A Taxonomic Approach To State Political Party Strength

• DOI: 10.1177/106591295801100305
• Date: 01 September 1958
• Published date: 01 September 1958
• Author: Robert T. Golembiewski
• Subject Matter: Articles

A TAXONOMIC APPROACH TO STATE POLITICAL
PARTY STRENGTH
ROBERT T. GOLEMBIEWSKI
Princeton University
INTRODUCTION
HE RECENT LITERATURE of American politics has displayed a
significant interest in the mine of comparative data which the several
states constitute. This is not to say that this mother lode of informa-
tion has lain virgin all these years under the unseeing eyes of academic sour-
doughs. Obviously this has not been the case,1 although the comparative
potentialities of the states certainly have not received the attention they
merit.
A large number of recent studies, however, seem to constitute a de-
parture from the usual scholarly approaches. Indeed, many of the writers
would separate themselves from the other students of the political systems
of American states on two grounds. The first point of isolation is the area
of concern. &dquo;[Previously] students of state government have found it
easier,&dquo; one recent contributor politely phrased the difference in scope, &dquo;to
make meaningful comparisons of administration or legislative organization
than to capture on paper the dynamic political forces which appear to be
unique in each state....&dquo; 2 The second point of isolation is the method
employed. In brief, the form of the comparative approach -
that is, the
comparison of phenomena common in some measure to two or more similar
units -
has been retained, but the language in which the form is increas-
ingly being applied is the language of quantification.3
3
To put the matter
baldly without misrepresenting the essential change, the approach cham-
pioned by Lowell 4 and Rice,5 after enjoying some popularity in the first
three decades of the century has succeeded in once again luring some
1
See, for example, James W. Fesler, The Independence of State Regulatory Agencies (Chi-
cago : Public Administration Service, 1943).
2
Malcolm E. Jewell, "Party Voting in American State Legislatures," American Political
Science Review, XLIX (September, 1955), 733.
3
See, for example, the emphasis in Interuniversity Summer Seminar on Political Behavior:
Social Science Research Council, "Research in Political Behavior," American Political
Science Review, XLVI (December, 1952), 1003-45.
4
A. Lawrence Lowell, "The Influence of Party upon Legislation in Europe and America,"
Annual Report of the American Historical Society for the Year 1901, I (1902), 319-542.
5
Stuart Rice, Quantitative Methods in Politics (New York: Crofts, 1928), and "The Identi-
fication of Blocs in Small Political Bodies," American Political Science Review, XXI
(August, 1927), 619-27.
6
Herman C. Beyle, Identification and Analysis of Attribute-Cluster-Blocs (Chicago: Uni-
versity of Chicago Press, 1931), is probably the most advanced of such studies. See
also the studies which Beyle cites, pp. 4-21.
494

---

495
students away from the &dquo;group approach&dquo; sired by Bentley7 and from the
approaches represented by men like Graves,8 Sait,9 and Anderson.1o
If the recent work is a function of a growing feeling among students that
previous approaches have failed to meet scholarly expectations, it is not
obvious that the new re-emphasis has been much more rewarding thus far.
If one examines recent studies in regard to the area of concern, one can
hardly claim that many, or even a goodly number, of the quantifiable
elements in the lives of the states have been included within its scope. In
fact, work has been concentrated on two rather narrow bands of the com-
parative spectrum.
One of these bands consists of the state &dquo;party cohesion&dquo; studies which
were pioneered at the turn of the century by Lowell and later somewhat
improved by Rice, which have lately made a reappearance in force in the
literature.&dquo; In general, these new studies have not yet expanded the scope
of the work of the technique’s pioneers, 12 although the possible relation of
levels of party cohesion to sociological, 13 psychological, and other classes of
variables14 clearly demands investigation.
The second of these bands of study-concentration may be more or less
arbitrarily labeled &dquo;interparty competition&dquo; or &dquo;party strength&dquo; studies, a
7
Arthur F. Bentley, The Process of Government (Chicago: University of Chicago Press,
1908). The "group approach" inspired many students, who left behind a spate of
national and state pressure group studies. See, for example, Dayton D. McKean, Pres-
sures on the Legislature of New Jersey (New York: Columbia University Press, 1938);
and Belle Zeller, Pressure Politics in New York (New York: Prentice-Hall, 1937).
8
W. Brooke Graves, American State Government (Boston: Heath, 1953).
9
Edward McC. Sait, American Parties and Elections (New York: Appleton-Century, 1939).
10
William Anderson and Edward Weidner, State and Local Government in the United
States (New York: Holt, 1951).
11
Jewell, op. cit.; W. Duane Lockard, "Legislative Politics in Connecticut," American
Political Science Review, XLVIII (March, 1954), 166-73; William J. Keefe, "Parties,
Partisanship and Public Policy in the Pennsylvania Legislature," American Political
Science Review, XLVIII (June, 1954), 450-64, and his "Party Government and Law-
making in the Illinois General Assembly," 47 Northwestern U. L. Rev. (March, 1952),
55-71, are representative examples.
12
The re-emphasis is narrower than the scope of Rice’s work. See the broad applications
of "group cohesion" and "likeness" he attempted in his "Farmers and Workers in
American Politics," Columbia University Studies in History, Economy and Public Law,
Vol. CXXIII (No. 2, 1924).
13
On the national level, the late Julius Turner’s Party and Constituency: Pressures on
Congress (Baltimore: Johns Hopkins University Press, 1952), has attempted to
broaden the scope of the legislative cohesion study. Turner related the degree of
cohesion on various roll calls to two broad categories of pressures: party (the party
label) and constituency (metropolitan-urban-rural, per cent foreign and native born,
and geographic section). A similar procedure is clearly applicable to state cohesion
studies. But the literature does not go beyond Jewell’s aside that the "degree of party
voting appears to be significantly higher in those two-party states which are larger
and more urban." Op. cit., p. 791.
14
Keefe in his studies cited above did relate levels of party cohesion to various classes of
legislation, a not uninteresting but limited use of a variable in cohesion study.

---

496
number of which have appeared recently. 15 With some notable exceptions,
these studies have concentrated on the development of classifications of
party strength which, while not uninteresting in themselves, have not yet
been significantly extended beyond the initial taxonomic stage. 16 As was
the case with the scope of the cohesion studies, possible links between the
various interparty competition classifications and the broader range of
potential variables -
sociological, psychological,17 and others18 -
have not
in general been the objects of investigation.
Examimng the literature of state politics in regard to the second point of
isolation, the method employed, one finds no significant improvement in the
general methodology of cohesion studies since Rice’s work ’19 while in
general the party strength studies cannot be said to be characterized by re-
fined techniques.2° This estimate may seem harsh. Undeniably, progress in
methodology should not be measured against the standard of a colossal
differential equation which completely sums up all political experience.
Admittedly the results of initial studies -
both in terms of methodology
and uniformities isolated -
can legitimately be expected to do little more
than prepare the rough problem-terrain for more careful landscaping by
later students. But it is not unfair to demand that subsequent studies do
15
Most clearly in this class are: Austin Ranney and Willmoore Kendall, "The American
Party Systems," American Political Science Review, XLVIII (June, 1954), 477-85;
Belle Zeller (ed.), American State Legislatures (New York: Crowell, 1954), pp. 200
ff.; and Joseph A. Schlesinger, "A Two-Dimensional Scheme for Classifying States
According to Degree of Inter-Party Competition," American Political Science Review,
XLIX (December, 1955), 1120-28.
16
Ranney and Kendall and Schlesinger are quite clear that their classificatory work is the
first step in analysis, but their work has thus far not been extended beyond that first
step.
17
A recent study by Warren E. Miller, "One Party Politics and the Voter," American
Political Science Review, L (September, 1956), 707-25, has shown how survey ma-
terial may be used in combination with interparty competition classifications to test
the relation of the classifications and certain psychological variables.
18
Zeller, American State Legislatures, pp. 200 ff., has in effect bridged the gap between the
party cohesion and the interparty competition studies. She attempted to relate rough
measures of "party cohesion" and "pressure politics" to her classification of party
strength. This extends the approach somewhat. See also below, footnote 33 and the
corresponding context.
19
Lowell or Rice’s methodology is generally used. See Jewell, op. cit., p. 774; Lockard, op.
cit., p. 168; and Keefe, "Parties, Partisanship and Public Policy in the Pennsylvania
Legislature," pp. 450-52, and "Party Government and Law-making in the Illinois
General Assembly," pp. 57-58.
Lowell used the percentage of "party votes," roll...