Evaluation of the commune reapportionment reform

In December 1978 the Parliament decided to appropriate funds for a scientific evaluation of the consequences of the reapportionment reform for local self-government. The evaluation was to be carried out in accordance with the research program entitled "Evaluation of the Commune Reapportionment Reform" (DsKn 1978:5) which was worked out by the Commission on Local Democracy. The research program was divided into three major sub-programs: THE CITIZENS AND THE COMMUNE Sub-program I was concerned with the input side of the analytical model. Three projects take up the significant channels between the people and the local political system, i.e. the mass media, political parties, and organizations. It has seemed natural to include in the mass media project a study of public information on local issues. A fourth project in this sub-program sums up public activity via the various channels. LOCAL GOVERNMENT ORGANIZATION Sub-program II was concerned with the center of the political system. It is there the demands and desires concerning communal activity are considered and decisions made. The principal actors are local politicians and administrative officers. Their relationship to the people and the opinion representativeness of the politicians are studied. Important questions concern the recruitment, composition, and work of politicians, the development of the administrative organization, and the relationship between politicians and administrators. The budgetary process is also studied. COMMUNAL SERVICE Sub-program III was concerned with what, from the point of view of the people, is the primary output of the local government system. Service was studied in relation to three main questions: how services develop and how they vary between communes, what changes have occurred in the availability of services after the reapportionment reform, and how services are assessed by the inhabitants of the commune. SAMPLE OF THE COMMUNES Given the design of the research program, it was clear from the start that the studies would have to focus on investigations of a large number of communes in a representative sample. One consideration in making this decision was the findings of the earlier Local Government Research Group showing that there are considerable variations between communes in a number of relevant aspects. Considering the fact that the major cities of Stockholm, Göteborg, and Malmö were only marginally affected by the reapportionment reform and that conditions in these communes are in many respects special and would demand a large share of the limited resources available in order to be studied, it was decided that the major cities should be excluded. One idea behind the selection of the sample design was that the commune sample should be used for all of the studies which might be conducted within the overall research program. This was motivated by methodological considerations. A common sample of communes would increase the possibilities for coordinating the planning and analysis of results of different studies. There was also a practical consideration related to the fact that a large amount of background data and archive materials could be utilized jointly in the various projects, and that travels to the communes could be coordinated, etc. One problem with the construction of the sample was that the various projects had different demands on sample size. For certain studies it was desirable to have as large a sample as possible, for others which were especially costly only a few communes could be studied. The sample was constructed according to the Chinese box principle. For the studies requiring a large number of communes in order to present a representative picture of the communes a sample of 50 communes was decided on. The sample was constructed so that it could be divided into two samples of 25 communes each. In one of the samples of 25 communes, five "mode" communes which could be used for more intensive studies were selected. The selection of the sample of 50 communes was based on a factor analysis about 30 different commune characteristics - demographic, economic, and political - among which several described developments over time. This information was used to divide the 274 communes in the country (with three of the major cities excluded), into 25 strata in such a way that the communes which were classified in each stratum were as similar to each other as possible. Two communes were then selected from each of these strata. In this way two samples of 25 communes were obtained.