Research in Brief 593 TABLE I DiKriminant Analysis Summary Of the Principal Variables in New Mexico Election Study Variable F to Enter Entered Removed or Remove Radio Political Advertising 15.3930* Campaign Spending 8,8462* Newspaper Endorxment 2.SI20 Newspaper Coverage .4246 City Unemployment .8SS0 Voter Turnout .9395 Newspaper Advertising .3955 Incumbency .6712 andidate's Sex .0000 *p < .01. using Wilks-lambda test. high degree of separation as indicated by the fmal Wilks' lambda score of .5321 and a canonical correlation of .68 for the only discriminant function between the two groups (winners vs. losers, X2=I1.67. d.f. 3,p<.0l). Discussion The question of what effect certain media variables might have on the outcome of local elections is an important research question. Students of communi- cation theory should be concerned about such questions since the rise of the electronic journalism age over the past 25 years of national political races. But at the local level of campaigning, is radio political advertising elTective? Although scores of studies have been completed on media effects at the state and national level, few studies have focused attention on individual variables or sets of variables that have acceptable reliability in Younf Yun Kim. "Towiid an interactive theory of Gonununicalion-aocultiinuon,* pp, 436-453. Dan Nimmo, ed. Communkatkm Yearbook i, 1979, > WilUam Petenen. "Wtao^ Whit: 1790-1980.' 7V mbon QiMr<er(y. 9:97-120. ' Such u: Richard J. Oucnbcrg. The Social lntegralioD and Adjustmem of PoMwar Immigrant! in Montical and Toronto," Canadian Review i4 Socktlogy and Anlhropohgy. 1:202-214 (1964); Young Yun Kim. "X^mmunication Patterns of Foreign Imougraott in the Procets of Acculturation." Human CommioucaKon Reitanh. 4:<6-77 (1977); Melvin DeFkur and Chang-Soo Cho, 'AuimiUtion of japaneie-born women in an American city,* Sociat Problems, I9S7; Andrew M. Greeley, "An Allemativt Penpective for Studying American Ethnicity," in EthiticUy in the United States (New York: John Wiley * Soni, 1974); S. Alexander Weinstock, "Sone Faaon That Retaid or Aocelerate the Rale of Acculturation," Human (Footnote continued) predicting which candidate might win mayoral or local office. This study indicates politicans cam- paigning at the local level of government should take radio political advertising into account in their quest for public ofTice. It seems quite apparent that 11 winning mayoral candidates in 1986 New Mexico mayoral elections paid attention to radio political advertising in their media mix. Ethnic Concentration as Predictor of Media Use By Pamela J. Shoemaker, Stephen D. Reese, Wayne Danielson and Kenneth Hsu • Acculturation has been defined as an ethnic individual becoming acquainted with and adopting the "norms and values of salient reference groups."' What makes a reference group "salient"? Many studies assume that the mainstream culture is the salient one, and that, for example, Hispanics will acculturate to Anglo traditions. Yet this does not always happen. Ethnic differences in the United States, far from fading away in this century as some social theorists predicted, have endured and even increased.^ This study examines one important factor which may limit the salience of the host culture and incentives to acculturate to it and which may increase the salience of the ethnic culture — the concentration of ethnics in a community. Most studies of acculturation see acculturation as influenced by individual- level variables such as education, media use or length of time in the host country.^ ^ This research was funded by a grant to the flnt three authors from the Gannett Foundation. An earlier version of this paper was presented to the Intercultural/Development Communication Division of the International Communication Association, Chicago, May 1986. Pamela J, Shoemaker and Stephen D. Reese are assistant professors; Wayne Danielson is a profeuor; and Kenneth Hsu is a doctoral sttident in the Univtnity of Texas at Austin, Department of Joumalism.