Madics: Data
The Henry A. Murray Research Archive at Harvard University currently holds the data for six waves of MADICS which can be obtained at http://www.murray.harvard.edu.
Data Sample. The sample is drawn from a county on the Eastern seaboard of the U.S. that consists of several different ecological settings including: low income, high risk urban neighborhoods; middle class suburban neighborhoods; and rural, farm-based neighborhoods. The sample is broadly representative of different SES levels, with the mean pretax family income of the participants in 1990 being normally distributed around a mean of $45,000-$49,000 (range $5,000-$75,000). White families reported significantly higher pretax incomes in 1990 than the African American families.
Wave 1. In the fall of 1991, 1,700 adolescents and their families were initially contacted to participate in the study; 1,482 families of those initially contacted participated in the study. At that time, each family had a seventh grader attending a public seventh- and eighth-grade junior high school. The male to female ratio in the adolescent sample was approximately equal (51% male, 49% female). One thousand four hundred eighty primary caregivers (92%) female, and seventh grade target youth, 431 older siblings (50% female) and 789 secondary caregivers (85% male) completed Wave 1. Fifty-four percent of the primary caregivers had graduated from high school, and another 40% had graduated from college. Wave 1 constructs focused on all the relevant social contexts of the youth including neighborhood quality, peer values and characteristics, family socialization, family management and values, family processes and relationships, the youth's perceptions of their parent, older sibling, and themselves, as well as their gender role beliefs, perceived discrimination and psychological capital. We also gathered considerable data on their perceptions of their schools and parental involvement in schooling. Parents gave us information on their neighborhoods, their economic concerns, their spiritual and social support, as well as their psychological capitol and many similar constructs as the youth, including a significant section on schooling.
Wave 2. In 1992, 1188 target adolescents (80% of the wave 1 sample), and 1223 parents (83% of the wave 1 sample) were surveyed by phone. This survey was done during the summer after the target youth's 7th grade and focused primarily on activity involvement, talents, and time use, as well as peer group membership, family monitoring and rules, and self and identity related beliefs. The parents were asked about their management of the child's activity involvement and their time use with the child, as well as their relationship with the child.
Wave 3. The next in-home survey took place in 1993 (during the adolescents' eighth grade year) when 1449 families were relocated and 1060 were reinterviewed (76% of those still living in Prince George's County). The Wave 3 sample does not differ from the sample at Wave 1 in terms of parents' education, income, race, and both marital and employment status. The constructs covered in Wave 3 were similar to those in Wave 1 for both target youth and parents, with a greater emphasis on race and ethnicity constructs and some additional measures of family processes and relationships. Parents were also asked about their expectations and aspirations for their child and about their perceptions of the child's psychological functioning and abilities. There was also a short telephone interview with the primary caregiver only in the summer after the target youth's 8th grade (Wave 3T) when we asked about the parent's involvement in schooling and again about the child's activities.
Wave 4. In the fall of 1996 through the winter of 1997 (during the adolescents' eleventh grade year), 1057 adolescents and their families (71% of the wave 1 sample; 99% of the wave 3 sample) participated in a face-to-face interview and completed a self-administered questionnaire. The focus here was on high school: relationships at school, studying and learning styles, and plans after high school; peer relationships, problem behavior, youth activities and values, and physical development were also measured. Parents contributed their expectations for the child after high school and involvement in those plans.
Wave 5. In 1998 (one year after high school), 912 participants (62% of the wave 1 sample; 86% of the wave 3 sample) completed a self-administered questionnaire or were surveyed by phone. Transition into work and college, romantic relationships, and family management at this stage are among the major categories of questions asked in Wave 5.
Wave 6. In 2000 (three years after high school), 899 participants (61% of the wave 1 sample; 85% of the wave 3 sample) completed a self-administered questionnaire or were surveyed by phone. School climate for college attendees and work climate for employed participants were added in this wave to complement constructs carried over from Wave 5.
Data Collection. In the fall of 1991, letters were sent to the homes of 1,700 seventh graders of select schools in Prince George's County, Maryland. The letter asked for parents' permission for their seventh grader (target child) and his/her parent, and older sibling, if applicable, to participate in the study. Another letter was sent to the secondary caregiver, asking for them to participate in the study as well, if they wished. The primary caregiver was asked to participate in a face-to-face interview that lasted approximately 50 minutes and to complete a self-administered questionnaire that took about 30 minutes. The secondary caregiver was asked to complete a similar 30-minute self-administered questionnaire. The target child was asked to participate in a 50-minute face-to-face interview and a 30 minute self-administered questionnaire. If the seventh grader had an older sibling who wished to participate, that sibling was asked to fill out a self-administered questionnaire similar to that of the target child. This procedure was repeated in the spring of 1993. During the intervening months, two brief confidential telephone interviews were conducted with the primary caregiver and the target child to assess the child's transition into the eighth grade. Between the months of July and October of 1993, the process was repeated with only the parent being interviewed by telephone to evaluate the child's transition into ninth grade.
Many of the questions asked during the face-to-face interview and in the self-administered questionnaire were pre-coded questions. However, the face-to-face interview also included open-ended questions to learn more about parental aspirations for the children, as well as the child's own aspirations. The questionnaires included a broad range of items about the family dynamics, family and peer relationships, resources, well-being and stressors, as well as a broad array of indicators of adolescent development.
Development and Testing of Measures. Many of the items come from other large-scale longitudinal studies. Items on family management styles, monitoring, and rules came from the Philadelphia Family Management Study (Furstenberg, 1992; Furstenberg, Cook, Eccles, Elder, & Sameroff, 1999) and from studies by Steinberg (Steinberg, 1981; Steinberg, 1990; Steinberg, Dornbusch, & Brown, 1992) and Dornbusch (Dornbusch, Ritter, & Steinberg, 1991). Items on neighborhoods, communities, and delinquency came from the National, and the Denver, Youth Studies (Elliott, Menard, Rankin, Elliott, Wilson and Huizinga, 2006). Items on family perceptions, mental health, problem-solving, and perceived economic situation came from Conger et al.'s Iowa Youth and Family Study (Conger, Ge, Elder, Lorenz, & Simons, 1994; Conger, Lorenz, Elder, Melby, Simons, & Conger, 1991), as well as the National Study of Children (Allen, Moore, Kuperminc, Bell, 1998). Items on self-concept of ability, values and importance placed on academic domains, and on gender-roles, and peer characteristics came from The Michigan Study of Adolescent Life Transitions (Eccles, Midgley, Buchanan, Wigfield, Reuman, & MacIver, 1993). Items used to assess sense of personal efficacy were developed by Bandura, Cook, and Eccles for the MacArthur Network on Successful Adolescent Development. Most items on family involvement in the school came from surveys designed by Epstein (Epstein, 1990; Epstein & Dauber, 1991). Markus and Oyserman's (Oyserman & Markus, 1990) techniques were used to assess possible selves. A slightly modified version of The Petersen Scale of Pubertal Development (Petersen, Compas, Brooks-Gunn, Stemmler, Ey, & Grant, 1993) was used to assess physical maturation; the Children's Depression Inventory (Kovacs, 1992), Achenbach's Child Behavior Check List for parents only (Achenbach, 1991), a reduced version of the Eating Disorder Inventory for Anorexia Nervosa and Bulimia (Garner, Olmstead, Polivy, & Garfinkel, 1984), and items from Derogatis's SCL-90-R (Derogatis, 1976, 1983) were added at Wave 3 to better measure the adolescents' mental health. Finally, because they were often entering relatively uncharted territory, many open-ended and projective type questions were included to allow for richer responses than possible with close-ended question formats (e.g., questions focused on school climate, parent involvement in school, race and gender discrimination, experiences of race/ethnicity, family management strategies to deal with discrimination, and identity).
At wave 4, we added the Attitude-Achievement Paradox scale created by Mickelson (Items 289, 291, 296-302 of youth self-administered questionnaire; Mickelson, 1990); items from the Multidimensional Inventory of Black Identity created by RM Sellers and colleagues (items 53 a- gg of youth self-administered questionnaire; Sellers et al., 1997); items from BK Barber on parent psychological control (items 110-118 on the youth self-administered questionnaire; Barber, 1996). The items from the MMBI were also used in waves 5 and 6 along with unique items designed by T. Chavous to assess college environments.
In addition, the California Achievement Test, taken during the target child's third and fifth-grade school years, was used as a measure of prior academic ability. The Revised Symptoms Checklist assessed adolescents' reports of how frequently in the past month they had experienced symptoms such as feeling very upset, destructive, or hopeless.
Data Sample. The sample is drawn from a county on the Eastern seaboard of the U.S. that consists of several different ecological settings including: low income, high risk urban neighborhoods; middle class suburban neighborhoods; and rural, farm-based neighborhoods. The sample is broadly representative of different SES levels, with the mean pretax family income of the participants in 1990 being normally distributed around a mean of $45,000-$49,000 (range $5,000-$75,000). White families reported significantly higher pretax incomes in 1990 than the African American families.
Wave 1. In the fall of 1991, 1,700 adolescents and their families were initially contacted to participate in the study; 1,482 families of those initially contacted participated in the study. At that time, each family had a seventh grader attending a public seventh- and eighth-grade junior high school. The male to female ratio in the adolescent sample was approximately equal (51% male, 49% female). One thousand four hundred eighty primary caregivers (92%) female, and seventh grade target youth, 431 older siblings (50% female) and 789 secondary caregivers (85% male) completed Wave 1. Fifty-four percent of the primary caregivers had graduated from high school, and another 40% had graduated from college. Wave 1 constructs focused on all the relevant social contexts of the youth including neighborhood quality, peer values and characteristics, family socialization, family management and values, family processes and relationships, the youth's perceptions of their parent, older sibling, and themselves, as well as their gender role beliefs, perceived discrimination and psychological capital. We also gathered considerable data on their perceptions of their schools and parental involvement in schooling. Parents gave us information on their neighborhoods, their economic concerns, their spiritual and social support, as well as their psychological capitol and many similar constructs as the youth, including a significant section on schooling.
Wave 2. In 1992, 1188 target adolescents (80% of the wave 1 sample), and 1223 parents (83% of the wave 1 sample) were surveyed by phone. This survey was done during the summer after the target youth's 7th grade and focused primarily on activity involvement, talents, and time use, as well as peer group membership, family monitoring and rules, and self and identity related beliefs. The parents were asked about their management of the child's activity involvement and their time use with the child, as well as their relationship with the child.
Wave 3. The next in-home survey took place in 1993 (during the adolescents' eighth grade year) when 1449 families were relocated and 1060 were reinterviewed (76% of those still living in Prince George's County). The Wave 3 sample does not differ from the sample at Wave 1 in terms of parents' education, income, race, and both marital and employment status. The constructs covered in Wave 3 were similar to those in Wave 1 for both target youth and parents, with a greater emphasis on race and ethnicity constructs and some additional measures of family processes and relationships. Parents were also asked about their expectations and aspirations for their child and about their perceptions of the child's psychological functioning and abilities. There was also a short telephone interview with the primary caregiver only in the summer after the target youth's 8th grade (Wave 3T) when we asked about the parent's involvement in schooling and again about the child's activities.
Wave 4. In the fall of 1996 through the winter of 1997 (during the adolescents' eleventh grade year), 1057 adolescents and their families (71% of the wave 1 sample; 99% of the wave 3 sample) participated in a face-to-face interview and completed a self-administered questionnaire. The focus here was on high school: relationships at school, studying and learning styles, and plans after high school; peer relationships, problem behavior, youth activities and values, and physical development were also measured. Parents contributed their expectations for the child after high school and involvement in those plans.
Wave 5. In 1998 (one year after high school), 912 participants (62% of the wave 1 sample; 86% of the wave 3 sample) completed a self-administered questionnaire or were surveyed by phone. Transition into work and college, romantic relationships, and family management at this stage are among the major categories of questions asked in Wave 5.
Wave 6. In 2000 (three years after high school), 899 participants (61% of the wave 1 sample; 85% of the wave 3 sample) completed a self-administered questionnaire or were surveyed by phone. School climate for college attendees and work climate for employed participants were added in this wave to complement constructs carried over from Wave 5.
Data Collection. In the fall of 1991, letters were sent to the homes of 1,700 seventh graders of select schools in Prince George's County, Maryland. The letter asked for parents' permission for their seventh grader (target child) and his/her parent, and older sibling, if applicable, to participate in the study. Another letter was sent to the secondary caregiver, asking for them to participate in the study as well, if they wished. The primary caregiver was asked to participate in a face-to-face interview that lasted approximately 50 minutes and to complete a self-administered questionnaire that took about 30 minutes. The secondary caregiver was asked to complete a similar 30-minute self-administered questionnaire. The target child was asked to participate in a 50-minute face-to-face interview and a 30 minute self-administered questionnaire. If the seventh grader had an older sibling who wished to participate, that sibling was asked to fill out a self-administered questionnaire similar to that of the target child. This procedure was repeated in the spring of 1993. During the intervening months, two brief confidential telephone interviews were conducted with the primary caregiver and the target child to assess the child's transition into the eighth grade. Between the months of July and October of 1993, the process was repeated with only the parent being interviewed by telephone to evaluate the child's transition into ninth grade.
Many of the questions asked during the face-to-face interview and in the self-administered questionnaire were pre-coded questions. However, the face-to-face interview also included open-ended questions to learn more about parental aspirations for the children, as well as the child's own aspirations. The questionnaires included a broad range of items about the family dynamics, family and peer relationships, resources, well-being and stressors, as well as a broad array of indicators of adolescent development.
Development and Testing of Measures. Many of the items come from other large-scale longitudinal studies. Items on family management styles, monitoring, and rules came from the Philadelphia Family Management Study (Furstenberg, 1992; Furstenberg, Cook, Eccles, Elder, & Sameroff, 1999) and from studies by Steinberg (Steinberg, 1981; Steinberg, 1990; Steinberg, Dornbusch, & Brown, 1992) and Dornbusch (Dornbusch, Ritter, & Steinberg, 1991). Items on neighborhoods, communities, and delinquency came from the National, and the Denver, Youth Studies (Elliott, Menard, Rankin, Elliott, Wilson and Huizinga, 2006). Items on family perceptions, mental health, problem-solving, and perceived economic situation came from Conger et al.'s Iowa Youth and Family Study (Conger, Ge, Elder, Lorenz, & Simons, 1994; Conger, Lorenz, Elder, Melby, Simons, & Conger, 1991), as well as the National Study of Children (Allen, Moore, Kuperminc, Bell, 1998). Items on self-concept of ability, values and importance placed on academic domains, and on gender-roles, and peer characteristics came from The Michigan Study of Adolescent Life Transitions (Eccles, Midgley, Buchanan, Wigfield, Reuman, & MacIver, 1993). Items used to assess sense of personal efficacy were developed by Bandura, Cook, and Eccles for the MacArthur Network on Successful Adolescent Development. Most items on family involvement in the school came from surveys designed by Epstein (Epstein, 1990; Epstein & Dauber, 1991). Markus and Oyserman's (Oyserman & Markus, 1990) techniques were used to assess possible selves. A slightly modified version of The Petersen Scale of Pubertal Development (Petersen, Compas, Brooks-Gunn, Stemmler, Ey, & Grant, 1993) was used to assess physical maturation; the Children's Depression Inventory (Kovacs, 1992), Achenbach's Child Behavior Check List for parents only (Achenbach, 1991), a reduced version of the Eating Disorder Inventory for Anorexia Nervosa and Bulimia (Garner, Olmstead, Polivy, & Garfinkel, 1984), and items from Derogatis's SCL-90-R (Derogatis, 1976, 1983) were added at Wave 3 to better measure the adolescents' mental health. Finally, because they were often entering relatively uncharted territory, many open-ended and projective type questions were included to allow for richer responses than possible with close-ended question formats (e.g., questions focused on school climate, parent involvement in school, race and gender discrimination, experiences of race/ethnicity, family management strategies to deal with discrimination, and identity).
At wave 4, we added the Attitude-Achievement Paradox scale created by Mickelson (Items 289, 291, 296-302 of youth self-administered questionnaire; Mickelson, 1990); items from the Multidimensional Inventory of Black Identity created by RM Sellers and colleagues (items 53 a- gg of youth self-administered questionnaire; Sellers et al., 1997); items from BK Barber on parent psychological control (items 110-118 on the youth self-administered questionnaire; Barber, 1996). The items from the MMBI were also used in waves 5 and 6 along with unique items designed by T. Chavous to assess college environments.
In addition, the California Achievement Test, taken during the target child's third and fifth-grade school years, was used as a measure of prior academic ability. The Revised Symptoms Checklist assessed adolescents' reports of how frequently in the past month they had experienced symptoms such as feeling very upset, destructive, or hopeless.