Carmela R. Nanton
Being all things to all (wo)men: culture and research on women of colour
Carmela R. Nanton
Palm Beach Atlantic University, USA
Paper presented at the 36th Annual SCUTREA Conference, 4-6 July 2006, Trinity and All Saints College, Leeds
The cultural impact and influence on research projects involving women of color is undeniable. This study examined how African diasporan women made the decision to participate in adult education programs, drawing on the theoretical frameworks of socio-cultural theory and research methodology in a qualitative case study design. The purpose of this paper is to address researcher challenges, limitations, and opportunities from an empirical study on women’s participation in adult learning programs. It discusses the significance of spirituality as a cultural phenomenon with influence on the research methodology; addresses researcher challenges arising from cross-cultural perspectives on identity, as the ‘outsider within’ from a different cultural background, and as a rater of the test instrument for ego development level.
Theoretical framework and rationale
The primary strategy of inquiry for this research project was undertaken from a qualitative, interpretivist, case study standpoint (Yin, 1994; Denzin and Lincoln, 1998). Additional strategies of inquiry for this research project included a constant comparative method (Cresswell, 1994) in order to compare the raw data to the provisional categories and emergent themes. The rationale for the qualitative methodology was to allow for the use of narratives for the individual women’s cases, and to ensure that rich data captured the stories as told by them. The essence of this research project hinged on the stories respondents chose to share from their experiences. It seeks to emphasize what Ladson-Billings (1997) describes as “the inner voices that are self-reflective about the cultural and educational world that has shaped them” (p. 52). This interpretive and descriptive form of research allowed for the subjects to share the meaning they have constructed from their experiences and perceptions.
An extensive body of literature exists on research methods and data collection (Denzin & Lincoln, 1998, Cresswell, 1994, 1998). The socio-cultural history of African diasporan women’s value of education; their experiences with access, and the challenges inherent in their educational pursuit are well documented (Allridge, 1999; Gregory, 1995; Kett, 1994; Ladson Billings, 1997). The link between education and generational, racial, and socio-economic advancement as a histo-cultural and contemporary motivator for these women is clear (Alfred, 2003; Brice-Baker, 1994; Gregory, 1995). Yet, research on these women’s participation in contemporary adult education programs fail to adequately discuss the cultural influences inherent in the research process, particularly the socio-cultural dynamics of data collection, usage of metaphors, incorporation of spirituality as part of adult education research and practice (Hunt, 2005), and the challenges of analysis and interpretation of findings.
Research methods
Student volunteers (n= 28) were selected by stratified purposeful, criterion based convenience sampling (Marshall and Rossman, 1999). The intent was to capture the diversity of African diasporan women, the women were from African, American, and Caribbean nationalities and cultures (Alfred, 2003; Comas-Diaz & Greene, 1994). The Christian Liberal Arts higher education adult evening program context constituted the study’s physical boundary; a common African heritage comprised the ethno-cultural boundary. The women had to be fulltime degree seekers, willing to complete all aspects of the study. The researcher met with the women twice: the first meeting explanations/instructions were given for completing the demographic survey, including marital status and cultural heritage; the critical incident (Russ-Eft, Berrey, Boone, &Winkle, 2000); the in-depth interview (Seidman, 1998); and the SCT sentence stems. The Sentence Completion Test (SCT) instrument (Loevinger, 1998), comprised of 36 sentence stems was incorporated because it originated from studying the way women dealt with life’s problems its potential linking of ego development level and pursuing an education. This aspect of the research involved quantification: the women’s responses to the sentence stems were converted to the corresponding ego development level. At the second meeting the women brought completed documents, the semi-structured in-depth interview was tape recorded, subsequently transcribed and analyzed from a qualitative, interpretivist, case study standpoint (Yin, 1994; Denzin & Lincoln, 1998).
This methodology was chosen to create narratives of the stories and to preserve the social nature of decision-making for these women (Marshall & Rossman, 1999). The research questions examined the steps of the decision process, influential factors/challenges of decision making, decision outcomes and resulting changes. Findings were analyzed by constant comparative method (Cresswell, 1994) that enabled the researcher to weave together the data from the in-depth interviews and critical incidents that captured the women’s experiences (Miller and Crabtree, 2000). Each interview transcript and critical incident was read to understand the woman’s story, followed by thematic identification, then a cross-case analysis was done for all transcripts to identify common patterns. Finally, the researcher used triangulation (Daley, 1999), peer reviewers, member checks (Marshall and Rossman, 1999) with a focus group of participants; all re-entry women, that examined and discussed data, findings and analysis with the researcher. Some of the socio-cultural influences that became evident in the research surfaced at this point, and will be discussed next.
Socio-cultural influences on research methods
The examination of socio-cultural influences begins with research methods. One research boundary was that respondents had to be a part of the African diaspora, the cohesive commonality being an African heritage. Two major categories within this boundary comprised respondents of Caribbean and American backgrounds. Several sub-groups within these two major categories were additionally identified. Early on in the data collection phase, the researcher observed that some respondents did not wish to be classified as having an “African” cultural heritage; others did not want to be classified as “African American”, choosing rather to claim their ethnic culture or country of national origin. The distinction was articulated by those from Caribbean and mixed backgrounds originating outside of the U.S. The women from mixed backgrounds originating in the U.S. made no such distinction and unquestioningly assumed the ‘African American’ label as a personal descriptor. Black (1996, p. 64) points out that “people from the Caribbean are less likely to think of themselves as minorities than African Americans because where they grew up they were the numerical majority” The use of the term “African diaspora” was utilized as a means to ameliorate this phenomenon in the development of the sample, to ensure maximum representation of diversity, and to reduce participant’s voluntary de-selection based on this terminology.
The social nature of the interview process was also significant: the women sought to build relationship. In fact, they expected the researcher to allow them to share their personal background, even as they sought some social or relational connection with the researcher to increase the comfort level and rapport between the researcher and participant. Additionally, the women assumed a commonality of experience based on the shared African cultural heritage. As a group, African diasporan women have suffered cultural oppression and discrimination as a result of historical enslavement, colonization, or cultural socialization. The assumption of commonality of experience for African, American, and Caribbean study participants was evidenced in the study’s findings (Nanton, 2005a), but that commonality did not connote the mutuality of experience assumed by the participants. The women assumed the researcher understood or identified with their socio-cultural historical experiences using such statements as “you know where I’m coming from” or “you know what I mean”, when in reality the researcher was not always able to identify with the insider knowledge that participant’s believed was the same as theirs. Contemporary discrimination still continues though the basis for that discrimination varies between groups and across societies (Brice-Baker, 1994; Collins, 1999; Sowell, 1983). This variation in experience contributed to the researcher’s being positioned as outsider to participant’s experience, and points to other potential sub-group differences within the sample.
Socio-cultural influences on research data analysis
The area of analysis where cultural influence was evident was the Sentence Completion test (SCT) instrument for determining ego development levels. The problematic implications for the study was directly related to Hall’s (1997) pessimistic view of the theoretical literature of psychology that “consistently under-represents U. S. subcultures”, indicating a lack of adequate attention to socioeconomic and minority status factors when they are included in study samples (p. 74). Hall argued that though the use of this instrument has extended across gender, age, languages and various cultures, many of those studies have not included women of color, nor have the constructed samples been broad enough to adequately address the complexity of culture, ethnicity, and socioeconomic variables.
The significance of this continued omission is the prospect of current psychological literature resources becoming potentially irrelevant and of less value for dealing with a population where people of color are projected to approach 50% within the next decade. Furthermore a study that has assessed the developmental levels of a similar sample of women, in particular one from this research perspective that also utilized Loevinger’s (1998) ego development assessment instrument is relatively unique. The E level assignment process included both manual and electronic computation of an Item Sum Score (ISS), the development of a Total Protocol Rating (TPR), and assignment of the final E level. The instrument results were then forwarded to the second rater electronically for rating, along with a copy of the respondent’s original written work. The sentences rated by the researcher and second rater, were compared. Rating discrepancies were resolved via discussion between raters. Final ratings given were those that both the second rater and the researcher agreed upon.
Inter-rater reliability was 0.89 (Ellis 2004, personal communication), allowing for margin of error and sensitivities. Areas of disagreement and discussion, in descending order of frequency related to level ratings (differentiation between unelaborated vs. elaborated responses); different within-level score assignments (agreed upon level ratings, but different categorical assignments); culturally influenced responses (as compared to the available normative manual responses); and inferential determination tendencies or cultural differences between the researcher and second rater. It was in the latter two areas that highlighted cultural values in relation to family and spirituality, and identified areas of manual responses that did not adequately represent the cultural worldviews or values of this group of women.
Emergent challenges in administering the instrument included the researcher’s mastery of the SCT rating skills, via practicing with manual samples: this was offset by the use of a second rater. Additional limitations for SCT E-level results were the potential for “experimenter effects” that would cause the respondent to respond the way they think the researcher expects them to, in the hopes of pleasing the researcher. Or, conversely, the respondent may be having a bad day causing them to revert to responses at a lower level than that which is normal for them.
Of note were the women whose scores did not cleanly fit a particular E-level; for example, Item Sum Scores were at a higher level, but they did not have sufficient responses representing the higher levels as required by the evaluation criteria. .
The four women (14.2%), whose SCT results did not cleanly fit established criteria, could have potentially been located in transition between stages. These protocols were placed in a designated level after inter-rater collaboration and re-assessment of SCT responses and Total Protocol Ratings (TPR’s) in light of the electronically weighted responses. Plausible explanations for those protocols included: they may be individuals in transition between stages; they may have been influenced by the circumstances, their mood on the day the instrument was completed, not operating at their optimal development level, or trying out new ways of coping with situations they had not experienced before. Individuals under certain circumstances have been known to operate at lower than their normal ego-development levels (Hy and Loevinger, 1998). The researcher proposes that individuals also operate at levels higher than were formerly optimal for them when former responses no longer suffice for meeting their situational need.
Regarding relationship responsibility, the survey revealed that a majority of the study participants (n = 22) had dependents, in the form of children, two of them were ‘parentified’ (Hines and Boyd-Franklin, 1996) at a young age; with the responsibility of taking care of their children, younger siblings, and a maternal parent. Within this group, the caretakers 86.3% (n =19) were responsible for a number of dependents ranging from one to six in number. Parentification and out-of-sequence role responsibility also provide plausible explanations for higher developmental levels evident at a young age (27yrs at E7; 33yrs at E8). Transitional movement or possible shifts in development level expectations and roles outside of the traditionally expected age-related adult development levels. These examples are areas ripe for future research into adult development and adult learning.
Culture and researcher bias
The researcher identified a paradox or hidden bias in the significance of the role that spirituality played in the decision making and participation process. In the interest of maintaining objectivity, the researcher did not initially follow up on the first evidence of this phenomenon in the pilot studies. Incidentally, the pilots were done with three women who fit the criteria, but who were outside of the physical boundary of the study. One of the lenses, through which the research data was examined and analysis and interpretation informed, was developed from the researcher’s family legacy of spirituality that goes as far back as the researcher can remember. As an ordained minister, the researcher was well aware of the significance of the use of spirituality as a coping mechanism, the important cultural role that faith and spirituality has played, as well as the gender biases and dichotomies inherent in that system for many women of color. However, in the interest of objectivity the researcher did not consider it as a factor to be overtly addressed in the research design, perhaps because the research data had already been collected at the time of ordination. Spirituality had become such a tacit epistemological orientation that it was not an explicit part of the research process. It was the research pilot studies that initially highlighted the significance of spirituality to the women in the study; however, it was not until this phenomenon again appeared in the first research interviews that it formally became a part of the research process with its own question. Participant references to faith/spirituality were so deeply embedded in the researcher’s psyche, that it was not viewed as something of thematic importance. As a normative response component it became a thematic component when it was identified in all but one respondent’s transcript, as was the liberal use of metaphors by study respondents.