WU!!!MINI!WIWIHlWWWWNWHIM .~ . ‘ ( '“fi'fla ' L J ." fi’ ~ Date 07639 RARIES miix‘igyxm iiiii‘iiiii\\\\\\\\\\\\\\\\\\ This is to certify that the thesis entitled Black Political Representation on City Councils in Michigan Cities with Populations of 10,000 or More presented by Stacy Lynn Mooradian has been accepted towards fulfillment of the requirements for Sociology and M' A' degree in Urban Affairs :BH “CC-‘1) “\UXLK L ____.—__ " Major professor April 8, 1994 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Mlchigan State University PLACE ll RETURN BOXtomnwombehockomm yournoord. To AVOID FINES Mum onorbdoro dataduo. DATE DUE , DATE DUE DATE DUE MSU IoAn Afflnndvo ActioNEqal Opportunity Im Wt BLACK POLITICAL REPRESENTATION ON CITY COUNCILS IN MICHIGAN CITIES WITH POPULATIONS 0F 10,000 OR MORE BY Stacy Lynn Mooradian A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Sociology and Urban Affairs 1994 ABSTRACT BLACK POLITICAL REPRESENTATION ON CITY COUNCILS IN MICHIGAN CITIES WITH POPULATIONS OF 10,000 OR MORE BY Stacy Lynn Mooradian The purpose of this inquiry was to determine the extent to which cities' structural, demographic and socioeconomic variables predict election results as measured by the percent black on city council for Michigan cities in January 1990. Methods used in the study were one way analysis of variance (ANOVA) , Pearson product moment correlation analysis, and stepwise multiple regression analysis. The primary finding is that the percentage of blacks in the population and the type of election system are significant predictors of the percent black elected officials in Michigan cities with more than 10,000 in the population. However, even in Michigan cities with sizeable black populations, blacks were found to be underrepresented on city councils. This thesis is dedicated to the loving memory of my father Carl Edward Mooradian His respect for education, integrity, and individuality will always guide me. iii ACKNOWLEDGEMENTS I extend my most profound and sincere thanks to the very special people in my life who have loved and supported me throughout my education. Special thanks and love to my mother Kathryn Thomas Mooradian, my sister Wendy Anne Mooradian, and my brother Douglas Edward Mooradian. To my friends who provided me with laughter and perspective thank you Ronald Nerio, Russell Labosky, Maura Huntz, and Karyn Ventrilla. I am eternally grateful to Dr. Joshua Gisemba Bagaka's for his guidance and patience during the arduous process of data collection and analysis. You gave me countless hours of your valuable time and offered much appreciated encouragement. I shall always have.the‘deepest.respect and affection for you. Thank you to my committee members Dr. Brendan Mullen, Dr. Joe T. Darden, and Dr. Cliff Broman for your assistance and support throughout the preparation of this thesis. iv TABLE OF CONTENTS List of Tables . . . . . . . List of Figures . . . . . . . INTRODUCTION . . . . . . . STATEMENT OF THE PROBLEM . . . . Research Questions . . . . PREVIOUS RESEARCH . . . . . . Type of Election Syste . . . Percent Black in the Population Socioeconomic Characteristics . DATA AND METHOD . . . . . . . Data and Variables . . . . Data Analysis . . . . . . FINDINGS . . . . . . . . . CONCLUSIONS . . . . . . . . RECOMMENDATIONS FOR FUTURE RESEARCH APPENDIX A Correlation Coefficients for all Demographic, Structural, and Socioeconomic Characteristics of Michigan Cities with Populations of 10,000 or More . . . . . . . . BIBLIOGRAPHY . . . . . . . . Page vi 25 26 Table LIST OF TABLES Michigan Cities with at Least One Expected Black Elected Official (BEO), January 1990 Pearson Moment Correlation Analysis for the Relationship Between Characteristics of Cities and Black Representation on City Councils in Michigan, January 1990 (N= 84) Stepwise Regression Model for the Prediction of Percent Black Elected Officials on City Councils in Michigan, January 1990 (N= 84) vi Page 13 19 20 LIST OF FIGURES Figure 1 Black Representation Ratio by Percent Black Population, January 1990 2 Black Representation Ratio by Type of Election System, January 1990 vii Page 15 17 INTRODUCTION Since the passage of the Voting Rights Act of 1965, the political enfranchisement of blacks has been a topic of race research. Studies have focused on politics primarily because black progress, leadership and power has been consistently improving in this arena (Darden, 1984; Delaney, 1983; Williams, 1987). Blacks have realized steady nationwide gains over the last twenty years as the number of black elected public officials has risen from 1,469 in 1970 to 7,370 in 1990 (Joint Center for Political and Economic Studies, 1991). Although the number of black elected officials has improved fourfold over two decades, the inclusion of blacks in electoral politics continues to be minimal. To fully appreciate the racial disparity in political representation, the percentage of the black voting-age population must be compared with the percentage of black elected officials. Blacks comprise 11.1 percent of the nation's total voting age population and only 1.5 percent of the nation's electediofficials (Joint Center for Political and Economic Studies, 1991) . Furthermore, of the 7,370 black officials, half were elected to municipal office (Joint Center for Political and Economic Studies, 1991). In some cases, local level representation has resulted in the appropriation 1 2 of tangible benefits to blacks. Reported increases in black municipal employment, political appointments, governmental contracts for businesses, and fewer reported incidences of police brutality suggest that political representation is more than symbolic (Eisinger, 1982; Keech, 1968; Pohlmann, 1990). Nevertheless, blacks continue to make far fewer gains in state and federal politics (Joint Center for Political and Economic Studies, 1991). There are regional differences in the distribution of black elected officials across the United States. IMuch of the research on political underrepresentation of blacks has heretofore been limited to the South for two major reasons: the barriers, to registration. and 'voting' have been 'very pronounced and until 1990, the South's share of the nation's black elected officials trailed the South's share of the U.S. black population by almost 15 percent (Joint Center for Political and Economic Studies, 1991). The Midwest, as a region, has experienced the most.dramatic loss in the share of the nation's black elected officials. In 1970, 29 percent of the nation's black officials were elected in the Midwest compared. to only’ 18 percent in 1990 (Joint Center for Political and Economic Studies, 1991) Moreover, in recent years Midwestern cities have increasingly become the home of sizeable black populations who are inadequately represented in politics. Several cities, as a result, have been found in violation of the Voting Rights Act because their election systems serve to dilute the votes of minorities (Kopecky and Schrof, 1987). STATEMENT OF THE PROBLEM Michigan was one of ten states with the largest increases in black population between 1980 and 1990. The black population grew by 93,000 during the decade increasing the total to almost 1,292,000 or 14 percent in 1990 (U.S. Department of Commerce, 1991). More importantly, in 1990 the black voting age population in Michigan was 878,000 (13 percent.of the total voting-age population for the state) with approximately 650,000 or 74 percent of blacks registered to vote (Joint Center for Political and Economic Studies, 1991). As of January 1990, there were 19,923 elected officials in Michigan and only 309 or 1.6% of those officials were black. Of the 309 black elected officials, 34% were elected to serve at the municipal level. The purpose of this inquiry was to determine the extent to which the city population characteristics predict election results as measured by the percent black on city council for Michigan cities in January 1990. Typically, either structural and/or socioeconomic characteristics have been tested. For this study, the dependent variable is black political representation computed as the percentage of black city councilpersons. Predictors in the model include: percent blacks in the population, percent black in the voting-age population, type of election system, median family income, 4 median household income, percent black high school graduates‘, percent black with some collegez, percent black unemployment’, and percent black below poverty level“. a ns: 1) Do cities with at-large election systems have lower black political representation? 2) Is there a statistically significant relationship between socioeconomic characteristics of blacks and black political representation? 3) To what extent does the percent black in the city population predict the black political representation? PREVIOUS RESEARCH Prior to the Voting Rights Act of 1965, the explanations ‘Universe includes persons 25 years of age and over with a high school diploma or equivalency (U.S. Department of Commerce, 1993). 2Blacks in the population 25 years and over including those without degrees as well as those with associate's, bachelor's, master's, doctoral, and professional degrees (U.S. Department of Commerce, 1993). 3Includes only black civilian employment of persons 16 and over. Those categorized as "not in the labor force" are not included (U.S. Department of Commerce, 1993). ‘The average poverty threshold for a family of four persons was $12,674 in 1989. Poverty thresholds were applied on a national basis and were not adjusted for regional, state, or local variations in the cost of living (U.S. Department of Commerce, 1993). 5 for minority vote dilution could be traced to the discriminatory practices that impeded minorities from registration and balloting (Edwards, 1985; Engstrom, 1985; Butler, 1985; Karlan, 1989). Although black voter registration and turnout has drastically improved since 1965, there.are structura1.mechanisms still in place that dilute the votes of minorities. The Citizens' Commission on Civil Rights states: "a variety of devices in the electoral process operate to deny blacks equal access to the voting process" (1988:63- 64). Pohlmann has shown that the racial disparities in voter registration and turnout will persist without considerable reformation of the electoral process (1990) . The most complete and farreaching reform is automatic registration of all citizens on the day they attain the age of eligibility (Pohlmann, 1990). Pohlmann admits that this reform would be a.hard sell for conservatives, and.that universal registration may not necessarily encourage turnout as long as archaic voting procedures prevail (1990). Instead, partial reforms have been attempted in some cities. For instance, Detroit has instituted a partial reform of the city's electoral process by administering a program that promotes voter turnout of senior citizens. Every senior in the city's electorate receives an absentee ballot in the mail for each major city election (Webb, 1994). However, targeting only the senior electorate is problematic because it unfairly advantages an historically strong voting-bloc. Some 6 critics have argued that such reform is insufficient. Myrna ‘Webb, a black city' council candidate in Detroit's 1994 primary, Detroit's senior program is implicitly biased because it presumes seniors are the only voters disadvantaged by the present election process (1994). e ec 'o stem It has been well documented that the type of election system may often be a structural impediment to the election of minority candidates (Teasley, 1987; Kopecky and Schrof, 1987; Parker, 1984). Marable writes, "In multicandidate, citywide elections, in which minority constituencies represent one- third of the total vote or less, it becomes virtually impossible to elect candidates who represent their interests” (1990:22). More than any other type of election scheme, at- large/citywide systems have been challenged in the courts and recognized as having "the probable effect of discriminating against minorities" (Note, 1982:144). In 1982, Congress amended section 2 of the Voting Rights Act and replaced the "intent test" established in City of Mobile v. Bolden with a "results test" (Blumstein, 1983). The amendment gave plaintiffs the. opportunity' to jpresent the court with. a "totality-of-circumstances test" as proof that an election system has had discriminatory effects on minorities in the political process (Teasley, 1987). When an at-large system is scrutinized by the court, a 7 totality-of-circumstances test allows several factors to be submitted as evidence of minority discrimination. Factors that. have commonly' been. submissible are ‘the history' of official discrimination, the extent of racially polarized voting, and the success of minorities in winning public office (Kopecky and Schrof, 1987; Parker, 1984) . In cases where at-large election systems have been found unlawfully discriminatory, single-member districting has almost exclusively been the judicial remedy (Note, 1982:144). Given residential segregation in. most U.S. municipalities, single-member districting plans have been criticized for their racial gerrymandering potential (Lyons and Jewell, 1988; Karlan, 1989; Parker, 1984). Lyons and Jewell surveyed (1988) American cities with populations of 50,000 to 1,000,000. All cities in the sample had some or all councilpersons elected from single-member wards or districts. The research found that generally, the cities in which black voters appear to be equitably represented have all or most black districts with very large black majorities (often over 75 percent). Therefore, the drawing of single-member districts must avoid either overconcentrating minorities in a few'districts or underconcentrating minorities across several districts whenever the minority community is geographically compact (Karlan, 1989; Lyons and Jewell, 1988). In the event that a minority population is not residentially concentrated, modifications of the traditional at-large multimember districts have been implemented as 8 alternatives to single-member districting (Note, 1982; Karlan, 1989). Karlan states that a favorable alternative is cumulative voting because voters are provided with as many votes as available seats, and may cast more than one vote for a single candidate (1989). "Cumulative voting, in short, allows voters to do more than choose among candidates, it allows them to express the intensity of their preferences as well" (Cole, Taebel and Engstrom, 1989:191). Clearly, neither single-member districts nor cumulative voting can guarantee the election of blacks at rates proportionate to their percentage in the population (Bullock, 1989) . However, when blacks are in the minority in the municipality, the cumulative voting system or the single- member district system would be preferable to an at-large system. e ce t ack the Po ul tio Previous research.has stressed the positive relationship between the election of black city councilpersons and the percentage of a city's black population (Darden, 1984; Engstrom and McDonald, 1982; MacManus, 1987). In fact, most blacks successfully elected to office represent majority black constituencies. (Engstrom and McDonald, 1981). Jeffrey R. Henig writes "the growth of black majorities in a number of large American cities is one of the signal demographic shifts of recent decades, and it has been paralleled by a striking increase in the number of black officials" (1993:544). "Thus 9 the geographic distribution of black elected officials has consistently followed closely the distribution of the black population" (Williams 1987:126). W In minority vote dilution court proceedings, the totality-of-circumstances test allows for socially depressed participation to be entered as evidence of the lingering effects of discrimination (Teasley, 1987) . Socially depressed participation of blacks is normally measured by socioeconomic variables such as levels of education, income, and employment (Teasley, 1987). Engstrom and MCDonald's frequently cited study (1982) compared. the structural and. socioeconomic explanations for differences in black representation on city councils in the South.and.Non-Southm ‘The finding of the study was that South and Non-South differences were explained more by electoral structure than socioeconomic disparities. However, Karnig and Welch have argued that socioeconomic disparities, especially lower income and education levels, provide a better explanation than regional location for the severe disproportionateness of black council inequities in the South (Karnig and Welch, 1980). 10 DATA AND METHOD a b es From the 1990 Census of Population and Housing (U.S. Department of Commerce, 1993) a total of 272 Michigan cities were identified. The 89 cities included in the study represent cities in Michigan with a population of 10,000 or more. Data on the characteristics of these cities by race‘was obtained from.the 1990 Census of Population.and.Housing' (U.S. Department.of Commerce, 1993). Additional data for this study was obtained from three sources: (1) the names of the black elected councilpersons sworn into office before February 1990, and the cities they represent were obtained from the 1990 National Roster of Black Elected Officials (Joint Center for Political and Economic Studies, 1991), (2) the total number of council seats for each city was obtained from the Directory of Michigan Municipal Officials (Michigan Municipal League, 1990), and (3) the type of election system implemented in the city was obtained through a telephone survey of the city clerks' offices in all 89 cities. Black representation ratio which measures the degree to which black representation on a city council equals the percent of the city's black population was computed for each of the 89 cities. ‘To calculate the black representation ratio for each city, the percentage of black elected officials on city council is divided by the black percentage of the city's population. If blacks have equal representation the 11 ratio will be 1.00. Underrepresentation is reflected in places ‘with a ratio of less than 1.00, whereas overrepresentation exists when the ratio is more than 1.00. Darden's nationwide survey of majority black places regards the representation ratio method to be reliable due to its ease of interpretation and greater utility for comparative purposes (Darden, 1984). Expected black elected officials was also found to be a useful parameter for judging how many blacks should be elected for blacks to have equal representation on city council. The number of expected black elected officials was computed by taking the product of the number of city council seats and percent of blacks in the population, rounded to the nearest whole number. Hence, whenever a city elects the expected number of black officials to city council, the black representation ratio will be about 1.00. ta sis In order to determine the extent to which the type of election system and the black population percentage in a city influence the black representation ratio, a breakdown of mean representation ratios by type and percentage was performed. One way analysis of variance (ANOVA) was used to test the statistical significance of the difference among these means. The Pearson product moment correlation analysis was used to examine the significance of the relationship between the cities' structural, demographic and socioeconomic variables 12 and the percent of black elected officials. The resulting correlation matrix was useful to detect multicollinearty among the predictors. Stepwise multiple regression analysis was utilized to determine the extent tO‘WhiCh cities' structural, demographic and socioeconomic variables predict the percent black elected officials. In stepwise multiple regression the best predictor is selected in Step 1. The predictor selected in Step 2 is the variable that has the highest correlation with black political representation when the predictor in Step 1 is partialed out. Each successive step progresses in the same way until no significant predictors are detected (Glass and Hopkins, 1984). FINDINGS Many cities never question the absence of minorities elected to city council because the minority constituency is relatively small. Table 1 provides meaningful descriptive information, and illuminates 26 of the 89 sample cities where at least one black was expected to be elected to city council. For example, Ann. Arbor ‘which. has an 8.9 percent black population is expected to have one black elected to city council for equal representation to be realized. The expected black elected official parameter has the same revelatory power in cities such as Detroit, where blacks are in the majority (76 percent of the total) and.have secured 13 coHuuHsmom gnu CH xoaam accouom\.uudom HHOCQOO no Hona=z\oun vo>h0cnovlodudm coHuousououmom xocam H nucom HHocsou uo uonasz.so«uuasaom on» CH xoaan uncouomloum nouoomxm _ .Avoodv IxuoHo huau no shabuau econmoaou use .AHmmH. uOHpsum 0Hsocoon use HuOHuHHom new Monaco usaou .AommHvosucoq HamHOHcsx chHnOH: .Anmmnv couoaaoo «o accsuudmon .a.D scum vochuno cunt undo: uonusa an vousmsoo unnumDOm on. n can: mm. n v n ouuuHuuu m.~o Hana vaquuHm Hm. p o a voxHa ¢.um uonuum coucom as. a m a oouanua s.ms uHouuoa mm. c v o ouuuHuuu «.op uuaoHcm coo-gun: mm. c a o couanua o.nm nouuxcH an. e v m uoHuu-Hv o.ov ucHHu no.H n n s uoHuu-Ho u.uv ouHucom mm. m n o Dunedin. v.0w sandman vs.H N v m ooudatud n.mn ocuoon no. N H o ouudHIud H.mn oosom uo>Hm oo. H o v ouuanun m.¢n guns xuo ~6.H n n o uuHuuuHu m.on coHan oo. u o o couanuc m.o« uHoHunusom om.H « n o vost o.on coo-gas: mH.H n n 0H uuHuuqu «.mu HucuHH-d» co. m o a counHuuu m.Hn usHssom oo. H o o nouuHuua «.mH oouasuH-n oh.H H a o cost o.mH .uHmam venue so. H H o vouHs o.mH ucHucuH nu. H H o oouanun o.sH acoscHo venom on. H H o uuHuu-Hv n.5H conga.» oo. H o o vost ~.oH guano oHuuam oo. H o m oouanun o.vH xoeuuuaus oo. H o o uuHuu-Hu s.HH .oHHz oo. H o o oouaHuun b.m .HHHm engage oo. H 0 6H uuHuu-Hu a.o wanna ace «canon a»... liven» nonvcusnom aoHuauaououaou .oau 6.63: HHoasoo scene-Ha on» aH xuan noun-nun copuouao no no HoaHn uuHo quIJMIIIIIIMmmm. uncounm. .oaaH Hausa.» .Aoau. HaHoHuuo noun-HI HoaHn cove-awn ago vac-a an aaHx .oHuHo aumHnoH: H oHnga 14 five out of the nine city council seats. It is expected, however, that there should.be seven black.elected.officials in Detroit for equal representation to occur. As a result, the black representation ratio in Detroit is .73 indicating underrepresentation. The mean.black.representation ratio is a mere .20 for the 88} sample cities. For the 88 cities, the percent black in the population ranges from 0.0 percent in Alpena to 92.9 percent in Highland Park. This severe underrepresentation can be attributed to the fact that in January 1990, Michigan cities of 10,000 or more had no black city councilpersons in any of the places where the black population was less than 17 percent (Table 1). Figure 1 shows that there is a positive relationship between. percent black. in. the jpopulation and. mean black representation ratio in Michigan cities. When the black percent of the population increases frem 10 to 50 percent, there is a steady increase in the mean representation ratio. Conversely, when blacks constitute a majority of the city's population“ (more than 50 percent), the mean representation ratio decreases. This finding suggests there is something other than percent black in the population predicting black ‘5 Black representation.ratio could.not be calculated for the City of Alpena due to zero percent black in the population. ° Sixty-five percent of Michigan's 1990 black population resides in the five majority black places (U.S. Department of Commerce, 1991). 15 Figure 1 Black Representation Ratio by Percent Black Population, January 1990 1 2Mean Black Representation Ratio 0.959 - 100035493956 20.00%-29.9996 Egg 3000%-5000% More than 50.00% 96 Black in Population 16 political representation in local government. The differential effects of three electoral schemes on black representation ratios is presented in Figure 2. An at- large election system is clearly most disadvantageous to black candidates running in city elections. Unfortunately, sixty- nine of the 89 cities in Michigan with a population over 10,000 have adopted at-large. electoral schemes. It is important to note that four of the five cities in the study having both majority black populations (Table 1) and black political underrepresentation on city council, also have at- large elections. Generally, it is in Michigan cities characterized by sizeable black populations and black political underrepresentation concurrently that at-large systems have had the most dilutionary effect on black votes. District elections stimulate the election of more blacks than either of the other two systems, yet mixed election systems realize a mean representation ratio of only 0.03 less than district systems (see Figure 2). Further, the mean representation ratio for district and mixed systems is about 5 times greater than the mean representation ratio for at- large systems (Figure 2). This finding suggests that improved black political representation in Michigan could be dramatically improved by the replacement of at-large systems with either district or mixed electoral schemes. The Pearson.product.moment correlation analysis revealed statistically significant relationships between all city structural, socioeconomic, and demographic variables and the _______ 876543210 00000000 18 percent black elected officials. The results are presented in Table 2. With the exception of median household and family incomes, the correlation was positive. Median household and family incomes were negatively correlated with the percent black on city council, which implies that as median incomes for cities increase, the black representation decreases. Since it is known that the strongest positive correlation is between percent black in the total population and black political representation (r = 0.897, P < 0.05), one may deduce that cities with considerable black populations have lower median incomes. Three pairs of predictors were found to be strongly correlated (see Appendix A). These pairs were: percent black in voting-age population and percent black in total population ( r = .999, P < 0.05); percent black high school graduates and percent black with some college ( r = .999, P < 0.05); and median household income and median family income ( r = .977, P < 0.05). For each pair, the predictor with a stronger relationship with the percent black elected officials was allowed to be entered in the regression model. Consequently, percent black in voting-age population, percent black with some college, and median household income were eliminated from the regression model. Stepwise regression analysis selected the percentage of blacks in the population and the type of election system to be significant predictors of the percent black elected officials. Table 3 presents the results of the stepwise regression model. .Ho>oH no.0 on» on uaaoHanuHm « .AvmmHv mxuoao huflo no nuo>u9u oconmoaou pad .AHmmHv noflvsum Udaogoom use Hoowuflaom haw Hougoo HQHOB .AommHv cannon Homwowcsx savanna: .AmmmHv oouoaaoo no usoauuomon .m.D ”mmumDom 19 «000.0 0mm.0l QEOOGH >AHadh scape: «000.0 MHm.0I oaoosH vaonouaom Guava: «mmo.0 0HN.0 mucom Hwocsoo mo Honfisz «m00.0 mHm.0 oooaaoo meow saws xooam ucoouom «N00.0 mmm.0 mouaspouo Hoonom swam xooam unmouom .ooo.o eme.o unassonsocn xouHm unmouom «000.0 0Hm.0 Ho>oa uuuo>om 3oaom xooam usoouom «000.0 mm0.0 owucm soauuugououmom xoaam «000.0 mmm.0 nowadasmom omdlmcwuo> cw Roman uncouom «000.0 500.0 cowuoasmom Hopes cw xouam usoonom Isadblm u cows-«uouoouoau .A¢o luv coon hugsuuh .gcm«£0dl n4 cadussou hudu no coauofisououmom locum Una nowudu «o cuwucwuouocuonu nonluon manageducdom on» How ewnhddfld nowalaouuou unolox nacuoom N OHA‘H 20 .Ho>oa 00.0 on» no uncowmwsuflm «« psoaoflmmooo cowuuouuou pouwpuopccuuso « .Av00H0 uxuoao 0900 no uhobusu ononmoaou can .AH00H0 uowpsum OHEosoon can HuOHpHHom no“ umucoo uaHos .AommH. oamuoq HamHoHcsz_aa0Haon .Amach oouossoo no ucosuunaoo .m.0 "0000000 00.0 000.0 00.0 ucofihoaqaoco gonam usoouom 000.0 00.0 I ofioocH haflfidh Guava: Ho>oq huuo>om 000.0 00.0 soaom msfl>flq xocam ucouuom sandstone 000.0 00.0: Hoonom £000 xoaam usaouom ««000.0 m0.Hu m¢.vu Ho.o smoumm coHuoon «0 «may a sowuoasmom ««000.0 00.0H 00.0 00.0 map a“ xouam usoouom H osdu>lm cadobla «usowowuuooo ouosva a uoaowvoum noun ecu-sane .A00 I30 000H huuaach .uoDHAOHI aw undossou hOdU go cacwuwuuo UO&OOHH sedan ugoouom mo noduuwvoum Gnu new Home: scan-oumou oewlnouu n Danna 21 The percentage of blacks in the population was entered in the regression.equation at Step 1 (B = 0.90, P < 0.05) and the predictor singly accounted for about 80.0 of the variance in the outcome. Addition of the type of election system (B = -0.10, P < 0.05) into the equation increased.the variance explained by the two predictions to over 81 percent. No other predictors were entered into the regression equation. The regression coefficients of the two predictors indicates that: Increasing the percentage of blacks in the population by one percent will increase the percent of the black elected officials by 0.85 percent; and changing the system of election from at-large to district or mixed will increase the percentage of black elected officials by about 4.5 percent. Other socioeconomic characteristics of blacks in the city like education, poverty status, income, and employment have a negligible effect on the percentage of black elected officials to city councils. CONCLUSIONS The general conclusion of this study is that blacks as a whole were politically underrepresented on city councils in Michigan cities with a population of 10,000 or more in 1990. The theoretical assumption guiding the research was that blacks are proportionately represented whenever the percent of blacks on city council equals the percent of blacks in the city population . In other words , pol itical 22 underrepresentation of blacks is realized when the black representation ratio equals 1.00. Because the percent black in the population predicts about 80 percent of the variance in the regression equation, there is justification for the racial polarization model, as posited by Bullock, (1984). "The racial polarization model assumes that voters support only candidates with whom they share a racial tie" (Bullock, 1984:239). While the present study does not submit that all black officials were elected to city council solely by black voters, the strength of percent black population as a predictor of percent black elected officials establishes that the racial polarization model could be relevant to Michigan cities. This study did not focus on models of voting behavior by race because analyzing the political preferences of voters minimizes the importance of structural impediments to minority representation in electoral politics. For instance, the type of election system predicts variance in the regression model, but if additional data from political opinion polls were added to the model, the discriminatory effect of at-large elections might go unnoticed. The type of election system was essential for predicting the variance in black representation in.majority black places where blacks are underrepresented on city council. The increase of Midwestern cities with at-large systems found in violation of the Voting Rights Act should serve as a warning to those Michigan cities where black political 23 representation is low. Cities with at-large systems such as Southfield and Oak Park which have 0.00 black representation ratios and percent black populations of 28.9 and 34.5 respectively, should.consider alternative voting schemes. 131 addition, cities with consistently low black political representation on city council need to be investigated, and should not be spared public scrutiny. The urban sociological relevance of this study is that blacks have realized gains in local politics since 1970, but today's gains are primarily attributable to demographics and geographic concentration. For example, blacks constitute 14 percent of Michigan's total population, but 65 percent of those blacks live in five majority black cities (U.S. Department of Commerce, 1991). In addition, black political representation on city councils in Michigan cities is positively related to the percent black in the population. This is not to say that the ratio of percent black in the population to percent black on city council is one to one for all cities. Clearly, blacks and other minorities continue to be disadvantaged. in. politics, but the ability to elect candidates of choice becomes easier at the municipal level in places with sizeable minority populations. Thus, minorities have made more political progress than economic progress, but with few exceptions, the integration of blacks in politics has continued.to‘occur at the local level where power and.prestige is recognizably lower. 24 RECOMMENDATIONS FOR FUTURE RESEARCH Although the results of this study found socioeconomic variables to be weak in predicting percent black elected officials, the intersections of class and race in the political sphere deserve attention in future studies. Despite improved black political enfranchisement, Williams concludes that "thus far black electoral power has been unable to change significantly one major component of black subordination, economic deprivation, which still severely limits the rights won.by the civil rights movement" (1987:97). Hence, a missing piece of data in this study was median family or household income by race. The gap between black and non-black median incomes across cities may explain differences in black political representation on city councils. The role of class in the political mobilization.of minorities deserves research consideration. Moreover, future research might also include a longitudinal analysis of the effects that increased black political participation has had on the material gains made by blacks. It should be understood, however, that black elected officials on city council can only influence municipal decisions and policy. Although economic gains or losses of blacks in any given city are not mediated solely by local governmental units, improvements in the municipal employment of blacks, treatment of blacks by the police, and city maintenance of predominately black neighborhoods are examples of variables that warrant future study. APPENDIX 25 .nmmH .ouuoaaoo no uaoauuamoa .m.: » ”Du—90m s¢uioz~ 0h1mm OOuJQEou on uoccau «covuvyuoou I 0* DouCHLQ m. 000. no H00 0 whbm. an. nu Ham 0 o®N—.i «mm 0000.— 000. no .00 0 mmmw.| 05.. no H00 0 00¢..- —vH. n0 A00 0 omn..u 000. no .00 0 0000.: 000. no H00 0 wNHm.i 000. no A00 0 NNNm.I 000. «0 H00 0 VBmQ.t 00D. ua Ham 0 mem.u mmraoza 000. na A00 0 nmN0.o 000. no A00 0 o—mm. 000. no .00 0 nu>oax0 5?». "a 0 H00 mmmr.o 0—0. In A00 homm. mn_. u “mm hovr.l A50 mmm—. Ram 0000.— 000. n Ame 00mm. Hv0 050p. 3001x000 mm—. no 0 V0 H_m. no 0 «Hp. no A00 0 005..u 0H0. «a H00 0 .NmN. .v.. "a A00 _ «00... mm". ua A50 0 VNmH. 000. no 0 ”00 00mm. wrixoom 000. ns avm 0 mmom.u .00. an new 0 mn—N. 000. no Hem _ 0000.- 000. no How 0 0000. 1.0. "a Ham 0 0500. map. no “cm 0 alzalxm A000 oo_.nu-~ \ “menace \ acoeu.ttooo. ooo. nu Ham 0 omnm.s 000. an Ame 0 ovmm. 000. u Amm mmmm.a vQ 000. A00 ompm. II we 000. ”mm omov. ll wCL 000. A00 050v. ll Va 000. u H00 0.0a. V0 000. «mm mumm. ll VQ Ram 0000.— 000. «a How 0 pnvo. 000. um H00 0 ommm. anxuasm 000. n Amm moww.n A00 mNNN. fine. "a 0 000. "a 0 “mm whmv.l 000. u A00 mama. Hmo. n A50 000—. 000. n “no Nwmp. —00. u Ava noun. VG 000. A00 mwmm. Va 000. A00 ”MVQ. VII Amm 0000.— 000. n Ram 0000. O~h(¢xm a 0 000. an Amm 0 mumm.u 000. "a H00 . 000v. 000. no How 0 MNHQ. 000. "a Ava . mvmv. 000. no H00 0 00mm. 000. nd A00 0 ommm. 000. no A00 0 memo. . no Amm 0 0000., aI0.Ix0 zauluzn mhdwm umxuuza an>oax0 4001x000 mrlxmom auzntxm enema auxu<50 O~h