Appalachian Self-identity Among Women in Ohio Appalachia

Many images exist about the Appalachian region and its residents, but little is known about how Appalachian residents view themselves. We aimed to determine both the prevalence of Appalachian self-identity and to ascertain factors associated with identifying oneself as “Appalachian.” Mixed regression models using data on 571 women from the Community Awareness Resources and Education (CARE) study were utilized to construct a multivariable model identifying correlates of Appalachian self-identity. Women who had lived in their current county for a longer period of time (odds ratio [OR]=1.94, p=0.0017), had a religious association or someone in their household with a religious association (OR=1.87, p=0.0050), had both parents raised in an Appalachian county (OR=1.65, p=0.0188), lived in a neighborhood with a name (OR=1.76, p=0.0076), had a private type of health insurance (compared to having Medicaid or Medicare only) (OR=2.31, p=0.0105), lived in the more southern regions of Ohio Appalachia (OR range=2.12-4.40, p=0.0001), and lived in rural Appalachian counties (versus urban) (OR=2.27, p=0.0015) all had significantly higher odds self-identifying as “Appalachian.” The results provide new insight into the prevalence of Appalachian self-identity and its correlates, which may be useful in planning future health promotion programs in the Appalachian region. Journal of Rural Community Psychology Volume E12 (1) INTRODUCTION The Appalachian culture and its residents are unique when compared to the rest of the United States. Early descriptions of the region depicted it as “a strange land” with its inhabitants being isolated, old-fashioned, and backwards (Harney, 1873; Kephart, 1913; PARC, 1964). Based on such descriptions, many images about the Appalachian region were formed and still persist. Terms frequently used to describe Appalachian culture and residents include: white, rural, low income, uneducated, religious, family-oriented, poor living standards, and unskilled. A great deal of research has been conducted by social scientists examining the Appalachian culture. One of the more influential studies was conducted by Ford (1962) and included a region-wide survey of Appalachian values among 190 counties in seven Appalachian states. The survey examined four dimensions commonly associated with Appalachians: individualism and self-reliance, traditionalism, fatalism, and religious fundamentalism. Results supported the existence of some dimensions (ex. religiosity), while being less supportive of others (ex. fatalism). Overall, Ford demonstrated the complexity of the Appalachian culture and its heterogeneity among various subgroups in the region. One aspect of the Appalachian culture that has not been studied extensively is Appalachian selfidentity. Self-identity has been previously described as “an integrated image of himself or herself as a unique person, which often includes an ethnic identity” (Bernstein et al., 1994). While it has been pointed out that Appalachian residents lack the church organizations, distinct language, or racial characteristics that usually define an ethnic group (Billings, 2006), there is little doubt that Appalachians are distinct from the rest of American Caucasians. It has even been suggested that Appalachians represent the “white other” to mainstream America (Harkins, 2004). The uncomplimentary nature of some Appalachian stereotypes may result in a sense of regional identity in some individuals, while also causing some people in the region to discard their regional heritage (Billings, 2006). Consequently, it becomes of extreme interest to determine how many people in Appalachia self-identify as “Appalachian” and what characteristics are associated with doing so. Attempts to determine the prevalence of Appalachian self-identity have found that a third or less of Appalachian migrants considered themselves to be “Appalachian” (Miller, 1978; Obermiller, 1982; Philliber, 1983). Individuals who self-identified as “Appalachian” tended to be older, more recent migrants out of Appalachia, have lower socioeconomic status, and not live in an inner-city (Philliber, 1983). These studies were, however, conducted among Appalachian migrants, as opposed to current Appalachian residents. Appalachian self-identity among current residents of Appalachia has been addressed in focus groups and workgroups. Terms mentioned in these groups describing Appalachia and its residents include friendly, God-fearing, proud, law abiding, hard-working, impoverished, rural, loyal, non-minority, self-sufficient, resourceful, having a high ethical and moral code, high levels of social cohesion, old-fashioned, few educational opportunities, older age, and having lived in the region for an entire lifetime (Coyne et al., 2006; Behringer et al., 2007; Keefe, 2000; Tribe, 1993). Journal of Rural Community Psychology Volume E12 (1) Appalachian self-identity is of interest because past research among other racial and ethnic groups has found that self-identity affects health and health behaviors. In studies of Latinos and Korean-American refugees, participants with increased levels of ethnic self-identity displayed higher levels of self-esteem and improved psychological well-being (Shrake & Rhee, 2004; Umana-Taylor & Updegraff, 2007). Among African Americans and Mexican Americans, increased levels of ethnic self-identity were associated with lower use of drugs and alcohol (Brook et al., 1998; Herd & Grube, 1996; Love et al., 2006). Before Appalachian self-identity can be examined for such associations with health and health behaviors, it is necessary to clarify both the prevalence and correlates of Appalachian selfidentity, which were the objectives of this report. It was hypothesized that women who were older, associated with a religion, had a lower socioeconomic status, had family ties to Appalachia, had lived in Appalachia for a longer period of time, and lived in a rural area of Appalachia would be more likely to self-identify as “Appalachian.” The results will provide new insight into the Appalachian culture and may be important for future health promotion programs in this region. METHODS Study Design This report uses data collected as part of a larger study, the Community Awareness Resources and Education (CARE) study, which has been described in great detail elsewhere (Hade et al., 2009). Briefly, it was designed to examine and test strategies to address elevated cervical cancer rates in Ohio Appalachia (Ohio Cancer Incidence Surveillance System, 2003). A total of 22 health clinics in Ohio Appalachia were approached to participate, of which 14 (63.6%) agreed. A monthly random sample of women from these clinics was selected and medical records were reviewed to determine eligibility. In order to be eligible, a woman had to be at least 18 years of age, reside in an Ohio Appalachian county, not pregnant, have no history of hysterectomy or invasive cervical cancer, and seen in that clinic within the last two years. If a woman was eligible and agreed to participate, the baseline survey was conducted, usually at the participant’s home. The current analysis analyzed data collected from the cross-sectional baseline survey. The study was approved by the Institutional Review Boards at The Ohio State University, the University of Michigan, and the Centers for Disease Control and Prevention (CDC). Measures The outcome variable was an Appalachian self-identity question, “Do you consider yourself to be Appalachian?” A participant could answer “Yes,” “No,” or “Don’t know” to this item, but responses were dichotomized into yes and no/don’t know for this analysis. Independent variables examined for possible associations with the outcome were selected based on their relationship to images about Appalachia or on results from quantitative and qualitative research addressing Appalachian self-identity. For example, locus of health control, trust in physicians, and living in a neighborhood with a name were examined because they relate to the Journal of Rural Community Psychology Volume E12 (1) images of fatalism, lack of trust in the healthcare system, and kinship associated with Appalachia. Independent variables were classified into larger categories based on the Social Determinants of Health Model (Marmont & Wilkinson, 1999), which was used as the theoretical framework for the CARE study. The categories included social and cultural factors, material factors, health and health behaviors, psychological factors, and county level factors (Tables 1 and 2). Existing instruments used in measuring independent variables included social cohesion level (Sampson, et al., 1997), Detroit Area Study Discrimination Questionnaire (DAS-DQ) (Taylor et al. 2004), Trust in Physician Scale (TPS) (Anderson & Dedrick, 1990), Center for Epidemiologic Studies Depression (CES-D) scale (Radloff, 1977), Beck Anxiety Inventory (BAI) (Beck & Steer 1990), Life Stressor Checklist-Revised (LSC-R) scale (Wolfe & Kimerling, 1997), Rapid Estimate of Adult Literacy in Medicine (REALM) (Davis, et al., 1993), and God Locus of Health Control (GLHC) measure (Wallston, et al., 1999). Additionally, the collective efficacy scale utilized was based on a similar concept used by Sampson et al. (Sampson, et al., 1997). County level factors were based on Census 2000 data. CARE region of residence was examined using the four previously explained CARE regions (Hade, et al., 2009). Region 1 is in the northeast part of Ohio Appalachia, Region 2 is located in the central part of Ohio Appalachia, Region 3 is in southern Ohio Appalachia, and Region 4 is the southwest part of Ohio Appalachia. Data Analysis Descriptive statistics were initially used to provide overall sample characteristics. A forward selection model-building procedure was then utilized to construct a multivariable logistic regression model with Appalachian self-identity as the outcome variable. Because women were clustered within clinics, mixed logistic regression models were fit to the data. Univariate models were first constructed and the independent variable producing the smallest p-value was selected into the model, provided the p-value was less than p=0.05. The process was repeated in a forward stepwise fashion until no variables were statistically significant at the p=0.05 level. Following the forward selection procedure, potential confounders and interactions between variables in the model were examined. The resulting final model was used to produce adjusted odds ratios (OR) and 95% confidence intervals (CI). Model fit and discrimination of the final model were assessed using the Hosmer-Lemeshow test and the area under the receiver operating characteristic (ROC) curve. A standard logistic regression model comprised of the same variables as the final mixed model was used for these features since they were not available for mixed models. The analysis was performed using SAS Version 9.1 (SAS Institute, 2002-2003). RESULTS A total of 571 women in the CARE study completed the baseline survey. Once potential eligible participants (n=801) were able to be contacted, the participation rate was 71.3%. Most participants were white (94.6%), married or living as married (61.2%), and had more than a high school education (55.9%). Of the 571 women, 201 (35.2%) identified themselves as “Appalachian,” 329 indicated they did not consider themselves “Appalachian,” and 41 responded “don’t know.” Journal of Rural Community Psychology Volume E12 (1) Descriptive statistics for all categorical and continuous variables by Appalachian self-identity status are listed in Tables 1 and 2, respectively. Univariate models indicated that women who were older, white, more educated, associated with a religion or had someone in their household associated with a religion, had both parents from an Appalachian county, lived in a neighborhood with a name, had lived in the same county their entire life, had lived in their current county for a longer period of time, lived in a rural county, had private health insurance, had lower depression scores, were not current smokers, perceived lower levels of social cohesion in their neighborhood, lived in counties with higher unemployment among women, and lived in less affluent counties had significantly higher odds (p<0.05) of self-identifying as “Appalachian.” Table 1: Frequencies and univariate results of categorical independent variables for Appalachian self-identity (n=571) (Note: DK=don’t know, OR=odds ratio) Appalachian Identity Yes No / DK OR p-value n (%) n (%) SOCIAL AND CULTURAL FACTORS Age (n=571) 0.0389 18-30 50 (27.2) 134 (72.8) 1.00 31-50 95 (36.1) 168 (63.9) 1.28 51+ 56 (45.2) 68 (54.8) 1.98 Race (n=571) 0.0144 White 197 (36.5) 343 (63.5) 1.00 Non-white 4 (12.9) 27 (87.1) 0.25 Education (n=571) 0.0205 HS degree or less 73 (29.0) 179 (71.0) 1.00 More than HS degree 128 (40.1) 191 (59.9) 1.57 Marital Status (n=570) 0.7117 Married / Living as Married 128 (36.7) 221 (63.3) 1.00 Never Married 30 (30.0) 70 (70.0) 1.24 Divorced / Widowed / Separated 43 (35.5) 78 (64.5) 0.98 Employment Status (n=570) 0.4010 Full-time / Part-time 139 (37.7) 230 (62.3) 1.00 Unemployed / Disabled 23 (29.9) 54 (70.1) 0.80 Other Employment 38 (30.6) 86 (69.4) 0.75 Occupation Type (n=569) 0.0953 Unskilled Labor / Never Employed / Other 84 (29.7) 199 (70.3) 1.00 Professional / Skilled Labor 116 (40.6) 170 (59.4) 1.38 Journal of Rural Community Psychology Volume E12 (1) Current Housing Status (n=570) 0.0778 Place Owned or Being Bought by Someone in Household 149 (38.3) 240 (61.7) 1.00 All Other (Rented / Occupied Without Payment / Other) 51 (28.2) 130 (71.8) 0.69 Consistent Residence (n=570) 0.0006 Participant Has Not Lived in Same County All Their Life 86 (28.7) 214 (71.3) 1.00 Participant Lived in Same County All Their Life 114 (42.2) 156 (57.8) 1.93 Years Living in Current County (n=571) <0.0001 26 Years (median) or Less 76 (25.6) 221 (74.4) 1.00 27 Years or More 125 (45.6) 149 (54.4) 2.51 Residence Location at Age 14 (n=570) 0.0606 In a Town or City 99 (31.2) 218 (68.8) 1.00 On a Farm / In the Country But Not on a Farm 101 (39.9) 152 (60.1) 1.42 County of Residence Type (n=571) 0.0413 Urban 53 (24.0) 168 (76.0) 1.00 Rural 148 (42.3) 202 (57.7) 2.01 Religion (n=570) 0.0001 Does Not Belong to Church / Religion 57 (25.0) 171 (75.0) 1.00 Belongs to Church / Religion 143 (41.8) 199 (58.2) 2.17 Education Level of Parents (n=503) 0.5955 Neither or Only 1 Parent Has HS Degree 69 (37.1) 117 (62.9) 1.00 Both Parents Have HS Degree or More 107 (33.8) 210 (66.2) 0.90 County Type Where Parents Were Raised (n=535) 0.0028 Neither or Only 1 Parent is From an Appalachian County 61 (26.9) 166 (73.1) 1.00 Both Parents are From an Appalachian County 133 (43.2) 175 (56.8) 1.81 Number of Siblings (n=570) 0.0965 2 (median) or Less 117 (37.1) 198 (62.9) 1.00 3 or More 83 (32.5) 172 (67.5) 0.73 Total Current Household Members (n=570) 0.2220 3 (median) or Less 132 (37.3) 222 (62.7) 1.00 4 or More 68 (31.5) 148 (68.5) 0.79 Number of Neighborhood Friends and Family Members (n=560) 0.2188 6 (median) or Less 100 (33.8) 196 (66.2) 1.00 7 or More 97 (36.7) 167 (63.3) 1.26 Neighborhood Name (n=564) 0.0320 Neighborhood Does Not Have a Name 109 (32.3) 228 (67.7) 1.00 Neighborhood Has a Name 90 (39.6) 137 (60.4) 1.51 Journal of Rural Community Psychology Volume E12 (1) CARE Region of Residence (n=571) 0.0517 Region 1 (Northeast Region) 30 (19.1) 127 (80.9) 1.00 Region 2 (Central Region) 54 (38.0) 88 (62.0) 1.87 Region 3 (South Region) 74 (44.0) 94 (56.0) 4.02 Region 4 (Southwest Region) 43 (41.3) 61 (58.7) 2.75 Health Insurance Status (n=570) 0.0065 Medicaid or Medicare Only 31 (23.7) 100 (76.3) 1.00 No Coverage (Self-Pay) 24 (27.0) 65 (73.0) 1.32 Private Insurance (Job or Purchased) 145 (41.4) 205 (58.6) 2.15 Usual Source of Medical Care (n=570) 0.4906 Doctor's Office 154 (35.3) 282 (64.7) 1.00 All Other (Health Clinic / ER / Doesn't Go for Healthcare) 46 (34.3) 88 (65.7) 1.19 MATERIAL FACTORS Poverty Income Ratio (n=543) 0.2404 Less than 2.00 93 (30.7) 210 (69.3) 1.00 2.00 and Higher 95 (39.6) 145 (60.4) 1.26 Current Satisfaction with Finances (n=570) 0.7409 Not at All Satisfied 60 (31.9) 128 (68.1) 1.00 Pretty Well Satisfied / More or Less Satisfied 140 (36.6) 242 (63.4) 1.07 Finances Compared with American Families in General (n=568) 0.6112 Far Below Average / Below Average 79 (35.1) 146 (64.9) 1.00 Average 86 (33.3) 172 (66.7) 0.90 Far Above Average / Above Average 35 (41.2) 50 (58.8) 1.17 Finances Compared to Parents at the Same Age (n=554) 0.4464 Worse Off / About the Same 73 (31.7) 157 (68.3) 1.00 Better Off 123 (38.0) 201 (62.0) 1.16 PSYCHOLOGICAL Depression (n=570) 0.0235 CES-D Score of 15 and Lower 151 (38.3) 243 (61.7) 1.00 CES-D Score of 16 and Higher 49 (27.8) 127 (72.2) 0.62 Anxiety (n=481) 0.5210 Beck Anxiety Inventory Score of 15 and Lower 129 (37.0) 220 (63.0) 1.00 Beck Anxiety Inventory Score of 16 and Higher 44 (33.3) 88 (66.7) 0.86 Journal of Rural Community Psychology Volume E12 (1) HEALTH AND HEALTH BEHAVIORS Alcoholic Beverages in the Last Month (n=570) 0.1311 4 or Less Total Drinks (Up to 1 Drink per Week) 164 (37.3) 276 (62.7) 1.00 5 or More Total Drinks (More than 1 Drink per Week) 36 (27.7) 94 (72.3) 0.70 Smoking Status (n=570) 0.0374 Never Smoker 112 (37.7) 185 (62.3) 1.00 Former Smoker 47 (40.2) 70 (59.8) 1.36 Current Smoker 41 (26.3) 115 (73.7) 0.67 Table 2: Descriptive statistics and univariate results of continuous independent variables for Appalachian self-identity (n=571) (Note. * mean + standard deviation, DK=don’t know, OR=odds ratio) Appalachian = Yes* Appalachian = No/DK* OR p-value Collective Efficacy (n=468) 12.84 + 4.67 12.81 + 4.32 0.99 0.6941 Social Cohesion (n=570) 11.54 + 3.82 12.62 + 4.00 0.94 0.0061 Discrimination (DAS-DQ) (n=570) 8.85 + 7.82 9.28 + 7.55 1.00 0.7370 Major Life Events (LSC-R) (n=519) 6.84 + 4.66 7.15 + 4.08 0.99 0.6941 Physical Activity (total weekly METs) (n=415) 12.05 + 12.90 12.74 + 15.68 1.00 0.7813 Locus of Health Control (GLHC) (n=442) 20.15 + 7.29 19.01 + 7.80 1.02 0.2491 Health Literacy (REALM) (n=570) 65.07 + 3.52 64.04 + 6.13 1.05 0.0879 Trust in Physician (TPS) (n=443) 40.50 + 7.32 40.97 + 7.49 0.98 0.0858 Census Education Level (n=571) (% females with more than a high school degree) 33.26 + 3.63 33.61 + 3.69 0.99 0.7200 Census Unemployment Level (n=571) (% females unemployed in labor force) 6.44 + 1.34 5.96 + 1.22 1.24 0.0442 Census Poverty Income Ratio (n=571) (% population with PIR greater than 2.00) 63.03 + 4.66 64.86 + 4.35 0.93 0.0187 Census Age Distribution (n=571) (% females age 65 and older) 16.62 + 1.86 17.21 + 1.96 0.90 0.1501 Journal of Rural Community Psychology Volume E12 (1) Census Household Composition (n=571) (% households with female head of household and children less than 18 years of age) 6.49 + 0.74 6.22 + 0.69 1.00 0.9858 The final model (Table 3) included number of years living in their current county, religious association, county type where their parents were raised, living in a neighborhood with a name, health insurance type, CARE region of residence, and county of residence type. According to this model, a woman who had lived in her current county longer than the sample median number of years (26 years) had about twice the odds (OR=1.94, 95% CI=1.28–2.94) of identifying herself as “Appalachian” when compared to a woman who was at or below the sample median number of years, even after controlling for all other variables in the model. Compared to women who lived in the northeastern CARE Region 1, women who lived in the more southern CARE regions had at least twice the odds of self-identifying as “Appalachian” (Region 2 adjusted OR=2.61, 95% CI=1.37–4.97; Region 3 adjusted OR=4.40, 95% CI=2.33–8.30; Region 4 adjusted OR=2.12, 95% CI=1.01–4.46). Furthermore, women who lived in rural counties had more than twice the odds of self-identifying as “Appalachian” compared to women from urban counties (adjusted OR=2.27, 95% CI=1.37–3.76). Additional factors significantly associated with self-identifying as “Appalachian” included being associated with a religion or having someone in their household who was (adjusted OR=1.87, 95% CI=1.21–2.89), having both parents from an Appalachian county (adjusted OR=1.65, 95% CI=1.09–2.50), and living in a neighborhood with a name (adjusted OR=1.76, 95% CI=1.16– 2.65). Lastly, women with private health insurance had increased odds of self-identifying as “Appalachian” compared to women with only Medicare or Medicaid (adjusted OR=2.31, 95% CI=1.34–3.99). Tests for model fit and discrimination did not indicate any problems with the final model. Table 3: Final model for variables significantly associated with Appalachian self-identity (n=530) (Note: OR=odds ratio, CI=confidence interval) OR 95% CI p-value Years Living in Current County 0.0017 26 Years (median) or Less 1.00 -27 Years or More 1.94 1.28 2.94 Religion 0.0050 Does Not Belong to Church / Religion 1.00 -Belongs to Church / Religion 1.87 1.21 2.89 County Type Where Parents Were Raised 0.0188 Neither or Only 1 Parent is From an Appalachian County 1.00 -Both Parents are From an Appalachian County 1.65 1.09 2.50 Neighborhood Name 0.0076 Neighborhood Does Not Have a Name 1.00 -Neighborhood Has a Name 1.76 1.16 2.65 Journal of Rural Community Psychology Volume E12 (1) Health Insurance Status 0.0105 Medicaid or Medicare Only 1.00 -No Coverage (Self-Pay) 1.71 0.85 3.46 Private Insurance (Job or Purchased) 2.31 1.34 3.99 CARE Region of Residence 0.0001 Region 1 (Northeast Region) 1.00 -Region 2 (Central Region) 2.61 1.37 4.97 Region 3 (South Region) 4.40 2.33 8.30 Region 4 (Southwest Region) 2.12 1.01 4.46 County of Residence Type 0.0015 Urban 1.00 -Rural 2.27 1.37 3.76 The primary aim of this research was to provide a quantitative analysis of data collected from women residing in Ohio Appalachia to further explore Appalachian self-identity. Results indicate that a sizable proportion of women in Ohio Appalachian consider themselves “Appalachian,” and multiple factors are associated with Appalachian self-identity. Some of these factors support previous research and this report’s hypothesis, while other factors provide new insight into Appalachian self-identity. The overall prevalence of Appalachian self-identity was shown to be 35.2%, which is higher than what has been reported elsewhere (Miller, 1978; Obermiller, 1982; Philliber, 1983). The observed increase could be attributable to a few reasons. First, these previous studies were conducted among Appalachian migrants instead of current residents of the area. There has also been a recent increase in the word “Appalachia” in the names of businesses, agencies, initiatives, and projects in northern Appalachia (Tribe, 2006), which may have resulted in an elevated sense of Appalachian self-identity among its residents. The length of time a woman had spent in Appalachia and her location within the region were shown to have the strongest associations with Appalachian self-identity. This finding supports the thoughts expressed by one participant from a past town meeting in Ohio Appalachia, who indicated time spent in the region helps determine if one is a true “Appalachian” (Tribe, 1993). Residing in Appalachia for a longer period of time would provide more exposure to the region’s culture, which may increase the likelihood of identifying oneself as “Appalachian.” Women outside of the northeastern CARE Region 1 were more likely to classify themselves as “Appalachian,” and this may be due to the characteristics of the counties in that region. Region 1 is the closest of all the CARE regions to the more urban areas of northeast Ohio, such as Youngstown and Cleveland. The other CARE regions are not located as proximally to these non-Appalachian urban areas, so the residents may not have as much exposure to them. Cincinnati is located near Region 4, but it is an urban area that is extremely close to the edge of Ohio Appalachia. Therefore, it may affect Appalachian self-identity differently than the urban areas of northeast Ohio. The other CARE regions contain multiple counties that share borders with West Virginia or Kentucky, both prominent Appalachian states. CARE Region 1, in Journal of Rural Community Psychology Volume E12 (1) contrast, has only one county that shares a border with West Virginia, and it is not a large segment of the border relative to other counties. It is possible that the simple lack of geographic contact with these two states played a role in the lower percentage of women self-identifying as “Appalachian” in this northeastern CARE region. Living in a rural county was independently associated with self-identifying as “Appalachian,” so it appears to be another important location component in addition to one’s general region in Appalachia. The observed results concerning rurality, a popular Appalachian image, concur with previous research (Philliber, 1983; Behringer, et al., 2007; Tribe, 1993). When combining this with the apparent regional effect, it suggests that both the general Appalachian region in which a woman lives and the characteristics (urban or rural) of that woman’s more immediate surroundings are important in determining Appalachian self-identity status. Results also indicate that religion, family, and community may play a role in Appalachian selfidentity. This coincides well with popular Appalachian images (religiosity, familism), as well as themes from previous Appalachian self-identity research (Coyne et al. 2006, Behringer, et al., 2007; Keefe, 2000; Tribe, 1993). Having both parents from an Appalachian county likely provides individuals with exposure to the Appalachian culture, its beliefs, and its norms at an early age. Thus, there is probably some degree of socialization that occurs among children raised in such households. Appalachian mountain religion has been described as a religious tradition with diverse worship practices, belief systems, religious experiences, and church traditions stemming from a mostly Christian perspective (McCauley, 1995). It was found to be important an important dimension of the Appalachian culture by Ford (1962), so it may be one characteristic thought to define being “Appalachian” by its residents. Other strong Appalachian images and characteristics, such as poverty, lack of education, unskilled occupation, and unemployment, were not associated with whether women selfidentified as “Appalachian.” In fact, higher income, increased education levels, being employed full-time or part-time, and having a professional or skilled occupation suggested an increase in the odds of self-identifying as “Appalachian” at the univariate level. These results are surprising since lower socioeconomic status has been has been linked to Appalachian self-identity both quantitatively and qualitatively (Philliber, 1983; Coyne, et al., 2006; Tribe, 1993). One possible explanation for our findings may be that women with higher socioeconomic status, particularly increased education levels, were simply more aware that southeastern Ohio is part of the Appalachian region. Our results have important implications for future research. Mainly, they provide a description of Appalachian self-identity that can be used as a starting point for researching the effects of Appalachian self-identity on health and health behaviors. Among other ethnic groups, increased levels of ethnic self-identity have shown positive effects on health and health behaviors (Shrake & Rhee, 2004; Umana-Taylor & Updegraff, 2007; Brook, et al., 1998; Herd & Grube, 1996; Love, et al., 2006). If Appalachian self-identity is shown to have similar effects, our results could then be used to identify target populations for future health intervention studies. The results of this research also underscore the need for future research comparing the effects of Appalachian self-identity and rural identity on health. It would be of great interest to determine Appalachian self-identity affects health and health behaviors differently than rural identity. Journal of Rural Community Psychology Volume E12 (1) Advantages of this analysis are that it included a large sample size, it examined a wide range of variables for associations with Appalachian self-identity, and it utilized quantitative methods to analyze the data. There are also a few limitations to this research that warrant discussion. With the format of the Appalachian self-identity question, it was not possible to assess the strength to which someone identified as “Appalachian.” Since only women who seek healthcare regularly and reside in Ohio Appalachia were eligible for the CARE study, the generalizability of the results is not known. There was concern about spurious associations occurring due to the large number of independent variables selected. However, it was hoped the forward selection modelbuilding procedure diminished the chance of such an occurrence, and the small p-values in the final model provide at least some evidence against spurious associations. In summary, it appears Appalachian self-identity is associated with several factors, including anindividual’s location within the region, length of time spent in one’s current county, beingassociated with a religion, and having family ties to Appalachia. The results also suggest thatsome of the most popular Appalachian images and characteristics have no association with thisregional identity. The results are important because they provide insight into the prevalence ofAppalachian self-identity and further our knowledge about what it means to be “Appalachian.”If Appalachian self-identity is shown to affect health and health behaviors in future research, theresults presented here could then be applied in designing health intervention programs aimed atreducing health disparities among Appalachian residents. REFERENCESAnderson, L.A. & Dedrick, R.F. (1990). Development of the Trust in Physician scale: a measureto assess interpersonal trust in patient-physician relationships, Psychological Reports, 67, 3,1091-1100. Beck, A.T. & Steer, R.A. (1990). Beck Anxiety Inventory Manual. San Antonio, TX:Psychological Corporation. Behringer, B., Friedell, G.H., Dorgan, K.A., Hutson, S.P., Naney, C., Phillips, A., Krishnan, K.& Cantrell, E.S. (2007). Understanding the challenges of reducing cancer in Appalachia:addressing a place-based health disparity population, Californian Journal of Health Promotion,5, 40-49. Bernstein, D.A., Alison, C.S., Roy, E.J., Srull, T.K. & Wickens, C.D. (1994). Psychology.Princeton, NJ: Houghton Mifflin Co. Billings, D.B. (2006). Identity, central and southern Appalachian. In: Abramson, R. & Haskell, J.(eds) Encyclopedia of Appalachia. Knoxville, TN: The University of Tennessee Press. Brook, J.S., Balka, E.B., Brook, D.W., Win, P.T. & Gursen, M.D. (1998). Drug use amongAfrican Americans: ethnic identity as a protective factor, Psychological Reports, 83, 3, 1427-1446. Journal of Rural Community PsychologyVolume E12 (1) Coyne, C.A., Demian-Popescu, C. & Friend, D. (2006). Social and cultural factors influencinghealth in southern West Virginia: a qualitative study, Preventing Chronic Disease, 3, A124. Davis, T.C., Long, S.W., Jackson, R.H., Mayeaux, E.J., George, R.B., Murphy, P.W. & Crouch,M.A. (1993). Rapid Estimate of Adult Literacy in Medicine: a shortened screening instrument,Family Medicine, 25, 391-395. Ford, T.R. (ed) (1962). The Southern Appalachian Region: A Survey. Lexington, KY: Universityof Kentucky Press. Hade, E.M., Ferketich, A.F., Lehman, A.M., Paskett, E.D., Wewers, M.E., Ruffin, M., Tatum, C.& Lemeshow, S. (2009). Background and design of the Community Awareness Resources andEducation (CARE) Project: reducing cervical cancer in Appalachia. Working Paper. Harkins, A. (2004). Hillbilly: A Cultural History of an American Icon. Oxford: OxfordUniversity Press. Harney, W.W. (1873). A strange land and a peculiar people, Lippincott's Magazine, 12, 429-438. Herd, D. & Grube, J. (1996). Black identity and drinking in the US: a national study, Addiction,91, 845-857. Keefe, S.E. (2000). Mountain identity and the global society in a rural Appalachian county.Presented at the Center for Ethnicity and Gender in Appalachia's National Conference. March 5,2000, Huntington, WV. Kephart, H. (1913). Our Southern Highlanders: A Narrative of Adventure in the SouthernAppalachians and a Study of Life Among the Mountaineers. Knoxville, TN: University ofTennessee Press. Love, A.S., Yin, Z., Codina, E. & Zapata, J.T. (2006). Ethnic identity and risky health behaviorsin school-age Mexican-American children, Psychological Reports, 98, 735-744. Marmont, M. & Wilkinson, R. (eds.) (1999). Social Determinants of Health. Oxford: OxfordPress. McCauley, D.V. (1995). Appalachian Mountain Religion: A History. Chicago: University ofIllinois Press. Miller, T.R. (1978). Urban Appalachian ethnic identity: the current situation. In Weiland, S. &Obermiller, P.J. (eds.) Perspectives on Urban Appalachians. Cinicinnati, OH: Ohio UrbanAppalachian Awareness Project. Obermiller, P.J. (1982). Labeling Urban Appalachians: The Role of Stereotypes in the Formationof Ethnic Group Identity. Unpublished Dissertation, Union Graduate School. Journal of Rural Community PsychologyVolume E12 (1) Ohio Cancer Incidence Surveillance System (2003). Cervical Cancer in Ohio, 1996-2000.Columbus, OH: Ohio Department of Health. PARC (1964). Appalachia: A Report by the President's Appalachian Regional Commission.Washington DC: US Department of Commerce. Philliber, W.W. (1983) Correlates of Appalachian identification among Appalachian migrants. InBuxton, B.M., Crutchfield, M.L., Lightfoot, W.E., Stewart, J.P. (eds) The AppalachianExperience. Boone, NC: Appalachian Consortium Press. Radloff, L.S. (1977). The CES-D scale: a self-report depression scale for research in the generalpopulation, Applied Psychological Measurement, 1, 385-401. Sampson, R.J., Raudenbush, S.W. & Earls, F. (1997). Neighborhoods and violent crime: amultilevel study of collective efficacy, Science, 277, 918-924. SAS Institute, (2002-2003). SAS Version 9.1. Cary, NC: SAS Institute. Shrake, E.K. & Rhee, S. (2004). Ethnic identity as a predictor of problem behaviors amongKorean American adolescents, Adolescence, 39, 601-622. Taylor, T.R., Kamarck, T.W. & Shiffman, S. (2004). Validation of the Detroit Area StudyDiscrimination Scale in a community sample of older African American adults: the Pittsburghhealthy heart project, International Journal of Behavioral Medicine, 11, 88-94. Tribe, D.L. (2006). Identity, northern Appalachian. In: Abramson, R. & Haskell, J. (eds)Encyclopedia of Appalachia. Knoxville, TN: The University of Tennessee Press. Tribe, D.L. (1993). Contemporary images of Appalachian Ohio: the view from within, Journal ofAppalachian Studies, 26-33. Umana-Taylor, A.J. & Updegraff, K.A. (2007). Latino adolescents' mental health: exploring theinterrelations among discrimination, ethnic identity, cultural orientation, self-esteem, anddepressive symptoms, Journal of Adolescence, 30, 549-567. Wallston, K.A., Malcarne, V.L., Flores, L., Hansdottir, I., Smith, C.A., Stein, M.J., Weisman,M.H. & Clements, P.J. (1999). Does God determine your health? The God Locus of HealthControl scale, Cognitive Therapy and Research, 23, 131-142. Wolfe, J. & Kimerling, R. (1997). Gender issues in the assessment of posttraumatic stressdisorder. In: Assessing Psychological Trauma and PTSD. New York: Guilford Press. Journal of Rural Community PsychologyVolume E12 (1) ACKNOWLEDGEMENTSThis research was supported by National Institutes of Health (NIH) Grant 1P50CA105632, Grant GRT8230100from the National Cancer Institute, and the General Clinical Research Center at The Ohio State University,Grant UL1-RR025755 from the National Center of Research Resources of the NIH.