¹ Notable among studies having interactional focus are: Lazarsfeld, P. F., The People's Choice (New York: Columbia University Press, 1948)Google Scholar and Berelson, B., Voting (Chicago: University of Chicago Press, 1954).Google Scholar The shift to attitudinal explanation is marked by the work of the Michigan theorists, especially Campbell, A. et al. , The American Voter (New York: Wiley, 1960).Google Scholar

² Campbell, et al. , The American Voter, p. 16Google Scholar, Chaps. 1 and 2.

³ The Michigan theorists specifically deny the importance of personal interaction as an influence on voting behaviour. See Campbell, et al. , The American Voter, p. 76, p. 297Google Scholar, and Chap. 12.

⁴ Galtung, J. S., Theory and Methods of Social Research (London: Heinemann, 1968), p. 15.Google Scholar

⁵ Jennings, H. H., Sociometry in Group Relations (Washington: American Council of Education, 1948), p. 11.Google Scholar The best general overviews of the sociometric method are Moreno, J. L., Who Shall Survive? (New York: Beacon House, 1953)Google Scholar and Moreno, J. L., ed., The Sociometry Reader (New York: Glencoe Free Press, 1960).Google Scholar

⁶ The earliest description of this phenomenon appears in Tingsten, Herbert, Political Behaviour (London: P. S. King, 1937), pp. 126–7.Google Scholar Other studies include Bealey, F. et al. , Constituency Politics (London: Faber, 1965), pp. 179–86Google Scholar; Lane, R. E., Political Life (New York: Glencoe Free Press, 1959). pp. 261–4Google Scholar; Putnam, R. D., ‘Political Attitudes and the Local Community’, American Political Science Review, L (1966), pp. 640–54CrossRefGoogle Scholar; Foldare, I. S., ‘The Effect of Neighbourhood on Voting Behaviour’, Political Science Quarterly, LXXXIII (1968), pp. 516–29CrossRefGoogle Scholar; Katz, D. and Eldersfeld, J. S., ‘The Impact of Local Party Activity upon the Electorate’, Public Opinion Quarterly, XXV (1961), pp. 1–24.CrossRefGoogle Scholar

⁷ The suggestion has been made that the phenomenon is general among all social classes, but little empirical data has been offered.

⁸ Blondel, J., Voters, Parties and Leaders (Harmondsworth: Penguin, 1966), p. 65.Google Scholar

⁹ The most explicit statement of these two models can be found in Putnam's article cited above.

¹⁰ For example Tingsten found that women, whom he assumed spent more time in their neighbourhood, were more responsive to the neighbourhood effect than were men; Tingsten, , Political Behaviour, pp. 170–2.Google Scholar

¹¹ See Litwak, E., ‘Primary Group Structures and their Functions’, American Sociological Review, XXXIV (1969), pp. 465–81;CrossRefGoogle Scholar and Mitchel, O. G., Neighbourhood and Community (Liverpool: Liverpool University Press, 1951), pp. 39–45.Google Scholar

¹² For example Bealey suggests both that individuals may be ‘influenced by the prevailing patterns of voting in the areas in which they live’, and that ‘socially homogeneous communities seem to put pressures on voters’, without apparent awareness of the difference or that the response mechanism must differ in the two cases; Bealey, et al. , Constituency Politics, pp. 180 and 185.Google Scholar

¹³ See Royal Commission of Local Government in England, Research Studies 9, Community Attitude Survey (London:HMSO, 1969), p. 15Google Scholar and Hampton, W. A., Democracy and Community (Oxford:Oxford University Press, 1970), p. 106Google Scholar

¹⁴ I.e., between 21 and 26th May 1970 and following 18th June.

¹⁵ Two residents of Archer Street could not be contacted.

¹⁶ In a small section of Nelson Street, houses had small front gardens measuring five feet from door to pavement. All the objective middle-class respondents lived in these houses.

¹⁷ The choice of streets of this kind was made on the a priori assumption that neighbourhood community was more likely to develop in such an environment as compared to streets with more dispersed housing. The choice was also conditioned by the desire to conduct research on a clearly finite canvas.

¹⁸ The choice of streets with such a social composition was deliberate and motivated by the evidence from previous studies that neighbourhood might be more salient for this group than others.

¹⁹ Source: unpublished data from the Partial Census1966.Google Scholar

²⁰ Source: unpublished Census data.

²¹ Source: Partial Census 1966, Parliamentary Constituency Tables (London: HMSO, 1969).Google Scholar This percentage is the lowest of all the Manchester constituencies.

²² Source: unpublished data from the Partial Census 1966.Google Scholar

²³ Source: unpublished Partial Census data and Partial Census 1966, County Tables, Lancashire (London: HMSO, 1968).Google Scholar

²⁴ The inter-election swing was 3.7 per cent in Withington and 4.2 per cent in Ashton (to the Conservative Party in both cases). The statistic is of course not directly comparable because of the difference in time scale; Butler, David and Pinto-Duschinsky, Michael, The British General Election of 1970 (London: Macmillan, 1971), pp. 362 and 365.Google Scholar

²³ For example, in the North of England as a whole (on the evidence of aggregated NOP Surveys conducted between 1963 and 1966) 63.5 per cent of the working class expressed a preference for the Labour Party; recalculated from data in Table 6.14, Butler, David and Stokes, Donald, Political Change in Britain (London: Macmillan, 1969), pp. 140–1.Google Scholar

²⁶ Bealey, et al. , Constituency Politics, p. 183.Google Scholar

²⁷ Except the Indian couple.

²⁸ The question was not asked on the Nelson Street schedule.

²⁹ This antagonism was expressed in response to the ‘sociometric test’ even though this was constructed in a positive form: ‘Now I wonder if you could tell me which of your neighbours you are most friendly with?’

³⁰ For a listing of ‘social trust’ questions see Almond, G. and Verbam, S.The Civic Culture (Princeton: Princeton University Press, 1963), Appendix B, p. 534.CrossRefGoogle Scholar The questions were not asked in Nelson Street.

³¹ Hampton discovered that as length of residence increases so does size of perceived neighbourhood, evidence which gives colour to this data; Hampton, , Democracy and Community, p. 106.Google Scholar It might be thought that the variation in length of residence was a reflection of the age structures of the streets. This was not so. When the streets’ populations were dichotomized between those older and those younger than 40 it was found that 50 per cent of Caesar Street's inhabitants were over 40, 52 per cent of Archer's and 63 per cent of Nelson Street's.

³² Comment by respondent R26.

³³ The question tapping subjective assessment of class was unstructured. The numbers designate respondents and are used in the street diagram above and the sociograms. The suffix A refers to the wife in a marital household.

³⁴ On both the pre- and post-election surveys, respondents were asked whether they had discussed politics with ‘anyone apart from your family’ recently.

³⁵ Respondents were asked whether they would avoid ‘talking politics with anyone’ and then asked to explain their answer.

³⁶ The term was coined by R. K. Merton; see Katz, E. and Lazarsfeld, P. F., Personal Influence (Chicago: Free Press, 1955), pp. 59–61.Google Scholar Moreno used the term ‘tele’ to describe the same phenomenon.

³⁷ The effect of distance on group formation has been examined in detail by Festinger; Festinger, L. et al. , Social Pressures in Informal Groups (London: Tavistock, 1959)Google Scholar, Chap. 3.

³⁸ The three items which respondents were asked to agree or disagree with were as follows: the trade unions have too much power in the country; the men who own the big businesses have too much power in this country; the upper classes in Britain have always tried to keep the working classes from getting their fair share.

³⁹ A detailed study of attitudes and voting to be found in Lipset, S. M., Union Democracy (New York: Anchor, 1956), Chap. 16Google Scholar, indicated that there could be major variation in the correlation between attitude and vote depending on the favourableness of the individual's immediate social context. It is argued here that attitudes themselves may also be modified by the social context.

⁴⁰ Comment by respondent R43 in answer to the question ‘Would you say that all the people round here are of the same social class as yourself?’

⁴¹ Unlike Caesar Street there was little evidence of variation between the supporters of the different parties on this dimension.

⁴² In comparison all respondents in Caesar Street described themselves as working class (using that term) in response to the unstructured question and none had to be prompted.

⁴³ This raises two possibilities with relevance to the street environment. It could be argued that political interest generated on the street, though having low partisan content, spilt over into other friendship relationships where partisan discussion was less inhibited. Secondly, it could be argued that fortuitously the inhabitants of this street were more interested in politics unrelated to street experience, but that this interest spilt over into neighbourhood contacts. The first argument seems the most plausible.

⁴⁴ The nine respondents who perceived political discussion as presenting a threat were R39A, R41, R43A, R44/44A, R46/46A, R48 and R49.

⁴⁵ R51A gave her reasons for change as support for Powell and reaction to rising prices.

⁴⁶ He became aggressive when the question was asked, ‘the upper classes in Britain have always tried to keep the working classes from getting their fair share.’

⁴⁷ Comment by respondent R2.

⁴⁸ Respondent Rio, a lady of 74, had lived in the street since birth.

⁴⁹ She also avoided discussion with R 10.

⁵⁰ Fortuitously, as a result of research on another project it is known that the employees of the firm he worked for were also almost wholly Labour voters.

⁵¹ In part he said ‘It's a funny thing – I kept myself to myself in Guide Bridge, but I often go drinking with the people in… (specifying R14/14A)’.

⁵² This raises the intriguing possibility, when considered in the light of other data presented, that street communities may have life cycles in which stability is achieved after all potential change has taken place; the cycle then being renewed when new inhabitants are incorporated. The relative position of Archer and Caesar Street on such a cycle is apparent.

⁵³ Galtung, , Theory and Methods of Social Research, p. 150.Google Scholar