Collaborative research and reproductive health outcomes among African American women.

Abstract: The cause of disparities in the reproductive outcomes among African American women continues to elude health researchers and professionals. While stress has been identified as a health risk factor, researchers have called for a reexamination of the measures and methods employed in investigations aimed at uncovering the link between stress and reproductive health outcomes. This paper describes an iterative methodology employed in community-based research conducted in Atlanta, Georgia Guided by a focus on the multiple and intersecting identities assumed by African American women (race, class, and gender), the research sought to locate the stressors confronted by African American women that are embedded in their lived experiences. Another aim was to develop a race and gender-specific stress measure drawn from the lived experiences of African American women, These objectives were accomplished through procedures that explored the conceptualization of community as experience by African American women, the interplay of qualitative and quantitative methods, and the translation of reciprocity, accountability, and sensitivity to research where the participated were seen as collaborators informing every phase of the research process.

**********

Bombarded with so many things, if it ain't the men, it's the children, it's society, it's something. So I'm not saying that, you know, we don't consciously take care of us, it's a thing that by the time you finish caring for everybody and deal with everything that happening on the outside of your body ... the inside is probably the last thing that you pay attention [to].

Jennifer, a college educated woman

This comment by Jennifer is indicative of the causal link between stress and poor health outcomes acknowledged by African American women. While the association between stress and health outcomes, in general, is well established, investigators have argued for a closer examination of the link between stress and reproductive health outcomes among African American women (Burchfield,1979; Dressler, 1991; Kreiger, 1996; Rowley, 1993). In light of finding revealing that for African American women education does not mediate reproductive health outcomes as expected, the search for answers has led to a reexamination of methodologies and measures employed in stress research (Kreiger, 1995; McClean,1993). In particular, it has been argued that while generalized stress measures provide certain information linking stress to poor health outcomes, they nonetheless are inadequate for assessing the specific stressors confronted by African American women. The acknowledgement of the necessity to "fill in the gaps" on race and gender specific stressors coupled with the promotion of community-based research has informed the development of iterative, triangulated approaches for discerning the link between stress and the disparate health outcomes among African American women.

In this paper, we will describe the processes deployed in a series of studies on stress and reproductive health outcomes that were conducted in metropolitan Atlanta. The intent is to simultaneously illustrate the procedures and the theoretical and conceptual frameworks that shaped the research process. Because of the focus on the identities of African American women linked to race, gender, and class the aim is to demonstrate the translation of a focus on multiple and intersecting identities to the procedures for conducting community-based research. Correspondingly the goal is to examine the notion of collaboration between researcher and study participants within the parameters of a research endeavor.

The aim of our research on stress and reproductive health outcomes was two-fold: 1) to locate the stressors and supports in the lives of African American women, particularly college-educated African American women through qualitative methodologies, i.e. focus groups and interviews and 2) to have what the women said about their lives and the stress they encountered in the those lives inform the development of a race and gender specific stress measure (quantitative). Those objectives were accomplished through a community-based approach that employed iterative procedures combining qualitative and quantitative methods. Here we demonstrate the application of a community-based approach to a research project that has evolved to development interventions for stress reduction (Jackson, 2002).

COMMUNITY-BASED RESEARCH

Community-based research has been widely endorsed as an effective approach for uncovering the complexity of factors that imperil health outcomes. (Braithwaite, 1992) Proponents assert that the research process is greatly enhanced by the insights and expertise of community participants. They bring to the process an authentic view of the perceptions and behaviors surrounding health and well-being. Correspondingly, a community based approaches can expose barriers to health and well-being while promoting critical analysis and activism aimed at health promotion and intervention.

Unfortunately, the concept of a community-based approach has been applied to research endeavors where participants are merely present without any real involvement in the research process. More ideally, community-based research can encompass a comprehensive collaboration where researchers and community participants form authentic partnerships characterized by the community's participation and monitoring of all phases of the research (Mullings, 2001). Best practices in community based research are characterized by community empowerment as the foundation for developing sustainable health promotion and intervention. Implicit in this goal is adherence to the tenets of community based research: reciprocity sensitivity, and accountability. Thus community involvement in the process and access to its findings is requisite of for community partnering efforts.

While there are no limitations on the methodologies employed under the "umbrella" of a community-based approach for conducting health research, conceptual models and associated methodologies exist that lend themselves to collaboration between researchers and community participants. Grounded theory as a conceptual model and methodology offers an opportunity for the voices of the community to be heard and provides a framework for organizing those voices into theory; theory that can guide future explorations. The objective of this approach is to create theory that is representative of the direct experiences under observation. Thus the theory is juxtaposed to the methodology and the interpretation of the data; the emergent theory is evaluated in terms of its authentic representation of the observed experiences (Richie. 1997; Glazer, 1992).

Proponents of grounded theory assert that investigators should restrict their immersion into the literature associated with the topic. Rather, the emphasis should be placed on the methodology as the source of information. It is improbable that investigators will enter a research endeavor, "naive" about the topic; yet the focus in grounded theory on capturing the experience is complimentary to a community-based research approach where the voices and experiences of the participants frame the research agenda.

Qualitative research methods provide the best approach for collecting the lived experiences of community participants in their own words. Focus groups, interviews, participant observation and other qualitative methods permit a systematic way of observing study participants in their communities, workplaces, or other settings. Through qualitative methods we gain perspective on the behaviors associated with health consequences and the multiple contexts in which they occur (Scrimshaw, 1992; Glesne, 1992; Peskin, 1993).

Qualitative approaches insure rich, contextualized data however the methods are limited in the ability to produce findings that can be generalized across populations. Hypothesis formulation and theory development may result from qualitative data but the determination of the wider application of theoretical and conceptual models require methods that are reliable (Weller, 1988). The goal of reliability and capacity to generalize findings across population unfortunately typically come with the loss of contextual information for interpretation of the data.

To resolve the dichotomy between qualitative and quantitative data researchers have advocated the combination of both approaches. Advocates argue that the integration of qualitative and quantitative methods enhances the researcher's ability to understand risk factors by cross validating the qualitative and qualitative findings (Carey, 1995). The appropriation of qualitative methods or the integration of qualitative and quantitative methods under the rubric of a community-based approach, however, does not in and of itself insure a process that will uncover the significant factors contributing to disparate health outcomes. In the case of research focusing on the health of African American women, what is needed is the deliberate articulation of the identities that African American women assume and those imposed upon them to gain a view of their lives and how their lived experiences impact health outcome.

Investigators acknowledge that experiences of oppression can have adverse consequences for health outcomes; it has been demonstrated that racism is a health risk factor (Harrell, 1998; Kreiger, 1993; Utsey, 1998; Williams, 1997). However, more information is needed to understand the pathways by which racism contributes to poor health. In the case of African American women examinations of the triple forms of oppression they encounter are warranted. That is a focus on oppression associated with race, gender, and class permits detailed analysis of the barriers to health confronted by African American women. Correspondingly, we assert that examinations of race, class, and gender, must be inclusive of how those identities are embraced by African American women. In other words, research endeavors involving African American women intended to explore the impact of racism and other forms of oppression should consider the polarized relationships that accompany notions of race (and gender). Examinations of oppression therefore must include way to explore the impact of the identities of African American women that are simultaneously embraced by and imposed upon African American women (Jackson, 2001).

To locate the connection between the identities of African American women and the stress in their lives demands methodologies informed by the voices of African American women. Our community-based research employed a practice and a tradition of African American women naming their experiences and making sense of their lives. Several scholars have analyzed the long history of African American women describing their lives and interpreting their history (Giddings, 1984; Carby, 1989; Hills-Collins, 2000, Hill-Collins, 1998, King, 1988, Higgobothan, 1994).

African American women have a long reputation of articulating their experiences associated with health. In this statement by an African American women, we gain a glimpse of complexities surrounding pregnancy and birth:

   To me having a baby inside me is the only
   time that I'm really alive .I know that I
   can make something, do something no matter
   what my skin is and what names
   people call me. When the baby gets born I
   see him and he's full of life, or she is
   and I think to myself it doesn't make nay
   difference what happens later at least
   now we have got a chance, or the baby does.
   (Ladner, 1993, p.314).

The voice of this woman points to the multiple and devalued statues occupied by African American women. Their discussions and descriptions offer analysis of the simultaneous experiencing of the race, class, sexuality, and gender. Understanding the cornucopia of risk factors and health mediators for African American women is best accomplished through processes that take into account the multiplicative interaction of the of the identities assumed by African American women.

UNDERSTANDING STRESS: THE RESEARCH PROCESS

The research on stress and reproductive health was guided by the desire to answer several questions:

* If a college-education and presumed access to prenatal care is not protective for reproductive health outcomes among African American women, why not?

* If stress is a risk factor for adverse birth outcomes among African American women, is it possible or desirable to measure the context and content of the particular stressors that impact their health.

* How is stress operationalized within the context of the lived experiences of African American women representative of diverse educational and income categories?

* How do the stressors that precede pregnancy impact the stressors that accompany pregnancy, birth, and motherhood for African American women?

* How should a community-based research project be implemented to promote the collaboration of African American women from diverse educational and income categories?

* How can the research process and the subsequent findings inform the development of stress interventions?

Because we acknowledged the authoritative voices of African American women for understanding their lives and those conditions that imperil health and well-being, we sought their collaboration in our quest for answers. Research participants were collaborators rather than subjects and as such their voices informed the research process. In total 545 African American women, from all walks of life but mainly college-educated participated in a series of studies to uncover the stressors confronted by African American women. (Table I, II). In actuality the number of collaborators exceeds the women who participating in any phase of the research; collaborators also included those individuals who assisted in identifying individuals and groups of women who completed the focus groups, interviews, and the administration of the instruments. During the first phase, 444 women from all walks of life but mainly college-educated participated in interviews, focus group and the pilot testing of a race and gender-specific stress measure. Subsequent to the development of the stress measure, 101 pregnant African American women replicated the process.

Our quest for answers proceeded through an iterative process that included focus groups, semi-structured interviews, a jury exercise, pilot testing of the stress measure, and the administration of a battery of validity measures (NHIS-depression, Spielburger state and trait anxiety, Spielburger state and trait anger, John Henryism Scale of Active. As Figure 1 indicates, each phase of the research process was informed by the women's voices. Beginning and ending with focus groups, the procedure was constructed so that we "checked in" with the women throughout the data collection phase to confirm their experiences of stress and support. The administration of the quantitative measures was preceded and followed by focus groups and/or interviews. After the completion of the initial focus groups and interviews, content analysis was performed to develop statements capturing the women's voices. Research collaborators were subsequently asked to participate in a jury exercise where they critiqued the statements examining their content and format. As a consequence of their assessment, exploratory factor analysis, and further cross validation using the qualitative data, we constructed a 71 item pilot stress measure.

The pilot measure and the validity scales were administered twice over a thirty day period. After the completion of the second administration of the stress measure and validity instruments, groups of women participated in discussions where they were again asked to critique the pilot measure and elaborate on their response to individual items. To confirm our impressions about the utility of the instrument and our thoughts about the contexts for responses to individual stress items, interviews were conducted with a sub-sample of the women who pilot tested the stress measure (n= 42). Specifically the women were asked to explain their response to items from the pilot stress measure.

Our elaborate and time-consuming research approach was deployed to "ground" the research and most explicitly the stress measure in the lived experiences of African American women. This integration of qualitative and quantitative methods allowed the cross validation of what the women said about stress and our measure of those experiences. The goal was to maintain a corresponding relationship between what the women said and the stress scale items. In other words we were committed to sustaining the integrity of the authority of the women's voices in the creation of a race and gender-specific stress measure (Figure 1).

Because of our adherence to the tenet of reciprocity and accountability, the research did not end once data collection was completed; rather once the data was collected we began a process of disseminating the research finding (Figure 1). This phase of the research was accomplished thought the publication of newsletters and the convening of a day-long gathering where the women were exposed to the findings from the research and intervention strategies for stress reduction.

BEGINNING WITH THE VOICES OF THE WOMEN

To pre-assess our methodology, we conducted an initial focus group where participants were asked a series of questions to elicit their experiences of stress embedded in their lives. In order for us to observe the process, we asked one of the women recruited for the research to facilitate the focus group. A corporate manager experienced in group facilitation, she introduced the discussion:

   African American women are being asked to
   participate in a study on our health issues.
   Previously we were engaged at the point of
   the questionnaire. Because these questionnaires
   were not designed by us sometimes
   issues were left out that are applicable to the
   problems that we see in our lives. So tonight
   we are trying to change that. This focus
   group is designed so that we can get the right
   answers, so that our daughters and granddaughters
   don't have to suffer the same ills
   that we are seeing.

As revealed in her introduction the research was conceived as a process where the research and study participants would interact as collaborators. In contrast to the designation as "subject" we were in partnership with the women. At the essence of our research was the perspective that African American women are authorities on their lives and therefore possess significant insights on those conditions that imperil their health.

The "conversation" during the pilot session was initiated by the women being exposed to health disparity data, particularly that which focused on African American women. The presentation of those data became the stimulus for research collaborators to share their own experiences as the avenue for offering explanation for the poor health seen among African American women. As expected the women provided the usual explanations: diet, lack of exercise, family history, and access and utilization of health care services. But of all of the factors they identified, the burdens of having so many expectations and responsibilities imposed upon them as women and African Americans--many of whom were established middle class or upwardly mobile--resonated for the women as the explanation for disparate health outcomes. They specifically identified stress as a significant contributor to their poor heath and the health of relatives, friends, and colleagues. Their "naming" stress and theorizing about its link to preterm delivery and low birth weight and other health concerns among African American women was the foundation of our search for answers.

RECRUITMENT: MORE THAN A TELEPHONE CALL

From our pilot focus group we preceded to recruit women for subsequent focus groups, interviews and the administration of the pilot stress measure and the validity scales. To operationalize stress within the context of the lived experiences of the women, we identified multiple sites for the recruitment of research collaborators. With the exception of the last phase of the research, women were recruited from non-clinical sites. During the last phase the women who volunteered were recruited from three Ob-Gyn offices located in metropolitan Atlanta.

Understanding the multiple and taxing demands in the lives of African American women informed our process for recruiting collaborators. We sought opportunities to enter the day-to-day involvements of the women which led in an examination of the configurations of "community" among African American women. The term community has been appropriated to symbolize residential enclaves where members of a racial or ethnic group and the poor reside. This limited perception of community obstructs the realities of how and where people of color live. Our understanding of the fluidity of community led us to look beyond the boundaries of location, education, income and employment when attempting to identify the "places" of association embraced by groups of individuals. Thus community in these studies was inclusive of locale and the interactions maintained among the women as they participated in clubs, churches and even as associates in the workplace. Adherence to an approach extending beyond locale insured stability of the populations of women in the study. Because of the transient characteristics of the women (relocation by women across SES categories), we encountered challenges maintaining contact with them throughout the study. Our chances were improved if the communities that we entered emerged from associations. Furthermore, we have hypothesized that "community" as a consequence of association offers advances for sustaining effective interventions.

In our effort to maximize the data collection process, we explored opportunities where the women would be more apt to share the experiences of their lives. Since we wanted to know about the lived experiences of stress and support for college-educated women, we predicted that contacts with groups of women during their college reunion activities would be opportune time to engage them in dialogue. Our hunch was correct; the focus groups during reunion activities served a means to organize the sharing that was already taking place around their experiences of stress and support.

The sampling approach was both purposive and opportunistic. Contacts with alumnae groups from HBCUs offered an avenue for us to locate study collaborators. Beginning the process with graduates of Spelman College, a historically black college for women, the recruitment of other collaborators was done through a "snowball" process; the women referred us to other individuals and organizations.

Recruitment of the collaborators for the pregnancy phase of the research was achieved through response to posters placed in medical offices and referrals by the physicians and office staff. A snowball sampling process was employed in the pregnancy phase where study collaborators identified other prospective participants.

WHERE WERE THE WOMEN?

Our intent was not only to hear what the women said about their lives but to also have some limited observations of what they were experiencing. The research was not ethnographic in the traditional sense, but it was nonetheless conducted in ways that validated our perceptions of the context for understanding stress and support. There were instances where we had multiple and sequential contacts with a sub-sample of the women who were involved in the post-measures interviews and had been participants in the administration of the battery of instruments. What allowed us to examine more closely the conditions shaping the experiences of stress for the women was the settings were we collected the data. Data collection, particularly the interviews took place in a variety of settings where the women lived and worked. Typically, the interviews were conducted in the homes of the collaborators. At other times we visited worksites or other non-residential settings. In one instance, a worksite interview required a member of the research team to accompany a florist as she traveled throughout the city purchasing supplies and making deliveries. Interviews were conducted in shopping mails and during times when women participated in volunteer and social activities. It was those approaches for entering the lives of the women that framed our layered view of what stress and supporting the lives of study collaborators.

It was also important to be mindful of the fact that the women lived in the Atlanta metropolitan area, and geographical location provided yet another important context in which the women lived their lives. Atlanta, long considered a "Mecca" for African American professionals, is a city--much like the women themselves--characterized by complicated multiple and intersecting identities. It is a city (and metropolitan area) of both a long-standing elite African American political class, a visible African American middle-class within which financial straggle is often masked, and a visible sector of the population that is poor. Atlanta is also a city with a history of negotiated "peace" between African American and white, primarily male, political and economic elites when the rest of the South was in open warfare during the 1960s and 1970s (Baylor, 1996; Rutheiser, 1966). In fact, this same negotiated peace has allowed for the recent gentrification of key neighborhoods of Atlanta. However, underneath the negotiated calm there are racial hostilities and suspicions that are the unspoken dimensions of the debates, for example, about MARTA's (Atlanta's public transportation system) extension into the suburbs, and the city's past affirmative action policies (Atlanta Journal and Constitution, July 23, 1999). The interesting irony of the racialized discourse (dichotomized into "black" and "white") about Atlanta and its "surrounding cides" is that some of those cities and counties have sizable African American, as well as immigrant populations. It is within this very layered and complex "Atlanta" that the women experienced many of their stressors.

WHAT THE WOMEN SAID: THE CONTENT AND CONTEXT OF THE PROCESS

Each phase of the research expanded our perception of what stress meant in the lives of the women. As indicated, conversation during the focus groups was stimulated by the women's exposure to health disparity data. In the focus groups following the pilot session, the women identified personal neglect and stress as risk factors. Janine, a 10-year college reunioner offered as explanation for the health among African American women:

   The most important thing is that we tend
   not to take care of ourselves for taking care of
   others.

   Lisa, a manager in the group said:
   You know what I think it is denial.... OK,
   I'm comfortable because I went to college
   and I know when I get to hurting I will go.
   And in the mean time I have too much too
   do and I don't have time and it does take a
   lot of time and money.

During this same focus group, Kay an attorney contributed:

   Stress, I fed that the reason is stress. Society is
   probably the biggest thing
   Probe: In what way?
   Even when you are educated, even when
   you are attorney you are still always
   fighting the battle of being Black first and
   the profession second ... It's always the
   next hurdle that you go over because you
   are black. That's weight-heavy.

At subsequent gatherings other collaborators continued to identify stress linked to racism and other forms of oppression and the context shaping those experiences. During the jury exercise collaborators confirmed or challenged the statements that were gleaned from what the women said during the focus groups and interviews. In response to the statements, "Everyone expects me to be strong for them" and "Everyone depends on me" one of the collaborators indicated:

   For the most part most of these statements
   relate to black women ... Well right
   now I have a friend--she works and her
   mother does not work. And her mom was
   in an accident, now she has to take care of
   her mom. So she goes to work and
   when she gets home she has to take care of
   her mother ... you expect to kind of
   have things that you want to do and what
   you want to do but now she has to think
   about someone else as opposed to thinking
   about herself. And that's great stress.

Because a majority of the women administered the pilot stress measure agreed to the statement "everyone expects me to be strong for them" we asked participants selected for the post-measure interviews to explain their response in the statement. Vanessa, in her mid thirties said:

   Um I think in any group of people you will
   have strong women. But I think as
   black women, from birth we are told we have
   to be strong because there will be a
   lot of obstacles that we will have to overcome.

In another post measure interview, Mary elaborated on the concept of the strong black women by saying:

   ... just a long history of strong black women
   who jus dealt with any number of
   things that would make us look like punks.
   The things that we whine and
   complain of, you know, knowing that you
   come from that you know, type of
   stock ... it makes everything seen possible.

The responses of these women and other collaborators demonstrated that stress is embedded in the multiple and overlapping contexts of women's lives. Their voices confirmed our assertion that the stress encountered by African American women is shaped by race, gender, class, social conditions, culture, history and region. Specifically, we uncovered that stress as the result of racism is exacerbated by the gendered roles that African American women assume as caretakers (Jackson, 2002). Our process allowed us to capture in each progressive stage the dimensions of race specific and gendered stress.

Comparable to the construction of the pilot measure, questions posed during the focus groups and the interviews were drawn from the prior sessions with the women. For instance, during the focus groups the women identified their responsibilities as caretakers as a stressful; therefore, during the interviews we asked the women to discuss the positive and negative aspects of their family life involving both immediate and extended kin. The post measure interviews were designed for the women to expound on their responses to the scale items. That is, collaborators detailed the events in their lives that informed their responses.

The women participating in the focus groups indicated that they saw those experiences as desired opportunities to interact with other African American women mound concerns impacting their lives. In this respect the focus groups became a form of intervention fostering the exchange of resources and experiential information. Our embrace of reciprocity and accountability as core to community based research led us to examine the merits of focus groups and group discussions as avenues for health interventions. Without question, support groups and other group processes are widely employed as approaches for health interventions and promotion. But our appropriation of a group process was for data collection and less explicitly for the purpose of intervention. Despite this distinction, led by the voices of the women we sought ways to continue the "dialogue" as a form of reciprocity for their participation in the research. The aim was to initiate an evidence-based approach for stress intervention as reciprocity for the women's participation in the research. The goal of facilitating interactions among the research collaboration resulted in our convening two day-long conferences where the women gathered in small "talking circles" to continue conversations that took place during the focus groups and interviews. These gatherings were also rimes when the collaborators were provided with the preliminary findings of the research and given information and demonstrations on stress interventions. As a consequence of our methodology including the conferences, there is evidence that the collaborators are forming new associations among themselves aimed at sustaining approaches for stress intervention.

CONCLUSION

Situated in the lived experiences of African American women, the methodology allowed us to simultaneously confirm and extend our conceptualization of stress. More importantly, as a consequence of the process (including the conferences), it appears the collaborators are forming new associations among themselves aimed at sustaining approaches for stress intervention. That is the methodology has fostered the development of a "community" posed to address their health concerns. The potential longevity of our collaboration with this community represents an evolution from research to an integrated approach for what we hope will be the beginnings of sustained interventions for eliminating health disparities among African American women.

Table 1. Selected Sociodemographic Variables for Total Sample (N=444)

Variable                                                             N

Age
  18-24                                                              13
  25-34                                                             154
  35-49                                                             192
  50-79                                                              77
  Missing data                                                        8
Education
  College educated                                                  321
  No college degree                                                 100
  Other degree                                                        1
  Missing data                                                       22
Marital status
  Married                                                           162
  Not married (includes single, separated, divorced, or widowed)    281
  Missing data                                                        1
Birth History
  No children                                                       173
  At least 1 child                                                  264
  Missing data                                                        7
Annual household income (in thousands of dollars)
  <10                                                                14
  10-30                                                              81
  31-50                                                              92
  51 and higher                                                     130
  Missing data                                                      127

Table 2. Selected Sociodemographic Variables for Pregnant Sample
(N=101)

  Variable                                             N

Age
  18-24                                                19
  25-34                                                64
  49-49                                                17
  Missing data                                          1
Education
  College educated                                     62
  No college degree                                    38
  Missing data                                          1
Marital status
  Married                                              58
  Not married (includes single, separated,
    divorced, or widowed)                              53
Birth History
  No children                                          50
  At least 1 child                                     42
  Missing data                                          9
Annual household income (in thousands of dollars)
  10-30                                                14
  31-50                                                31
  51 and higher                                        48
  Missing                                               8

Figure 1. From the Voices to Scale Items

Women's Voices                       Scale Items

"Always there for everyone else      Burden item
... you are like a barrel that       I taking care of everyone else,
gets water and if it never rains     but no one is taking care of me.
you are going to get empty"

"I would work long hard hours        Racism Item
because I wanted to make a good      Individuals at my work assume that
appearance ... but it didn't make    I am incapable of performing the
a difference I was still a black     job because I am African American.
person and how dare I have the
right to be there."

REFERENCES

Baylor, R. (1996). Race and the shaping of 20th Century Atlanta. Chapel Hill: University of North Carolina Press.

Braithwaite, R.L., Resnicow, K. (1992). Coalition partnerships in health promotion. In Braithwaite, R.L., Taylor, S.(eds). Health issues in the black community. San Francisco: CA: Jossey Bass Publishers.

Burchfield, S. K. (1979). The stress response: A new perspective. Psychosomatic Medicine, 41,661-672.

Carby H.(1990). Reconstructing womanhood: The emergence of the afro-american novelist. New York: Harpers Collins.

Carey, J. (1995). Linking qualitative and quantitative methods: Integrating cultural factors into Public Health. Qualitative Heath Research, 3(3), 298-318.

Dressier, W.W. (1991). Lifestyle, stress, and blood pressure in a Southern black community, Psychosomatic Medicine, 52(2), 182-198.

Giddings, P. (1984). When and where l enter: The impact of black women on race and sex in American. New York: Random House.

Glazer, B.G. (1992). Emergence vs. forcing: Basic of grounded theory. Mills Valley: CA: Sociology Press.

Glesne, C., Peshkin, A (1992). Becoming a qualitative researcher: An introduction. New York: Longman

Harrell, J.P., Merritt, M.M., Kalu, J. (1998). Racism, stress, and disease. In Jones, R. Editor, African American mental health. Hampton, VA: Cobb and Henry Publishers, 247-78.

Higgobothan, E. (1994). African-american women's history and the metalanguage of race. Signs, 17:251-274

Hills-Collins (2000). Black feminist thought: knowledge, consciousness, and the polities of empowerment. New York: Harper Collins.

Hill-Collins (1998). Fighting words: Black women and the search for justice. Minneapolis: University of Minneapolis Press.

Jackson, FM., Phillips, M.T., Hogue, C.J.R., Curry-Owens, T.Y. (2001). Examining the burdens of gendered racism: Implications for pregnancy outcomes among college-educated African American women. Maternal and Child Health Journal 5(2), 95-107.

Jackson, FM (2002). Considerations for community-based research with African American women. American Journal of Public Health, 92 (4), 561-564.

King, D. (1988). Multiple jeopardy, multiple consciousness: The context of black feminist ideology. Signs Journal of Women in Culture and Society, 14, 42-72.

Kreiger, N., Rowley, D.L., Herman, A.A., Avery, B. Phillips, M.T. (1993), Racism, sexism, and social class: Implications for studies of health disease and well-being. American Journal of Preventive Medicine, 9, 82-122.

Kreiger, N., Sidney, S. D. (1996). Racial discrimination and blood pressure: the CARDIA study of young Black and White adults. American Journal of Public Health, 69, 1370-1378.

Ladner, J. (1993). Black women in white America. New York: Vintage Press

McClean, D.E., Hartsfield-Timajchy K., Wingo DA, Floyd, RI (1995). Psychosocial measurements: Implications for the study of preterm delivery in black women. American Journal of Preventive Medicine, 9(suppl 6), 3981.

Mullings, L., Walt, A (2001) Stress and Resilience. New York: NY: Kluwer Academic Publishers

Peskin, A. (1993). The goodness of qualitiative research. Educational Researcher, 22(2), 23-29.

Richie, B.S. (1997). Persistence, connection, and passion: A qualitative study of the career development of highly achieving African American-black and white women. Journal of Counseling Psychology, 44(2), 133-148.

Rowley, D., Hogue CRJ, Blackmore, C et al (1993). Preterm delivery among African American women: A research strategy. American Journal of Preventive Medicine, 9(supp 6), 1-6.

Rutheiser, C. (1996). Imagineering Atlanta: The politics of place in the city of dreams. New York: Verso Press.

Schoendorf, K., Hogue, C.J.R, Kleinman, J.C., Rowley, D. (1990). Mortality among infants of blacks as compared to white college-educated women: A research strategy. American Journal of Public Health. 80,70-78.

Schrimshaw, N.S., Gleason, G.R. (Eds) (1992). Rapid assessment procedure, Qualitative methods for planning and evaluating health related programs. Boston, MA: International Nutrition Foundation for Developing Countries.

Utsey, S.(1998). Assessing the stressful effects of racism: A review of instrumentation. Journal of Black Psychology, Aug, 42(321), 269-288.

Weller, S.C., Romney A.K. (1988) Systematic data collection. Newbury Park, CA: Sage.

Williams, D. (1997). Racism and health: Basic questions, emerging decisions. American JournalofEpidemiology, 7, 322-333.

Wooten, J. (1999). "Mayor's words incite racial divide" Atlanta Journal and Constitution, July 23.

HEALTH EDUCATION RESPONSIBILITY AND COMPETENCY ADDRESSED

Responsibility IV--Evaluating Effectiveness of Health Education Programs

Competency C--Interpret results of program evaluation.

Subcompetency 2--Translate evaluation results into terms easily understood by others.

Adress all correspondence to Fleda Mask Jackson, Ph.D., Rollins School of Public Health, Emory University, 1158 Clifton Road, Atlanta, Georgia 30322, PHONE: [phone], FAX: [phone], E-MAIL: [e-mail].