ll 11 . o C ...1.1 33...... .........n........_.1....d1..: “ax... ....m . _ ... .0 . . .0 _. ., . 0 .L . .1 c . .. . .. . s u .. .. ....0..... o 00 ...» 00000.. .0.. 0. . ”.....0..4_0 ....u. . ... ......H ... u". 4.loo..¢a¢.0000 . .00.“ ....”ovro .0. 0.oo./.0. 0. 3/0.. 0" . . . . . . . . .. . . . .. . . . . . _ . . . . I . _ . ..-I .- ll. . .. o.. . . . . to... .... . . .. . .. . . . .0. . . .. . ... . ... . . _ . . . ... ...... Pa... .... .....u.... .. ....ono.o.o..... on oo.Ta../0.0./..o.00orvo ..uouaonvo 0o.o .00....”40 . o“ I. 0 ”WWW” . .o 0"” . . o . ..1 It... ...-.... .t.l... ... . .. . ..l” o . . . . . ... . . . 0. o. . 0... . ..... . .. ...} . .a.. .. . . ... ......... .... o . . . 0. ......q ... ...0 on. .... n. o .-.... .oo.o.0..0.00...d ofo.¢/of....ooof “Griffin... .0" ..Jo0.'00a ’finfi/a r a) . ... 0- .0.‘I..\v.o‘.0.l_ .0 .....H.. . . o o . .. I . . F I. 0 I .. .. ... .... . .....oo . . .....0 .. ... r . ...... o... ..70oao. .0.Io‘ . .010. o.... 00 .0.00'. a . ....0‘0... .. n . . . . . . -0 - . . .0 . a 0. .. . ... . .... . ... . .o....,.o........ ooav 0.0.0.. .000 0A9 9%. .0 no .....t 09 ‘o. . .. ... . .. . . . . ._ ... . . .. .... .. _. .. . . . . ...... ... ...ooo. « . .0 0. ....... ...... . . . . .... . . . - t 0 . s. . . . . . ...u .. . . . . .. . ”J . . ...- _ . . . . . . . . . ...... .. ..... .... . . . . ... .. .. .....o... ...... ......p.. 0o........ 0.0. . .....U. ......U...... .. (...... 2...... .W/.0 ....KF0..0W. .0 I ..0. . 0 0 . . . . ... . . .. . . . . .. .. ...... .. ...... .... .0 0. ...... . . .. 00 .. . 0060'. .00 . c. . ... . . a. . . . . ... . .. .... .... . o .o.... o.0....o..o.o.. 0.0....ouo. 0 .o . I. . .. ... .-. v. . . 1 . a 0 . . ... 0 .. ... . o .. m o a. o ..n... 00" O. n o o.¢ .G. .-. . . - . . .. 2... . . .. . ..z .... ..(l .... ......o... .. ...r. 0.9....405 .oo. 0180...!!!»019 “WWW.” _ . .-. 4. S . . o . ... v.0 . . .c . . .... . . .. {Noun/4.2.0.300 . . 0..If....' .0.0..0. 1.0.40”... 0... J?!m 0.. A . f . ~ . . . . . o . . .. . ... o J . ..0 . .. , . fr..00.o. 0! 001.0va 0000/" Iva-ogaf... .00. _ .. 0 . . A R. . I . . . .. . .... ... . . .0 I .1 .. . .a . 0... .. .. . 0.0.0 2”. .. If, lino-f..0r1 0 I .. I. I00f0.!h..o“ovhovo ... v o P . . .. .. . . o . 0 . . . o. . u . I L. o 050. . . . t . '0 _ 0 d {..QJI 0‘0 .00 o..... .lofW... a. 0‘. .fv .I. .. . .. . a . . . J ,. . . . _ .. . . a . . o . .c _ o... 0. .0 0’.:. ... ... 00.} 0.. 00 I... /.II 003...... . . ...l' . . . . . . a . . . oI. . o .. v .. . . L . ... n “V . .. . . . . . .. . .. _. . .4 .. .. I . £00..-. .F. .. 3...... 4.4 ... on I ....J... . . , . , I 0 . .. u o o 0 I 0 00 .v- o f . 0.. F 0‘" . .0 02b. . . . - n u. . . . .. 0 . 0 Of 0.1.." 1 0.0 _ Yb. . . 0 . . o n I I. I o .o . . .0 I i o 0f}. o (slug £000.! 0" 0h” “0900...? . . . . . . . . . ..r.u..n ... 0.. .. . :u. .. ......ot...0.....¢.o .... - . .a . . . . .. o. 0 p . r. .k. r... ,0. or... 0. .........AI‘007..00/' ‘0 . It.) .0400. v3.0..0. )0 1...!“ N. . . 0 . . ..0 Io. .10 . . .00 00.005...’ 7» . .'..Po¢ . e 0 p . 0... . . I 00000 .V'J .0 I. 0:. ’I.Q.¢ .Jr.ov.{ .ylr vylut"'0.0.m . U o . . . . . . . . . . n. 0... .. .o .0 ... I. . o/ 00.1... 01.1... 0.00. .l..0...00000....0 {from}? o 0. Jur..onlw?'l...o1l. . r . . . . . . . .... o. . . .0 . ..n .00.. 1......r.f. .r... n(..tl. Iva-Hahn. ..P..£.'f.l900l. ‘ r . . . .. . . . . . . . A .. ..0. 0.1000. .00.. 0 .2.va ..\ .r0..... I‘ll... . .... . .. . ... . I . ..l. .. . .. . . .. t , “00.“.o... I ... I. .7..000 00.4. v ‘ g . . . . . 0 ... . . . a o .0 a . ... 0 000 0-.0 o . 0" .0... f . 00. o [AWOVHHJ’06’+..’.V o"' I. ofiff... 00"? 03‘. v .. . c a .- . 0 . . . . I ... ..o... . . . .o 0.0 (.....f' aooortlr. o nlltw ’30500wr . I .. ...r . ...o.v._ . J. 0.0 .. 00(..0__I . 0 II 0? Ar - 90. I...o ..0 w . . . . - _ . . .. . . o . . 0. 0 f.. 90 \0 0 . J 041 0.:"....0.U.fllf..uu4....tr.0..l. a“ v ... -. . - o . ) . . o. 0 0 (00o ’ 0/0. 0o..0 00-...0. :0..1.' Cloveiurlyrao. .0'rl0— ”I” . .I .W' . O . . . . o 0 . o ’.. .. 00.. 0 J 0.. .0. 0 0.9- .0.10 Iu"’0'.l JIL '(ONIIIpW. .00 v . This. . . 0 I . . ..0 4.. .o .4 I...“ 03.... .0000: )...r.l.0l.9»00 .0 . .o... . e r. . . . .4 1.. .. .r. . . LU/nfnuv. .ou...... .., (1003”...0... .- 5.0. 0/? ”.m'f“... ). .rrm _. .. . 0 n I 0 . . t O o. . I ..1 .0. . I . . I h _ . w . . . . . 0 _ . .. . .0! d0.0v.fl. .0 I 0 v 00."0f"'”‘r3«. -. ..Whodl 0-... ’l . ”$00! 0.0,] 0 . t . . a a .10 . . . ..0. . . . r ..0 f . .. ..0 0.. .r.n....o0.l “le— 0'100.000!!...'.‘r.qf.'rll."o>0.r.'l0'_0 v N .. . u . .. .. . .. . .. _ .0 .. . . . r.....-:..r0.. .. 5.0.100! vlppfl.i........p. 70f!" 0 . . . .. . . 0. I..I. . I. 1.0.. .’.. ft W r... ...- I . . .00 c. . I. 00o h. I 00 .00.. ’ ..00’.’ ...- ’0..(YI 000.4"0. .‘ 'l'Ht'“ 0. An .. . . . 5.... .31.. 1.... ......- . 0 . . .. .. .. .. . . .. .40 ...0........Ir... r0 13! 0 j 0 . f. .. . .. . I . . . .0 .0: 00.9..0. I 00. .l ..l . 0.! 0"..I.l..... _ .107? M 0 ’1. I’lvf.’ II. /'.0 M f . . . ... .. . _. . _. . ...-.. . .....rp ........r.I..... .1. 7.1.1:- .1 ...}. ..p I 3 ZIP... . . . . . . . .... . ”...-...... ...! ... ..I.I¢.r.-. 00.0.0.3 ...rIIP. _ . . . . .0 .-. I v u. ..0 0 . .0 0 0 0L . ill... m w . o . .. 0 .. . . . .- .0- n. 0 IN... 0. .. . a ".0... o (......dcnu ..rl.. . .0 . . . . . . I. .l . a . r OI. I0 . S . 1. . . . . . . . a . .. .I.. I v . i n ..0. $.00 s'. “02704”!!le 0 .Il. . I. I . . '1 u I I ’Ul. ll ' e 0 0 l c .0 .00 0 000'I {#10 "I“ 001.10]. . O 'l’II O 90') . . . . 4 o .0. . .. I r 00 000-10. ( 0- .oo‘ . . . o . . .. . 0 . . . o 0 . . v.0 . n r .lr'fi. 0‘ .. . . u . I l. . .v o . ‘ .10 .n..'.0..0." - I . . 0. . v I! 0 .0 .0 I 00 3 . _ O ..n. 0 . . v. I 1.: o 000: ”4,4900r915 00- . TI. . a . r r . . .. .(uII-... .01.: A. . . . o o. o I A V A 0 . . . at J J 0. -I . 0 . I... I 000 0. clf-O .t‘. 4 I . f (1.0 0!!! .J.l'.| .I L00 .00000. Ft, 0'“! .IA . 0 I 0 . I 0 4 Or. . ’.'00"9.0000 ..‘ 0’]. I . . . . . . . «0.00! .0 . .oi'l..1.¢.....lrl. “Nun-II. ... 0 ... . . . 0 O .. .. .t 0’. . ....0”..AJH'O..I.Q.000..... 0'}! 000-- .. . . . _ a ...-.... In. (.5... .-r.!..r.v........r.. ....-. .. :- . . . . .. .. ..r.. ......2... .4. ......»wu up .. . w . . . 0 0 .I . . . o u. . ... .- .VI' O .10.0 10"). 0’.“ .( I . .. . I . . 0 . . .. y. . ..II . .ID 0.. 0...: n . _ . I I . . . . . r . . .. . 1 01.700) .0.: - o . J . .00 o 0 0 0. (.0. I.‘v'f.10. ..00H000. 0 a 0 u v c. . Il.’. v 0 “Jr.-. 1 g I 0’0. o.. I00. 7 1 . . 0 o - . v.-.tl«n1.v..x .. - . . . . . . .0 .f0. . . QIIII . ." 0‘..'010 01’; . . ..I"_. 0 o . . . .. ..' 00”.Iao I, I A 0 . . . .0 l...’ 0.0II.OI.Ud JIL'H 0 . -1 o. .o. I... .0. a 0 .1. ...!0’0. . . .’ Q . a u 0. O 7. 0.0 . o ...-.0 0.0 ’ftu'. . 09” . I :0 .u 0 I. O -' 0 T . . O 0 cl, 0 0 o n t I ' t It, .0 . . . . ... .hrn0aI .1. 4.0. ... I 00 0 0 0 ta 0 0 0 '9. W . 0 I. 00 f A. v n. o o s ‘0 . .. V o - . . . . . - . . . v ...... . .I...¢.....N.. \... . .JW... ......ootojllwucloscvo. 0...... 01......dnx‘ ..0....I.|..0.A. .. .cuTk.m‘0....f . 6 x. ..- u I . o . “Sacra“ 0.“? .‘a .. .\_0 ... n. .... o 3|. I .0... 0 03.0. 0 hy...‘......... as} In .. 10.0 0.. o..&0.d.0\..0L.L.0Lc1 a“. - . . . 41.1.. . - . . ... . . . - .... ...".nr... 113.3%...3 ......xs «.14.. ...-.55 .ts... .39.. . .u .... I .... t... huh... ... 7 a... pir. 0 .IJJJ. .01..“I010Pv?40$5.§.¢008..9..1 ..0...’.oh.flu.......Lf. 0 . O IIIO‘II..- I ‘ RES-ST REACT INTER-PARTY VOTING AND INTRA—PARI‘Y ooaasron by Gretchen Albrecht This thesis deals with the factors which contribute to party cohesion within the.Michigan House of Representatives in 1964-65 on the issue of migrant labor legislation. vot- ing on eight bills was analyzed-~four of which both parties favored and four of which the Democrats favored and the Republicans Opposed. The concept used in this analysis was the conflict between party position and presumed interest of constituency as it affected the relative cohesion of the rural legislators and urban legislators. The rural legis- lators were, in addition, divided on the extent to which their district utilized migrant labor. The indices of like- ness and cohesion devised by Stuart A. Rice were used to ascertain the extent of inter-party voting and intra-party cohesion. The Specific hypotheses tested in this study are as follows: (1) Rural representatives, regardless of party, vote more alike than is eXpected on the basis of party affi- liation, (2) The majority party is more cohesive than the minority party, (3) The degree of homogeniety of a political party in terms of rural-urban characteristics increases Gretchen‘Albrecht cohesiveness of votes, and (4) representatives from typical districts, that is, the type of districts from which a majority of the party support is derived, are more cohesive in their voting than representatives from atypical districts. The first hypothesis was not supported by this study. There appeared to be no rural coalition or urban coalition formed in this particular instance. Migrant labor legisla- tion, therefore, appears to be essentially a partisan issue. The next two hypotheses were established although it was not possible to ascertain whether majority position or homogeniety of party contributed most to the relative cohesiveness of the Danocratic Party. This question can only be answered through comparison of other situations in which the majority party is heterogeneous and the minority party is homogeneous. The "typical district“ hypothesis was proved to be of little value in the analysis of legislative behavior. It is essentially tautological in that it states that the group which controls the party and formulates the policy of the party will vote more in accord with the party. It appeared to be more important to ascertain the degree to which the typical district representatives controlled the party, that is, the degree of homogeniety of the party, than to analyze legislative behavior in terms of typical district representatives versus atypical district representatives. Thus rurality of constituency played no part in the formation of legislative coalitions. It was, however, a Gretchen Albrehht factor in intra-party cohesion in that if party position conflicted with the presumed interests of constituency the representatives subject to this conflict voted less cohesively. It was discovered, however, that the extent to which this conflict influenced voting behavior was dependent upon the legislative position of the party and the homogeniety of the party. In other words, the Republican Party in this instance was heterogeneous and the minority party and as such the Republicans were more influenced by the interests of their constituency. This led to a less cohesive voting pattern. Legislative behavior, moreover, was explained in terms of the ability to reward. The majority party was able to reward its members through passage of favorable legislation which allowed the rural Democrats to vote against constituency interests on this series of bills in anticipa- tion of future legislation which could be of value to their constituency. The Republicans, on the other hand, were not able to count on this type of reward for party cohesion. Thus the Republican legislators were more apt to vote in accord with the presumed interests of their constituency. Thus when the Republican Party position was Opposed to the interests of the rural constituency, the rural Republicans voted less cohesively: but when the two positions were in agreement, the rural Republicans voted more cohesively. Gretchen Albrecht Thus the amount of influence that constituency interests exert on legislative behavior is dependent upon the position of the party in the legislative system and the internal characteristics of the party. Constituency interests are more important when the party is in the minor- ity and is relatively heterogeneous. INTE R— PARTY VOT ING AND INT RA-P ARTY COH ES ION BY Gretchen Albrecht A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ART 8 Department of Sociology 1967 Acknowledgements I would like to acknowledge some of the many persons who gave willingly of their time and energy throughout the thesis process. Dr. James Hundley was generous with his time and energy in supervising the thesis process. Grateful acknowledgement is also extended to Dr. Harvey M. Choldin for becoming substitute chairman of the committee. I also wish to thank Dr. Denton.Morrison and Mr. Grafton D. Trout, Jr. for their assistance on this thesis. ii Page General statement of the problem . . . . . . . . . . . 1 Conceptual Framework . . . . . . . . . . . . . . . . . 3 Research Design . . . . . . . . . . . . . . . . . . . 9 AnalysisofData...................15 Interpretation . . . . . . . . . . . . . . . . . . . . 38 Bibliography 0 O O O C O O O O O O O O O O O O O O O O 47 iii TABLE 1. INDEX OF TABLES Indices of likeness between various groups in the Michigan House of Representatives, 1964-650.oooooooooooooooo Indices of cohesion within political parties in the Michigan House of Representatives, 1964’6500000000000ooooooo Indices of cohesion of “typical district" representatives in the Michigan House of Representatives, 1964-65 . . . . . . . . . Indices of cohesion within political parties on bills that both parties favored in the Michigan House of Representatives, 1964-65 Indices of cohesion of various groups within the Republican Party on bills that both parties favored in the Michigan House of Representatives, 1964-65 . . . . . . . . . . Indices of cohesion of various groups within the rural Republicans on bills that both parties favored in the Michigan House of Representatives, 1964-65 . . . . . . . . . . Indices of cohesion within political parties on hills which the Democrats favored and the Republicans Opposed in the Michigan House of Representatives, 1964-65 . . . . . . . . . . Indices of cohesion on various groups within the Republican Party on bills which the Eamocrats favored and the Republicans Opposed in the Michigan House of Representatives, 1964-65 . . . . . . . . . . Indices of cohesion on various groups within the rural Republicans on bills which the Democrats favored and the Republicans Opposed in the Michigan House of Representatives, 1964-65 . . . . . . . . . . iv Page 17 18 22 24 27 28 30 31 33 INDEX TABLE 10. 11. OF TABLES Continued . . . . Page Indices of cohesion of various groups within the Democratic Party on bills that both parties favored in the Michigan House of Representatives, 1964-65 . . . . . . . . . . . . 35 Indices of cohesion of various groups within the Democratic Party on bills that the Democrats favored and the Republicans Opposed in the Michigan House of Representatives, 1964-65 . . . . . . . . . . . . 36 INTER-PARTY VOTING AND IhTRA-PIARI‘Y COHESION Generalystatement of the_prob;§m This thesis is concerned with the issue of inter- party voting and intra-party cohesion. The question raised is what factors contribute either to cross party voting or to the relative frequency with which it occurs? More Specifically, is it possible to isolate any situational factors which increase the incidence of representatives vot- ing in accord with party position on a selected issue? The main eXplanatory concept utilized in this study is the cross-pressure or conflict eXperienced by the legislator be- cause of the incongruence between party position and pre— sumed interest of constituency. Because the issue selected was chosen on the basis of a potential rural-urban Split, the interest of constituency variable will be based on the rurality or urbanity of the legislative district. The particular issue that was considered in this study is the series of roll calls dealing with migrant labor in the years 1964;1965. This legislative series was chosen for its importance to the rural areas of Michigan so that if any rural-urban coalitions are present they will be evidenced in regards to this particular issue. Agricultural CDOps necessitating employment of migrant labor are big business 2 in Michigan, and hence it is assumed that constituency pressure will be relatively strong for the rural represent- atives. In addition, migrant labor legislation is a dis- tinct issue which can be subsumed neither under the general rubric of agricultural legislation nor general welfare legislation. Thus past voting patterns would not be the major determining factor in the voting behavior of the state legislators. Therefore, this thesis will be concerned with the inter-party voting and the intra-party cohesion of the Michigan legislators on the specific issue of migrant labor legislation. It is this issue that will be utilized as the arena in which various hypotheses will be tested and from which conclusions will be drawn. More Specifically, what effect does belonging to the majority or minority party have on the tendency for the legislator to vote in accord with party position? What is the validity of the hypothesis that legislators from typical districts vote more often with party position. The definition of a typical district is the type of district from which a majority of the representatives of a particular party are elected. Does the alignment of the parties around a particular bill have any effect on party cohesion? How do the rural-urban characteristics of the constituency affect the legislator in his voting behavior? Can the voting pattern of the Michigan House of Representatives be eXplained in terms of 3 the croSSpressure of party and presumed interests of con- stituency? Conceptual Framework Much sociological research emphasizes only one side of social behavior. The position.taken by Lipset is an example of this type of thinking. "Emphasis on the integrative aSpects of electoral behavior would not only fill lacunae in our understanding of democracy as a system, but would necessarily be more sociological in orientation that cleavage”. (Lipset, 1959, p. 93) In studying integrative or cohesive behavior of any kind, however, it is necessary to analyze the factors which account for the variation in cohesive behavior. An approach that can account for both cohesion and cleavage offers more insight than a one-sided analysis of either. It is from this position that analysis of the voting patterns will be- gin. Migrant labor legislation, when passed, becomes public policy in that it is a ”. . . structure of confluence of values and behavior involving a governmental prescrip- tion". (Kroll, 1962, p. 363) In other words, the bills once passed become law and as such define the situation for the participants in the system. More importantly, as was indicated by Morton Kroll ”one cannot separate agnblic policy from the milieu in which it is set“. (Kroll, 1962, p. 364) Thus it is plausible to assume that one cannot separate the legislator and hence his behavior from the milieu in which he participates. Therefore, the legislative 4 system in which he acts while in Lansing influences his be- havior as well as the district in which he was elected. Moreover, the expectations of each of these systems need not always be congruent; the party and the district may be Opposed on a particular issue. The representative, there- fore, is exposed to conflicting eXpectations and this con- flict should be evidenced in his voting behavior. Thomas.A. Flinn maintains that “inter-party conflict and intra-party cohesion describe essentially, but with varying degrees of imprecision, many roll calls in state legislative bodies“. (Flinn, 1964, p. 70) Parties, there- fore, are more or less cohesive, and the question raised is what factors contribute to cohesion or straight party voting and to low cohesion or cross party voting? William Kornhauser states that common economic interests often lead to joint political action for limited goals. (Kornhauser, 1958, pp. 575—576) Kroll indicates that with- in a Specific policy field groups may align and realign to the extent to which they relate their interests to the series of issues on which decisions must be made at a given time. (Kroll, 1962, p. 375) Thus on the basis of joint economic interests it is eXpected that certain coalitions will form around the issue of migrant labor legislation, for as Eulau points out -- state legislators perceive representation of the interests of some geographical area as a prOper function of their legislative activities. (Eulau, 1962. p. 288) 5 Therefore, it is posited that the division of the representatives on the basis of rural or urban constituen- cies should offer some insight into their voting behavior. As Froman states, “Since shared attitudes about various problems confronting people are often the result of sharing similar environments . . . it is not surprising to find peOple located in similar environments choosing up sides in similar ways on matters of public policy and differing from those who do not share the same environment“. (Froman, 1963, p. 57) Thus, according to this position, coalitions should be formed on the basis of a rural-urban Split. More Specific- ally, rural legislators should vote more alike than would be eXpected from the party affiliation of the legislator. Analysis of coalitions, therefore, should be useful in ex- plaining cross-party voting. What, then, contributes to intra-party cohesion? Malcolm E. Jewell posits that members of a legislative party are most likely to vote in the same manner when they represent similar constituency interests, for the represent- ation of Similar constituencies leads to a decrease in the number of serious conflicts to which the individual legis- lators are eXposed. (Jewell, 1966, pp. 90—91) In other words, if a large majority of a particular party represent similar areas--that is, the party is relatively homogeneous-- it is more likely that party position on a particular issue will be consistent with the interests of that area. This in turn will eliminate one area of conflict--party versus 6 constituency--in which the legislators may be “squeezed" in cross—pressure. Therefore, it follows that if a party is relatively homogeneous it will tend to be relatively cohesive. Jewell also posits that there is a higher level of unity in the majority party. (Jewell, 1959, p. 127) This is eXplained by Jewell and Patterson on the basis of ability to reward. More Specifically, they state that if a party is to have any effect on consensus it must be able to diSpense rewards. (Jewell and Patterson, 1966, p. 424) The majority party has more ability to reward.merely through the fact that it can and does control the “pork barrel” and in itself can control the flow of legiSlation. It thus has the power to reward and hence should be more cohesive. In addition Kurt W. Back states that an indivi- dual belongs to a group (and, I feel, follows the mandates of a group) because the group has the ability to mediate goals. The party which holds the majority position in the legislature can through cohesive voting pass legislation which is important to the district of the representative. Thus the legislator is more apt to vote in accord with the position of his party in most instances if that party is in the majority in anticipation of party support on bills of greater interest to his constituency. The data, therefore, are eXpected to Show a higher degree of cohesive voting in the majority party than in the minority party. If a certain political party exhibits a certain degree of cohesion, the next question that arises is 7 whether or not there are certain constituency characteris- tics of the legislators which influence their conformity with party position. In other words, is there a way to ex- plain the variation which is evidenced within the party? Who are the legislators who break with party position? Duncan MacRae maintains that “those representatives who come from districts that are most typical of their parties tend to Show highest party loyalty". (MacRae,l956, p. 323) This finding is supported by Jewell when he states that disunity in a party with widely scattered representation is caused by legislators representing areas where the party is not dominant and by Flinn when he main- tains that members from districts typical of their party support the party position more often than others. This position can be stated in different terms. For instance, it seems logical that in a majority of issues the party position will be consistant with the interests of the con- stituency of districts from which their strength is derived. Thus these legislators in most instances will not be con- fronted with a conflict between party position and the interests of their constituency. These legislators, there- fore, will not be the ones who vote against party position. The question that arises is what occurs when the party position is in Opposition to the interests of these con- stituencies. Obviously, the number of times that this occurs will be relatively small, but it does occur, which affords an Opportunity to test the typical district hypothesis. 8 Therefore, this study is based on the idea that a representative perceives himself to be reSponsible to different social worlds--two of which entail representation of his district and allegiance to his party. As David R. Derge indicates, however, party loyalty exerts a greater pull on the legislator than does his constituency. (Derge, 1959, p. 223) More Specifically, it will be assumed that a representative will vote in accord with his party unless he is subjected to a conflict between party position and presumed interest of constituency. This will be evidenced in the voting patterns of the legislatureas a low degree of cohesion because more of these representatives will break with party position. In other words, if party position con— flicts with the presumed interests of a Specified type of district, the representatives from those districts will vote less cohesively than if party position and interest of constituency coincide. Thus from the above literature certain assumptions can be extracted. They are as follows: 1. All groups have the ability to reward. (Parties control legislative rewards while districts reward in terms of election to office.) 2. Adherence to group eXpectations is dependent on the relative ability of the group to reward. 3. Groups with factors in common tend to have Similar interests. From the above assumptions there are certain ideas which can be derived which are on a less abstract level. They are as follows: 1. Similar interests tend to lead to similar voting patterns. Legislators feel allegiance to both party and district. The majority party has the greatest ability to reward. Congruence of interests between party and con- stituency tends to reduce the total conflicts to which the legislators are eXposed. Thus the Specific hypotheses to be tested are: 1. Rural representatives, regardless of party, vote more alike than is expected on the basis of party affiliation. The majority party is more cohesive than the minority party. The degree of homogeniety of a political party in terms of rural-urban characteristics in- creases cohesiveness of votes. Representatives from typical districts are more cohesive in their voting than representatives from atypical districts. Research Desigp The setting selected as an arena in which to test the above hypotheses was the Michigan House of Representatives in the years 1964-1965. The House at this time was comprised of 110 representativeS--73 of which were Democrats and 37 of 10 which were Republicans. Michigan at this time had a Republican governor. Thus the minority party had control of the executive branch. The series of roll calls dealing with migrant labor was chosen for analysis for several reasons. First of all, any rural-urban Split in the House should be evident on these bills. Because the prOposed.migrant labor legislation would increase costs to the rural areas employing migrant labor, the rural representatives should tend to be Opposed to passage while passage of this particular series of bills would entail no great additional costs for the urban dis- tricts. Secondly, the governor had established a committee to study needed legislation which indicates executive pres- sure for passage of some legislation. The Republicans, in particular, probably felt this pressure: and this, if any- thing, should increase the cohesive voting of the Republican legislators. Thirdly, the issue of migrant labor was of national interest as was evidenced by the legislative activity concerning migrant labor in Washington. Thus migrant labor legislation was an important series of roll calls in the Michigan House of Representatives. The Specific bills analyzed include all of the bills concerning migrant labor voted upon in the legislative sessions of 1964-1965 on which there were at least four negative votes. Only final votes were analyzed and no record was kept of absences. There were four bills analyzed on which both parties were in favor of passage. Because the Republican Party is ll comprised mainly of rural representatives, this seems to indicate that these four bills were not as antagonistic to the interests of the rural legislators as were the other bills analyzed. The Republican Party, therefore, was able to assume a position that these bills should be passed. The Specific bills are: 1. Senate Bill 926 dealt with the transfer of reSponsibildty from the Department of Agriculture to the Department of Commerce for the regulation of minimum safety requirements for vehicles used to tranSport migrant labor. Obviously, this is a relatively innocuous bill in that it entailed no great threat to the rural areas of the state. On this bill there were four negative votes and 83 affirma- tive votes. 2. House Bill 3516 was concerned with the establish- ment of permanent and mobile supervised overnight rest camps and information centers for migrant laborers. Again, this bill posed no serious threat to the rural areas. There were four negative votes and 81 affirmative votes on this particular hill. 3. Senate Bill 114 dealt with workmen's compensa— tion but for the first time included farm Operators. Accord- ing to an eXpert in the field the rural legislators thought that this bill was relatively innocuous. Now, however, much controversy surrounds this bill. There were 11 nega- tive votes and 92 affirmative votes on this bill. 12 4. Senate Bill 118 dealt with the establishment of summer schools for migrant children. This bill was relatively unimportant in that the total apprOpriation was only $15,000. There were 7 negative votes and 79 affirma- tive votes on this particular bill. The remaining four bills afforded instances in which the two parties assumed opposite positions, that is, the Republicans were against passage while the Democrats were in favor of adOption. These bills, therefore, are probably more antagonistic to the interests of the rural constituencies. The Specific bills are: 1. House Bill 2238 was concerned with the registra- tion and licensing of farm labor contractors. A.similar bill was passed by the Federal Government the previous year. On this particular bill there were 26 negative votes and 69 affirmative votes. 2. House Bill 2973 dealt with the establishment of the Rural Manpower Center at Michigan State University. An eXpert in the field described this bill as essentially a partisan issue. The bill was introduced by the Democrats and the subsequent vote, he maintains, was a manifestation of partisan politics. This bill was passed by the House by a margin of 59 to 30. 3. House Bill 2125 was concerned with the licens- ing and regulation of agricultural labor camps and dealt with their minimum health requirements. This bill entailed added eXpense for the rural constituencies and was in 13 Opposition to their interests. This bill, according to a member of the Governor's commission, was probably the most important migrant labor bill to come before the house dur— ing these sessions. There were 22 negative votes and 73 affirmative votes on this bill. 4. House Bill 2707 established minimum safety regulations and requirements for vehicles tranSporting migrant laborers. This bill was described as a "do-gooder“ bill although there is now a question of its constitution- ality. This bill passed the house by a margin of 62 to 33. Assignment of the legislators into the dichotomous categories of rural or urban was accomplished through reference to the 1960 Census Report. About 3% of the total employed working force in Michigan is agricultural: the counties, therefore, were divided on this basis. If a legislator represented only one county, assignment created no difficulties. However, often a legislative district in- cluded several counties or parts of several counties. In this instance assignment was more arbitrary. It was done on the basis of averages of the counties and through refer- ence to the district map versus the county map. Thus representatives from districts of 3%lor more agricultural employment were assumed to be elected from a rural district while those representative from counties of 0.0 to 2.9% were assigned to the urban category. 37 legislators were designated as rural representatives while 73 were considered to represent urban districts. 14 For purposes of analysis the rural districts were broken down into two other categories--those districts which Spend a relatively high percentage on agricultural hired labor and those which spend a relatively low percent- age. This percentage is based on the total amount Spent on agricultural hired labor out of the aggregate income. The average Spent on agricultural hired labor in the state was .3%.of the aggregate income although for purposes of analysis this was increased to .5%. In treating rural districts as two groups on the basis of percentage Spent on hired labor it is assumed that a high percentage Spent represents a high degree of utilization of agricultural hired labor with- in that district which should increase the importance of any migrant labor legislation for the constituency of the particular representative. Thus the voting pattern of these representatives should exhibit an intensification of the pattern found when the legislature was divided on the basis of rurality or urbanity of of constituency. Thus the representatives were assigned on this basis following the same procedure as was discussed above. The votes were analyzed with the method develOped by Stuart A. Rice in his article "The Behavior of Legisla- tive Groups" in 1925. Rice formulated two distinct measures: the index of likeness and the index of cohesion. The index of likeness is the complement of the arithmetic difference between the percent of affirmative votes on a Specific issue. It is utilized to Show how similar the voting behavior of 15 two Specific groups are. For example, if one group votes 93% aye and another group votes 82% aye the index of like- ness will be 89. More Specifically, the difference between 93 and 82 is 11: and when 11 is subtracted from 100, the resultant figure is 89 which represents the index of like- ness. Moreover, the index of likeness according to Rice may be computed using the arithmetic mean of the indices derived from the separate roll calls in the series which en- ables the analyst to study the voting pattern on a series of bills rather than merely one bill at a time. The index of cohesion measures whether or not the members on Group A are more like each other than are the members of the more inclusive Group B which includes not only the members of Group A but also non-A members as well. In this analysis a vote Split 50%.aye and 50% nay would be represented by 0.0 cohesion whereas an instance in which all members of Group A voted alike would be indicated by an index of 100. For example, a vote Split 20% aye or 80% aye would be represented by an index of 30/50 or an index of 60. Analysis of Data In order to test the first hypothesis which states that rural representatives, regardless of party, should vote more alike than is eXpected on the basis of party affiliation, Rice's index of likeness will be employed. The results are Shown in Table 1. This table does notgro- vide evidence that supports the idea that constituency l6 characteristics are the major determining factors in legislative voting patterns. The group which is of most interest is, of course, the representatives from rural areas because the issue of migrant labor was chosen because of its ramifications for rural areas. As is indicated by Table 1 there is no in- crease or a very slight decrease between rurality of con- stituency and coalition formation on that basis. More Specifically, the rural legislators vote in accord with party position: there is no increase in inter-party voting. Therefore, over all, the issue of migrant labor was, in essence, a partisan issue. This finding supports the thesis of Derge in which he states that party loyalty exerts a greater pull on the legislators' loyalty than does geographical base. (Derge, p. 223) Therefore, even on an issue with great import for rural areas legislative voting patterns do not indicate the presence of a coalition formed on the basis of rural or urban characteristics of the con- stituency. For the total of the eight roll calls the Democrats had an index of cohesion of 97 whereas the Republicans had an index of cohesion of 69 as is shown in Table 2. The data presented in this table will be used to ascertain t.he validity of hypotheses 2 and 3. The hypothesis that the party in the majority ex- hibits more cohesion is re—affirmed by Table 2. As was in- dicated previously if the party is to have any effect on 17 Table 1. Indices of likeness between various groups in the Michigan House of Representatives, 1964-65 M Groups Compared Republicans Rural RepUblicans Urban Republicans and and and Roll Calls Democrats Rural Denocrats Urban Democrats S-118 78 81 75 S-926 91 76 98 H-3516 99 100 96 S-114 68 52 80 H-2238 10 6 15 H—2973 6 18 13 H-2125 38 35 60 H-2707 8 8 13 Total 50 47 56 18 Table 2. Indices of cohesion within political parties in the Michigan House of Representatives, 1964-65 _- ...—n. , .- .,_.._—_.__ -—. .w— -... ..~...-._....._. -..-..- -..... .--— Groups Compared Roll Calls Democrats (73) Republicans (37) 5-118 100 56 S-926 96 78 H-3516 9O 92 S-114 100 35 H-2238 100 80 H-2973 94 94 H-2125 96 28 H-2707 96 88 Total 97 69 l9 consensus it must be able to reward, and it follow 8 that the majority party is in a position to more readily dis- pense rewards. If a group controls the reward structure ,of the system and places emphasis on conformity, an indivi- dual will conform more readily than if the group has little control over the distribution of rewards. A political party to achieve any of its legislative goals must be relatively cohesive. Reward for voting in accord with party position could be thought of as entailing the passage of “pork barrel” legislation that will enhance the legis- lator's possibility of being re-elected. The majority party has within itself enough votes to pass any legislation and therefore possesses a greater ability to reward. The results from this table also add credence to the third hypothesis which states that when one party is more homogeneous than another party it tends to exhibit a more cohesive voting pattern. In this instance the Democratic party is more homogeneous in that 92% of the Democratic representatives come from areas with a low percentage of agricultural employment whereas the Republican Party is com- prised of representatives 35%.of which come from areas of low agricultural employment and 65% from areas of high agricultural employment. Thus, obviously, on the particular characteristic the Democratic party is more homogeneous. According to Kurt w. BaCK the factor which is common to all members is larger in highly cohesive groups than in less cohesive groups. (Back, 1958, p. 183) In 20 other words, the fact that the Democratic Party is character~ ized by a higher percentage of urban representatives should be evidenced in its voting pattern as a higher index of cohesion. More Specifically, there is no sizeable rural- urban Split within the Democratic Party whereas the Republican Party is Split internally on the basis of rural— urban characteristics. Thus it seems plausible that representation of similar constituencies leads to a decrease in the number of serious conflicts to which the individual legislators are eXposed which in turn enhances the probability of cohesive voting. In other words, the party position will probably not conflict with the interests of the urban Democrats: and since there are few rural Democrats, the absolute number of representatives whose interests are Opposed to party policy is small. This will enable the Democrats to have a relatively cohesive voting pattern. ‘As was indicated previously 65% of the Republicans are rural and 92%vof the Democrats are urban. Therefore, the rural Republicans and the urban Democrats will be assumed to represent typical districts-~that is, districts from which a majority of the party support comes. The data are shown in Table 3. Since the index of cohesion for the total group of Democrats is 97 and the Democrats from typical districts Show an index of 98, there does not seem to be any signifi- cant increase in the cohesion of "typical“ Democrats. 21 Rural Republicans show only a slightly greater tendency to vote more cohesively than the total group of Republicans. Thus the hypothesis of typical district representatives does not seem to enhance eXplanation of party cohesion. In summary, coalitions formed on the basis of rural or urban constituency do not seem to be evidenced in the voting patterns on the issue of migrant labor. It appears to add little to the factor of party affiliation as a pre- dictor of voting behavior. Party loyalty thus overrides any tendency to vote in accord with interests of constituency. However, it is necessary to analyze the variation within the parties to ascertain the importance of rural or urban characteristics. In other words, the relatively high co- hesion of the Democratic Party might obscure any influence of constituency interests. It is possible, on the other hand, to analyze legislative behavior in terms of conflict and in this way attempt to explain the variation in intra—party cohesion. The conflict that will be analyzed is represented by an in- consistency between party position and presumed interests of the constituency on the basis of rural-urban character— istics. Because party loyalty is assumed to exert a greater force in the determination of voting behavior than the presumed interests of constituency, conflict between the two will be Operationalized as a low degree of cohesion within the group under analysis. In other words, represent- atives will vote in accord with their party unless they Table 3. 22 Indices of cohesion of “typical district" representatives in the Michigan House of, Representatives, 1964-65 W Groups Compared Roll Calls Urban Democrats (67) Rural Republicans (24) 5-118 100 62 S—926 98 52 H-3516 92 84 S-114 100 4 H-2238 100 88 H—2973 100 100 H-2125 100 66 H-2707 100 100 Total 98 72 23 eXperience conflict between their party and the interests of their constituency. However, because the constituency pressure is less important in voting behavior than party pressure the resultant voting pattern will be less cohesive. There were four bills concerning migrant]abor-- either directly or indirectly-~on which the majority of both parties was in favor of passage. S-118 dealt with the establishment of a summer school for migrant children; but because only $15,000 was apprOpriated, this bill was rela- tively innocuous. S-926 concerned the transfer of reSpon- sibility from the Department of Agriculture to the Department of Commerce for minimum safety regulation for vehicles used to tranSport migrant laborers. H-3516 concerned the estab- lishment of permanent and mobile rest camps and information centers for migrant laborers. S-ll4 dealt with workman's compensafion and included farm Operators. The data on these bills are shown in Table 4. According to the data in Table 4 the dissenters are most frequently found in the minority party, and the res- pective indices of cohesion for these four bills do not differ to any great extent from the overall indices.for the entire series of eight bills. However, on this series of bills the party position is in Opposition to the interests of the rural representa- tives. For the present the analysis will deal only with the Republicans in that the high degree of cohesion within the Democratic Party may obscure any variation which might be 24 Table 4. Indices of cohesion within political parties on bills that both parties favored in the Michigan House of Representatives, 1964-65 Groups Compared Roll Calls Democrats (73) Republicans (37) % aye 100 78 S-118 cohesion 100 56 % aye 98 89 S-926 cohesion 96 78 % aye 95 96 H-3516 cohesion 90 92 % aye 100 68 5-114 cohesion 100 36 Total Cohesion 97 65 25 explained through reference to the rurality of the representa- tive's constituency. When the Republicans are divided into two groupS—-urban Republicans and rural Republicans--and the index of cohesion is computed, the results are as shown in Table 5. This table, first of all, indicates that it is the rural Republic ans--the typical Republicans--that deviated most often from party position. But more importantly, the Republicans from typical districts were less cohesive, not more cohesive, than the total group of Republicans. This is in opposition to hypothesis 4 or the “typical district" hypothesis. Why is this the case? \ It follows that if the pressure to vote in accord with the interests of constituency was the most powerful factor in the determination of legislative voting behavior division on the basis of rural-urban characteristics should result in a higher index of cohesion for both groups than the index for the total group of Republicans. As is indi- cated in Table 5, however, this occurs only in relation to the urban Republicans. It, therefore, seems plausible that the interests of the constituency did exert some pressure on the rural legislators, that is, the rural legislators eXperienced a conflict between party position and the in- terests of their constituency which precluded a highly co- hesive voting pattern. More Specifically, some of the rural Republicans voted in terms of party position while others voted in terms of constituency interests which lead 26 to a low index of cohesion for the entire group. The urban Republicans, on the other hand, were not subject to the conflict between party position and interests of their constituency: therefore, they voted rather cohesively or in accord with the position of their party. To add further support to the ideas presented above the group of rural Republicans were divided into two diSparate groups--those from districts in which there is a high percentage Spent on hired labor in the agricultural sector and those from districts which Spend a low percentage. The data are shown in Table 6. As is shown in the above table the Republicans from areas which Spend more on agricultural hired labor have a lower index of cohesion than do Republicans from rural areas which do not Spend a large percentage on agricultural hired labor. Following the same argument that was presented above, therefore, Republicans from rural areas which Spend a low percentage of their aggregate income on migrant labor are not subjected to the same degree of conflict between interest of constituency and party position. Hence they vote more in accord with party position and, consequently, evidence a higher index of cohesion. More Specifically, the tendency of legislators to vote in accord with party posi- tion decreases as the party position conflicts with con- stituency interests although it remains the major factor in legislative voting. 27 Table 5. Indices of cohesion of various groups within the Republican Party on bills that both parties favored in the Michigan House of Representatives, 1964-65 _ I Groups Compared Urban Rural Roll Calls Republicans (13) Republicans (24) % aye 75 81 S-118 cohesion 50 62 % aye 100 76 S-926 cohesion 100 52 % aye 100 92 H-3516 cohesion 100 84 % aye 80 52 S-114 cohesion 60 4 Total Cohesion 77 50 28 Table 6. Indices of cohesion of various groups within the rural Republicans on bills that both parties favored in the Michigan House of Representatives, 1964-65 ...." w.-. ...... ”--.... ... . . ..-w— .....— ...“...1. ...... Groups Compared Percentage Spent on Agricultural Hired Labor High percentage Low Percentage Roll Calls Districts (15) Districts (9) % aye 77 100 S-118 cohesion 56 100 % aye 78 88 3-926 cohesion 56 76 % aye 100 86 H-3516 cohesion 100 72 % aye 46 8 8-114 cohesion 56 12 Total Cohesion 55 65 29 A majority of the Democrats were in favor of pas- sage and a majority of the Republicans were against passage on four bills included in the series of roll calls that was analyzed. H-2238 required the licensing and registration of farm labor contractors. H-2973 established the Rural Manpower Center at Michigan State University. H-2125 was concerned with minimum health standards necessary for the licensing of labor camps. H-2707 established.minimum safety regulations for vehicles tranSporting migrant laborers The data for both parties is shown in Table 7. This data shows that although the Republicans still have a lower index of cohesion than the Democrats it is higher than on the previous series of bills. On these bills the position of the Republican Party is consistent with the interests of the constituency from rural areas. There- fore, it is not surprising that the Republicans Show a higher index of cohesion than on the former series of bills because the majority of the Republicans represent rural areas. There is less conflict between the interests of the rural constituency and the position of the “rural“ party. When the Republicans are divided into two groups on the basis of rurality or urbanity of constituency, the results are as shown in Table 8. This table shows that the rural Republicans are much more cohesive than Republicans from urban districts. Thus in this instance because there is no conflict between party position and interests of the 30 Table 7. Indices of cohesion within political parties on bills which the Democrats favored and the Republicans Opposed in the Michigan House of Representatives, 1964-65 GIOUpS Compared Roll Calls Democrats (73) Republicans (37) % aye 100 10 H-2238 cohesion 100 80 is aye 97 3 H-2973 cohesion 94 94 % aye 98 36 H-2125 cohesion 96 28 % aye 98 6 H-2707 cohesion 96 88 Total Cohesion 97 73 3L Table 8. Indices of cohesion on various groups within the Republican Party on bills which the Democrats favored and the Republicans Opposed in the Michigan House of Representatives, 1964-65 W Groups Compared Urban Rural Roll Calls Republicans (13) Republicans (24) fiéaye 15 6 H-2238 cohesion 70 88 % aye 13 5 H-2973 cohesion 74 90 % aye 60 17 H-2125 cohesion 20 66 iéaye 13 4 H-2707 cohesion 74 a 92 Total Cohesion 60 84 — 32 constituency, the rural Republicans are able to vote more cohesively. Therefore, interests of constituency do have an effect on legislative voting in that it can reinforce the probability of voting in accord with party position. In other words, for the rural legislators there is no factor in this situation which tends to decrease the importance of party as a determinant of legislative voting. The data necessary for the analysis of the low degree of cohesion of the urban Republicans was not collect- ed so the reasons offered for their behavior can only be a result of Speculation. To deny the "political“ of legis- lative voting is to overlook an important aSpect of legis- lative behavior. It is commonly known that final votes en- tail much give and take on the part of the individual legis- lators. In addition, if the Specific bill has no Special relevance to the home district of the individual legislator he is in a position of bargaining--that is, he can vote against the party on one issue in anticipation of gather- ing support on another issue more important to his con- stituency. Moreover, if any “deals“ are made they will be made in instances of this type. Therefore, this factor could possibly be a determinant in the voting behavior of the urban Republicans on these four bills. To add support to the interpretation of the above data the rural legislators were again treated as heterogeneous. The results are shown in Table 9. 33 Table 9. Indices of cohesion on various groups within the rural Republicans on bills which the Democrats favored and the Republicans opposed in the Michigan House of Representatives, 1964-65 Groups Compared Percentage Spent on Agricultural Hired Labor High Percentage Low Percentage Roll Calls Districts (15) Districts (9) % aye 9 0 H-2238 cohesion 82 100 % aye 0 6 H-2973 cohesion 100 72 % aye 8 63 H-2125 cohesion 84 26 % aye 0 11 H-2707 cohesion 100 78 Total Cohesion 92 69 34 AS is shown in the above table those Republicans representing areas which Spend a high percentage of their aggregate income on migrant labor exhibit a larger index of cohesion because, again, there is no conflict between the interests of their constituency and the position of their party. This increases the probability that these Republi- cans will vote in accord with party position. Thus the hypothesis that representatives from typi- cal districts more often vote in accord with party position seems to be dependent upon the stance that the party assumes on a particular issue. If a majority of the party favors a bill which is in conflict with the presumed interests of their constituency, the representatives from typical dis- tricts will be subjected to party-constituency conflict and hence will exhibit a lower degree of cohesion. If the party position, on the other hand, coincides with the in- terests of the constituency from areas which elect a majority of that particular party--as seems likely in a majority of the bills--then and only then do the representa- tives from typical districts exhibit a higher degree of cohesion than representatives from atypical districts. Because the Democrats exhibit such a high degree of cohesive voting, the analysis of the Democrats cannot be expected to Show as large a difference as the Republicans. Moreover, because only twelve Democrats are from rural areas these representatives will have to be treated as homo- geneous-~that is, not be divided into areas Spending a high 35 Table 10. Indices of cohesion of various groups within the Democratic Party on bills that both parties favored in the Michigan House of Representatives, 1964-65 Groups Compared Rural Urban Roll Calls Democrats (‘12) Democrats (61) % aye 100 98 S-926 cohesion 100 96 % aye 92 96 H-3516 cohesion 84 92 % aye 100 100 S-114 cohesion 100 100 % aye 100 100 s-118 cohesion 100 100 Total Cohesion 96 97 36 Table 11. Indices of cohesion of various groups within the Democratic Party on bills that the Democrats favored and the Republicans opposed in the Michigan House Of Representatives, 1964-65 W Groups Compared Rural Urban Roll Calls Democrats (12) Democrats (61) % aye 100 100 H-2238 cohesion 100 100 % aye 82 100 H-2973 cohesion 64 100 % aye 82 100 H-2125 cohesion 64 100 % aye 100 98 H-2707 cohesion 100 95 Total Cohesion 81 n v 99 37 or a low percentage on agricultural hired labor. Table 10 shows the data for the Democrats when they are divided into rural Democrats and urban.Democrets for the bills on which a majority of each party was in favor. .As is shown by Table 10 there is no difference in the cohesive voting pattern on the basis of a rural-urban Split. More Specifically, even though the party position was inconsistent with the interests of their constituency the rural Democrats tended to vote in accord with party position. In Table 11 the data showing rural Democrats and urban Democrats on bills which the majority of Democrats favored and the majority of the Republicans Opposed is presented. From this table it appears that the Democratic representatives from rural areas were subject to conflict between position of their party and interests of their con- stituency as is evidenced by the lower index of cohesion. Why should there be a difference between Table 10 and Table 11? The most logical explanation is that in the latter case the bills had more negative impact for the rural districts. This is plausible for two reasons. First of all, the conflict between their party position and the presumed interest of their constituency for the rural Democrats appears in their voting behavior only on these four bills. More Specifically, in regard to the former series of bills the Democrats from rural areas followed the 38 position of their party with only one exception. Secondly, the fact that the Republican legislators--e majority of which came from rural areas--supported the former series of bills seems to indicate less Opposition from the rural con- stituencies. In other words, even the party which is pre- dominantly rurel could gather a majority of its members to be in favor of passage. Thus it seems plausible that members of the majority party are more likely to vote accord- ing to party position, and it is only extreme conflict which can cause them to abandon their party. Even with this pressure for cross-party voting, however, the members of the majority party who are from “atypical districts" vote more cohesively or in accord with their party then do "typical district" members of the minority party at any~ time. Ipterpretation From the above results it seems that the major factor in the prediction of legislative voting behavior is party affiliation. There appears to be no large incidence of coalition-formation across party lines on the basis of rural-urban constituency. More Specifically, the pressure for conformity to party position is the most significant force in the determination of voting behavior. If no rural coalition across party lines is formed on the migrant labor issue which is very important to the rural districts, it seems to indicate that there is no rural coalition in the Michigan House of Representatives. The common factor of 39 rurality, therefore, does not lead to inter-party voting. However, the data does indicate that constituency characteristics, although not the most important factor in legislative voting behavior, do influence intra-party cohesion. The extent of constituency pressure, however, depends on the position of the party in the legislative system. This can be explained through reference to the 'reward structure of the legislative system. More Specific- ally, the majority party has the greater potential for the distribution of rewards in that it can control the legis- lative input and output. In other words, the majority party controls not only the voting in the legislature it- self but also the bills which reach the floor of the house becauseof their control of the committees. Thus the Democratic Party possesses a greater ability to reward, and hence the willingness of the legislators to vote in accord with the party position is greater. Unfortunately, assessment of the relative weight of the pressure to conform on the basis of majority position and the degree of homogeniety cannot be ascertained using merely the above data. A comparative group of a majority party which is heterogeneous and a minority party which is homogeneous is necessary. However, data on intra-party co— hesion do suggest that homogeniety of party does increase the probability of cohesive voting in that it decreases potential conflict within the party. The data also indi- cate that being a member of the majority party is one of the 40 most important factors in legislative voting as is Shown on the data concerning rural Democrats. More Specifically, even those Democrats who are from atypical districts had a higherlindex of cohesion than the Republican from typical districts. Of course, it is also possible that the homo- geniety of the Democratic Party tended to increase the rela- tive strength of the party position in that the party itself was more united on the ultimate decisions. Being in the majority gives a party control over the reward structure but being homogeneous increases unity over the distribution of the rewards within the party. Both of these factors con- tribute to the ability of the party to reward. Therefore, it Seems plausible to state that majority position has major import for cohesive voting; the homogeniety of the majority party, in turn, increases the already strong tendency for this party to vote cohesively. Thus homo- geniety of party decreases the extent of conflict between the party position and the interest of the constituency of the party members. According to the results the typical district hypo- thesis seems to have little eXplanatory or predictive power and, therefore, should either be modified or discarded. The extent to which representatives from typical districts vote cohesively is determined not only by the margin of the party it comprises but also the position of the party in the legislative system. Typical district representatives are defined as those members of a particular political party 41 who represent areas from which a majority of their party are elected. Other than the fact that in most instances the party position will be consistent with the interests of the constituency who give the party its support, the typi- cal district hypothesis seems to add little to the knowledge or the analysis of legislative voting behavior. To talk of the typical district representative is to talk about the group which controls the party. The extent to which they control the party is represented by the degree of homo- geniety of the party. Cohesiveness of party votes, however, depends upon the legislative position of the party and the degree of homogeniety. It, therefore, seems superfluous to talk about typical district representatives, for what in essence is being said is that the group which controls the party will vote more often in accord with the decisions of the party. It was stated previously that the interests of con- stituency do not seem to be a factor in coalition formation, rather party position in the legislative system seems to be the major determining factor in party cohesiveness or inter-party voting. However, in the analysis of intra- perty cohesion or the degree to which certain groups within a particular party adhere to their party's position the presumed interests of their constituency on the basis of rural-urban characteristics appears to be relatively import- ant. More Specifically, the votes of the legislators which are cast in Opposition to party position can be Shown 42 to be patterned on the basis of rurality or urbanity of constituency. Therefore, it is possible to use the rural- urban dichotomies as a factor in the eXplanetion of intra- party cohesion. The variation in cohesiveness can, in this instance, be Specified as a result of conflict between party position and interests of constituency. It is necessary, however, to assess the relative strength of the party position be- fore anything explicit can be stated about the effect that conflict has on voting behavior. Strength of party position is increased, as was indicated above, by being the majority party in the legislature and by the probable strength of agreement on party position as indicated by homogeniety of party. Deviation from party position in the majority party is much less frequent and voting in accord with the pre- sumed interests of the conS'tituency depends on the relevence of the issue to the district. It is only when the issue is most relevant that the legislators of the majority party Show any cross party voting. The issues on which the two parties were in Opposition to each other indicated strong pressure from the rural districts. The rural Democrats, however, even when faced with an apparently strong issue that contradicted the presumed interest of their constituency Show less evidence of conflict, for their index of cohesion is still larger than the minority party with no conflict between party positions and interests of constituency. 43 Thus knowledge of party position is the most powerful pre- dictor of the votes of the majority legislators: the con- flict between party position and presumed interests of their constituency is much less important. The minority party, on the other hand, is not able to offer the same incentives for party cohesion, and it is this factor which affords analysis in terms of conflict be- tween perty position and interests of their constituency. Knowledge of constituency characteristics seems to afford the possibility of predicting relative cohesiveness in voting patterns. More Specifically, since the party is un- able to control the reward structure and hence the distri- bution of reward in payment for party cohesion, constituency interests become a more powerful determinant in legislative voting. The Republican Party either by reason of minority position or by its relative heterogeniety does not exert as great a pressure on legislative voting behavior. The Republican Party members are more rSSponsive to theinterests of their constituency which increases conflict when these interests are not in accord with party position. Another indication of the relative weakness of the minority party in the control of the voting behavior of its members is that when the issue does not have import for their constituency-~as in the case of the urban Republicans-- there is an increase of cross party voting. This was ex- plained as a result of the necessity of "making deals" 44 with the majority party to increase the possibility of the passage of other bills more important to their constituency. To be sure this could also be a result of the relative weakness of party position because of the heterogeniety of the party. The important fact, however, is that the con- stituency of the legislator in the minority party is a r factor in the extent to which the legislator adheres to ' party policy. Cohesion, therefore, seems to be a function of the degree of conflict SXperienced by the legislators. The #1 less the conflict the greater will be the tendency to vote according to party position. If there is a conflict be- tween party and constituency interests, the voting pattern will be less cohesive. Legislators will vote in accord with party position if the pressure for conformity is great enough and the ability to reward conformity is large enough. Thus when attempting to analyze legislative voting behavior, it is necessary to ascertain the relative ability of the party to reward conformity and the rigidity or strength of party position. Constituency interests seem meaningful only to the extent with which they provide in- sight into the variation in .begislative voting patterns. In other words, they are useful to explain the voting be- havior within the political parties but not between them. Thus cohesive voting or consensus within a parti- cular party is the usual feature of legislative voting behavior. In other words, integration characterizes the 45 party as a whole in that the prediction of votes can be fairly accurate merely through knowledge of party position. The legislators of a particular political party vote more like members of their own party then like members of the other party. As a whole, therefore, the legislative party achieves consensus among its members. To look at the vot- ing pattern of the entire legislature in terms of partisan activity appears to be a fruitful method of analysis. To assume integration without first attempting to show why and how the sys tem achieved this integration Opens the social scientist to the criticism that his assump- tions not only predetermined his results but also that he employs his assumptions as his eXplanation. It is neces- sary to Show why the consensus exists and how this con- sensus is maintained when employing integration as a con- ceptual framework. Conflict, on the other hand, can be used to ex- amine the variation in consensus. Given that political parties stress the reSponsibility of the individual members to adhere to party policy, consensus is valued. Analysis in terms of conflict can afford insight into the reasons for the relative lack or presence of consensus. However, as was indicated by the data conflict must be analyzed in the context in which it occurs. In legislative parties consensus is valued, but the ability of the party to reward is related not only to the larger legislative system but also to the party itself. Therefore, consensus and conflict 46 cannot be attributed to the system which is being analyzed but must first be located and then eXplained in teams of variation. Assuming that either or both phenomena are the defining concepts of social behavior is to explain in terms of the "givens" of the analysis. BIBLIOGRAPHY Beck, Kurt W. “Influence Through Social Communication" in Readings in Social ngchology ed. by Eleanor E. Maccoby, Theodore M. Newcomb and Eugene L. Hartley, iggw Yprk: Holt, Rinehart, and Winston, Inc., 1958) -19 . Becker, Robert W., Frieda L. Foote, Meltrias Lubega, and Stephen V. Monsma. “Correlates of Legislative VOting: Michigan House of Representatives, 1954-61“, Midwest Journal of Political Science, VI, (Nov. 1962 , 384-396. Dawson, Richard E. and James A. Robinson. “Interparty Competition, Economic Variables, and Welfare Policies in the American States", Journal of Politics, 25, (May, 1963), 265-289. Derge, David R. “Urban-Rural Conflict: The Case in Illinois" in Legislative Behavior ed. by John C. Wahlke and Heinz Eulau, (Illinois: The Free Press of Glencoe, 1959), 218-227. Epstein, Leon D. “Electoral Decision and Policy Mandate: An Empirical Example,“ Public Opinion Quarterly, XXVIII, (Winter, 1964), 564-572. Eulau, Heinz. “The Legislator and His District: Areal Roles“ in The Legislative System ed. by John C. Wehlke, Heinz Bulau, William Buchanon, and Leroy C. Ferguson, (New York: John Wiley and Sons, Inc., 1962) 287-310. Flinn, Thomas A. “Party RSSponSibility: Some Causal Factors,“ American Political Science Review, LVII, (March, 1964), 60-71. Froman, Lewis A. "Inter-Party Constituency Difference and Congressional VOting Behavior", American Political Science Review, LVII, (March, 1963 , 5 -6l. Jewell, Malcolm E. ”Party VOting in American State Legislatures", in L islative Behavior ed. by John C. Wahlke and Heinz Eulau, (Illinois: The Free Press of Glencoe, 1959), 121-132. 47 48 Jewell, Malcolm E. “The Political Setting”, in State Legislature in American Politics ed. by Alexander Hearp, (New Jersey: Prentice-Hall, Inc., 1966), 0-9 . Jewell, Malcolm E. and Samuel C. Patterson. The Legislative Process in the United States, (New YOrk: Random House, 1966). Kornhauser, William A. "Political Sociology“ in SociologyI ed. by Leonard Broom and Philip Selznick, (Illinois: Row, Peterson and Company, 1958), 558-599. Kroll, Morton. “Hypotheses and Designs for the Study of Public Policies in the United States“, Midwest Journal of Political Science, VI, (Nov., 1962) 363-375. Lipset, Seymour Martin. “Political Sociology“ in Sociolggy Today ed. by Robert K. Merton, Leonard.Broom and Leonard S. Cotterell, Jr., (New YOrk: The Free Press of Glencoe, 1959), 121-132. MacRae, Jr., Duncan. "The Relation Between Roll Call Votes and Constituencies in the Massachusetts House of Representatives", in gglitical Behaviog, ed. by Heinz Eulau, Samuel J. Eldersveld, and Morris Janowitz, (Illinois: The Free Press of Glencoe, ‘ 1956), 317-323. Rice, Stuart A. “Measuring Cohesion in Legislative Groups" in Legislative Behavior ed. by John C. Wahlke and Heinz Eulau,fi(Illinois: The Free Press of Glencoe, 1959). 372-377. Zeriski, Raphael. “Party Factions and Comparative Politics: Some Preliminary Observations“, Midwest Journal of Political Science, IV, (Feb., 1960) 27-51. MICHIGAN STATE UNIV R ITY Hll IIIIIIH 1 3 LIBRARIES 293 03056 0605