Student Projects
Research Apprenticeship

Social Support in Adolescent Pregnancy: From the Voices of African American Teens

Denise Fryzelka

Denise K. Fryzelka, CNM
MW21 Critical Inquiry II,v2,


Adolescent pregnancy is a priority health issue in the Unites States. This is directly due to the increased concomitant maternal and neonatal pathological sequela, such as pregnancy-induced hypertension, severe anemia, preterm labor, low birth weight, admissions to the neonatal intensive care unit, and maternal and infant death, and the increased negative psychosocial sequela for both the teen and her child, such as depression, infant neglect and abuse, and inadequate maternal-infant bonding. Of all developed countries, the United States has the highest rate of adolescent pregnancies, at approximately one million per year. With a pregnancy rate of 19.8%, African Americans teens are a population hardest hit by this epidemic. This is greater than a 2-fold increase over teen Caucasian rates of 8.8% and approximately 1.25 fold increase over the teen Hispanic pregnancies rate of 16.2%.

Social support networks directly influences how vulnerable populations, such as adolescents, and different sub-cultures, such as African Americans, are affected by the consequences of pregnancy. Social support refers to the number of people or resources that an individual can turn to in times of life changes and stressors. Social support can have both positive and negative influences on health. While the topics of social support, adolescent pregnancy and African American teens have been studied extensively, a review of the literature demonstrates that very few researchers have explored these areas together. Little is known regarding social support systems of African-American adolescents in terms of cultural differences from other pregnant adolescents. Exploration into the ideas voiced by pregnant African American teens regarding the sources of social support that they deem most valuable during this transitional time in their lives presents the focus of this triangulated study. Additionally, the introduction and influence of social support provided by doulas will be examined. A doula is a non-medical trained person who is present during labor and delivery and serves as a support person, providing both emotional and physical support, as well as acting on the woman’s behalf as her advocate if needed. The conceptual perspectives that guide this study include Norbeck's work regarding social support, critical social theory, and the concept of an Afrocentric approach to pregnancy presented by Mim. This paper describes the research apprenticeship to this project and the learning derived from this experience.


Chapter 1: Introduction, Scope, and Significance of the Problem
Research Questions

Introduction to the problem

Over 1 million American teenagers have continued to become pregnant yearly, despite recent data suggesting a downward trend in the teen birthrate (Coley & Chase- Lansdale, 1998). Although there have been documented recent declines in the rates of teenage pregnancy for all ethnic and racial groups over the last decade, it still remains one of the major health concerns for the adolescent population of the United States today (U.S. Department of Health and Human Services(USDHHS), 2000; Ventura, Mosher, Curtin, Abma & Henshaw, 2001).

Teen pregnancy is defined as a pregnancy that occurs in a young woman of 19 years or less. A 19% decrease in the pregnancy rates for teens 15-19 years old occurred from 1990 to 1997 when a drop from 116.3 to 94.3 pregnancies per 1000 teenagers occurred (Ventura et al., 2001). In 1999, the rate of pregnancy for this same age group was 49.6 / 1000. Although the rates for 15-19 year olds have decreased, this is not the case for adolescents less than 15 years old. Additionally, of all adolescent pregnancies 25% are repeat pregnancies, 95% are unintended, and up to 50% end in elective abortion (USDHHS, 2000).

The rates of adolescent pregnancy are problematic due to the associated economic, medical, and psychological risks and complications. The public costs for teenage childbearing as reported in the Morbidity and Mortality Weekly Report (1997) totaled $120 billion from 1985-1990. It also suggested that $48 billion could have been saved if each birth had been postponed until the mother was at least 20 years old, with these savings resulting from a decrease in seriously ill infants. Younger women of maternal ages between 10-13 years are approximately 2.5 times more likely to have a low birth-weight infant and 3.4 times more likely to have a preterm birth than women of "prime" childbearing age (DuPlessis, Bell, & Richards, 1997). Within the African American populations, there are greater incidences of perinatal morbidity and mortality as well as other medical and psychosocial concerns for the teen mother and her infant. DuPlessis, Bell, and Richards (1997) found that African American women are 1.7 times more likely to have a low birth weight infant and 2.0 times more likely to have a preterm birth than white or Hispanic women.

The pregnancy- related mortality (PRM) rate calculated for the years 1987 - 1990 was 9.2 deaths per 100,000 live births. This did not meet the Healthy People 2000 objective of no more that 3.3 maternal deaths per 100,000 live births overall. Among African American women, the PRM rate jumped to 26.5 per 100,000 live births. The difference in PRM ratios between black and white women increased from 3.4 times to 4.1 times greater in this same time period. Hispanics experience an average 10.3 PRM rate for a 14-year period. The cause-specific PRM ratios for all causes of death were 3-4 times higher for black women than white women or those of other races with hemorrhage, pulmonary embolism, and pregnancy-induced-hypertension the leading causes of death for each race group (Morbidity and Mortality Weekly Report, August, 1997).

Social support is an important factor in how the African American teen deals with both the pregnancy and the transition to motherhood. There is no specific definition that has been agreed upon in the literature for social support and social networks and how it functions to protect health and buffer stressors. However, Vaux (1988) identified the three elements of social support as social network, support behavior, and subjective appraisal of support. He suggested that social support consisted of two basic terms: (a) the perception that there is a sufficient number of available others to whom one can turn in times of need, and (b) a degree of satisfaction with the available support. Social network is a term related to social support that usually refers to relationship ties, people in one's life, or the structural aspects of support (Secco & Moffatt, 1994).

There are a variety of support systems that can be identified from family, parents, friends, partner and institutional sources, e.g. school, church. Wayland and Rawlings (1997) reported that adolescent mothers commonly identify their own mothers as their major source of support. Paskiewicz (2001) concluded that not all teens report positive relationships with their mothers, nor do the teens wish to assume total responsibility for their healthcare and that of their child. Social support can consist of both positive and conflictual components. Unger and Wandersman's study (1985) demonstrated that important outcome variables such as birth weight, health care of babies, knowledge, education, and parenting behaviors were influenced by the effectiveness of supportive interventions for the adolescent mothers. According to Perrin and Mc Dermott (1997), when positive family support is available for adolescent mothers, their chances of returning to school, entering the labor force, and finding employment improves.

Every culture has unique characteristics and types of social support systems. Within the African American community, young people are supported by families, primarily matriarchal in nature. Other sources identified are friends, schools, churches, and by the community at large; however there are locations "out there" in which there is an absence of support (Bolla, De Joseph, Norbeck & Smith, 1996). Cultural values alone, according to Norbeck and Tilden (1988) do not account for the whole picture of social support, as situational stressors, such as poverty, and other variables may influence the integrity and effectiveness of an individual’s social network.

Scope of the problem

Although there has been a recent decline in the rate of teenage pregnancy over the last decade, it still remains one of the major health concerns for the adolescent population of the United States today. The rates of pregnancy within the adolescent African American population are significantly higher than those of Caucasian or Hispanic teens. The overall adolescent pregnancy rates for 2000 are 11.8% of all births. Hispanic adolescents had a 16.2%, Caucasians, 8.8%, and African Americans, 19.8%. One of the outcome variables, low birth weight also shows a disparity for the different races. 12.9% of African American infants versus 6.4% of Hispanic and 6.6% of Caucasian infants are born to adolescent mothers with birth weight < 2500 grams. 6.7% of African American adolescents, 6.4% of Hispanic teens, and 2.3% of Caucasian teens received late or no prenatal care (Martin, Hamilton & Ventura, 2001). Specialized prenatal care for pregnant adolescents can have an impact on the outcome of the pregnancy. Fewer complications for the infants can impact favorably on the future of this generation (Smoke & Grace, 1988). The negative consequences of pregnancy within these populations have shown a decrease when there exists adequate social support systems to which these adolescent mothers can turn (Koniak-Griffin, Anderson, Verzemnieks & Brecht, 2000; Norbeck, De Joseph, & Smith, 1996).

Increasing social support may improve perinatal and psychosocial outcomes for the pregnant African American adolescent, but first an investigation into the current strengths and weaknesses of their social support systems should be done. With this information, programs can be designed that are culturally respectful and more effective. According to Norbeck and Tilden (1988), cultural differences in family ties and friendship patterns might be a vehicle to make discoveries about how social support functions. While it is possible that those whom poverty strikes the hardest of the African American community would benefit from increased social support networks from within their communities or made available to them, the generalization that support is necessary for this population as a whole should not be made. Connelly (1998) conducted a study that found that the adequacy of social support had no correlation with the number of adolescent pregnancies, which further gives importance to understanding the many components that make up social support and that increasing levels of support may not be the most effective solution.

Research Questions

The purpose of this study is to understand, identify and describe effective types of social support networks that exist or may be implemented within vulnerable African American adolescent populations as reported by the pregnant African American teens themselves. Research has shown that perceived social support is directly related to well-being (Yarcheski, Scoloveno & Mahon, 1994). This study will answer several inquiries:

v How do pregnant African American adolescents describe social support?
v What are their perceptions of the existing social support networks that are available to them?
v Are they adequate and beneficial or at least not counterproductive support systems?
v What sources or types are seen as most supportive?
v Do they believe that the characteristics of amount, type, and adequacy of social support affect them today and in their future?
v Do they feel at risk for future pregnancies during their adolescence?
v What are their goals for the future, school, employment, and their family?
v What are the necessary components that the teens describe as helpful or critical to assist or improve social support when gaps are identified in existing networks? The perceptions of African Americans on social support have been studied (Burke & Liston, 1994; Bolla, De Joseph, Norbeck & Smith, 1996) but with each study there were limitations that prevent generalization to the African American adolescent populations. Secco and Moffatt (1994) who did a review of social support theories and instruments used for research related to adolescent mothers stated that the most effective types of socially supportive interventions and the outcomes most reflective of treatment effects are still relatively unknown. Norbeck (1988) reviewed a number of studies that demonstrated the effects of social support among a vast variety of population groups and recommended that qualitative research be done to understand how social support functions for specific groups or situations. Connelly (1998) suggested adequacy of social support and its development for pregnant adolescents as a topic for research. Raisler (2000) recognizes that the care of adolescents is a population that midwives often serve and she prioritizes it as an area for midwifery research efforts to be focused.

Significance to women's health

The decisions that an adolescent makes during the time of her pregnancy will affect her health for life. These same decisions will also affect the health of her child. Many of these decisions are mediated through the avenues of social support. From physical to psychological, the results are innumerable. It is important that healthcare providers are able to assess and intervene when there are insufficient social support networks for the pregnant adolescent potentially placing her health and that of her child in further jeopardy. The African American pregnant adolescent is no exception. There is an even greater need to become aware of the needs of this clientele as many health care providers have not received formal exposure to this culture and are perhaps oblivious to culturally competent care. As evidenced in clinical practice, the number of African American health care providers appears to be greatly disproportionate to the number of clients within this same population (Declercq et al., 2001). This creates potential barriers and lack of adequate cultural appreciation, awareness, and knowledge for the African American social networks and ways that they can be used to assist pregnant teens. Access to and greater knowledge regarding social support in relation to pregnancy and other life changing events will be advantageous to health care providers in their provision of care to these adolescents. A heightened awareness of the need for intervention during critical times in the adolescents' exposure to major stressors and the ability to reinforce or equip her with adequate coping techniques is crucial (Perrin & McDermott, 1997). Identification and implementation of appropriate interventions to fill gaps in or within existing social networks or to augment those already in place within the community will not only be beneficial for her but others in her same situation.

Significance to midwifery

Caring for women is the foundation of midwifery care. Midwives ascribe to a philosophy of care that includes among its tenets, acknowlegment of the right to safe, satisfying health care, the physical care of the individual and family, as well as, the emotional and social support, and active involvement of significant others according to their cultural values and personal preferences. They serve as advocates, encourage active participation in health care choices and provide safe, competent and non-interventive care in normal processes. In addition, they promote health education for women through the childbearing cycle (American College of Nurse Midwives(ACNM), 2001b). Deeply imbedded within this philosophy and ethics is the care of all women and their families without discrimination, regardless of any other descriptive characteristics of their lives, such as religion, life-style, sexual orientation, class, race, healthcare problem, and socio-economics status (ACNM, 2001a) They also promote community...efforts to ensure access to quality care and to meet the health needs of women and their families. Midwives in particular are a group of health care providers frequently involved in the care of vulnerable populations (Raisler, 2000; Declercq et al., 2001). Many pregnant African American adolescents receive midwifery care as their sole form of healthcare (Bruckner & Muellner,1985; Scupholme, Paine, Lang, Kumar & DeJoseph, 1992).

The midwife tends to focus on the holistic dimensions of health, including the social and psychological components of health, rather than just surveillance for pathology. In order to provide health guidance to pregnant African American adolescents, the midwife often addresses the social support systems and networks of the women she/he serves. The midwife can build on this to either augment care to women and their unborn children, alter the delivery or content of healthcare or to address concerns or devastation caused by inappropriate support. Further research into the understanding of strengths and deficiencies in social support systems of African American communities, the important types of social support for pregnant women, and their perception of social support will help to assist the midwife in ensuring optimal health for the women and children of these communities (Bolla, De Joseph, Norbeck & Smith, 1996). The purpose of this study is to understand, identify, and describe effective types of social support networks that exist and what degree they are perceived and reported by the pregnant African American teens themselves. These results will be eventually utilized in order to design and implement interventions within vulnerable African American adolescent populations in order to reduce national rates of perinatal morbidity and mortality.

Chapter 2: Review of Literature & Conceptual Framework

Problem statement

Pregnancy rates among African American adolescents have declined in the past decade (Coley & Chase-Lansdale, 1998) but the lack of decrease in complications and the increased cost associated with these complications and social services support to these families continue to remain a priority health issue for the United States (Ventura, Mosher, Curtin, Abma, & Henshaw, 2001). There are several studies (Koniak-Griffin, Anderson, Verzemnieks & Brecht, 2000; Norbeck, De Joseph, & Smith, 1996; Yarcheski, Scoloveno & Mahon, 1994) that support the idea that adequate social support in pregnancy among vulnerable populations improve the rates of complications to the infant and mother, improve the psychological well-being of the mother, and demonstrate a significant decrease in the cost associated with the care of infants.

Article identification and retrieval

A review of the literature was performed through word searches via several different search engines individually and simultaneously. The key words and word groups searched were adolescent pregnancy, teen pregnancy, African American, African American adolescent pregnancy, and social support. Predominantly, Pubmed, Medline, PsycINFO, and Cinahl were utilized through Silver Platter at the Kansas University Medical Center library. The years that were reviewed in the searches were 1980-2001. A review of the list of hits from these sources and the abstract review assisted in the determination as to relevance to the topic matter. The selected resources were retrieved, reviewed, and reproduced. Analysis of the articles was made to determine as to whether they would provide useful insight to the topic that was chosen for research. The references of the articles also were reviewed to determine if other sources would be helpful in further researching and in the presentation of the current information base for social support and its relationship to African American adolescent pregnancy and health outcomes.

Review of the literature

In a qualitative study focusing on the unique social support needs of lower-income African American women during pregnancy, Bolla, De Joseph, Norbeck, and Smith (1996) carried out focus group interviews. The sample population consisted of 6 African American women, all of who worked with and were familiar with the needs of low-income women of their ethnicity. They discussed experiences encountered in working with their clients, the issues that these low-income pregnant women experienced, and their perceptions and ideas of culturally appropriate support strategies and interventions from their own personal and professional experiences that could benefit this population of women. Six major themes were identified from the content analysis of the transcriptions resulting from the discussions. These included strengths of the African American communities, deficiencies in social support for many low-income African American women (see Table 1), the meaning of pregnancy for African American women, important types of social support, "making it" as descriptions of strengths of low-income pregnant women, and social support as the means for helping to facilitate and maintain a connection to society for the pregnant adolescent.

Table 1: Strengths and Deficiencies of the African American Support Networks
Strengths of the African American CommunityDeficiencies in social support resources- areas "out there" where "class of lost souls" reside
Family- essential source of social support
Community at large:
Support to excel
Foster a sense of anticipation and growth
Create an excitement for and expectation of achieving important things
Supportive Factors from the family - once readily available but because of "changing times" may now be absent: giving directions, instilling values and goals, providing social guidance, and conveying love and acceptance
Lack a nurturing home or "nest"
Lack of guidance from parents, schools, and churches
Lack of long-term goals
Lack of hope
Lack of self-esteem
Lack of positive role models
Lack of perceived options
Lack of self-efficacy
Lack of marketable skills
Lack of values
Lack of emotional bonds
(Bolla, De Joseph, Norbeck, & Smith, 1996)

Some of the limitations of the study are the small sample number. There were as well only 2 focus group meetings, which did not allow for clarification of concepts after analysis. It was uncertain whether the participants were low income but they were employed in some capacity within the healthcare setting. As well, the sample did not include pregnant women or adolescents specifically and participants were from one geographic location. These may all serve to limit the generalizability of results to all African American women, and even to low-income women of the same ethnicity. The major strengths of this study are the introspection of African American women in describing the needs of their culture and the familiarity that the interviewees had with low-income issues and women. The discussions and findings directly relate to the inquiry of social support needs of low-income African American women consistent with this study as well as begin to present an understanding of the issues of low-income women from an African American perspective that has been difficult to find in previous literature. This study provided some of the groundwork for development of the social support intervention discussed in the next study reviewed.

Norbeck, DeJoseph, and Smith (1996) did a two part study, first part, qualitative and second part, randomized trial measuring the outcome variable of low-birth weight (LBW) among African American women. Several decisions by the researchers made this study unique from many others addressing social support and health outcomes. The first of these decisions was the incorporation of the population at risk, those who reported inadequate social support, into the sample rather than inclusion of participants who reported adequate social support which could serve to skew the results and reality of the findings. Secondly, was the implementation of an intervention based on empirical or theoretical data.

In the first part of the study, qualitative research collecting narrative accounts in individual and focus group settings was used to establish an appropriate intervention by learning about the life situations of low-income African American pregnant women and about the characteristics that they needed during this life event. The sample size was not indicated in the journal article but the participants granted permission for use of video taping that was utilized to gather both observational and narrative data. It found that for African American women, support from her mother or male partner was related to the social support that validates emotionally significant aspects of pregnancy, in contrast to unsupported women who expressed the dilemma of bringing a new life into the world and having no one who cared about it. These results validate those from a previous study by the same authors that found a relation between adverse pregnancy outcomes and lack of support from the mother or male partner of the pregnant African American (Norbeck & Anderson, 1989).

The randomized clinical trial then analyzed data for the intervention group and the control group retrieved from the medical charts by reviewers unaware of the group assignment of the participants. The sample consisted of second trimester pregnant, low-income adult African American women who were invited to participate from several different prenatal clinics on the west coast. An assessment of social support adequacy using the Norbeck Social Support Questionnaire (NSSQ) was completed by each individual. Other measurements of established reliability and validity, such as the Rosenberg Self-Esteem Scale (RSES), the Hassles Scale, and the Spielberger State-Trait Anxiety Inventory were also used as for secondary multivariate analysis purposes. Scores were evaluated and 36% of the original 319 women were deemed to have low or inadequate social support and these women were randomized into the control or intervention group. Both groups received standard prenatal care but those in the intervention group were invited to participate in face-to-face sessions focused on different aspects of the woman's life and the associated social support she had available at those times, the meaning of her pregnancy to her, response to a focus group video from the qualitative part of the study, and discussions of self-esteem and types of relationships that inhibit or foster self-esteem. Demographic data for the two groups did not differ statistically. Results were correlated from the medical chart data retrieval and found support for the original hypothesis that social support interventions were effective in decreasing the rate of LBW when applied to appropriately identified low-support women and when an empirically supported intervention addressed the qualities that have been associated with pregnancy outcomes. Among the different ethnicities, the effectiveness of the identified main support person may hold similarities although this was not the focus of this study; but the contrary may also hold true. Cultural differences according to previous work by the primary author (Norbeck & Anderson, 1989) have been found to exist in determining who is considered an effective support person during pregnancy.

Limitations of this study include the possible inability to generalize the results to other groups or populations, as this group was intentionally sampled to determine the specific needs and results for this African American population. Another stated limitation within this study is the inability of the study due to its design to uncover the cause-effect relationships between social support and maternal-fetal outcomes. Small sample size was another limitation to carrying out multivariate or stratified analysis. One of the major strengths that this study supports is that appropriate supportive interventions can improve health outcomes for newborn especially in low-income African American populations. The main object of the research to discover the perspectives of the at-risk populations that interventions are meant to serve or assist has not only been carried out in this project but an implemented intervention has been tested as well. The research at hand contains similar components to this study with a focus on this specific population sample and in the evaluation of a social support intervention. The results may lend to a greater credence of this study’s conclusions.

Family social support and prenatal care among single African American pregnant teens was studied by Cosey and Bechtel (2001). A sample of 25, 15-18 year old African American primiparas living at home were studied. A comparison was also made between non-family support system strength and that of family support systems. A very important factor that led to this research study was the recognition that rarely has an ethnocentric perspective been given in carrying out an examination into the health beliefs and practices of the African American culture. They recognized that social support has been studied within the African American teenage population but that the type of support and integration into their family or community structures has not been implemented. The basis for their study was their belief that family social support is instrumental to the health and well-being of both mother and infant but the duration and type and extent to which this influences prenatal care and health outcomes is unknown.

Questionnaires were given to the participants 1-2 days following delivery prior to discharge from the hospital pertaining to the extent of family and non-family social support networks, and perceived social support. Prenatal care initiation as well as number of visits was determined from the demographic data as well. Forty-eight percent received adequate prenatal care versus 52% which was defined as inadequate without differences found in demographic variables of age, income, or education. No statistical differences were found between the two groups of care. All family support system networks were scored higher than non-family support systems. More social support was received by 17-18 year olds than 15-16 year olds. The implication of these findings in reference to the present study are that African Americans identify family as a source of support over non-family sources. These findings support previous study results and their implication that the major sources of support for different cultures can be found and utilized to incorporate family into intervention programs and assist in health care decisions for the improvement of healthcare outcomes and other variables. The authors encourage future research to clarify between received support and perceived support, which is one of the objects of this research study.

Schaffer and Lia-Hoagberg (1997) reported on a descriptive, correlational quantitative study addressing the effects of social support from partners and others to the adequacy of prenatal care and health behaviors of low-income women. The sample included 101 low-income women, primarily single, in their early twenties, between 28 and 40 weeks gestational age, and of various ethnicity, Caucasian, African American, Hispanic, and Native American. Self reports, including the NSSQ, the Prenatal Health Questionnaire (PHQ), and the Demographic/Pregnancy Questionnaire, as well as, a medical records review were utilized as sources of data. Support systems were identified in number and amount; adequacy of prenatal care and substance use prior to and during pregnancy was determined. Demographic data was included in the analysis. Results from a comparison made between partner and others' (except partner) support demonstrated that there was a positive correlation between partner support and adequacy of prenatal care in low-income women with the perception of love and value as relevant factors comprising the quality of this support. There was also a positive correlation between others' (mothers, sisters, friends, extended family) support described as advice, information, and talking and prenatal health care behaviors (see table 2).

Table 2
Supportive Health Behaviors- primarily with a prenatal focus- by social support network members
Obtain and maintain prenatal care
Talking / information-sharing / advice about pregnancy, progression, and bodily changes
Participation in prenatal education
Nutritional intake- what are food and beverages choices and patterns of eating?
Use of tobacco, alcohol, illegal / narcotic drugs
(Schaffer & Lia-Hoagberg, 1997)

Strengths of this study include the isolation and identification of partner support and the importance of studying and comparing the individual types and sources of social support. The use of an ethnically diverse sample population is impressive as to the generalizability of results to low-income women of different ethnicities but due to the relative ease of access to prenatal care for this urban population, the generalizability may be impeded to others who experience this as a significant barrier to care. The author who based her study on a family stress framework believes that eliciting only the women's perspective presents a limitation. Content validity was determined via the review and critique by four healthcare and academic professionals. Reliability had not been estimated for the PHQ's first time use in this study as internal consistency was deemed inappropriate in regards to this tool but test-retest reliability was recommended for future use. Finally, it does not focus on African Americans or adolescent populations specifically in terms of the cultural differences that may have an influence on the amount and types of support. For the purposes of relevance to the study at hand, this study serves to reinforce the idea that all components and sources of social support need to be assessed and how these individually can affect reception of care and consequently serve to improve outcomes.

Paskiewicz (2001) studied the relationships between African American adolescent mothers and their mothers. Mothers are often identified as the major source of support for adolescent. The study was based the belief that there is a dynamic and interactive relationship between mother and daughter and this is transformed with the occurrence of pregnancy. Interviews in a semi-structured format with open-ended questions were conducted, recorded, and analyzed in this descriptive qualitative study. A convenience sample of 15 low-income primipara African American teen mothers and their mothers / mother figures were utilized from another study related to adolescent mothering. Additional criteria were that the teen mothers had received prenatal care through a teen obstetric clinic and had delivered and were raising healthy term infants. Separation of the pair for interviewing purposes occurred to maintain confidentiality and freedom of expression at approximately 12 months postpartum. Data reduction, category coding and definition, data display via grids, and conclusion/verification drawings were components of the data analysis. Experts were consulted throughout the process of data collection and analysis and specific steps were adhered to in order to maintain credibility of the investigator.

From the content analysis of the data, four major themes emerged. Two of these themes were symbolic: communication between the mother and daughter and roles changes. The other two, classified as interactive themes, were conflict and social isolation. The author reported that lack of direct communication about the pregnancy between the mother and daughter was significant until others confirmed suspicions regarding the reality of the pregnancy. Denial, anger, and guilt were responses that both the mothers and daughters experienced. School attendance and completion often became the bargaining tool for acceptance of the pregnancy between the two. Grandmothers defined their role primarily through child-care, newborn care education, and giving advice. Adolescents identified with the mothering role predominantly through child care activities and decision-making regarding the child. The role of grandmother was identified either with ambivalence or satisfaction in terms of future expectations of the grandchild. Some of the same ambivalence was included in the adolescence description of their role as a mother. When relationships were described as positive between the two, so were the new roles and changes in their lives. The converse also was true. Conflicts that were most noted were child-care and housekeeping issues and responsibilities, school attendance, and friends. Some degree of social isolation was reported by all the participants.

Limitations of this study are the small sample size and the possible inability to generalize to all populations and ethnicities due to the sample composition of low-income African Americans adolescents, although some of the results seem to relate to all adolescents. It does nonetheless for this same reason directly relate to the topic of research of this study and provides a greater understanding of the relationships between adolescent mothers and one of their frequently cited and greatest sources of support, their mothers. The author alludes to the potential bias of responses created by the inclusion of the participants in the other study from which these participants were recruited. The author concludes that among other benefits, interventions that strengthen the mother-daughter relationship could improve infant health, mother and daughter psychological well-being and adolescent development.

Koniak-Griffin, Anderson, Verzemnieks, and Brecht (2000) studied the effects of a public health early intervention program for adolescent mothers during pregnancy and compared it to a control group that received traditional public health nursing care. The dependent variables for this study were various social and health outcomes of adolescent mothers and their children and the quality of mother-child interaction.

The early intervention program (EIP) is the theoretical framework upon which the study is based and was designed to improve prenatal health behaviors and perinatal outcomes, maternal-child interaction and health, maternal educational achievement and social competence. It utilizes a public health nursing model "to help the childbearing adolescents to gain skills in managing their internal world (e.g. perceived stresses, feelings of self-esteem, depression) and in interacting with their external environment so as better to manage their daily lives". Eligible referrals to the county health department that desired to participate were randomly assigned to two different groups from computer-entered data using specific classifications. While participants in both groups were followed from pregnancy through 6 weeks postpartum, the experimental group, the EIP received greater home visits by public health nurses for 12 months postpartum at an approximate ratio of 17:2 for the control group provided traditional public health nursing (TPHN). The adolescents in the EIP group also received motherhood preparation classes. Five major areas of intervention that were addressed for the EIP group were health, sexuality and family planning, life skills, maternal role, and social support. For both groups, these visits were in addition to their routine prenatal care.

The final sample size of 121 participants were adolescent primiparas, from primarily underserved and impoverished minority populations, with no major medical or obstetric health problems including chemical dependency, and at a gestational age less than 27 weeks. Data was retrieved from medical records to measure health outcomes of gestational age at delivery, infant birthweight, number of hospitalization days, complications of pregnancy. On intake and repeated at 6 weeks postpartum, a structured interview of closed-format, multiple-choice and fill-in questions and an instrument packet containing various scales such as the RSES, the Community Life Skills Scale, and the Pearlin's Sense of Mastery Scale was implemented. The objectives of these data collection methods served to assess among others, social competence, education and participation in school, and use of health care services and health related behaviors. Video recordings of the mothers and infants during a standardized teaching using the Nursing Child Assessment Satellite Training (NCAST) procedure were made as well. Results indicate reduced premature birth and low-birth weight rates were found in both groups of mothers although the premature gestational age in the EIP group averaged 1 week greater than the TPHN group. Descriptive statistics revealed that 92% of the adolescents experienced a normal vaginal delivery of a full-term healthy infant, 5.8% of the adolescents delivered a preterm infant, and 8% of the infants born to the adolescents had a low birth weight. No significant statistical differences were found between the groups for type of delivery or infant birth weight. Infants of the EIP group had fewer hospitalization days for birth and re-hospitalization during the first 6 weeks of life. More positive educational outcomes were demonstrated in the EIP group than in the control group.

The large sample size of 121 low-income participants, as well as the criteria of pregnancy and adolescent, provided some of the strengths for this study and related it to the current research underway. Language-specific questionnaires were used for the various ethnicities represented. Randomization eliminated concerns of selection and bias to the two groups. Although probably not cost effective, the positive effects of social support that the social and educational visits provided with the implementation of this intervention was a strength of the study independent if its results. Reliability and validity were accounted for throughout the study by the use of sound instruments, nurse recruiters were not involved in delivering health care to the participants, evaluator-coders of the NCAST were unaware of the group's assignment, and abstraction of data had an inter-rater agreement of 95-100%.

Few limitations could be identified. Recruitment occurred from adolescents that were already seeking health care so in a sense the results cannot be generalized for the population at even greater vulnerability that does not seek prenatal care. Implementation would become difficult if clients could not be identified though. In terms of the relevancy to the undertaken research the ethnicity of the participants, being primarily Caucasian, provides information for comparison perspectives but the ability to generalize these results may be limited due to probable variation in cultural aspects of African Americans.

Maton, Teti, Corns, Vieira-Baker, Lavine, Gouze and Keating (1996) use a cultural specificity framework to make a comparison between three studies that were carried out. The studies focus on sources of support in terms of parental, peer, partner, and spiritual support. Levels from the various sources as well as cultural differences in comparison to variables of psychological well-being areas or academic adjustment were made across three age groups and between two ethnicities, African Americans and Caucasians. The psychological variables include depression symptoms and self-esteem. The first study focuses on teen pregnancy, the second on academic achievement and the third is essentially a control group in that it does not focus on a particular life event as did the former two. It compares an historical cultural approach or evaluation of the life events from a extensive review of the literature after describing the correlation between the cultural acceptance and level of support for the different life events.

For purposes of relevance to this research only the first study focusing on adolescent pregnancy and the third study, the control group will be discussed. Neither the topic of academic achievement nor the age group addressed in the second study provide relevancy to this research. Several hypotheses were generated based on the cultural meaning of events framework and historical cultural resources framework: a) parental, peer, and spiritual support will be perceived as more available for African Americans versus Caucasian pregnant teens; b) that parental support will be more critical for well-being and partner support less critical for African American versus Caucasian pregnant teens.

A large sample of 102 adolescents was recruited from 2 urban teen mother clinics, 59 African Americans and 43 Caucasians. Adolescents were primiparas, less than 18 years of age, greater than 12 weeks gestational age, and planning to keep their babies. Both a questionnaire and interview process was conducted in an apparently ethical manner that addressed perceptions of support and well-being. Various scales of widely demonstrated reliability and validity were utilized, such as the RSES, Brief Symptom Inventory (BSI), and the parental social support scale. The data were tabulated, and analyzed for the different types of support and variables of well-being. Results revealed that no statistical differences could be found for age, current opposite-sex relationship, living at home, romantic relationship with the father of the baby, or socioeconomic status. Slight differences of greater gestational age, less likelihood of marriage, and more likely to have remained enrolled in school existed for the African American adolescents than their Caucasian counterparts. Statistical differences in depression and self-esteem did not exist between the two groups. Higher levels of spiritual support were reported for African Americans than Caucasians but no differences in parental, partner, or peer support. Parents were identified by the majority of both groups as the most supportive person, with the opposite-sex partner next, and finally some other person. When psychological variables were analyzed with types of support, peer support was inversely related to depression in African Americans and partner support was inversely related to depressive symptoms in Caucasians. The opposite was found in a positive correlation for both groups. Neither parental nor spiritual support was differentially related to well-being for either group.

The third study with a sample size of 215 participants, 118 Caucasian and 97 African American looked at ethnic differences among a diverse group of African American and Caucasian adolescents and young adults in terms of ages, academic status, and different life contexts. The prediction for this study based on a historic cultural resources framework was that parental and spiritual support would be perceived as more available for African American than Caucasian adolescents and young adults. Secondly, parental and spiritual support would be more critical for well-being for African American than Caucasian adolescents and young adults. Three components, regularly experiencing God's love and caring, a relationship with God, and having a religious faith that helps with coping in difficult times were utilized on a scale format to assess spiritual support. These same variable were assessed in the first study. Results after MANCOVA analysis demonstrated that there were no statistical differences in age, gender and socioeconomic status between the groups. African Americans reported higher levels of parental support, spiritual support, well-being and self-esteem and marginally lower levels of depressive symptoms than Caucasians. No significant differences for peer support were found for the two groups. Parents were identified almost 2 times greater as the most supportive person for African American while Caucasians reported almost 3 times greater an opposite sex partner as most supportive. For both groups, parental support was significantly and inversely related to depressive symptoms while parental and peer support was positively significant for self-esteem. Spiritual support was positively related to self-esteem in African American but negatively related in Caucasians.

Limitations of the study identified by the authors were that the partner social support scale created for the purposes of the study may lack validity and that only Caucasians were utilized as interviewers. Additionally, participants were not evaluated as their identification with the culture of the group nor their norms and values associated with specific life events of social supports systems. The studies participants were from eastern US city schools and health clinics; this may limit the generalizability of results. For the most part, the study applied sound research principles of ethics, data collection, analysis, application to framework and hypotheses. The major benefit in relation to the research of interest is the perspective of cultural specificity that was utilized within this study. This supports the idea that social support interventions should involve those populations affected or assisted by the interventions. Support type, amount, source, and expression are dependent on and specific to the culture and this is an important factor to assess and utilize in effective research process.

Social Support, relationship quality, and well-being among pregnant adolescents was studied by Stevenson, Maton, and Teti (1999). This was a correlational study of 110 single, pregnant teens, 67 black and 43 white recruited from 2 East Coast prenatal teen clinics of similar ethnic and demographic characteristics. They were in their second and third trimesters of pregnancy, having their first baby with the intention of keeping the infant. This sample is similar in criteria and composition to the previous study which had some of the same authors leading to the speculation that the same sample was used but with different variable outcome investigated, although no mention was made as to whether this was the case. Questionnaires were utilized for data collection including the instruments such as, Procidano and Heller's parents and friends scales, the Marital Adjustment Scale (MAS), Hollingshead's four-factor index, and the BSI.

Unlike many of the related studies this one focused on the pregnant teens prior to delivery rather than post-delivery and included an investigation into the two way exchange of support between these pregnant teens and their parents and peers rather than on just the reception of support. Additionally, the quality of the partner relationship and social support were examined in relation to the five measures of psychological well-being of depression, anxiety, self-esteem mastery, and life satisfaction. The results found that the exchange of support between parents and teens was positively correlated with increased mastery and life satisfaction and decreased depression and anxiety. This same correlation for well-being did not occur with friends in the same manner. Pregnant teens, to a large degree, also equally provide and receive high or low levels of support rather than just the one or the other with their parents and friends and thus can be interpreted as having active participation in their support networks. Increased self-esteem, higher mastery, and lower depression among pregnant teens dating the father of their child were associated with a high quality relationship with a significant other.

This study is relevant in that it provides insight into the bi-directional characteristics of support and the impact that this has on health outcomes. Additionally, it presents the idea that interventions that are of a social support nature should be based on two-way versus that support the pregnant and parenting adolescent should be given alone. Incorporating the source of the support, partner or parental appears to have evidence that this component may have an important role to play as well in outcomes for the adolescent mother and consequently her infant. Limitations that are significant are generalizability as to location of the participants geographically, those who seek care, and single, pregnant Caucasian and African American teens. The study does not specifically address low-income women, which is pertinent to the research at hand and evidenced to be related to inadequate support systems. Additionally, it does not focus on other types of support, such as tangible or informational aspects, which could have a role to play beyond merely emotional components (see Table 3).

The perceptions by adolescent mothers of social and community support and the impact of parenting on their lives were the focus of the research by Burke and Liston (1994). The Adolescent Parent Perception Inventory was self-completed by 78 non-urban, first-time adolescent mothers who were enrolled in community-based programs in the eastern US who made up the sample population of this descriptive survey. Excluded were physically, cognitively, or psychologically impaired mothers or those with developmentally delayed infants. They were questioned regarding the importance, availability, and helpfulness of the services in the community, to identify social or professional persons that provided tangible support (advice, information, emergency help), and the impact of parenting in their lives as well as their future projections and advice to pregnant friends. Both open and close-ended questions were included in the survey that was then subjected to content and frequency distribution analysis. With a focus on non-urban community-based programs, the results from this study determined that the most important community service to this population was infant health care, followed by visits to the home by a nurse. For social support sources, the mother and the father of the baby were most consistently identified, with some reference to a friend for more limited support. Finally, the pediatrician was most highly rated as the source of professional support followed by the mother's own doctor, and a social worker. Categories from the content analysis about parenting revealed that the most common surprise since becoming a parent was the increased responsibility and personal growth, restrictions on their lives were cited by most as the worst change and self-improvement as the best change that occurred. Future live projections included increased finances, a car, a job, followed by marriage, and another child or pregnancy. Advice to a friend was related to decision-making activities and the encouragement to think, know options, and stick to one's own decisions.

The limitations that this study presents are the non-randomized sample that consisted of 87% Caucasians, 12% African Americans, and 1 Hispanic. It would not be possible to accurately generalize the results to large populations of adolescent mothers of all ethnicities and socioeconomic levels, although for the latter consideration, 75% reported receiving Medicaid. Demographically, other inconsistencies existed such as living arrangements and years of education. The tool used to gather information had some potential concerns of reliability and validity as it was created for the purposes of the study, although it did undergo pretesting and revisions.

The findings of this study contribute to the knowledge that not only parents, partners, and family are important sources of support but that community and professional services have some important attributes to contribute as well to adolescent mothers and their families. This gives credence to the continued implication from this literature review that all sources, types, amounts and characteristics of support should be evaluated for the consideration of the different populations that they will serve and focus on utilization of most frequently identified sources. This study also supports the belief of this researcher that adolescent mothers themselves should be assessed for whether existing services or support systems are sufficient, desired, and beneficial to her needs or lacking in certain components or access.

Connelly (1998) explored the association between perceived hopefulness, self-esteem, and social support with pregnancy status in adolescents. The convenience sample included 91 non-pregnant and 58 pregnant adolescents between the ages of 14-18 from high schools and social service agencies. Included in the study were five different ethnicities from urban and rural settings. Three instruments were utilized in this cross-sectional design study. The Hopefulness Scale for Adolescents, the RSES, and the NSSQ were analyzed by multiple regressions whereby each of the three variables was associated with pregnancy status. When socioeconomic status and age were controlled neither of the variables of self-esteem, social support, nor hopefulness were found to be associated with pregnancy. Prior to controlling for these two components, tangible support and self-esteem were associated.

Whereby the variety of ethnicities and the even distribution of participants from rural and urban settings is impressive, the overwhelming percentage of Caucasians within this sample group totaled 84% pregnant and 91% non-pregnant. The non-randomization may bias the findings as well. The limitations for generalizability of results is significant to a Caucasian adolescent population alone taking into account the geographical and socioeconomic boundaries. The three instruments used were appropriate for adolescent populations and have significant validity and reliability ratings. Although the findings did not support the association between social support and pregnancy in adolescents this study did suggest limitations and was nonetheless useful in addressing the issue and presenting another perspective. Adequacy of social support in this study was not facilitated via the instrument utilized. This would have been an aspect of relevance to the research at present. The investigation of whether these variables lead to an adolescent women seeking pregnancy are important issues to understand although a cultural perspective would seem an appropriate avenue for this research.

Conceptual framework

With the understanding that for a qualitative study, a theoretical / conceptual framework is difficult to utilize and sometimes ill-advised, the purpose of the presentation of this framework should serve to define and explain the concepts of social support and its relation to adolescent pregnancy in African Americans, as well as, provide a perspective to guide the flow of the interaction of concepts including the importance of an Afrocentric approach. The clarification of these concepts and the specificity that the research will focus on can be seen in the conceptual map. The research study will explore the perspectives of the low-income young pregnant African American women in terms of their social support networks and their perceptions of both adequacy and identification of gaps within these systems. Through this exploration and thorough assessment social support interventions that are culturally derived may be developed and implemented for supplementation of the needs of adolescents within these communities. An evaluation of the effectiveness of the interventions will lie in maternal and infant health outcomes statistics and an evaluation of health care cost reduction. The conceptual perspectives that guide this study include Norbeck's work regarding social support, critical social theory, and the concept of an Afrocentric approach to pregnancy presented by Mim.

Social support

Due to the extensive research by Norbeck though the last 20 years, specifically, in relation to low-income pregnant African American adolescents, the conceptual framework for this study seems appropriately based on her work and the focus that she has placed on social support (DeJoseph, Norbeck, Smith, & Miller, 1996; Norbeck, De Joseph, & Smith, 1996; Norbeck, & Anderson, 1989). She has contributed to the advancement of social support research through numerous research studies and the creation of a social support questionnaire, named after herself, the Norbeck Social Support Questionnaire (NSSQ). The basis for this instrument is the definition of social support by Kahn (1979) who defined social support as "interpersonal transactions that include one or more of the following: the expression of positive affect of one person toward another; the affirmation or endorsement of another person's behaviors, perceptions, or expressed views; the giving of symbolic or material aid to another" (p.85). In short, the three components of supportive transactions are affect, aid, and affirmation, which also comprise the functional component of Norbeck's instrument. Kahn further describes social support in terms of "convoy" in how those that an individual relies on for support and likewise those that rely on him are constantly changing over time, sometimes joining and at other times separating from the network (Norbeck, Lindsey, & Carrieri, 1981).

According to Norbeck, Lindsey and Carrieri (1981), three main concepts are included in the framework of social support systems: functional components, social support network, and total loss. Functional components include emotional (affect and affirmation) support and tangible (aid) support (See Table 3). Properties of the "convoy" social support network that are included in the NSSQ are size, the number of persons involved, stability, the duration of the relationships, and availability, the frequency of contacts. The relationships between people may or may not actually provide social support. Total loss is a characteristic of the number and perceived amount of recent support loss one experiences within the social support network because it is ever-changing. Perceived support is also measured in the NSSQ based on the reciprocal support referred to in Kahn's definition (Norbeck, Lindsey, & Carrieri, 1981; Norbeck, Lindsey, & Carrieri, 1983). What is not included in the questionnaire or conceptual framework that is an important aspect for this study is the component of adequacy of support, which is essentially the subjective perception that the support that an individual has is sufficient. The other aspects of amount, type, source, and contribution by the source are otherwise answered.

Table 3: Socially supportive behaviors: Tangible, Informational, and Emotional
*Material items including clothing, supplies
*Money / Financial assistance
*Medical / Dental insurance
* Other forms of resources
*Problem solving
*Components of limit/boundary-setting and discipline
*"Being there for me"-present
*"Can call anytime"-available
*Willing (to help)
*Affirmation / Value
*Affect / Love
*Inclusion-feeling of belonging
*Components of limit/boundary-setting and discipline
(in part: Norbeck, Lindsey, & Carrieri, 1981)

Additional concepts referred to by Norbeck and Tilden (1988) and described as shared assumptions in the social support literature are that a) social support refers to interpersonal interactions and relationships; b) these interactions provide emotional support or actual help with tasks or problems; c) this support or help is usually provided by members of the individual's social network, not by strangers, professionals, or casual acquaintances; and d) social support is both given and received by the members of the network, and the members try to share equally in the giving and receiving. In later work Norbeck suggested that social support serves as a buffer for stress on health outcomes and among pregnant women (Norbeck, 1988; Norbeck & Anderson, 1989) as well as directly affecting stress and health. As has been seen from the review of the literature, increased social support for those who have inadequate resources improves physical and psychological health outcomes for mothers and infants. The form and style of social support networks also seems to be strongly influenced by cultural differences (Norbeck & Tilden, 1988). In later work, as discussed in the literature review, (Bolla, De Joseph, Norbeck, & Smith, 1996; Norbeck, De Joseph, & Smith, 1996) focus groups, including a professional advisory group, community workers, and pregnant women and individuals interviews were conducted to explore the supportive needs of African American women. The themes that were identified were the importance of self-esteem as an African American and as a pregnant woman, the need for personal resources, especially with the mother and partner, and a strong social support network, as well as how to access services and environmental support. This qualitative data was then analyzed and used to develop a social support intervention for low-income African American pregnant adolescents with beneficial results (Norbeck, De Joseph, & Smith, 1996).

Critical Social Theory

Research efforts among specific populations, especially those that historically have been oppressed and the belief that oppression is pervasive, is central to critical social theory. Critical social theory places an emphasis on the subjective, the social construction of reality, the socio-political and economic influences, and the prevalence of racism and sexism in scientific and social activities (Gortner,1993). It takes a critical stance towards the social reality that it investigates, by providing grounds for the justification and criticism of all components (i.e. institutions, mentalities, practices) that make up that reality, with the purpose of exposure and revision. In so doing, the need for political and social action may arise, which may lead to social change and thereby increase the power and influence of the oppressed groups. In applying critical social theory, the underrepresented or oppositional views for a given problem are specifically represented (Gortner,1993) in a more collective rather than individual or personal manner.

Critical social theory can be used to address and expose oppressive sociopolitical conditions influencing health and health care (Boutain, 1999; Browne, 2000). These oppressions, according to Boutain (1999), in respect to health care are believed to affect individuals, families, communities, and populations in American society. The relationship between social support and teenage pregnancy is significant as seen in the review of the literature. Among low-income women who are deemed to often have less resources of social support the significance is augmented. Adequacy in effective support systems has been shown to improve health and social outcomes for vulnerable populations. Teenage pregnancy is a pregnancy that occurs during the adolescence of a woman under the age of nineteen. Low-income African American women who are teenage and pregnant fit into this category of vulnerability.

Critical social theory encourages the representation of this and similar populations, especially in light of the social and economic concerns related to the issues and the reality of adolescent pregnancy. Furthermore, a great need has been identified for African American studies about health with the inclusion of diverse perspectives of African Americans (Boutain, 1999; Mims,1998) and from an Afrocentric approach. Boutain (1999) encourages the exploration of how power relationships inform the perspectives of African Americans and operate to account for the social realities in American society and the conditions that impede human health and fulfillment.

Pregnancy among African American Adolescent Women of Low Socioeconomic Status

According to Mims (1998), it is not only important but necessary to understand pregnancy within African American populations from an African American perspective, especially when it occurs in adolescents. The issues of significance can be best understood from an Afrocentric approach when consideration into oppression, differences in communication, view of motherhood, social class, and societal norms are evaluated. Over time, the structure of the African American family has changed as a result of oppression, whereby the infants of adolescent mothers are typically absorbed into the existing family unit. This contributed to emotional difficulties and the economic disparity that many of these families are already subjected to as a result of long-term societal oppression. Societal norms for African Americans have been influenced greatly by other cultures (Mims, 1998), predominantly by Caucasians, which has led to the initiation of early onset premarital sex although not sanctioned by the community at large. From the perspective of social class, there are few opportunities for the future and therefore few advantages to delay starting a family. As motherhood is generally viewed as a representation of maturity and adulthood is, parenthood is sometimes viewed as the only viable passage into adulthood. Not only is identification as a mother validated but for many adolescents from lower socioeconomic groups it becomes the only visible opportunity for social mobility (Mims, 1998). Bolla, De Joseph, Norbeck, and Smith (1996) discuss the importance of understanding the strengths of the African American communities in order to promote culturally sensitive care and preventive health care.

Understanding adolescent pregnancy from an Afrocentric approach is essentially the focus of the research for this study. How social support affects this subpopulation of low-income women is the topic that drives the exploration into the culture itself. The perspectives that the women voice are critical for clarification of the issues that continue to lead to poorer outcomes for this population than for others. Understanding the constitution of their social networks and the functionality of them will provide insight into the gaps that may exist within them. This development of culturally appropriate social support interventions will help to alleviate some of the stressors or crises that low-income African American pregnant adolescents experience and buffer them in such a way as to improve health outcomes for them and their infants.

Conceptual Map based on the Theories of Norbeck for Social Support, Critical Social Theory, and Afrocentric Perspective by Mims
Additional Concepts Related to the Research Apprenticeship Project

Several other concepts were reviewed and presented as a major focus of the research project, "A description of social support and hope in pregnant and parenting teens receiving care from a doula" by the principle investigator (PI) and author (Breedlove, 2001) that provided an opportunity for participation as a research apprenticeship. In addition to the concepts of social support and adolescent pregnancy in the African American population, the concepts of hope and hope theory, and doulas also contributed to the context of the study's purpose. Participation in the undertaking and analysis of data for this study provided an opportunity to understand to a greater level the concepts of social support in the African American teen population presented in the first two chapters of this paper. Insight into hope and hope theory, as well as, the impact and benefits of doulas as a unique form of social support was an additional benefit received in engaging in the role of research apprentice.


Doulas are described as lay persons that have been trained to provide labor support to women throughout the intrapartum portion of their perinatal period. This support can be emotional, physical, or educational in nature and usually contains a combination of these components. Doulas or other female companions provide support that is said to be significantly different, in behaviors, encouragement and otherwise, than that provided by fathers and male partners (Kennell & McGrath, 1999). They do not discount that components of the care can be similar, positive, and beneficial but they pose the question as to whether male support persons should be expected to fulfill the role of sole supporter because of these differences. Doulas, in providing information and education to the laboring mother and her other support persons not only serve in a supportive and educational role but there is also room to establish a relationship with the woman and possibly her family in this transitional time and event in her life. Klaus and Kennell (1997) report results that lend to the confirmation that participation in the labor and birth form or increase a bond between those present at birth and the infant. There is reference to the probability of a reciprocal care that occurs in future relationships between the doula, mother, and infant (Klaus & Kennell, 1997).

Zhang, Bernasko, Leybovich, Fahs, and Hatch (1996) performed a meta-analysis that reviewed four randomized clinical trials on the effects of continous labor support by a doula for low-income nulliparous women. Results of the analysis found a significant decrease in cost of care due to fewer medical interventions and complications, such as a reduction in the number of cesearean section and vacuum or forcep deliveries experienced as part of the labor and delivery process. Additionally, a reduction in the number of hopsital days and use in hospital supplies was also demonstrated. Improvement was likewise seen in maternal and infant outcomes, including the psychological health of mother and baby. Some of these outcomes were shorter lengths of labor, decreased neonatal intensive care unit admissions, increased maternal satisfaction, increased breastfeeding, and enhanced maternal-infant bonding (Hodnett, 1999; Kennell, Klaus, McGrath, Robertson, & Hinkley, 1991; Kennell & McGrath, 1999; Klaus & Kennell, 1997).

It is in light of the numerous benefits that the above referenced studies report regarding the support that doulas provide, as well as, the experiential influence that the PI has evidenced, which has given rise to the choice of doula support as a viable option to provide and investigate support to pregnant African American adolescents. She recognized that continuous doula-provided care for pregnant teens in the prenatal, intrapartum and postpartum period has not been explored in the published studies available. Furthermore, that understanding from the perceptions of the adolescent mothers as to the specific components of social support relationships that were beneficial or helpful for them presented a solid basis for consideration of this concept.

Hope and Hope Theory

Hope is described as the desire or expectation for a positive future (Snyder, Cheavens, & Sympson, 1997). Goal conception and actualization appear to be central to the concept of hope theory. These same researchers suggest that first, individuals can perceive one or more ways to achieve goals, called named pathways thinking. Secondly, individuals evaluate their capacity or ability to utilize the pathways they have envisioned to move toward those goals, called agentic thinking. The foundation of hope lies in the ability to make linkages between previous desires and to use strategies that meet those needs (Hinton-Nelson, Roberts, & Snyder, 1996). The Hope Scale for Children was created based on this definition which requires that the child develop both an understanding of the self as being able to move towards a goal (mental agency) and a plan to reach the goal despite obstacles (pathways) (Hinton-Nelson, Roberts, & Snyder, 1996).

Additional research has been done to define hope and hopefulness in regards to specific populations; for the purposes of this paper the focus will relate to the adolescent population predominantly. Hinds (1988) conducted a study based on grounded theory methodology in order to generate and verify a definition of adolescent hopefulness. Results from the study delivered a definition of adolescent hopefulness as the degree to which an adolescent possesses a comforting or life-sustaining reality-based belief that a positive future exists for self and others. This definition had four dimensions representing four degrees of hopefulness, from least to highest (see Table 4). Hinds (1988) also found in her analysis that hopefulness contains an acknowledgment of the difficulties present in a situation and a desire for those difficulties to subside as well as a recognition of the personal efforts needed to affect a desired change. Included in this definition is the underlying focus on self, concern for one’s future, and recognition of the seriousness of a situation, reality orientation, concern for and a focus on others in addition to self.

Table 4: Dimensions of adolescent hopefulness:

Forced effortPersonal possibilitiesExpectations of a better tomorrowAnticipation of a personal future
* the degree to which an adolescent tries to artificially take on a more positive view* the extent to which an adolescent believes that second chances for self may exist* the degree to which an adolescent has a positive although nonspecific future orientation* the extent to which an adolescent identifies specific and positive future possibilities for self
(Hinds, 1988)

Herth (1998) conducted a study of adolescents describing the meaning of hope from which emerged five themes: connectedness, internal resources, cognitive strategies, energy, and hope objects. One of the pertinent findings from this study relates to the first theme, in that adolescents describe that connectedness to at least one other person was of great importance. An additional important factor was the hope that others harbored for them. Snyder (2000) described the need for significant others in the lives of children and adolescents, such as teachers, parents, or mentors who provide encouragement and role-modeling as a critical component to their coping abilities and in fostering the achievement of high hope. According to Hinds and Gattuso (1991) hopefulness in adolescents is dynamic and sensitive to changing situations. Hopefulness is believed to directly and indirectly influence an individual’s health state by helping the individual to work at maintaining, regaining, or augmenting his or her health (Hinds & Gattuso, 1991). A scale was created and tested based on this conceptualization named the Hopefulness Scale for Adolescents (HSA). According to Hinds and Gattuso (1991), it is able to accurately measure how hopeful the adolescent is and what is hoped for.

The influence that hope has on health behaviors and outcomes has been referred to in the literature (Foote, Piazza, Holcombe, Paul, & Daffin, 1990; Hinds, 1988; Hinds & Gattuso, 1991; Lesser & Escoto-Lloyd, 1999; Yarcheski, Scoloveno, & Mahon, 1994) and has been studied in reference to several different sub-populations. Foote, Piazza, Holcombe, Paul, and Daffin (1990) and Yarcheski, Scoloveno, and Mahon (1994) report that hope is highly related to a sense of well-being and is thought to be a key factor in acquiring a state of optimum health. The former study found that hope is related to social support and self-esteem, as well as, the latter two also being inter-related. "Because hope is a positive, motivating force and positively correlated with self-esteem and social support, care can be planned toward increasing self-esteem and social networks and create an environment which would make health care more effective"(Foote, Piazza, Holcombe, Paul, & Daffin, 1990). The latter study of middle age adolescents (Yarcheski, Scoloveno, & Mahon, 1994) found a significant correlation between perceived social support, hopefulness, and general well-being. The researchers concluded that individuals with higher hope cope better in times of crisis, especially in the presence of supportive networks and conversely, persons with ineffective social support may be at a greater risk to experience adverse health outcomes when exposed to high degrees of stress and low hope. Carrera (1996) reports that disadvantaged teens facing a crisis pregnancy are often left with little hope and frequently with few adults to support or encourage them. Health-related problems such as depression, suicide attempts, domestic violence, poor birth outcomes, and sexually transmitted diseases can negatively impact the teen mothers and their children and be greatly exacerbated by a lack of resources (Lesser & Escoto-Lloyd, 1999).

The inclusion of hope as a concept of importance in this study involving pregnant adolescents focuses on the meaning that hope has for these teens and the future goals that they envision for themselves and their families as a result of participation in a doula supported intervention program. The PI hopes to add to the current literature on hope research by providing an understanding of the perception of hope and the presence of hopeful behavior in teen pregnancy. In understanding this perspective from pregnant and parenting adolescents, additional hope engendering and promoting strategies can be created to improve not only the health behaviors, perinatal outcomes, and well-being of teen mothers and their infants but also of other at-risk pregnancy populations."Interventions must be grounded in these young women’s reality and in their hopes and dreams for themselves and for their children" (Lesser & Escoto-Lloyd, 1999).
A detailed summary of the research project:

The purpose of the research project was to explore and describe the perception of social support and hope in pregnant and parenting teens that have received a unique form of social support during pregnancy, delivery, and the postpartum period. The support provided for these teens was through the services of doulas. Doulas are trained persons who provide support to women in labor and delivery. They do not have a medical education background but do have training to provide emotional and physical support, such as comfort measures, reassurance, physical contact encouragement, and explanations about progress in labor. They can, as well, assume the role as advocates and educators during the intrapartum and immediate postpartum periods. Various studies confirm the benefits that doulas have in regards to non-interventative care and healthcare cost reduction, such as increased breastfeeding at 6 months, decrease in maternal fever, decrease in prolonged infant hospitalizations greater self-esteem reported, reduction in several areas: cesearean deliveries, the duration of labor, the use of pain medications, and operative vaginal deliveries (Kennell, Klaus, McGrath, Robertson, & Hinkley, 1991; Klaus & Kennell, 1997; Zhang, Bernasko, Leybovich, Fahs, & Hatch, 1996).

The study was conducted using a descriptive cross-sectional design. The subjects for this research project consisted of two groups; the first was comprised of 12 pregnant teen mothers (Group A) and the second of 12 parenting teen mothers (Group B). A convenience sample was used. Purposive sampling for homogeneity and sample size was done, as participants were of African American teens participating in a community agency doula-supported project in a major metropolitan city in the Midwestern States of the US. Doulas were paired up with the teens one-on-one throughout the duration of their involvement in the community-based program whilst pregnant and this relationship continued during the postpartum period. Eight research questions were designed to answer several aspects related to the purpose of the project. They are listed in Appendix 1.

Based on specific inclusion and exclusion criteria, the adolescents were invited to participate by staff employed within the agency. The inclusion criteria for pregnant teens for participation in the study required that the participants: a) must be at least 14 years of age and up to their 18th birthday, b) must be able to read and speak English as a primary language, c) must be a recipient of State Medicaid services for pregnancy, d) must be of African-American descent, e) must be less than 36 weeks gestation by estimated date of confinement (EDC), f) must be described as having a normal, uncomplicated pregnancy, and finally, g) must have had a minimum of three visits by the assigned doula prior to the interview. Exclusion criteria for the pregnant teen group were those participants that a) had a change in assigned doula during pregnancy and b) had a known mental health illness.

The inclusion criteria for participants in the group who are mothering their infants required the participant to a) be at least 14 years of age and up to their 18th birthday, b) must be able to read and speak English as a primary language, c) must have received State Medicaid services for pregnancy, d) must be of African- American descent, e) must have delivered an infant without diagnosed complications, f) infant must at time of interview be between the ages of three months and up to their first birthday, g) assigned doula must have been present for birth, and h) participant may not be currently pregnant. Exclusion criteria for this same group of parenting teens are a) that the teen is not currently parenting her infant and b) those teens with a known mental health illness.

The data has been collected and completely analyzed at this time using Statistical Program for Social Sciences (SPSS), a computer software statistical program, for analysis of the demographic data and content analysis according to Krippendorf (1980) of the semi structured interview data that was retrieved from the participants. The semi-structured interview consisted of open-ended questions. Probing techniques, as well were utilized as needed throughout the interview process.

Tables were produced and formatted to display the contextual themes from the tertiary analysis, the third step in the analysis of the data. A comparison of similarities and differences in the contextual themes between the groups were conducted in order to examine a cross-sectional view of the concepts of support, hope, and the doula experience in this sample.

Results revealed few differences between the two groups in response to the eight research questions. Both groups identified the doulas as key contributors to their social support needs. The doulas provided support during the pregnancy, delivery, and postpartum period that had both emotional and physical components. In addition, they added to the knowledge base of the teens on aspects of pregnancy and newborn care and safety education. They helped in an advocacy role, as well as were instrumental in breastfeeding assistance and facilitation of bonding between the teens and their infants. Family was also identified as instrumental components of the social support networks of the adolescents but differences in when their support was offered and relied upon pointed to before versus after the delivery. In addition, the pertinent needs for support that the adolescents identified were not always the ones that the family was able to or did provide for. The major focus of hopes for the future that the teens verbalized was for a better and brighter future for themselves and their children. This included the reality that to achieve these hopes they would need to "move" beyond their current social situations, which they believed could be facilitated with hard work, education, and the assistance of others.

The principle investigator (PI) identified several limitations of the study. One of these was the use of participants from only one of the three nationwide programs utilizing the doula model of support for pregnant and parenting adolescent. Secondly, due to time constraints in collecting the data, the PI petitioned her committee to change the inclusion criteria to include those adolescents who may have already experienced a previous childbirth, originally including only primiparas, whilst retaining all other previously established criteria. Thus, the possible limitation exists that those adolescents having a second or third infant may have had previous knowledge related to pregnancy, parenting, and resources to assist young mothers that first-time mothers may not have had.

The PI identified several implications for future research as a result of the study findings. She recognized a need for replication of similar studies to validate the current findings. These studies should explore the relationships between doula care and adolescent pregnancy, specifically, in relation to models in which the doula and adolescent share the same culture, as well as investigate the specific characteristics that make up the positive aspects of this type of support. Additional recommendations for further research should focus on hope and the implementation of needed strategies to bring about better and brighter futures for families with young mothers.

Finally, the PI related the findings and their implications to the nursing profession by giving recognition to the value of the role of the doula in providing social support to pregnant and parenting adolescents as an adjunct to nursing care both in the clinic and hospital setting. Encouragement to nurses and other health care professionals in terms of hope and future goal setting could be improved for adolescents in various health care settings if the decision to address these issues is made and implemented.

Credentials of the Principal Investigator

The principal investigator, Ginger K. Breedlove, is a Certified Nurse Midwife who completed her midwifery education at the University of South Carolina, her Masters in Nursing through Case Western University and a PhD candidate* after completing her doctoral studies at the University of Missouri - Kansas City. Her anticipated graduation is December 2001.
* At the time of submission the PI was notified of her change of status to PhD

She chose to conduct this study as part of her doctoral program of studies and as the subject for her dissertation. She completed the process of the research proposal, obtained the IRB approval and began the data collection and transcription on several of the participants included in the sample prior to my becoming involved in the summer of 2001. Although data collection was begun in March of 2001, the majority of her data collection, transcribing, and data analysis was conducted this summer and very recently completed.

The goals and work agreed upon between the PI and research apprentice

In June of 2001, the PI and the research apprentice met and set goals and work objectives for a research project that the PI was currently undertaking. The initial assistance included an extensive and updated literature review on the main topics of adolescent pregnancy, African-American adolescent pregnancy, hope and theories of hope, social support concepts, theories and intervention programs implementing unique forms of social support and their results, and finally, doula and doula-supported intervention programs.

Secondly, the decision to make a trip to the metropolitan city in which the interviews were conducted was made. The rationale for this excursion was in order to assist in the assessment of the site, take field notes, make observations of the agency and classes for the teens, and informal interviews of the teens. The principal investigator was to simultaneously conduct interviews over the two days of the participants that were present at the facility at that time and wished to participate in the study.

Following the completion of the interviews and transcriptions, the next goal was to assist in the verification of the transcription, word-to-word, to prevent loss of information and potentially influential data segments, as well as, to correct errors and misinterpretations. This also was to serve as a crosscheck for the content of the data by several different facilitators. Data and file cleaning was also planned as an appropriate objective for the apprenticeship project.

Analysis and cross-analysis of the data was the major goal and work in terms of time, information, and pertinence to the role of an apprentice that was agreed upon. Content analysis was the primary method of analysis performed on all of the transcribed interviews in a three steps series: primary, secondary, and tertiary. Each step has specific requirements for performing the analysis and these were to be performed and reviewed either by the primary investigator or myself for accuracy and reliability when the other was the primary facilitator of this process.

Other work to assist the PI that had not been initially discussed but completed was a Power Point presentation, editing of the written work, cross- reference and citation check, and formatting of the manuscript. The Power Point presentation was created to present the research purpose, process, analysis, results, and conclusions of the study that was undertaken. The rest of the work listed described the tedious but important finishes of the research work in its written presentation. Final retrieval of articles for the dissertation defense contributed to the final work of the research apprenticeship which specifically focused on additional theories and work related to social support which has been the component of the PI’s work that I have specifically chosen for my own efforts these past few months.

The observation of the dissertation defense became the last objective, not initially considered in the original timeline discussion due to absence of consideration that the project would be completed prior to the completion of this class session. This also concluded the work of the research apprenticeship role that was undertaken.

Letter of agreement from the principal investigator

Letter of agreement signed and sent June 21, 2001 upon initiation of research apprenticeship work. (Appendix 6)

Research apprenticeship diary:

June 21, 2001 (4 hours)

The initial meeting between the PI and myself, the research apprentice occurred today. We discussed the reason for my interest in the research project, as well, as my previous interest and work with adolescent pregnancy and involvement in a mentor-doula project in both a clinical setting, as well as, the focus thereupon in undergraduate work and my desire to continue this work through the research pathway. The reason for choosing this format for the final project to fulfill course requirements versus other options were considered and deliberated upon. The expectations of a research apprenticeship as stated in the class curriculum and syllabus were shared with the PI in order to evaluate if both the needs of the PI and the apprentice would be met in working together. This was determined to be satisfactory to both parties.

Goals, objectives, and work to be completed, individually and jointly, contributed to the content of remaining discussion. The work objectives were also prioritized as to their importance for assistance with and how personal and professional schedules could best be merged to accomplish these efforts.

The PI explained the research project in full detail and what had already been accomplished up to the current date. She informed me that some of the data collection for Group A and B had been completed but were awaiting transcription completion from the audio taped interviews that were conducted. Upon questioning regarding the total of participants in the sample size, the PI referred to what Krippendorff (1980) acknowledged for sample size in that "there is no set number" and that the number that is chosen "depends largely on how the property to be generalized is distributed in the sample" (p. 69). She anticipated a total of 24 at that point in time. The timeline for desired completion of the entire project as well as the specifics for data collection were discussed as they necessitated preplanned travel engagements out of the area.

Discussion of the type of research project that the PI described as "triangulation" ensued. She explained that due to the combination of quantitative and qualitative components, the study could not be designated as solely one or the other. "While qualitative is generally the heading for a study that asks open-ended questions, leaves room for stories and examples and is analyzed by content analysis" (Sandelowski, 1995), this study also consists of pertinent quantitative demographic data and leaves room for analysis by quantitative methods, in this case the frequency of responses. "Triangulation has come to mean virtually any more-than-one instance of one or more elements of the research process within a study: for example, more than one data collection strategy, method, investigator, and theoretical perspective, or any combination of qualitative and quantitative techniques"(Sandelowski, 1995).

Finally, a book search for the Krippendorf reference was undertaken that day and located for use by the PI to study, understand, and utilize for the purposes of qualitative data analysis for the study. It was decided that a review of content analysis and the manner in which it was to be conducted would follow in another meeting.

The insights gleaned from the business of the day stemmed primarily from the importance of assessing whether a research project or any project is a good "fit" for all those involved. This includes whether there has been any previous knowledge or work experience related to the concepts involved, whether there is a true interest in the project and ability to meet the time commitments, and whether the needs and expectations of those working together will be met. A meeting of the minds in this fashion promotes the successful initiation and work-through on the joint and individual work involved in the larger scheme of the research project.

July 3, 2001 (7 hours)

An initial literature search was conducted for updated information and publications pertinent to the main concepts of the study for the duration of the last year. These main concepts were hope and hope theory, social support, African- American adolescents, pregnancy within this population, and doula. This search was performed via search engines, library databases, and journal browsing. In addition, the articles that were deemed pertinent to the study at hand and of probable utility to the researcher were retrieved after reviewing the abstract of each individually.

This initial step in the review of the literature provided an excellent opportunity to become oriented to the process in and of itself. The actualization of how to slim down the searches in order to reduce the time and resources that on-line and engine searches retrieved for pertinent and updated literature on the major themes being researched and described in the research project and dissertation was realized as well. Finally, browsing through the abstracts and the purpose of many of the articles impressed upon me the possibility to further “weed” out unnecessary or unrelated sources of information.

July 11, 2001 (1 hour)

Today the PI and I spent some time discussing content analysis and reviewed the literature regarding how it is described and how to conduct the three levels of analysis. Most of the focus revolved around the description by Krippendorff (1980), as this was the reference and choice of content analysis that the PI chose to utilize due to the comprehensiveness of description. "Content analysis seeks to understand data not as a collection of physical events but as symbolic phenomena and to approach their analysis unobtrusively (p.7). "It has evolved into a scientific method that promises to yield inferences from essentially verbal, symbolic, or communicative data." (Krippendorff, 1980, p. 20). Polit, Beck, and Hungler (2001) define content analysis as "the process of organizing and integrating narrative, qualitative information according to emerging themes and concepts; classically, a procedure for analyzing written and verbal communications in a systematic fashion, typically with the goal of quantitatively measuring variables" (p. 495).

July 18, 2001 (10 hours)

A trip to the site of the research data collection site with the PI began today. Prior to arrival, she briefly described the site and the doula support program that had been implemented there as one of only three sites in the United States with a similar doula support intervention program. She explained the volunteer status of some of the employees, and the overall purpose and management of the program that was being run at this facility. She enumerated the many objectives and services that the program offered as well as some of the characteristics of the recipients and employees.

Upon arrival at the facility, I had an opportunity to meet the manager of the program, some of the employees and adolescents, and tour the facility layout. Field notes were taken of the site, the private conference room for the interviews, and from the observation and participation in one of the parenting classes. The PI simultaneously went about the process of data collection, obtaining three interviews throughout the day.

Later that evening, the PI reviewed the steps involved in performing and completing content analysis. She described the first level of analysis, primary analysis as extrapolating units of words and sentences that characterize a set of attributes described through participant responses in order to categorize responses for each research question (Breedlove, 2001).

Primary analysis was thus initiated for Group B. The PI and I worked through the first primary analysis together and then developed a computer template in order to categorize all the responses to each of the research questions and thus retain organization for the presentation of the analysis results. This template was utilized for both Group A and B participant interviews. Essentially the template consisted of identifying responses to the 18 open-ended questions (Appendix 4) presented to Group A participants and the 20 open-ended questions (Appendix 5) to Group B participants. These were already predetermined to relate to one of the 8 research questions (Appendix 1). The rest of the process involved coding the verbatim transcription to the research questions 1-8 and copying and pasting these into the template. In the case of semi-structured interviews the questions can be coded and the responses categorized by number of the question. "From the initial sort, the investigator then reads all of these responses and may conduct a content analysis of these data. If desired, if the sample size is adequate, responses may be numerically coded and nonparametric statistics conducted to look at relationships between items or variables" (Morse & Field, 1995, p.141).

For the initial 3 interviews that had already been taken, transcribed, and verified, the PI and I reviewed the transcriptions and compared the extracted responses from the interview questions. This process was used as a cross check for accurate interpretation and implementation of analysis, as well as, to insure the reliability of results.

The practical application of performing content analysis led to a greater understanding of the process of content analysis. Performing and comparing the results of primary analysis were helpful in establishing a guideline to correctly performing the process and a confidence level to continue through the rest of the transcriptions as they were completed.

July 19, 2001 (9 hours)

Day 2 of the site visit consisted of ongoing interviewing by the PI and continued field note taking on my part. Upon invitation I had the opportunity to participate in a doula-training workshop that lasted the duration of the day. Observational field notes were taken for the PI of the meeting, as well as, listening to and partaking in discussions with the doulas who were sharing their experiences and interactions with the adolescents in the antepartum, intrapartum, and postpartum periods of their pregnancies. They focused on problem-solving techniques. They described interactions and interventions that were both successful and supportive for the adolescent mothers as well as reactions of the pregnant and mothering adolescents. They discussed ways to further educate and care for themselves and for each other so as to better provide education and support for the adolescents.

Having had an opportunity to visit the site and record field notes, talk with the teens, and attend parenting classes of the agency, gave me the opportunity to learn about some of the characteristics and services of the program. The pregnant and parenting teenagers were given the possibility to attend prenatal and parenting classes every week, assistance and counseling with high school or GED completion, job fairs or training, provision of baby supplies and equipment, and a doula assignment for antepartum, intrapartum, and postpartum support. Attending the site of the doula program and the workshop helped to establish a foundation for understanding the work that doula can do and the benefits of their continual support in a variety of settings.

July 22, 2001 (3.5 hours)

Primary analysis was continued today for the next 5 completed transcribed interviews for Group B. The process is the same as described in detail above. The same review and comparison process was completed between the PI and myself.

July 24, 2001 (3.0 hours)

Primary analysis was continued and completed today for the last 4 of the Group B participant interviews. The process is the same as described in detail above. The same review and comparison was completed between the PI and myself.

The completion of the primary analysis for the first groups of completed transcribed interviews (Group B) was a success not only in the finalization of this step of analysis for this group but also in providing validation that the interviews had all been completed for the sample size determined to be sufficient for this group. It promoted an attitude to continue goal-setting for timely completion of the rest of the data collection and analysis not only for myself but the PI as well.

July 26, 2001 (6 hours)

Today marked the beginning of the primary analysis process for the second group of participants, Group A (pregnant adolescents.) The transcription process was completed for 8 of the Group A participant interviews. The same process was utilized in coding and creating a template for each of the analysis for this second group of interviews as previously described.

August 2, 2001 (4 hours)

Completion of Group A primary analysis was accomplished today. Following this process I was able to assist with some of the verification of the transcription of interviews. The original tape-recorded data was replicated by verbatim transcription in order to facilitate data processing. Regarding the importance of verifying transcriptions, Krippendorff (1980) remarked "noise, ambiguities, and counterintuitive perspectives in the semantics of a data language make subsequent interpretation of the findings difficult" (p.75). "The tape should be replayed, with the researcher listening carefully to the content as well as the questions asked and the participant’s responses". It is crucial that the tape be transcribed exactly from the interview and not paraphrased"....The transcription is then checked against the tape for accuracy." (Morse & Field, 1995, p.131)

The process of verbatim transcription is not only important to the reproducibility of data as I found out in partaking in the task which revealed some minor and other more significant alterations to the text; but it is a lengthy process that is nothing short of time-consuming. Astute attention must be maintained in the process of transcription as well as verification so that no segment of data is lost that might lend to the significant alteration of the participant’s message.

August 5, 2001 (4 hours)

Comparison of grouping and extracting for primary analysis results were accomplished today for the 12 interviews from the second (Group A) group with the PI. Krippendorff (1980) remarked on the importance of this review and comparison process: "usually the inter-rater agreement is based on two people who have worked together intimately in the development of a coding scheme, and who have engaged in much discussion of definitions and disagreements". He further acknowledged " the worse practice in content analysis is when the investigator develops his recording instructions and applies them all by himself or with the help of a few close colleagues and thus prevents independent reliability checks" (p. 74). The chair of the dissertation committee revalidated the work of analysis as a third party when there existed any discrepancies or disagreements.

August 10, 2001 (7.5 hours)

The PI presented the SPSS (Statistical Program for Social Sciences) that is to be used for analysis of the demographic data. This is a computer statistical software program that among other measures calculates the central tendency measures, as well as, correlations and frequency distributions. After a brief introduction and orientation to this statistical package, the work of data entry ensued. Demographic data entry for all of the participants included in the two groups of parenting (Group A) and pregnant (Group B) adolescents for 7 different variables was accomplished. This data included age, marital status, number of children, intendedness of pregnancy, age in weeks of gestation of the pregnancy or weeks post-delivery, years of education completed, attendance in school and employment status at the time of the interview. One open-ended question was utilized to determine the frequency and amount of doula contact that the adolescent received from the doula on a weekly basis both before and after delivery. The PI explained that the data would be analyzed with the SPSS and reported in means, percentages, and frequencies. The PI then demonstrated how specific variables for the two groups could be measured and compared along several different lines.

The work of today presented concrete and logical information that was not only appealing in terms of presentation of the data but also provided a resource for acknowledgement that the many hours and trips that the PI put into travel and data collection had come to an end and the work of analysis on all levels was coming together. Krippendorff (1980) explained the utility of statistical programs by stating that "data are coded to conform to the input requirements of the technique and computational options are specified by the user. In the course of the computation, data are sorted, reorganized, transformed, and accounted for by numerical indices that the user must then interpret in view of what the technique does" (p. 121). In terms of content analysis, it is more usual to apply statistical tests after inferences have been obtained (Krippendorff, 1980).

August 13, 2001 (4 hours)

The initiation of the secondary level of content analysis was begun today. The PI reviewed the process of this step (see Table 5) and explained it as identifying and clustering attributes to partition co-occurrences of participant responses for each research question (Breedlove, 2001). The templates from the primary analysis were analyzed in this process of identifying and clustering attributes. The templates as previously discussed include the entirety of responses to the research questions asked as part of the semi-structured interview under the 8 research questions addressed for the purposes of this study. Krippendorf (1980) uses the terms "clustering" or "lumping together" when referring to secondary analysis. He explains the process as a narrowing down of responses to the pertinent research question into clusters of words that merge conceptualization of data as well as separating when necessary variables or sets of objects that represent "mutually exclusive classes" (Krippendorff, 1980). A new template was created for each of the interviews with this second level in the analysis process.

The process of secondary level of analysis appeared much more critical in terms of reviewing and comparison with another validator. The level in which the loss of sensitivity with the reduction of data in this phase is possible appears to be substantially greater than with primary analysis.

August 16, 2001 (8 hours)

Secondary level analysis of the Group B interviews was completed today according to the same process as described in detail above.

August 30, 2001 (6 hours)

Data file cleaning was also performed to verify findings from the demographic data sheets. Also data file organization was also performed on all of the existing data files and articles. The process of secondary level analysis of Group A participants was begun today in the same manner as has been previously described. .

September 7, 2001 (8.5 hours)

Secondary level of analysis was completed today for the Group A participants. The analysis was reviewed and verified by the PI in a similar manner as has occurred throughout the rest of the analyses. Repeating themes were identified from the participant interviews. Theme clusters were determined and verified by the PI and myself. As a cross check, the PI and I re-coded portions of the same material twice at different points in time, independently, to determine if disagreements or inconsistencies were found.

September 10, 2001 (1hour)

Other articles that were deemed absent in the process of the several years work on the research project were identified from the data and file cleaning process. These articles were retrieved for the researcher today via a literature search in the University library and then photocopied. Inter-rater validation between the PI and myself for the secondary analysis was performed for all the participant templates. This was accomplished by comparison of the grouping and extracting results that were found between both of us. Additional inter-rater validation occurred between the PI and her dissertation chairperson via a comparison of the emergent themes for accuracy of interpretation. Krippendorff (1980) acknowledged that research results to be valid must be based on valid data, valid analysts, and reliable analytical processes.

The continual process of checking and re-checking each other and clarification of disagreements by a third party gave credence and stability to the importance of establishing reliability and validity in the methodology and analysis of data. As well, it demonstrated the bias that can occur with the unique interpretation of one researcher and the results that can be consequently different without several validators.

September 12, 2001 (8 hours)

The PI reviewed the process for conducting the third and final level of content analysis. Essentially, tertiary analysis includes the counting, classifying, and naming of identified units of commonality into contextual meaning that describes an underlying conceptualization of participant experiences (Breedlove, 2001) sometimes referred to as contextual classification. This process works to eliminate the redundancy of conceptual themes while maintaining the relationships between different variables shared by the participants. The process works by analysis of the clustering of attributes encompassed within the templates from the secondary analysis of both Group A and B, including the frequency of responses that were similarly classified for content of response, and then naming the category. This was performed for each of the initial eight research questions that were separated out in the process of creating templates for the primary analysis.

Tertiary analysis was begun today as a follow-up step to the secondary level of analysis. The completion and comparison of contextual themes derived from the tertiary analysis of the Group B participants was completed today with the PI. Beginning efforts were given to commencing the analysis of the second groups of participants (Group A).

As a component in this third level of analysis the frequency of like responses for each of the research questions is determined. According to Krippendorf (1980), the greater the frequency of occurrence, whether a subject matter or a symbol, exists the more attention, emphasis, or importance is given that idea (p.40). Frequencies can be used as direct indicators to represent data about underlying phenomena and as a means to establish correlations and associations between two or more variables. In regards to content analysis, Krippendorff (1980) extends this utilization of frequency to represent data when he acknowledges that relations can be made either within the results of a content analysis or between the results of a content analysis and data obtained independently.

The process of performing the tertiary analysis not only helped in the understanding of the process but as well began to lead to the comprehension of what the results and frequency of responses were important for. The process of counting the frequency of responses and categorizing them was probably the most interesting component of content analysis. It gave the impression that this was the information that was actually sought for by the research questions. Granted, the efforts of the primary and secondary analysis were vital to achieving this last part of analysis but they appeared to have less glamour than the contextual conceptualization and the frequency of responses expressed by the participants.

September 19, 2001 (8 hours)

Tertiary analysis continued to take the majority of the efforts for the day. The templates from the secondary analysis of the Group A participants were utilized and the tertiary analysis for this second group of participants completed in the same manner as for the first group as described above. The process of reviewing and comparison again occurred between the PI and myself for reliability in interpretation.

Jointly, the PI and I then produced and formatted several different tables for presentation of the steps involved in analysis or of the data obtained. Table 5 presents a detailed description of the three levels of analysis that were performed on the interview data to arrive at a list of contextual themes. Other tables were created to present the contextual themes from the tertiary analysis for the two different groups.

Level of analysis Steps in analysis (Breedlove, 2001)

Demographic data tables were also created to present the means for the two groups of some of the variables for comparison. These included such variables as mean age of participants, number of years of education completed, age of infant or of gestation at time of interview, and the intendedness of pregnancy. Several of these were created for inclusion in the manuscript of the dissertation.

Finally, I had an opportunity to compare the contextual themes from the two groups to evaluate similarities and differences between the two groups for each of the eight original research questions. These were scrutinized for specific details for the PI who then reported these results in her findings of the research.

September 27, 2001 (8 hours)

The work of today encompassed the tedious scrutiny of manuscript and reference list comparison of references and citations. In addition, editorial citations checks were completed of each of the citations within the text and each of the references within the reference list. The APA format was utilized for this process. In addition, specified formatting was begun for the manuscript in accordance to the requirements of the University receiving the doctoral dissertation manuscript for review. This formatting included a combination of Terabian and APA styles.

The importance of this exercise lies in understanding and accepting the reality of the world of publishing and academic expectations for the presentation of scholarly work and research. Every separate entity within these domains have specific guidelines that must be followed in order for research work and results to be presented in print for others to evaluate and learn from. Lack of adherence to the specification can result in undue delay and possible rejection of valuable research results and implications.

October 3, 2001 (3 hours)

The final literature review that was obtained today focused on the negative impacts of social support. Although considered in the initial process of creating a purpose for the research, the PI deemed that this literature and references would be imperative to the comprehensive presentation of the research project and of utmost important in the defense of the reason for performing this specific type of project. The process of this literature review included a general focused literature review specifically on negative aspects of social support interventions as well as the location and retrieval of the work of known authors who had addressed the issues surrounding this topic. The importance of not only focusing on the concepts of the research project but also acquiring the knowledge and remaining informed of the work that has been done by others in the field is important to not only lend credibility but also value to one’s own work. This became the lesson for the work of the day.

October18, 2001 (4 hours)

The bulk of the work completed today included orientation to the Power Point computer software and the production of a Power Point presentation. This presentation consisted of multiple slides including among other topics, the definitions of the major concepts, the conceptual framework basis for study, the current statistics for teenage pregnancy as found in the literature, the inclusion and exclusion criteria for the study, the design and sample of the participants, the study questions researched, the analysis and results, the limitations of the study, the implications and suggestions for future research.

The process of involvement in the Power Point presentation production led to a review of the fundamental and critical components of the research process. This form of presentation lends to a simplistic yet comprehensive approach to presenting the data and results from years of work and research by the PI to others, and thus, bypassing the need to read numerous pages of description.

October 26, 2001 (2 hours)

The conclusion of the research apprenticeship occurred with the attending and observation of the defense of the dissertation by the PI. The goal for attending the defense was strictly imbedded in the value of the observation and orientation to the process of the defense and the presentation of the research findings.

The PI presented the Power Point Presentation that had been the focus of previous work together although edited from the original number of slides. In the presentation, she presented the purpose, the design, sample, and research questions, some of the pertinent results, the conclusion, and implications for further study. She was then addressed questions related to the research concepts and findings from the audience of which I had the opportunity to address one myself. The general audience left the room while the dissertation committee of which 3 were physically present and 2 present via speakerphone then addressed their questions and gave their impressions and opinions. The PI later explained that the concepts of the study and the relationship between them, as well as, the results, their significance, and every aspect of the research process, analysis, editorial components, etc could and some were the focus of the dissertation committee examination. The PI was requested to leave the room after the questions were exhausted and a vote as to whether the PI would proceed onward to PhD status was conducted by the dissertation committee.

The insight that this experience provided was multifold, including, the importance of the presentation of the research process and results pertinent to individual projects in an audio-visual versus uniquely a printed format. It also served as an opportunity to share with the public that was present including a large majority of health care related individuals the current research endeavors that are occurring. It impressed upon me the stringency with which research must not only be conducted; but also that the purpose and the process be reliable and validated, the results and conclusions calculated and well stated, and the knowledge surrounding the concepts be up to date and accurate.

Total hours: 119.5 hours June 21, 2001

To: Cindy Farley, CNM
From: Ginger Breedlove, CNM
Re: Denise Fryzelka, CNM
Critical Inquiry 1

Dear Cindy,

I am pleased to work with Denise on her research requirements to complete her Masters degree at Philadelphia University. I am completing my doctoral dissertation at the University of Missouri, Kansas City School of Nursing. The title of this study is "A description of social support and hope in pregnant and parenting teens receiving care from a doula". I have completed half of my data collection, and have invited Denise to go with me to *(name of major metropolitan city in Midwest) mid-July for on-going data collection. Feel free to call at any time if you have questions. Glad to know you are still in midwifery education. I am now the program director for the University of Kansas Midwifery Program. We are graduating our first five students this weekend. Hope to see you sometime at convention.


Copy of signature unavailable - Original signature received by Cindy Farley, PhD via USPS

Ginger Breedlove, CNM, PhD(c)
University of Kansas, Midwifery Program Director
(o) 913-588-1723
3901 Rainbow Blvd
KU School of Nursing
Kansas City, KS 66109

* name of city eliminated to protect the confidentiality of study site and participants

American College of Nurse-Midwives (ACNM). (2001a). Code of Ethics. Quickening 2001 ACNM Membership Directory. Washington, DC: Author.

American College of Nurse-Midwives (ACNM). (2001b). Philosophy. Quickening 2001 ACNM Membership Directory. Washington, DC: Author.

Bolla, C.D., De Joseph, J., Norbeck, J.S., & Smith, R. (1996). Social support as road map and vehicle: An analysis of data from focus group interviews with a group of African American women. Public Health Nursing, 13 (5), 331-336.

Breedlove, G.K. (2001). A description of social support and hope in pregnant and parenting teens receiving care from a doula. Unpublished doctoral dissertation, University of Missouri, Kansas City.

Brett, K.M., Schoendorf, K.C., & Kiely, J.L. (1994). Differences between black and white women in the use of prenatal care technologies. American Journal of Obstetrics and Gynecology, 170 (1), 41-6.

Browne, A.J. (2000). The potential contributions of critical social theory to nursing science. The Canadian Journal of Nursing Research, 32 (2), 35-55.

Bruckner, M., & Muellner, M. (1985). Nurse-midwifery care of adolescents. Journal of Nurse Midwifery, 30 (5), 277-279.

Boutain, D.M. (1999). Critical nursing scholarship: Exploring critical social theory with African American studies. Advanced Nursing Science, 21 (4), 37-47.

Burke, P.J., & Liston, W.J. (1994). Adolescent mothers' perceptions of social support and the impact of parenting on thier lives. Pediatric Nursing, 20 (6), 593-599.

Carrera, M. (1996). Working with teens when the topic is hope: Lessons for lifeguards. New York: Donkey Press.

Coley, R., & Chase-Lansdale, P.L. (1998). Adolescent pregnancy and parenthood. Recent evidence and future directions. American Psychologist, 53 (2), 152-166.

Connelly, C.D., (1998). Hopefulness, self-esteem, and perceived social support among pregnant and nonpregnant adolescents. Western Journal of Nursing Research, 20 (2), 195-209.

Cosey, E.J., & Bechtel, G.A. (2001). Family social support and prenatal care among unmarried African American teenage primiparas. Journal of Community Health Nursing, 18 (2), 107-114.

De Joseph, J.F., Norbeck, J.S., Smith, R.T., & Smith, S. (1996). The development of a social support intervention among African American women. Qualitative Health Research, 6 (2), 283-297.

Declercq, E.R., Williams, D.R., Koontz, A.M., Paine, L.L., Streit, E.L., & McCloskey, L. (2001). Serving women in need: Nurse-midwifery practice in the United States. Journal of Midwifery and Women’s Health, 46 (1), 11-16.

DuPlessis, H.M., Bell, R., & Richards, T. (1997). Adolescent pregnancy: Understanding the impact of age and race on outcomes. Journal of Adolescent Health, 20 (3), 187-197.

Foote, A.W., Piazza, D., Holcombe, J., Paul, P., Daffin, P. (1990). Hope, self-esteem and social support in persons with multiple sclerosis. Journal of Neuroscience Nursing, 22 (3), 155-159.

Gortner, S.R. (1993). Nursing's syntax revisited: A critique of philosophies said to influence
nursing theories. International Journal of Nursing Studies, 30 (6), 477-488.

Herth, K. (1998). Hope as seen through the eyes of homeless children. Journal of Advanced Nursing, 28 (5), 1053-1062.

Hinds, P.S. (1988). Adolescent hopefulness in illness and health. Advanced Nursing Science, 10 (3), 79-88.

Hinds, P.S., & Gattuso, J.S. (1991). Measuring hopefulness in adolescents. Journal of Pediatric Oncology Nursing, 8 (2), 92-94.

Hinton-Nelson, M.D., Roberts, M.C., & Snyder, C.R. (1996). Early adolescents exposed to violence: Hope and vulnerability to victimization. American Journal of Orthopsychiatry, 66 (3), 346-353.

Hodnett, E. (1999). Caregiver support for women during childbirth. The Cochrane Library (Issue 1). Oxford: Update Software.

Kahn, R.L. (1997). Aging and social support. In M.W. Riley (Ed.), Aging from birth to death: Interdisciplinary perspectives (pp. 77-91). Boulder, CO: Westview Press.

Kennell, J.H., Klaus, M.H., McGrath, S., & Hinkley, C. (1991). Continuous emotional support during labor in a US hospital. JAMA, 265 (17), 2197-2201.

Kennell, J.H., & McGrath, S. (1999). Commentary: Practical and humanistic lessons from the third world for perinatal caregivers everywhere. Birth, 26 (1), 9-10.

Klaus, M.H., & Kennell, J.H. (1997). The doula: An essential ingredient of childbirth rediscovered. Acta Paediatr, 86, 1034-1036.

Koniak-Griffin. D., Anderson, N.L.R., Verzemnieks, I., & Brecht, M.L. (2000). A public health nursing early intervention program for adolescent mothers: Outcomes from pregnancy through 6 weeks postpartum. Nursing Research, 49 (3), 130-138.

Krippendorff, K. (1980). Content analysis: An introduction to its methodology. Beverly Hills: Sage Publications.

Lesser, J., & Escoto-Lloyd, S. (1999). Health-related problems in a vulnerable population: Pregnant teens and adolescent mothers. Nursing Clinics of North America, 34 (2), 289-299.

Martin, J.A., Hamilton, B.E., & Ventura, S.J. (2001). Births: Preliminary data for 2000. National Vital Statistics Report 2001 July 24; 49 (5), 1-7.

Maton, K.I., Teti, D.M., Corns, K.M., Vieira-Baker, C.C., Lavine, J.R., Gouze, K.R., & Keating, D.P. (1996). Cultural specificity of support sources, correlates and contexts: three studies of African-American and caucasian youth. American Journal of Community Psychology, 24, (4), 551-587.

Mims, B. (1998). Afrocentric perspective of adolescent pregnancy in African American families. The ABNF Journal, 9 (4), 80-88.

Morbidity and Mortality Weekly Report (MMWR). (1997, August). Summary: Pregnancy-related mortality surveillance-United States 1987-1990. MMWR, 46, SS-4. Retrieved from the world wide web, 8/01/01:

Morse, J. M., & Field, P. A. (1995). Qualitative Research Methods for Health Professionals (2nd ed.). Thousand Oaks, CA: Sage Publications.

Norbeck, J.S. (1988). Social support. Annual Review of Nursing Research, 6, 85-109.

Norbeck, J.S., & Anderson, N.J. (1989). Psychosocial predictors of pregnancy outcomes in low-income black, Hispanic, and white women. Nursing Research, 38, (4), 204-209.

Norbeck, J., De Joseph, J., & Smith, R. (1996). A randomized controlled trial of an empirically-derived social support intervention to prevent low birthweight among African American women. Social Science Medicine, 43 (6), 947-954.

Norbeck, J.S., Lindsey, A.M., & Carrieri, V.L. (1981). The development of an instrument to measure social support. Nursing Research, 30 (5), 264-269.

Norbeck, J.S., Lindsey, A.M., & Carrieri, V.L. (1983). Further development of the Norbeck Social Support Questionnaire: Normative data and validity testing. Nursing Research, 32 (1), 4-9.

Norbeck, J.S., & Tilden, V.P. (1987). International nursing research in social support: theoretical and methodological issues. Journal of Advanced Nursing, 13 (2), 173-8.

Paskiewicz, L. (2001). Pregnant adolescents and their mothers. Maternal-Child Nursing, 26 (1), 33-38.

Perrin, K.M., & McDermott, R. (1997). Instruments to measure social support and related constructs in pregnant adolescents: A review. Adolescence,32 (127), 553-557.

Polit, D.F., Beck, C. T., & Hungler, B.P. (2001). Essentials of Nursing Research: Methods, Appraisal, and Utilization (5th ed.). Philadelphia: Lippincott.

Raisler, J. (2000). Midwifery care research: What questions are being asked? What lessons have been learned. Journal of Midwifery & Women’s Health, 45 (1), 20-36.

Sandelowski, M. (1995). Triangles and crystals: On the geometry of qualitative research. Research in Nursing and Health, 18, 569-574. Schaffer, M.A., & Hoagberg, B. (1997). Effects of social support on prenatal care and health behaviors of low- income women. Journal of Obstetric, Gynecologic and Neonatal Nurses, 26 (4), 433-440.

Scupholme, A., Paine, L.L., Lang, J.M., Kumar, S., & DeJoseph, J.F. (1992). Nurse-midwifery care to vulnerable population. Phase I: Demographic characteristics of the national CNM sample. Journal of Nurse Midwifery, 37 (5), 341-348.

Secco, M.L., & Moffatt, M.E.K. (1994). A review of social support theories and instruments used in adolescent mothering research. Journal of Adolescent Health, 15 (7), 517-527.

Smoke, J., & Grace, M.C. (1988). Effectiveness of prenatal care and education for pregnant adolescents: Nurse midwifery intervention and team approach. Journal of Nurse Midwifery, 33 (4), 178-184.

Snyder, C. (2000). Handbook of hope: Theory, measures, and applications. San Diego: Academic Press.

Snyder, C.R., Cheavens, J., & Sympson, S.C. (1997). Hope: An individual motive for Social Commerce. Group Dynamics: Theory, Research, and Practice, 1 (2), 107-118.

Stevenson, W, Maton, K.I., & Teti, D.M. (1999). Social support, relationship quality, and well-being among pregnant adolescents. Journal of Adolescents, 22 (1), 109-121.

Unger, D.G., & Wandersman, L.P. (1985). Social support and adolescent mothers: Action research contributions to theory and application. Journal of Social Issues, 41 (1), 29-45.

U. S. Department of Health and Human Services (USDHHS). (2000). A national strategy to prevent teen pregnancy: Annual Report 1999-2000. Washington, DC: U.S. Department of Health and Human Services.

Vaux, A. (1988). Social support: Theory, research and intervention. New York: Praeger.

Ventura, S.J., Mosher, W.D., Curtin, S.C., Abma, J.C., & Henshaw, S. (2001). Trends in pregnancy rates for the United States: An update. National Vital Statistics Report 2001 June6;49 (4),1-9.

Wayland, J., & Rawlings, R. (1997). African American teen mothers' perception of parenting. Journal of Pediatric Nursing, 12 (1), 13-20.

Yarcheski, A., Scoloveno, M., & Mahon, N.E. (1994). Social support and well-being in adolescents: The mediating role of hopefulness. Nursing Research, 43 (5), 288-293

Zhang, J., Bernasko, J.W., Leybovich, E., Fahs, M., & Hatch, M.C. (1996). Continuous labor support from labor attendant for primiparous women: A meta-analysis. Obstetrics and Gynecology, 88 (4), 739-744.

. . . . . . .