Skip to content

Professor Ron JohnstonM.A.(Manc.), Ph.D. (Monash), LL.D.(Monash), D.U.(Essex), D.Litt(Sheff.), D.Litt(Bath), AcSS, F.B.A., OBE, BA (Hons) Manchester

Professor of Geography

Ron Johnston

Professor Ron JohnstonM.A.(Manc.), Ph.D. (Monash), LL.D.(Monash), D.U.(Essex), D.Litt(Sheff.), D.Litt(Bath), AcSS, F.B.A., OBE, BA (Hons) Manchester

Professor of Geography

Member of

Research interests

Ron’s academic work has focused on political geography (especially electoral studies), urban geography - much of the work in those two fields involving innovative use of multivariate statistical methods -  and the history of human geography.

Electoral studies

In this field, Ron has published many innovative analytical and empirical studies of electoral systems and voting behaviour, with particular reference to the United Kingdom, New Zealand, and the United States. Initial work was done while he was still in New Zealand, but the work flourished after his return to England in 1974 in a productive partnership with Peter Taylor with whom he wrote The Geography of Elections (1979). Major contributions have been made in five main areas:


Geographical variations in voting patterns. In the 1960s and 1970s the conventional wisdom in British voting studies was that partisan choice was predominantly influenced by social class – all else was ‘embellishment and detail’; there was no significant, country-wide spatial variation in support for the various political parties that could not be accounted for by the geography of class. Adapting a mathematical method developed for transport studies (entropy-maximising), Ron and Alan Hay devised a means of deriving maximum likelihood estimates of support for the main political parties across British constituencies, and of variations in the flow-of-the-vote there between elections. These made it very clear that there must be very wide variations across the country in which type of people voted for which party, and where; the results were published in a large number of papers and two monographs – The Geography of English Politics (1985) and (with Charles Pattie and Graham Allsopp) A Nation Dividing? (1998). The method was later applied to studies of ticket-splitting in the United States and of voting in the new (MMP/AMP) electoral systems adopted in New Zealand, Scotland and Wales in the 1990s; the mathematical basis of the method was extended with colleagues from the University of Mannheim.

Neighbourhood effects. One widely-discussed feature of voting patterns at local scales is the neighbourhood effect, a particular example of a contextual effect whereby people vote for a candidate or party because of the influence of neighbours. In his earliest work in electoral studies, Ron identified one type of such an effect in voting for local governments in New Zealand – the friends-and-neighbours effect, with candidates performing better close to their home than they did elsewhere. Evaluating the argument in the United Kingdom was difficult because of the paucity of relevant, micro-scale, data, but this was eventually overcome when he Danny Dorling, Charles Pattie and David Rossiter, assisted by Helen Tunstall and Iain MacAllister, developed the use of ‘bespoke neighbourhoods’. These were constructed by combining survey and census data to create a unique area (with a set radius or minimum population) around the home of each survey respondent, using census small-area data to characterize the population there. With such data it was possible to test for contextual effects whereby more people voted for a particular party in an area where its supporters dominated than would otherwise be expected. The results were very convincing – there is greater spatial concentration of support for a party than there is of the types of people who generally vote for it, largely because people who socialize with their neighbours tend to vote like them: the results were confirmed using more sophisticated modeling strategies in a large research project with Kelvyn Jones, Simon Burgess, Carol Propper, Rebecca Sarker and Anne Bolster. Such bespoke neighbourhood contextual data now accompany the survey data obtained by the regular British Election Studies, Ron having overseen the first such exercise in 2005.

Local campaign spending and election outcomes. Another item of conventional wisdom regarding British voting patterns in the 1960s-1970s was that local campaigns had no impact on general election results: the national campaign was the predominant determinant. In the late 1970s, Ron pioneered the use of candidates’ campaign spending returns as an index of constituency-level campaign intensity to demonstrate a clear relationship between the amount spent and the outcome; the more that candidates spent during the official campaign period, the better their performance there, and the poorer their opponents’. This argument was promoted in a number of papers, in The Geography of English Politics, and in a specialist monograph – Money and Votes (1987); later studies of more recent elections have substantially extended the findings. The argument was slowly taken up by others, using a range of indicators of campaign intensity alongside the amount spent, and is now generally accepted as a clear description of how parties and their candidates operate.

From 2005 on, these analyses of spending data have been combined with data from large panel studies of voters to identify where parties concentrate their campaigns and whether their targeting of particular voters there has a significant impact on the electoral outcome. Results show that those contacted by a party - especially face-to-face - are more likely to vote for it, and that such contacts are greater in volume the more that is spent on the campaign locally.

Associated research has linked the amounts spent on local campaigns to the financial health of constituency political parties and the amounts they receive in grants from the central party organisations. The healthier the local party, the more that is spent and the better the electoral performance. A book summarising these findings (with Charles Pattie) will appear in spring 2014 (Money and Electoral Politics.)


Boundary Commissions. In the 1970s, as part of a more general investigation of the modified areal unit problem (a key issue in statistical spatial analysis using aggregate data for defined areas), Peter Taylor and Graham Gudgin had shown that different election results could be produced in a city depending on how the constituency boundaries were drawn, even where this was undertaken by independent, non-partisan commissions with no possibility of gerrymandering (as in the UK). With David Rossiter, Ron built a computer program to identify all the possible ways in which the required number of constituencies could be defined in an area, which showed that very different election outcomes were possible depending on where the constituency boundaries were drawn. As well as publishing the results of this work, Ron and Dave used the program to suggest an alternative configuration of constituencies for the city of Sheffield to that proposed by the Boundary Commission for England in 1981; they presented this to the Local Inquiry held there, and it was adopted by the Commission to replace the original scheme. As a consequence of that work, Ron and Dave were expert witnesses in a case brought by senior members of the Labour party against the Boundary Commission for England, seeking to overturn its entire set of proposals for England in 1982 on the grounds that other sets of constituencies – with less inequality in their electorates – could be generated using computer algorithms. (The program was also used by the Australian Electoral Commission.)


When the Boundary Commission began its next redistribution programme in the early 1990s, Ron was commissioned by six Essex local authorities to produce an alternative scheme for part of that county to that proposed by the Commission; this was presented at a public inquiry in 1994, and accepted by the Commission. With David Rossiter and Charles Pattie, Ron then commenced a major study of that redistribution, which included interviews with many of the actors as well as statistical analyses of the procedure. This culminated in the publication of a standard work on The Boundary Commissions (1999), which considered both the history of constituency-definition in the UK and the conduct of the then most recent redistribution.


In 2010, the newly-formed UK coalition government proposed a major change in the rules to be employed by the Boundary Commissions when drawing up new constituency maps. With others, Ron had previously been consulted by the Conservative party on this, and had evaluated their proposals. To inform the Parliamentary and public debates about them, he was commissioned by the British Academy – with Michel Balinski, Iain McLean and Peyton Young – to prepare a paper on the issues. Published as Drawing a New Constituency Map for the United Kingdom (2010), this was widely used and cited by politicians from all parties during the debates in both Houses, and had an impact on some of the amendments to the original Bill. Ron twice gave evidence on the Bill to the House of Commons Select Committee on Political and Constitutional Reform, was consulted by officials and members of all three main parties on various aspects of the evolving legislation, and gave a number of media and other presentations. For this work whe was named the Political Studies Association's Political Communicator of the Year for 2012. Following passage of the Parliamentary Voting System and Constituencies Act, 2011, Ron, Charles and Dave undertook a comprehensive study of the procedures applied by the Boundary Commissions, work that was reported in a number of academic papers and the subject of Ron's Leonard Shapiro Memorial Lecture at the Annual Meeting of the Political Studies Association in April 2014 (to be published in the journal Government and Opposition),


Disproportionality and bias. It is widely appreciated among political scientists that the UK’s electoral system tends to produce disproportional results, whereby some parties get a much larger share of the votes than of the seats. What was not widely appreciated until the late 1990s was that it can also produce very biased results, in that the disproportionality tends to favour one of the largest political parties much more than the other. Using a mathematics developed by a New Zealand political scientist, Ron, David Rossiter and Charles Pattie analysed the result of the 1997 general election as part of their work on the Boundary Commissions, and identified very substantial bias: Labour was very much advantaged by the interaction between the geography of its support and the map of constituencies, and the Conservatives disadvantaged – so much so that if the two parties had won an equal share of the votes cast in 1997 nevertheless Labour would have obtained 82 more seats than their main opponent. Having identified this unexpected outcome, they then analysed all general election results since 1945 and found a major shift in the volume and direction of the bias, towards Labour, which continued at the next three elections (with equal vote shares, it would have won 142 more seats than the Conservatives in 2001, 112 in 2005, and 54 in 2010). They adapted the mathematics to decompose the sources of that bias, publishing the results (with Danny Dorling) in From Votes to Seats (2001). The work continues in a variety of ways, including (with Michael Thrasher, Colin Rallings and Galina Borisyuk) developing an extension of the method for application in three-party situations.


All of this work in British electoral studies has been published in more than 300 academic papers, and was brought together in a single overview volume (with Charles Pattie) – Putting Voters in their Place: Geography and Elections in Great Britain (2006).

Alongside those major concentrations Ron has investigated two other main themes in electoral studies.

  • The measurement of power in electoral systems. This has involved adapting and extending various indices for measuring power (initially used for studies of voting power in committees) to inform debates about electoral reform and proportional representation. One of the indices developed - and published in one of Ron's most-cited papers - has been named by others as the Johnston Index.


  • Studies of the political pork barrel in the United States, showing that the geography of the distribution of public goods reflected the geography of political power.


Ron has been a consultant on aspects of electoral systems and electoral reform to the governments of Bermuda, Jersey and Portugal. He was appointed a Deputy Electoral Commissioner and Member of the Local Government Boundary Committee for England (2008-2009) and was appointed a Trustee of the Arthur McDougall Fund in 2009.

Urban social geography


Ron’s main early research interests were in urban social geography, mapping the patterns and processes underpinning the creation of city residential fabrics and townscapes. He did original work on this for his PhD thesis on Melbourne and then on New Zealand cities, pioneering the use of multivariate statistical methods (classification and factorial ecology) for neighbourhood characterization and the use of mental mapping for uncovering the ways in which people understood residential environments and migrated between them. The evolution of shopping centres, residential building fabric and street patterns was linked to the changing social map. Much of this work was summarized in Urban Residential Patterns (1971) and an edited volume Urbanization in New Zealand (1973) and the methods used were set out in a pioneering textbook - Multivariate Statistical Analysis in Geography (2008).


Although other work, notably in electoral studies, attracted most of Ron’s attention after his return to England in 1974, some research along these lines continued – as in his monograph on the use of planning regulations to promote segregation, and court cases based on the equal rights amendment to challenge it, in American cities (Residential Segregation, the State and Constitutional Conflict in American Cities, 1984). Two texts were also completed during this period – City and Society (1980) and The American Urban System (1982), plus a series of six volumes (edited with David Herbert) on Geography and the Urban Environment.


Interest in this area was revived in 1998 during a visit to Macquarie University in Sydney. Ron began working with Michael Poulsen and James Forrest on problems in the measurement of ethnic residential segregation. Dissatisfied with the standard method, based on single-number indices, they developed alternative procedures for characterizing an area’s population composition that could be used for comparative studies. The goal was to establish procedures that could be used for international comparative studies, which, after a number of exploratory analyses, was undertaken for 2000-2001 covering Australia, Canada, New Zealand, the United Kingdom, and the United States (published in the Annals of the Association of American Geographers, 2007). This established that US cities were the most segregated and those in Australia and New Zealand the least. A linked procedure for studying changes over time (more relevant to the contemporary situation than those associated with models of extreme segregation that characterized US cities 1920-1980), using local statistical methods, was also developed and published in final form in 2011 (Geojournal and Environment and Planning B).

In 2004, Ron started collaborative work with members of the University of Bristol's Centre for Markets and Public Organisation on ethnic segregation if English schools, which was linked to resaerch on student performance. Alongside the substantive studies this has also generated contributions (with Richard Harris, Kelvyn Jones and others) to the continuing debate within UK social science on the measurement of segregation. The work on segregation has also informed collaborative work with Nabil Khattab of the Hebrew University of Jerusalem on ethnic penalties in the British labour market.

The history of human geography

When he returned to England in 1974, Ron was asked to teach a course on the recent history of human geography, charting its extensive growth and change in subject matter and methodologies, especially since 1945. By 1979 this material had been formulated into Geography and Geographers: Anglo-American Human Geography since 1945, which was selected by the American Association of College and Research Libraries as ‘one of the outstanding academic books for 1980/1981’. Further – fully revised and expanded – editions appeared in 1983, 1987, 1991, 1997 and 2004 (the last with James Sidaway), and various editions were translated into Bahasa, Chinese, Japanese, Portuguese, and Russian; it was accompanied by Philosophy and Human Geography (1983 and 1986).

As a consequence of publishing this book, Ron was invited to be the lead editor for the Dictionary of Human Geography (1981); further editions appeared in 1986, 1993 and 2001 and a fifth (with Derek Gregory as the lead editor) in 2009. He also coedited A Century of British Geography (2001) with Michael Williams for the centenary of the British Academy, wrote the essays on ‘Geography’ and ‘Electoral Geography’ for the International Encyclopedia of the Behavioral and Social Sciences (2001) and that on ‘Geography’ for Encyclopedia Britannica Online (2006).

Apart from those and other textbooks, plus monograph treatment of major themes (On Human Geography, 1986; A Question of Place, 1991) research into the nature and history of human geography has concentrated on biographical studies, including Leslie Curry, Robert Dickinson, Jean Gottmann, Leslie Hepple, David Hooson, and Emrys Jones.


Structured keywords and research groupings

  • Cabot Institute City Futures Research
  • Cabot Institute Environmental Change Research

View research connections