Adolescents’ Scripts 1
Life with Fathers and Mother: Adolescents’ Scripts
For Their Relationships with Parents and Stepfathers
William V. Fabricius
Arizona State University
Amy A. Weimer
University of Texas - Pan American
Jeffrey T. Cookston
San Francisco State University
Delia Saenz
Arizona State University
Ross D. Parke
University of California, Riverside
Scott L. Coltrane
University of California, Riverside
Sandford L. Braver
Arizona State University
Running Head: Adolescents’ Scripts
Abstract
When asked to “describe your relationship with your [parent],” young adolescents (7th grade) produced scripts of what generally happens in that relationship, spontaneously evaluated what happens rather than simply described it, and uniformly chose the same set of dimensions on which to evaluate what happens. The three primary dimensions reflected the amount of time the parent spends in interaction with the child, the emotional quality of the relationship, and the reliability of the parent’s responsiveness. The dimensional structure was similar for four types of parents (resident biological mothers, resident biological fathers, resident stepfathers, and non-resident biological fathers), and for European American and Mexican American boys and girls. This suggests that these young adolescents shared a common schema of what a good parent does. The coding scheme developed to analyze these scripts had good inter-rater reliability, and demonstrated construct and predictive validity. Measures derived from the scripts characterized the specific parent-child relationship and specific aspects of the relationship, and were significantly related to parent and teacher reports of adolescents’ internalizing and externalizing problems, parent reports of adolescents’ positive behaviors, and adolescent self-reports of delinquent behavior (smoking, substance use, and sexual activity).
Life with Fathers and Mother: Young Adolescents’ Scripts
For Their Relationships with Parents and Stepfathers
There is currently some debate about what constitutes good parenting, and in particular whether good parenting in middle-class, Western families is similar to good parenting in other cultural and socio-economic groups (Parke & Buriel, 2006). However, any consensus that does exist is largely founded on research that examines mothering, and commentators have pointed out that “much greater consensus exists about ‘good mothers’ than about ‘good fathers’.” (Marsiglio, Day, Braver, Evans, Lamb, & Peters, 1998, p. 109). Popular notions about “good fathers” are a reflection of the traditional and contemporary role demands on fathers. Among those role demands are to work, to spend time with their children, to show support for the mother, to provide guidance and discipline, to plan for and arrange to meet the child’s social, educational, and medical needs, and to set an example for their children of what a good father does. But what does a good father do? We know comparatively little, especially during adolescence, about what constitutes good fathering even in middle-class, Western families (Marsiglio, et el., 1998).
In the present study we developed a new measure designed to reveal the child’s point of view about what it means to be a good father. The child’s point of view is likely to be important, because if children think that their fathers are acting as good fathers they are likely to feel loved, respected, and secure, and that is likely to benefit them. We focused on the period of early adolescence because that is likely to be an important time to study the child’s view of parenting. During the transition to adolescence, earlier parenting behaviors and the meanings children attach to these behaviors might undergo revision, and children’s perspectives might thus become more conscious and available for study. For example, how much the parents are able and willing to provide financially during the teen years, or their willingness to re-negotiate old rules and impose new ones might carry new meanings for adolescents about how important they feel to their parents, and how much they think their parents accept, respect, and trust them. Early adolescence also might be a time when boys and girls expect different things from mothers and fathers, and when different behaviors come to convey good mothering versus good fathering. For comparative purposes, we applied our new measure to parents in four different roles: resident biological fathers, resident step-fathers, non-resident biological fathers, and resident biological mothers. Half of the children lived with their biological mothers and fathers, and the other half lived in step-father families. Finally, because we know little about cultural similarities and differences in adolescents’ views of good parenting, we recruited European American and Mexican American families representing a range of socio-economic levels.
The measure we developed was open-ended because we wanted to capture parent behaviors that adolescents might see as uniquely important for fathers, and that might not be captured by existing measures. Existing measures of older children’s perceptions of parenting assess pre-determined categories of parent and child behaviors derived from attachment theory (Bowlby, 1969/1982; 1973, 1980), the typology of parenting styles suggested by Baumrind (1971) and Maccoby and Martin (1983), or relationship schema theory (Baldwin, 1992). While some of these measures have shown no differences between mothers and fathers (Kenny, 1987; Lamborn, Mounts, Steinberg, and Dornbusch; 1991; see also Baumrind, 1991), it is unknown whether existing measures overlook important behaviors unique to fathers.
With minimal prompting and direction, we asked young adolescents to describe their relationships with their mother and resident father (biological father or step-father), and also, if applicable, their relationship with their non-resident biological father (see Appendix). Children described the typical things they and their parents do, invariably using the present tense (e.g., “He’s always saying, ‘Let’s go to Castles and Coasters, or let’s play a game.” “He tries not to ignore me.” “She treats me nice.” “I get mad at her when she doesn’t listen to me.” “He wants me to get good grades to have a better life.” “We like to be together.” “Most of the time if I ask for something and if I’ve been good, he gets it for me.”). They seldom made statements about specific events marked by the past tense (e.g., “Last Christmas we went to see his family.” “He got me a dog a couple of weeks ago.”). Thus children’s descriptions resembled scripts (Nelson, 1986) for “what happens in my relationship with my [parent].” We focused on the behaviors they chose to include in their scripts, under the assumption that they considered those behaviors to be the important and meaningful things their parent do with them.
One part of our coding scheme involved classifying these behaviors. The basis of classification was how the child framed the behavior. In the example given above, “Most of the time if I ask for something and if I’ve been good, he gets it for me,” the child frames the behavior as how reliably (in terms of frequency and under what conditions) the parent responds to the child’s requests. If the child had prefaced this remark by saying, “Dad tries to get me to be good, so most of the time if I ask…” then that would indicate the child also framed it as discipline behavior. Our focus was on children’s framing or interpretation of what the behavior represented. We did not attempt to classify the parents’ behaviors from our point of view. The coding scheme allowed us to identify the common frames, or dimensions (such as responsiveness and discipline), that adolescents chose to focus on.
A second part of our coding scheme involved rating the child’s evaluation of the adequacy and quality of these parent behaviors. We did not attempt to evaluate them from our point of view. We were able to ascertain the child’s evaluation because, as suggested by the examples above, children almost always used language that expressed one or more of the following: evaluation (e.g., nice), generalization (e.g., always), attribution of causality {if I ask and if I’ve been good), and attribution of mental states such as motives (to have a better life), intentions (tries), emotions, (mad), desires (like to), etc. Children’s use of these expressions revealed their evaluations of the adequacy and quality of the parent’s behavior. For example, one child might express satisfaction with going bowling with his father on Saturdays, while another might express dissatisfaction with only seeing him on Saturdays. One child might express understanding that her mother is strict when she needs to be, while another might feel her mother is too strict. Children seldom made purely factual statements (e.g., “He makes me go to bed at 8:00.” “I see him in the morning.” “We play ball.”) that did not convey their evaluations of the adequacy or quality of the parent’s behavior.
We wanted to determine how adolescent boys and girls from European American and Mexican American families might differ in their views about what it means to be a good mother, father, stepfather, and non-residential father. To do so we tested for gender, family-type, and parent-status differences in the dimensions adolescents focused on, and in the relative emphasis they placed on those dimensions. The coding scheme also allowed us to obtain scores of how well adolescents thought their parents were doing on each dimension and, by averaging over dimensions, how well they thought their parents were doing overall. We tested for similar differences on these scores. Finally, we assessed the validity of the instrument by examining the relations between script scores and other measures of adolescents’ relationships with their parents, their internalizing and externalizing problems, positive behaviors, and delinquent behaviors.
Method
Participants
Participants were part of a five-year, three-wave, two-site (Phoenix, AZ and Riverside, CA) longitudinal investigation of the role of fathers in adolescent development. Data for the present study were collected during the first wave of the project. Approximately equal numbers of participants came from each location, and included 392 (48% boys) self-identified European American (n = 199) and Mexican American (n = 193) adolescents, ages 11 to 14 years (M =12.93 years), and their resident parents. Data were collected in the Spring (n = 200) and Fall (n = 192) semesters of children’s 7th grade school year. All three family members were of the same, self-identified ethnicity. Approximately half (n = 217) of the children were living in intact families, the rest (n = 175) were in step-father families. For purposes of recruitment, step-father families were defined as families in which the target child’s biological mother had been living for at least the past year with a man who was not the child’s biological father, and in which the target child lived with the mother more than half time. All participants were interviewed individually in their language of preference (369 English- and 23 Spanish-speaking adolescents; 204 English- and 108 Spanish-speaking mothers; 280 English- and 112 Spanish-speaking resident fathers).
Recruitment of families. Recruitment strategies varied between sites due to differing laws and school district policies. In Arizona, adolescents were recruited from eight ethnically diverse schools in the Phoenix metropolitan area. Teachers administered a short survey to all 7th graders asking about the students’ ethnic background and family composition, in return for a small donation of equipment (scanner, fax machines) to the school. A total of 2,459 families appeared eligible. A staff member at each school was employed by the research project to telephone families according to a random selection scheme to ascertain eligibility, explain the project, and ask for consent to have research staff call the family. A total of 640 families were contacted. Research staff then called families to explain the details of the project, offer a monetary reward for participation, and obtain consent as per university Institutional Review Board procedures. In Arizona, 204 families were both eligible and initially agreed to participate.
In California, families were recruited from two school districts. School staff used emergency contact cards and enrollment data to determine families that appeared eligible. They then contacted these families to explain the project and screen for eligibility. If the families agreed to participate and met eligibility requirements, research staff called families to explain the details of the project, offer a monetary reward for participation, and obtain consent as per university Institutional Review Board procedures. In California, a total of 540 families were contacted, and of these 192 were both eligible and initially agreed to participate.
Recruitment of teachers. Children were asked to provide the names of two teachers that research staff could contact. At both sites, a letter describing the project and a copy of the written consent from the parents were mailed to each teacher requesting that he or she complete the enclosed questionnaire about the adolescent’s behavior. In Arizona, an incentive of $5 cash, and a pre-paid envelope were enclosed along with the letter, and if necessary teachers were reminded by follow-up phone calls from research staff. In Arizona, 387 (97%) were completed by teachers; 197 (98%) adolescents had a questionnaire from at least one teacher, and 190 (95%) had questionnaires from both teachers. The procedure in California was similar, except that a non-monetary incentive was used in place of cash (e.g., movie gift certificates), and no follow-up phone calls were placed, due to state policies. In California 261 (68%) were completed by teachers; 169 (88%) adolescents had a questionnaire from at least one teacher, and 92 (48%) had questionnaires from both teachers.
Procedures
In Arizona, all three family members were interviewed in the home during one visit by three trained research staff members. The interviews were conducted concurrently in different rooms of the house so that privacy could be maintained. Interviewers were instructed to ensure that no participant could overhear the responses from any other participant. In California, the family traveled to the research site and all members were similarly interviewed in separate rooms. The interviews lasted approximately 2 ½ hours. As part of the larger battery, interviewers presented all the measurers analyzed here verbally to family members, except the delinquent behavior items. The delinquent behavior items were presented to adolescents in a self-administered, paper-and-pencil questionnaire mid-way through the interview.
Measures
Relationship Scripts. Children were asked to describe their relationships with their residential fathers, mothers, and if applicable, non-residential fathers. The requests always came in this order, and they were spaced widely apart during the interview. In the case of non-residential fathers, children were first asked if they were comfortable providing a description of their relationship with him “like we did with your mom and step-dad.” Children had to have a relationship with their biological fathers in order to describe it, and we were concerned that some children might not feel able to talk about their relationship with him if they had not seen him for a long period of time. Those children who had indicated that their biological fathers were no longer alive were not asked to describe their relationships with them. Children were asked four questions (see Appendix) to elicit their descriptions of their relationship with each parent. The questions were designed to be as generic, conversational, and open-ended as possible. Interviewers’ questions and children’s responses were audio-recorded and transcribed. Coding was done on the transcriptions. The 23 scripts that were produced in Spanish were translated into English by trained coders who were bi-lingual, and coding was done on the English transcriptions.