¹ This approach to the empirical study of representation has been suggested by Wahlke, John in his “Public Policy and Representative Government: The Role of the Represented,” a paper prepared for the Seventh World Congress of the International Political Science Assocıation, Brussels, Belgium, 09, 1967 Google Scholar, and published as Report #9 from the Laboratory for Political Research, University of Iowa, 1967.

² Symbolic or expressive satisfactions are discussed at length in Edelman, Murray, The Symbolic Uses of Politics (Urbana: University of Illinois Press, 1964)Google Scholar.

³ This sample is also discussed in an earlier report by the author, “Correlates and Consequences of Beliefs in the Legitimacy of Regime Structures,” Midwest Journal of Political Science (August, 1970).

⁴ Wahlke, “Public Policy and Representative Government …”

⁵ Wahlke, , “Public Policy and Representative Government…,” p. 21 Google Scholar.

⁶ This research is summarized in Wahlke, “Public Policy and Representative Government…”

⁷ Expressive satisfactions are certainly of some importance also: for example, see Anton, Thomas J., “Roles and Symbols in the Determination of State Expenditures,” Midwest Journal of Political Science.” (1967)CrossRefGoogle Scholar.

⁸ Wahlke, , “Public Policy and Representative Government …,” p. 33 Google Scholar.

⁹ Warren Miller and Donald Stokes' model of the relationships between constituency policy preferences and legislators' roll-call behavior permits a statistical estimate of such congruence; see Miller, Warren E. and Stokes, Donald E., “Constituency Influence in Congress,” this Review, 57 (1963)Google Scholar; and Miller, Warren E., “Majority Rule and the Representative System of Government,” in Allardt, Erik and Littunen, Yrjo (eds.), Cleavages Ideologies and Party Systems, Vol. 10 of Transactions of the Westermarck Society (Finland, 1964)Google Scholar. However, data from the same study suggest that most citizens would hardly be aware of any congruence that might obtain statistically; see Stokes, Donald E and Miller, Warren E., “Party Government and the Saliency of Congress,” Public Opinion Quarterly, 26 (1962)CrossRefGoogle Scholar.

¹⁰ This does not imply that the behavior of the representative is of any diminished importance. The relationship between representative and represented is still the phenomenon of interest; and the behavior of the representative is, of course, an integral component of that relationship. Focusing on the represented simply means that the representative-represented relationship is investigated in terms of the represented's perceptions.

¹¹ Davidson, Roger H., “Congress and the Executive: The race for Representation,” in de Grazia, Alfred (ed.), Congress: the First Branch of Government (New York: Doubleday & Company, Inc., Anchor Books Edition, 1967), p. 373 Google Scholar.

¹² Prewitt, Kenneth and Eulau, Heinz, “Political Matrix and Political Representation: Prolegomenon to a New Departure from an Old Problem,” this Review, 63 (1969), p. 441 Google Scholar.

¹³ See Eulau, Heinz, Wahlke, John C., Buchanan, William, and Ferguson, Leroy C., “The Role of the Representative: Some Empirical Observations on the Theory of Edmund Burke,” this Review, 53 (1959)Google Scholar; and, of course, Wahlke, John C., Eulau, Heinz, Buchanan, William, and Ferguson, Leroy C., The Legislative System (New York: John Wiley & Sons, Inc., 1962)Google Scholar.

¹⁴ The quote is from Prewitt, and Eulau, , “Political Matrix…,” p. 430 Google Scholar.

¹⁵ Prewitt, and Eulau, , “Political Matrix…,” p. 430 Google Scholar.

¹⁶ A comprehensive history of the development of direct representation theory, with particular emphasis on the American setting, is in de Grazia, Alfred, Public and Republic (New York: Alfred A Knopf, 1961)Google Scholar.

¹⁷ Representation does have an identifiable meaning, which is, in Hanna Pitkin's words, “that in representation something not literally present is considered as present in a nonliteral sense.” Pitkin, Hanna Fenichel, The Concept of Representation (Berkeley; University of California Press, 1967), p. 9 Google Scholar. However, the substance of this meaning can be defined in various different ways. For an explication of the various ways in which the concept of representation has been applied to different contexts, see Pitkin, The Concept of Representation, passim; also see the Introduction to Pitkin, (ed.), Representation (New York: Atherton Press, 1969)Google Scholar.

¹⁸ Of course, as Wahlke notes, such an approach has been implied in a variety of ways by various analysts. See the discussion in Wahlke, , “Public Policy and Representative Government…” pp 29–33 Google Scholar.

¹⁹ Easton, David, A Systems Analysis of Political Life (New York: John Wiley & Sons, Inc., 1965), Ch. 17Google Scholar.

²⁰ Easton, A Systems Analysis …, chs., 11–13.

²¹ Easton, , A Systems Analysis…, p. 278 Google Scholar.

²² Easton, , A Systems Analysis…, p. 268 Google Scholar.

²³ Easton, , A Systems Analysis…, p. 268 Google Scholar.

²⁴ The notion of a scope sample is discussed in Willer, David, Scientific Sociology (Englewood Cliffs: Prentice-Hall, Inc., 1967), Ch. 6Google Scholar. The discussion presented in the following paragraphs relies on his explication.

²⁵ The quotes are from Willer, , Scientific Sociology, p. 114 and p. 113, respectivelyGoogle Scholar.

²⁶ A preliminary report on these data, written shortly after the fieldwork was completed, appeared under the title “Political Stability and the Habit of Legitimacy,” Report #23 from the Laboratory for Political Research, University of Iowa, 1968. The present analysis superscedes that report, which suffered from various conceptual and methodological inadequancies and general lack of refinement.

²⁷ Legitimacy beliefs are the focus of analysis in Muller, “Correlates and Consequences…”

²⁸ On intensity of affect toward Congressman, Congress, and the Supreme Court, approximately one-fifth of the respondents scored high or low, three-fifths in the middle (+ or −2); intensity of affect toward President Johnson was distributed two-sixths high, one-sixth low, three-sixths in the middle.

²⁹ The intervals along this continuum have been labelled, for consistency and as a mnemonic device: Represented; + Unrepresented; Arepresented; −Unrepresented; Nonrepresented. These are not particularly felicitous terms; they are intended simply as shorthand references to operationally defined characteristics.

³⁰ “Referential symbols are economical ways of referring to the objective elements in objects or situations; the elements identified in the same way by different people. Such symbols are useful because they help in logical thinking about the situation and in manipulating it … Condensation symbols evoke the emotions associated with the situation. They condense into one symbolic event, sign, or act patriotic pride, anxieties, remembrances of past glories or humiliations, promises of future greatness, some one of these or all of them.” Edelman, , The Symbolic Uses of Politics, p. 6 Google Scholar. A legislator's vote on a civil rights bill is a referential symbol. Party affiliation for, most citizens, is one of the most salient condensation symbols in political life. It is an economical way, not of referring to the objective elements in objects or situations, but of organizing a whole congeries of basically abstruse and uninteresting information into a coherent entity which has some meaning for the individual whose major concerns in life are not political. The difference between referential and condensation symbols is on the order of more or less; yet it would appear to be of sufficient clarity to permit relatively uniform differentiation between responses indicating perception of outputs characterized mainly by a high degree of referential content, by a high degree of condensation content, or by a high degree of extraneous content.

³¹ For example, the following categories were employed for coding the range of perceived stimuli with respect to Congressman:

Perception of Instrumental Outputs. (In each category, some policy content is implied or specified.)

I favor (do not favor) his position or voting behavior on political issues.

He has not carried out his campaign promises.

He has worked for needed projects in the state or district.

He has had many legislative acts passed which have benefitted his district.

His voting behavior has emphasized concern with the problems both of his district and of the nation.

His voting behavior has emphasized national interests above local interests.

His voting behavior has represented (not represented) the majority feelings of his district.

He has voted with his party in Congress.

His voting behavior is too liberal (conservative).

He has served the nation as a watchdog over the economy, defense establishment, etc.

His interests are strongly urban (rural) and he has little understanding of rural (urban) problems.

Perception of Expressive Outputs. (In each category, policy content is unspecified).

He is a member of the same party as I am; he is a member of the other party.

He is a liberal (conservative)—socialist (fascist).

He communicates (does not communicate) often with the people in his district.

He has done a personal favor for me.

Unlike his predecessor, he has been able to get the cooperation of many of the people in his district.

He performs his job well: e.g., he has introduced important legislation; he speaks out well on important problems.

He does not perform his job well: e.g., he is too concerned with petty problems.

He is intelligent and capable: e.g., he went to Harvard; he is a lawyer.

His background suits him for our district: e.g., he is a member of the farm bureau.

He is concerned with great national problems.

He does not follow the party line, but stands up for his own opinion even if it is unpopular.

He is a pawn of those who contributed to his party fund—not independent enough of special interests.

He is carrying out (not carrying out) my beliefs.

He has been an active member of his party in national politics.

He supports (opposes) the present administration.

Perception of Extraneous Outputs.

He has a pleasant (unpleasant) personality—he is a friendly (unfriendly) man.

I like (dislike) him because of the influence of family and friends.

He keeps getting reelected so he must be doing a good job—he does not do anything.

I have not heard many favorable (unfavorable) remarks about him.

He is an admired (not an admired) man.

³² Unfortunately, screening questions comparable to the Congressman item (e.g., “Have you been following the activities of Congress lately?”) were not included in the other question-sets. Thus, the assumption must be made that respondents who selected the “Don't Know” option when requested to make a summary evaluation, did so because they were unaware of the performance of that object. In general, examination of the protocols indicates that this assumption is tenable; however, particularly in the case of President Johnson (judging from comments written in the margins), some respondents who selected the “Don't Know” option did so because their evaluation was neutral, not because they were unaware of the performance of the object.

³³ Three cases could not be coded because respondents reported “favorable” evaluations with the qualification “in terms of the interests of the ruling class,” or in terms of the interests of Capitalism [Imperialism].”

³⁴ See Stokes, and Miller, , “Party Government…,” p. 540 Google Scholar; the data reported at p. 340 of Malcolm E. Jewell and Samuel C. Patterson, The Legislative Process in the United States; Eldersveld, Samuel J., Political Parties: A Behavioral Analysis (Chicago: Rand McNally & Company, 1964), p. 493 Google Scholar.

³⁵ See Stokes and Miller, “Party Government…”

³⁶ Stokes and Miller, “Party Government…”

³⁷ See Cronbach, L., Essentials of Psychological Testing (New York: Harper & Row, 2nd ed., 1960)Google Scholar.

³⁸ See Guilford, J. P., Fundamental Statistics in Psychology and Education (New York: McGraw-Hill Book Company, 4th edition, 1965), pp. 308–317 Google Scholar.

³⁹ The formulais: (η ² − r²) (N − k)/(1 − η ²) (k−2)

⁴⁰ A good introductory discussion of this procedure appears in Ferguson, George A., Statistical Analysis in Psychology and Education (New York: McGraw-Hill Book Company, 2nd edition, 1966), Ch. 21Google Scholar. For a more comprehensive treatment see Hays, William L., Statistics (New York: Holt, Rinehart and Winston, 1963), esp. pp. 539–562 Google Scholar The computing formula used here is that suggested by Hays.

The use of orthogonal polynomial weights, and of course the use of linear regression, is based on the assumption of equal intervals. One good reason for assuming interval measurement, even if such is only approximately the case, is that considerably more information can be gained from the data. Also, even if the data is “truly” ordinal, order-based measures of association leave much to be desired. Gamma always overestimates the degree of association, and is a measure of one-way association rather than co-relation. Tau-B and Tau-C have no simple interpretation; also, like r, they range between +or − 1 only if the bivariate distribution is perfectly congruent. Interval statistics have been used in this analysis because it was felt that they would best describe the data.

A word is also in order with respect to the use of tests of statistical significance in this analysis. Strictly speaking, they do not apply to this sample. But, they do serve as a useful—if arbitrary—guide to the acceptance of some relationship, however slight; or to the rejection of a relationship, as being too slight to matter, in terms of what might be expected from further replication.

⁴¹ A 4-point General Political Knowledge scale was constructed from the sum of correct responses to the following three items:

(1) Do you happen to remember the names of any of the Chairmen of the Congressional Committees?

(2) Which political party elected the most members to the House of Representatives in 1966?

(3) If you remember how many justices there are on the Supreme Court, write the number in the following space.

In order to measure the respondent's awareness of Congressional policy-making activities, a list of eleven policy items was posed and respondents were asked to report whether Congress passed or did not pass each item during its 1967 session (e.g., “a bill authorizing the appropriations requested by the President for Rent Supplements,” etc.); a similar procedure was followed in order to measure the respondent's awareness of Presidential policy-making, in this case, respondents being asked to report whether President Johnson did or did not propose each item (e.g., “a resolution favoring the principle of greater economic assistance to Latin American,” etc.). The items were developed from sections reporting the policy activities of the Presidency and the Congress in the Spring, 1968 Congressional Quarterly Guide to Current American Government. There was no overlap between the Congressional and Presidential policy items. The resulting Congressional Policy Knowledge and Presidential Policy Knowledge scales were 4-point scales based on quartiles of correct responses. The measure of interest in politics is a 4-point scale based on respondents' reports of how often they read about public affairs in magazines: “Nearly Every Day” through “Very Rarely.”

⁴² See Harmon, Harry H., Modern Factor Analysis (Chicago: University of Chicago Press, 2nd edition, 1967), pp. 350–360 Google Scholar.

⁴³ Since unequal numbers of cases are located at each interval on these sense of representation scales, only one trend component can be tested by the procedure of orthogonal polynomials. However, this is no constraint here, because the purpose is to test only one trend component (of a higher order than the linear); the possible power of any cubic and quartic trends can be assessed from inspection of the other trends mean square—but there is no need to separately test such trends. Trend analysis with unequal n's is discussed at pp. 351–352 of Ferguson, Statistical Analysis…

⁴⁴ Easton, , A Systems Analysis…, p. 156 Google Scholar.

⁴⁵ A useful discussion of these dimensions is in Gurr, Ted, “Psychological Factors in Civil Violence,” World Politics, 20 (1968), 272–277 CrossRefGoogle Scholar.

⁴⁶ These dimensions are discussed in Chs. 18 and 19 of Easton, A Systems Analysis…

⁴⁷ See Dahl, Robert A., “The American Oppositions: Affirmation and Denial,” in Dahl, Robert A. (ed.), Political Oppositions in Western Democracies (New Haven: Yale University Press, 1966)Google Scholar.

⁴⁸ The construction of these variables is discussed at length, and the exact wording of the questions reported in Muller, “Correlates and Consequences…”

⁴⁹ See, for example, Easton, David and Hess, Robert D., “The Child's Political World,” Midwest Journal of Political Science, 6(1962)CrossRefGoogle Scholar; Easton, David and Dennis, Jack, “The Child's Image of Government,” 361 The Annals, (1965)Google Scholar.

⁵⁰ See Lane, Robert E., Political Ideology (New York: The Free Press, 1962), p. 146 Google Scholar; Easton, and Dennis, , “The Child's Image of Government,” pp. 48–49 Google Scholar.

⁵¹ See Muller, “Correlates and Consequences…”