Ch-1: The Study of Voting and Elections
*According to a view: Elections allow citizens to choose the government, and they also restrain political leaders who behave in a way that maximizes their chances of reelection. Elections are means of linking public attitudes with governmental policy. Electing a government is a way of legitimizing its authority. They provide a peaceful means for political change, while permitting individuals and groups to resolve their conflicting needs peacefully. Voters choose intelligently among the candidates. Although no one would argue that all voters are well informed, this view claims voters as a whole make careful and informed choice. *An opposite view: Elections are just symbolic in character. They are a secular ritual of democracy, and voting makes citizens consider themselves participants in the nation’s governance. Voters feel they have fulfilled their civic duty by voting. Voters do not necessarily make intelligent, informed decisions. Elections are effective instruments of popular control. They don’t necessarily affect the course of government policies. *In terms of policy consequences or leadership choices, elections retain their importance in democracies. They make a difference, at least in the short run. Voters to some degree make choices on the basis of this difference. Methodology:
*Modern voting studies rely heavily on survey research. Most important in conducting a survey is the way in which the sample is selected. The best scientific sampling procedures give everyone in the population an equal probability of being part of the sample. Probability theory can then be used to estimate how close results for the sample are to those for the population of interest. *The validity of a survey depends on the wording of the questions that are asked, and for this task mathematical principles are less of help. *Experiments are increasingly being used in voting studies. *Surveys and experiments each have advantages and disadvantages. Surveys excel in their representativeness, which allows ready generalization from samples to the mass public. Surveys can also mismeasure political reality, as when the National Election Studies (NES) post-election surveys find exaggerated turnout rates and overstated votes for winning candidates. Surveys may misrepresent attitudes on sensitive issues because individuals choose ‘don’t know’ rather than revealing a politically incorrect response. It is also difficult to assess causation in surveys, since all variables are measured simultaneously. Experiments are better able to establish causation, as the researcher manipulates the experimental variable and controls for the effects of extraneous variables. Experiments are a useful addition to the toolbox of voting researchers, although we expect political surveys to remain the dominant mode of studying elections because they are able to employ representative samples in a real world setting. *In addition to surveys and experiments, voting studies make use of several other types of data. Aggregate election data have been analyzed from the earliest voting studies. State legislative research has also relied heavily on aggregate data because of the sheer number of races involved and the variability of institutional settings, which is absent in congressional elections. *Another important type of data is written and electronic material, including party platforms, campaign information, newspaper articles on campaigns, and television ads. Methods of content analysis help digest these large volumes of material. *Though surveys remain the dominant mode of studying elections. The Data Base:
*This book is generally based on special surveys rather than on commercial polls, because those organizations forecast election results for their newspaper customers. Newspapers want to publish interesting studies, and prediction of election outcomes. However political scientists are...