Common use of Data Collection and Processing Clause in Contracts

Data Collection and Processing. The dependent variable describes the evolution of forty environ- mental policy items in twenty-four countries over thirty years. This vast amount of data was collected with the help of expert surveys. This way, a comprehensive database was compiled, offering unique opportunities for studying patterns of environmental policy conver- gence between the countries involved. In the following, we discuss the measures taken to ensure data validity and reliability during data collection and data processing. Expert surveys The necessary output data could not be compiled from existing sources. The ECOLEX database on environmental law, developed by the World Conservation Union, the UN Food and Agriculture Organisation and the UN Environment Programme comes closest to our requirements, but even here ‘historical’ information on early environmental legislation is often lacking. To obtain this information, it proved necessary to access the original laws, decrees and ordinances. In so doing, we relied on expert surveys. Expert surveys are generally seen to have several crucial advantages over alternative approaches. Precisely because they reflect the judgements of experts, they acquire a certain weight and legitimacy. Furthermore, as expert judgements are reasonably comprehensive, they permit the collection of comparable and standardised data. Despite these advantages, how- ever, problems of validity and reliability of the data may occur. For instance, the same question may be interpreted in different ways by different experts. Measurement errors could occur for example when experts overlook legislation. However, as we asked the experts for facts (introduction of laws, use of instruments, setting of limit values etc.) and not for opinions or attitudes, most of the problems regarding the validity and the reliability of data that are common to social science surveys did not apply here. An important step in guaranteeing the quality of data was to make a careful selection of experts to carry out the work. First, in order to address adequately the wide variety of questions in the survey, the experts had to have a comprehensive overview of environ- mental policy development from 1970 to 2000. Second, the group of experts should be as homogeneous as possible, so as to ensure the comparability of data (Frankfort-Nachmias and Nachmias 199ł). In order to meet both criteria, we limited our selection to academics with a background in political science or law and preferably working at research institutes or universities. The design of the questionnaire and the manual The questionnaire had a standardised set up so as to limit the room for interpretation to a minimum (see table A4.1 in the annex for an example). At the same time, open questions (‘remarks section’) were included to allow for the description of item-specific characteristics. This facilitated the checking of data comparability. The questionnaire was accompanied by a detailed manual, presenting general instructions and definitions as well as background information on each individual policy item to ensure the validity and reliability of the data (see below). As set out above, the selection of policies addressed by the project covered one or more of the three dimensions: presence-of- policy, policy instrument and setting. The first type – ‘presence- of-policy’ questions – asked for the mere existence of certain environmental policies in 1970, 1980, 1990 and 2000 and, if they existed, the year of their introduction. The second type – ‘policy instrument’ questions – addressed the kinds of instruments used to tackle environmental problems as well as their application over time. Experts were asked to select one ‘dominant’ instrument type from the following list: (1) obligatory standard, prohibition or ban (2) technological prescription (3) tax or levy (4) subsidy or tax reduction (5) liability scheme(s) (ł) planning instrument