but two of the safety practices were reported by 85% or more of the parents (Exhibit 5-1). The only items receiving proportions lower than 85% were “having a first-aid kit in the home” (67.7%) and “keeping the poison control center number and other emergency numbers by the telephone” (67.6%). Parents, on average, reported following 7.8 activities out of the nine possible, and their reported use of these practices remained high throughout the study (Mean change score = .13 from fall 1997 to spring 1998). By the spring of 1998, parents improved their practice of the two least frequent household safety items. Parents having a first-aid kit in the home increased to 78.3% in the spring, while the proportion of parents keeping emergency numbers by the telephone increased to 76.3% by the spring.
|
During the spring 1998 interview, parents were additionally asked about keeping their firearms under lock and key. While most parents (62.1%) indicated that this question was not applicable to them, one third (33.1%) did report that they always kept their firearms under lock and key, and 2.6% noted that they did not. One half of the parents who lived in rural locations (50.3%) acknowledged having firearms in their homes compared to 27.9% of the parents who were urban dwellers. Over one third of the parents who lived in non-subsidized housing (37.9%) indicated they had firearms in their homes while slightly lower proportions of parents who lived in subsidized or public housing (28.5 %) acknowledged having a firearm in their homes. Almost one half of the parents of White children had firearms (46.2%), more than parents of African American children (36.2%), and twice as often as parents of Hispanic children (21.3%).
5.3 Social Support
Families need outside sources of support in raising young children. In the fall of 1997, parents were asked about the people or groups in their lives that were helpful to them during the past six months in raising their Head Start children. Even at the beginning of the school year, almost all of the parents (87.1%) reported that Head Start was helpful (25.2%) or very helpful (62.0%) as a source of support (Exhibit 5-2). Overall, Head Start was considered slightly more helpful than relatives (84.3%) and much more helpful than other parents (67.9%), friends (64.0%), people from religious or social groups (46.5%), child care staff (31.5%), professional help givers (23.3%), or co-workers (21.3%). By the spring of 1998, 94.0% of the parents indicated that Head Start had been helpful or very helpful to them in raisin g their children.
|
A summary score measuring total support1 was created for each parent who responded to the questionnaire. Interestingly, parents who were employed reported significantly higher levels of overall support, t(2957) = 16.65; p < .0001, than parents not in the workforce, even though coworkers were least likely to be mentioned by the parents as a source of support in raising their children. Levels of overall support significantly decreased among those parents who lost employment from the fall of 1997 to the spring of 1998, t(207) = 4.44; p < .0001, and increased for parents who gained employment from fall of 1997 to the spring of 1998, t(299) = 8.70; p < .0001.
Differences in the overall level of support also varied by ethnicity. Parents of African American children reported having significantly higher levels of support than parents of White children and parents of Hispanic children, F(5, 2948) = 6.24; p < .0001. No significant differences were found in reported levels of support between parents who were married or not married at the time of the fall 1997 interview. However, those parents who were no longer married by the spring of 1998 reported increased overall support for raising their children during the second interview, t(77) = 2.96; p = 0.004. No change in support was found among parents who were not married in the fall of 1997 but were married in the spring of 1998. Parents who lived in rural locations reported significantly higher levels of support than parents who lived in urban areas, t(2929) = 3.20; p < .001. Finally, a small but statistically significant relationship was found between the number of individuals who lived in the household and the level of overall support reported by the parents. The more individuals living in the household, the less overall support the parents reported, (r = -0.08; p < .001).
5.4 Depression
Because depression is a frequent phenomenon in low-income families with young children (Belle, 1982), depression among the Head Start parents was measured using the CES-D Depression Scale2 (Radloff, 1975). Overall, parents had a mean score of 7.2 in the fall of 1997, which is in the mildly depressed range. While most parents were classified as not depressed (41.9%) or only mildly depressed (27.7%), close to one third of the parents (28.4%) were classified as moderately depressed (15.6%) or severely depressed (12.8%). From fall to spring, there was a small decline in the overall mean depression scores (spring 1998 score of 7.0), but the difference was not statistically significant.
Levels of depression varied by ethnicity. Larger proportions of parents of African American children were classified as moderately or severely depressed (35.2%) than parents of White children (30.1%) or parents of Hispanic children (23.5%).
Exhibit 5-3 presents a series of zero-order correlations between depression and other factors such as educational attainment, discipline methods, safety practices, or activities with their children. Findings indicate that parents who were more depressed were also those who had a need for (r = 0.25; p < .0001) and used (r = 0.20; p < .0001) more social services, had a more external locus of control (r = -0.35; p < .0001) and reported less social support (r = -.05; p < .001), reported a lower household income (r = -0.11; p < .0001), and engaged in fewer safety practices (r = -.11; p < .0001). When asked about activities with the children, parents who were more depressed were more likely to report that the mothers in the households participated in fewer activities with their children (r = -0.06; p < .01). A higher proportion of mothers living without a father in the home were classified as moderately or seriously depressed (32.7%) than those who had a father present in the home (22.9%). Parents who identified smokers in their households, t(2965) = 6.37; p < .0001, and problem drinkers in their households, t(2974) = 4.06; p < .0001, were signif icantly more depressed. As expected, levels of depression significantly increased for those parents who lost employment between the fall of 1997 to the spring of 1998, t(208) = 2.38; p < .0001, and decreased for parents who gained employment between the fall interview and the spring interview, t(301) = -2.19; p < .0001.
 |
|||||
| Child Problem Behavior |
Child Behavior - Aggressive |
Child Behavior - Withdrawn |
Child Behavior - Hyperactive |
Positive Social Behavior |
Number of Mother's Activities with Child |
| r = -.28 p < .0001 |
r = .22 p < .0001 |
r = .22 p < .0001 |
r = .20 p < .0001 |
r = -.08 p < .0001 |
r = -.06 p < .01 |
| Social Support |
Number of Social Services Needed |
Number of Social Services Received |
Household Income |
Home Safety Practices | Educational Attainment |
| r = -.08 p < .001 |
r = -.06 p < .0001 |
r = .20 p < .0001 |
r = -.11 p < .0001 |
r = -.11 p < .0001 |
r = -.05 p < .01 |
|
|||||
Findings also revealed that parental depression was significantly related to children’s behavior ratings. Parents who were more depressed reported children with higher ratings of problem behavior3 (r = .28, p < .0001), including aggressive4 (r = .22, p < .0001), hyperactive5 (r = .20, p < .0001), and withdrawn6 (r = .22; p < .0001) behavior. Two small but significant negative correlations were also found between parental depression and children’s positive social behavior ratings7 (r = -.08; p < .0001) and emergent literacy8 (r = -.04; p <.05), indicating that less depressed parents reported having children with better social and academic skills.
5.5 Household Rules
To learn about the structure of children’s activities in the household, parents were asked about rules or routines for their children. A large majority of parents (90.2%) reported that their children had a set time to go to bed each night, while 86.0% reported having rules about what types of television programs their children watched. Fewer children, about three quarters (76.2%), had responsibilities for helping with household chores, while less than two thirds of the parents reported that they had rules or routines for how much television their children could watch (63.2%) or what types of foods their children were allowed to eat (64.7%).
In fall 1997, two thirds of the families used at least 4 of the 5 rules, and the mean number of reported rules or routines was 3.9. Almost one tenth of the parents (9.3%) reported having only two rules or routines, and 4.3% reported having only one rule or routine for their children. Among the families who had the same respondent complete the spring 1998 interview, there was a significant increase in the number of rules used by most families (mean change = .17; t = 7.2; p < .0001) over the Head Start year.
5.6 Discipline Practices
Parents were asked about their use of two discipline practices: time out and spanking. In fall 1997, 69.4% of the parents reported that they had used a time out with their children during the previous week, while 46.2% reported that they had spanked their children during the same time. The parents who reported using time outs had given an average of 3.0 time outs during the previous week, while parents who indicated they had spanked their children had done so, on average, 2.1 times during the previous week.
Among the sample of parents who were respondents in both fall 1997 and spring 1998, there was little change in the reported use of either discipline practice. For this group, the proportion of parents who spanked their children in the previous week went from 45.7% in the fall to 44.2% in the spring. The reported use of time outs in the previous week went from 67.7% in fall 1997 to 71.0% the following spring. Paired t-tests showed that from baseline to the spring 1998 follow-up, there was no change in the number of spankings or time outs used.
When asked if Head Start had taught them any new ways to discipline their children, 42.3% of the parents replied that Head Start had helped them. Given that 72.0% of the parents who reported they had learned a new way to discipline their children were from the families in which the children (and most parents) were new to Head Start in fall 1997, this suggested that Head Start may have some influence with parents new to the program. Therefore, further analyses were conducted looking at parents with no previous Head Start experience. The parents who were new to Head Start spanked and used time outs at similar rates to the overall sample of parents. In fall 1997, 42.7% of the parents reported giving spankings, and 67.7% reported using time outs. These numbers showed little change in spring 1998, with 43.4% issuing spankings and 70.1% giving time outs. The proportion of parents who reported learning new information from Head Start about discipline was 39.1%.
In assessing parents’ use of discipline across time, an unusual pattern of change emerged. As seen in Exhibit 5-4, a majority of these parents (69.7%) were consistent in their use or non-use of time outs or gave the same reports of spanking or not spanking in the fall of 1997 and the spring of 1998 (71.5%). However, for both disciplinary practices, about equal percentages of parents reported adding or dropping the behavior (see the shaded cells in Exhibit 5-4). For example, in spring 1998, 13.3% of the parents who did not report using time out at baseline now reported having used time outs with their children. In contrast, an almost identical proportion of parents, 13.4%, who had reportedly used time outs in the fall did not report them the following spring. Similarly, while 12.9% of the parents who had spanked their children in the fall did not report doing so in the spring, virtually the same percentage, 12.5%, reported spanking their children in the spring after not reporting so in the fall. The implication of this finding is that while it was noted that almost two fifths of the parents new to Head Start reported learning new discipline styles from Head Start, this seems to have had little effect on reported behaviors. Where 68.1% of the parents reported use of time outs in the fall, 71.5% reported use of time outs in the follow-up interview. There was virtually no change in spanking behavior within the group of parents new to Head Start. In the fall, 41.4% of the parents spanked their children, while 41.6% reported the same in the spring
| Use of Time Out | Use of Spanking | ||||
|---|---|---|---|---|---|
| Fall 1997 | Spring 1998 | Fall 1997 | Spring | 1998 | |
| No time out | Time out | No spanking | Spanking | ||
| No time out | 15.5% | 13.3% | No spanking | 41.7% | 12.5% |
| Timeout | 13.4% | 54.2% | Spanking | 12.9% | 29.8% |
Among the larger sample of parents who participated in fall 1997 and spring 1998, discipline practices appeared to vary by ethnicity. Time outs were most likely used by parents of White children (77.3%), followed by the parents of African American children (62.2%) and parents of Hispanic children (57.7%). In fall 1997, more than one half of the parents of African American children (53.9%) reported using spankings, about two fifths of the parents of Hispanic children (41.9%) and one third of the parents of White children (32.8%) reported having spanked their children in the previous week. Only slight changes in these percentages were reported in spring 1998 (53.5% African American, 40.5% Hispanic, and 38.8% White). Using paired t-tests to assess change over time in the number of spankings by parents, no significant differences were found within any of the ethnic groups. Regardless of the ethnic group, between one third and one half of the parents reported learning a new discipline practice from Head Start. The highest report of learning new discipline practices came from parents of Hispanic children (45.5%), followed by parents of White children (40.4%) and parents of African American children (33.7%).
The use of spanking as a disciplinary practice was explored more fully through univariate and multivariate logistic regression models. Crude odds ratios and 95% confidence intervals for the independent variables (depression, single parenthood, and educational attainment) included in the models are presented in Exhibit 5-5. These estimates indicate that being depressed, regardless of the severity, and having only a mother in the home increased the likelihood of children being spanked. Parents who were mildly depressed were 1.6 times more likely to spank their children (OR = 1.63) than parents who were not depressed, while parents who were moderately depressed (OR = 1.71) or severely depressed (OR = 1.65) were approximately 1.7 times more likely to use spanking as a discipline method than non-depressed parents. Single mothers were 1.4 times more likely to spank than mothers who had the children’s fathers present in their households.9 There appeared to be no relationship between educational attainment and the parent’s use of physical punishment. Parents who had a high school diploma, GED, or some college were no more likely to spank their children than parents who had less than a high school diploma.
It is widely accepted that cultural norms influence the type of discipline practices used by parents (Pinderhughes, Dodge, Bates, Pettit, & Zelli, 2000; Whaley, 2000; McGroder, 2000; Kilgore, Snyder, & Lentz, 2000; Kelly, Power, & Wimbush, 1992). Univariate logistic regression models indicated that, compared to parents of White children, parents of Hispanic children were 1.3 times more likely to use spanking as a discipline method (OR = 1.31) and parents of Afric an American children were two and one-half times more likely to have reported spanking their children in the past week (OR = 2.49). In a multivariate logistic regression model, the adjusted odds ratios suggest that even after controlling for the effects of income, education, age, single parenthood, and use of social services, parents of African American children (OR = 2.30) as well as parents who were mildly (OR = 1.63), moderately (OR = 1.55), and severely depressed (OR = 1.63) were more likely to spank their children than parents of White children or parents who were not depressed (Exhibit 5-5), respectively.
| Use of Spanking as a Method of Discipline | ||||
|---|---|---|---|---|
| Crude OR |
95% CI | Adjusted OR | 95% CI | |
| Depression | ||||
| Not depressed (referent) | -- | -- | -- | -- |
| Mildly depressed | 1.63 | (1.37 - 1.95) | 1.63 | (1.32 - 2.01) |
| Moderately depressed | 1.71 | (1.38 - 2.13) | 1.55 | (1.19 - 2.01) |
| Severely depressed | 1.65 | (1.31 - 2.08) | 1.63 | (1.23 - 2.16) |
| Single Parenthood | 1.39 | (1.20 - 1.60) | 1.06 | (0.88 - 1.29) |
| Educational Attainment | ||||
| Less than high school (referent ) | --- | -- | -- | --- |
| High school diploma or GED | 1.08 | (0.90 - 1.20) | 1.14 | (0.92 - 1.42) |
| Some college or more | 0.94 | (0.86 - 1.03) | 0.87 | (0.77 - 0.97) |
| Ethnicity/Race | ||||
| White(referent) | --- | --- | --- | --- |
| African American | 2.49 | (2.07 - 3.01) | 2.30 | (1.82 - 2.90) |
| Hispanic | 1.21 | (1.10 - 1.34) | 1.14 | (1.02 - 1.30) |
5.7 Exposure to Violence and Crime
Neighborhoods have long been recognized in theory and research as important contexts for child development. Children who are exposed to neighborhood violence are at increased risk for lower social competence and negative emotional or behavioral functioning (Brooks-Gunn, Duncan, & Aber, 1997). Parents were asked about the violence they knew to occur in their neighborhoods, and were asked additional questions about their own personal exposure to violence, as well as the exposure to violence experienced by their Head Start children.
Family Exposure to Violence and Crime
More than one fourth of all parents (27.0%) reported seeing nonviolent crime such as selling drugs or stealing in their neighborhoods in fall 1997 (18.8% more than once), while even more (32.1%) reported being a witness to violent crime (17.2% more than once). Almost one quarter (23.2%) of the parents knew someone who was the victim of a violent crime in their neighborhood, bringing the reality of violence very close to many of the Head Start families. Being a victim of violent crime in the neighborhood was reported by 6.3% of the parents while 7.0% of the parents reported being victims of violence in their homes (Exhibit 5-6).
|
Exposure to violence varied across ethnic groups. Among parents of African American children, 43.5% reported seeing nonviolent crimes in their neighborhoods, a figure that was about twice the rate reported by parents of White children (16.2%) or parents of Hispanic children (24.2%). In contrast, just under one quarter of each group of parents reported exposure to violent crime in their neighborhoods. For reports of victimization, parents of African American children were again highest, with 10.2% indicating they were victims of crime in their neighborhoods, and 11.9% reporting they were victims of violent crime in their homes. These figures were generally twice as high as comparable reports for parents of White children (5.6% in the neighborhoods; 3.5% in their homes) and for parents of Hispanic children (1.0% in the neighborhoods; 3.5% in their homes).
Child Exposure to Violence
As for the Head Start children, almost one fifth (17.4%) were reported by parents to have witnessed a crime or domestic violence in their lives, and 2.7% had actually been the victims of domestic violence or crime. As with the parents’ self-reports, exposure to violence varied by ethnic group. About one fifth of the African American children (19.6%) and the White children (21.3%) had some exposure to violence, almost twice as high as the reported rate for Hispanic children (11.0%). In the spring of 1998, 7.5% of the children had witnessed domestic violence and 3.8% witnessed a violent crime during the Head Start program year. Less than one percent of the children were reported by their parents to have been victims of violent crime (0.5%), while slightly more were victims of domestic violence (1.5%) during the program year.
Effects of Violence on Child and Family Outcomes
Exposure to neighborhood violence may have direct and indirect effects on child outcomes. In order to test the relationships between exposure to violence and other family and child factors, a summary score measuring total exposure to violence10 was created for each parent who responded to the questionnaire. Small but significant positive correlations were found between neighborhood violence and the child problem behavior subscales (r = .09; p < .0001), including aggressive (r = .10; p < .0001), hyperactive (r = .05; p < .001) and withdrawn behavior (r = .04; p < .05). Stronger correlations were found between neighborhood violence and parental depression (r = .26, p < .0001) and between parental depression and overall child behavior problems (r = .28; p < .0001), suggesting that depression may mediate these relationships by serving as the mechanism through which exposure to neighborhood violence leads to problem behavior in children. A mediational model similar to that displayed in Exhibit 5-7 was tested using linear regression. In path c, neighborhood violence was a significant predictor of child behavior. However, once paths a and b were controlled, this relationship was no longer significant, confirming that depression did mediate the relationship between violence and child behavior.
|
A series of linear regression models tested whether the impact of exposure to violence on child behavior and depression varied as a function of (or was moderated by) social support, family size, presence of father, locus of control, and family activities, as well as Head Start support, satisfaction, experience, and involvement. Exhibit 5-8 presents the interaction terms found to be significant moderators of exposure to violence in the regression equations, and indicates the important role Head Start can play in moderating these relationships.
|
| Child Problem Behavior as the Dependent Variable |
| Among Hispanic Families: |
| Neighborhood Violence x Head Start Satisfaction (p = .006) |
| Among White Families: |
| Neighborhood Violence x Head Start Experience (p = .04) |
| Among All Families and African American Families: |
| No significant interactions |
| Depression as the Dependent Variable |
| Among All Families: |
| Neighborhood Violence x Locus of Control (p < .0001) |
| Among Hispanic Families: |
| Neighborhood Violence x Locus of Control (p < .0001) |
| Neighborhood Violence x Head Start Experience (p = .04) |
| Neighborhood Violence x Head Start Involvement (p = .01) |
| Among White Families: |
| Neighborhood Violence x Locus of Control (p < .0001) |
| Neighborhood Violence x Head Start Experience (p = .05) |
| Among African American Families: |
| No significant interactions |
|
Satisfaction with Head Start, as well as having a positive experience at Head Start, significantly modified the relationship between exposure to violence and problem behavior in children for parents of White and Hispanic children. The negative impact of exposure to violence on depression was moderated by Head Start experience for parents of Hispanic and White children, moderated by Head Start involvement for parents of Hispanic children, and moderated by an internal locus of control for all parents.
5.8 Involvement with the Criminal Justice System
Parents were also asked if they, another household member, or a non-household biological parent had been arrested or charged with a crime since the birth of their Head Start children. Almost one fourth (22.6%) of the parents reported that someone had been arrested and charged with a crime and 17.5% reported someone who spent time in jail. When asked who was arrested or charged with the crime, 93.9% of those identified were fathers or mothers. Almost one fifth of all fathers (17.1%) and 5.4% of all mothers were arrested and charged with a crime since the birth of their Head Start children. Parents of Hispanic children had the lowest proportion of reports of having someone in their family who was arrested (12.9%), about one half of the percentages reported by parents of White children (26.7%) or parents of African American children (27.6%).
Univariate logistic regression models were used to determine estimates of risk among those families who had someone close to them who was involved in the criminal justice system. Crude odds ratios and 95% confidence intervals are presented in Exhibit 5-9. These risk estimates indicate that parents who reported that they, another household member, or a non-household biological parent had been arrested or charged with a crime since the birth of their Head Start children were almost twice as likely to be depressed (OR=1.71), approximately two and one half times more likely to be single mothers (OR=2.33), and over four times more likely (OR=4.23) to have been a victim of violent crime in their homes than parents who did not have someone significant in their lives involved in the criminal justice system.
Children in families from which someone had been arrested were at great risk for witnessing or being a victim of violence compared to children in families where no one had been arrested or charged with a crime. These Head Start children were almost five times more likely (OR = 4.74) to have been a witness to violent crime or domestic violence and four times more likely (OR = 4.04) to have been a victim of violent crime or domestic violence.
| Arrested or Charged with a Crime | ||
|---|---|---|
| Crude OR | 95% CI | |
| Parent depressed | 1.71 | (1.43 - 2.04) |
| Parent single mother | 2.33 | (1.95 - 2.78) |
| Parent victim of violence in home | 4.27 | (3.13 - 5.82) |
| Child witness to violent crime or domestic violence | 4.74 | (2.20 - 5.80) |
| Child victim of violent crime or domestic violence | 4.04 | (2.62 - 6.21) |
5.9 Household Cigarette, Alcohol, and Drug Use
Almost one half of the Head Start children (45.9%) lived in households with at least one individual who smoked cigarettes. Although most of the parents reported being in good health, those parents who lived in non-smoking households reported better health than those parents who lived in smoking households, t(2966) = 3.15; p <.01. Only 4.7% of the parents reported that someone in their household had a drinking problem and even fewer (1.2%) reported living with someone who had a drug problem.
Reported substance use varied by ethnicity. Cigarette smoking was reported less frequently in households where Hispanic children lived (28.9%) than in households of African American children (42.4%) or White children (64.4%). Also, parents of White children reported living with someone who had an alcohol problem (6.4%) or drug problem (2.0%) more often than parents of African American children (3.3% alcohol problem; 1.2% drug problem) or parents of Hispanic children (3.8% alcohol problem; 0.3% drug problem). Smoking households were more often located in rural areas (55.0%) than in urban locations (41.4%).
5.10 Family Risk Factors
Recent research has focused more on how multiple occurrences of some family characteristics may predict negative outcomes for children (Huston, McLoyd, & Garcia, 1997; McLoyd, 1998; Vandivere, Moore & Brown, 2000). A particular approach, taken in the 1999 Kids Count Data Book (Annie E. Casey Foundation, 1999) looked at how six particular family characteristics affected child development and well-being. These six characteristics, labeled as risk factors, were:
- The child was not living with two parents;
- The household head was a high school dropout;
- The family income was below the poverty line;
- The child was living with a parent(s) who did not have steady, full-time employment;
- The family was receiving welfare benefits; and
- The child did not have health insurance.
Although the available data from the FACES parent interview do not allow an exact match with these categories, close approximations are possible. For example, as outlined in the Kids Count Data Book, the actual effects of the household heads not completing high school were mostly centered on the mothers. Given that FACES has much more complete information about mothers, the risk factor was adjusted to mothers who did not complete high school. As noted earlier, FACES collected information on household income, not family income. Therefore, the classification of families as being below the poverty line was based on household income. Exhibit 5-10 shows the percentage of families in FACES determined to have each risk factor, as well as a breakdown of each risk by urbanicity and ethnicity. Across all families, the most prevalent risks were being in a family that was below the poverty level (66.7%) and being from a single -parent household (52.8%). None of the other risks were reported for more than one third of the families.
| Weighted Percentages | |||||||
|---|---|---|---|---|---|---|---|
| All 2983 |
Urbanicity | Ethnicity | |||||
| Urban n=2011 |
Rural n =972 |
African American n=1,050 |
White n=826 |
Hispanic n=662 |
Other n=421 |
||
| Risk Factors | |||||||
| Single parent household | 52.8 | 52.7 | 53.0 | 67.7 | 52.5 | 39.8 | 44.9 |
| Mother did not complete high school | 29.4 | 30.6 | 27.1 | 27.5 | 20.5 | 42.2 | 27.1 |
| Household income below the FPL | 64.9 | 66.7 | 61.3 | 74.2 | 54.8 | 69.5 | 63.8 |
| No household parent with a job | 25.6 | 27.0 | 22.9 | 34.5 | 21.8 | 21.1 | 24.0 |
| Family receives welfare | 30.3 | 34.3 | 22.1 | 46.7 | 23.0 | 22.8 | 21.4 |
| Child not covered
by health insurance or Medicaid |
18.2 | 16.4 | 21.9 | 11.8 | 17.3 | 25.0 | 19.6 |
| Number of Risk Factors | |||||||
| Family has one risk factor | 23.8 | 21.6 | 28.2 | 17.8 | 28.9 | 24.3 | 26.0 |
| Family has two risk factors | 25.8 | 25.1 | 27.2 | 22.9 | 25.2 | 30.2 | 26.4 |
| Family has three risk factors | 19.1 | 20.6 | 16.2 | 20.9 | 15.9 | 20.2 | 19.1 |
| Family has four or more risk factors | 20.6 | 21.9 | 17.9 | 31.2 | 14.8 | 17.4 | 15.2 |
| aRisk factors adapted from Kids Count Data Book, 1999(back) |
As noted in the Kids Count Data Book, increases in the number of risk factors, particularly counts of four or more risks, increase the likelihood of negative child outcomes. Exhibit 5-10 also shows the percentage of families with multiple risk factors. One fifth of the families (20.6%) were found to have four or more risk factors. The level of risk did not vary by urbanicity. Less than one fifth of the families of White children (14.8%) and Hispanic children (17.4%) had four or more of the risk factors, but almost one third of the families of African American children (31.2%) were found to have this highest level of risk.
Those families who reported four or more risk factors had children who scored significantly lower on the emergent literacy scale than parents who reported fewer than four risk factors, t(2977) = 2.92; p< .001. Interestingly, while the mean scores for child behavior problems were slightly higher for those children who were members of families with four or more risk factors than families with fewer risk factors, the differences were not statistically significant. However, parents from families with four or more risk factors were more depressed, t(2977) = -8.13; p < .0001, reported less social support, t(2971) = 8.12; p < .0001, and had a more external locus of control, t(2979) = 4.88; p < .0001, than parents with fewer than four risk factors.
The percentages in Exhibit 5-10 show baseline data (fall 1997). It was also possible to assess change in risk from the fall 1997 to the spring 1998 data collection. The overall level of reported risk decreased among the families (mean change = -0.3; t = -14.02; p < .0001) over the year. The proportion of families encountering four or more risk factors fell to 12.4%, while the proportion of families facing one or fewer risks went from 36.3% in the fall to 44.6% the following spring. As shown in the chart in Exhibit 5-11, the largest proportion of families (40.5%) had no change in their reported level of risk. For 39.0% of the families, the level of risk declined, while the total number of risk factors increased for about one half that number of families (20.5%).
|
5.11 Summary
Findings from this chapter have contributed to a more complete understanding of how Head Start families function in their daily lives. The following is a summary of the key findings.
Home Safety Practices
-
Almost all parents reported engaging in safety practices such as using a child safety seat or seat belt for their children, keeping medicines in childproof bottles, having an operating smoke detector, and having a first aid kit.
-
Parents were also asked about keeping firearms under lock and key. While most parents indicated that this question was not applicable to them, more than one third acknowledged having firearms in their homes.
Social Support
- Even at the beginning of the school year, almost all of the parents reported that Head Start was helpful to them as a source of support in raising their young children. Overall, Head Start was considered slightly more helpful than relatives, and much more helpful than other parents, friends, people from religious or social groups, child care staff, professional help givers, or coworkers.
Depression
-
Close to one third of the parents were classified as moderately or severely depressed. Parents who were more depressed had a greater need for and reported use of social services, had a more external locus of control, had less social support, had a lower household income, engaged in fewer safety practices, and participated in fewer activities with their children.
-
Mothers living without their children’s fathers in their homes more often reported being depressed than those with fathers present.
-
Parents who were more depressed reported children with higher ratings of problem behavior, including aggressive, hyperactive, and withdrawn behavior. Parents who reported less depression also reported children with higher ratings of positive social behavior and emergent literacy.
Household Rules
- Two thirds of the families used 4 out of 5 household rules. A large majority of parents reported that their children had a set time to go to bed each night and rules about what types of television their children watched.
Discipline Practices
-
Over two thirds of parents reported that they used a time out with their children during the week prior to the fall 1997 visit, while almost one half of the parents reported spanking their children during the same period. There was little change in the reported use of either timeout or spanking from fall 1997 to spring 1998.
-
Being depressed, regardless of the severity, and having only a mother in the home increased the likelihood of parents spanking their children.
-
Compared to parents of White children, parents of Hispanic children were one and one third times more likely to spank their children and parents of African American children were two and one half times more likely to spank their children.
Exposure to Violence
-
More than one fourth of all parents reported seeing nonviolent crime in their neighborhoods, while close to one third reported seeing a violent crime in the same area.
-
Almost one quarter of the parents knew someone who was a victim of a violent crime in their neighborhoods, bringing the reality of violence very close to many of the Head Start families.
-
About one fifth of the children were reported to have witnessed a crime or domestic violence in their lives, and three percent had actually been victims of domestic violence or crime.
Effects of Violence on Child and Family Outcomes
-
Findings suggest that exposure to neighborhood violence did negatively impact child behavior, even in children as young as 3- and 4-years old.
-
Since depression was found to mediate the relationship between neighborhood violence and child behavior, the effect of neighborhood violence was most likely indirect, with depression serving as the mechanism through which exposure to neighborhood violence led to problem behavior in children.
-
It appears that being involved in and having a positive experience at Head Start may have served as protective factors against exposure to neighborhood violence.
-
Cultural differences seemed to play a role in determining what factors moderated negative outcomes of depression and problem behavior in children.
Involvement with the Criminal Justice System
-
Almost one fourth of the parents reported that they, another household member, or a non-household biological parent had been arrested or charged with a crime since the birth of their Head Start children and almost one fifth indicated that someone spent time in jail.
-
Children from families who had involvement with the criminal justice system were almost five times more likely to have been exposed to violent crime or domestic violence and four times more likely to have been a victim of violent crime or domestic violence.
Household Cigarette, Alcohol, and Drug Use
- Almost one half of the Head Start children lived in households with at least one smoker. Less than five percent of the parents reported that someone in their household had a problem with alcohol or drugs.
Family Risk Factors
-
Across all families, the most prevalent risks were being in a family with overall household income that was below the Federal Poverty Level and being from a single -parent household.
-
One fifth of the families were found to have four or more of the identified risk factors. Children in these families had significantly lower scores on the Emergent Literacy Scale and may be considered to be at risk for developmental problems.
-
Parents in families with four or more risks were more depressed, had less social support, and were more external on the locus of control scale.
1Summary support score is based on respondents’ ratings of how helpful individuals were in helping them raise their Head Start children over the past six months. Each of nine categories of individuals was rated on a 3-point scale ranging from “not very helpful” to “very helpful.” Summary score ranges from 0 to 27, with higher scores representing more support. M = 13.5; SD = 5.2.(back)
2The CES-D Scale (12-item version) measures levels of depression among parents. Score range 0-36. Zero-4 = Not depressed; 5-9 = Mildly depressed; 10-14 = Moderately depressed; 15 or more = Severely depressed. M = 7.2; SD = 6.7.(back)
3An adaptation of the Achenbach Child Behavior Checklist (Total Problem Behavior Index). Each of 12 behavior items, based on parent report, is rated on a 3-point scale ranging from “not true” to “very true or often true.” Summary score ranges from 0-24, with higher scores representing more frequent or severe negative behavior.(back)
4A subscale of the Total Problem Behavior Index, each of four items is rated on a 3-point scale ranging from “not true” to “very true or often true.” Items include parents’ reports of whether child hits and fights with other children, has temper tantrums, doesn’t get along with others, and is disobedient at home. Subscale score ranges from 0-8.(back)
5A subscale of the Total Problem Behavior Index, each of three items is rated on a 3-point scale ranging from “not true” to “very true or often true.” Items include parents’ reports of whether child can’t pay attention for long, is very restless, and is nervous, high-strung, or tense. Subscale score ranges from 0-6.(back)
6A subscale of the Total Problem Behavior Index, each of five items is rated on a 3-point scale ranging from “not true” to “very true or often true.” Items include whether child is unhappy, worries, feels worthless, has difficulty making changes, or acts too young. Subscale score ranges from 0-8.(back)
7A summary score of 7 parent-reported behavior items rated on a 3-point scale ranging from “not true” to “very true or often true.” Score ranges from 0-14, with higher scores representing more positive behavior.(back)
8A summary score of 5 parent-reported child skills including whether child can identify all of the primary colors, recognize most or all letters of the alphabet, count to twenty or higher, write rather than scribble, and write own name. Scores range from 0-5.(back)
9Data were not collected to determine whether or not the father in the household spanked.(back)
10Violence score is based on the frequency of respondents’ exposure to five items of neighborhood and personal violence. Frequency ranged from never, once, or more than once. Total score range was from 5-15. M = 6.1; SD = 1.8.(back)
| Table of Contents | Previous | Next |