What do you think?
Rate this book
Cambridge Studies in Public Opinion and Political Psychology
How Voters Decide: Information Processing in Election Campaigns
This book attempts to redirect the field of voting behavior research by proposing a paradigm-shifting framework for studying voter decision making. An innovative experimental methodology is presented for getting 'inside the heads' of citizens as they confront the overwhelming rush of information from modern presidential election campaigns. Four broad theoretically-defined types of decision strategies that voters employ to help decide which candidate to support are described and operationally-defined. Individual and campaign-related factors that lead voters to adopt one or another of these strategies are examined. Most importantly, this research proposes a new normative focus for the scientific study of voting behavior: we should care about not just which candidate received the most votes, but also how many citizens voted correctly - that is, in accordance with their own fully-informed preferences.
366 pages, Paperback
First published June 29, 2002
2 people are currently reading
42 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 of 1 review
August 24, 2016
This book may be helpful to certain readers looking within a narrow focus. The question is something like what methods do voters' minds use to sift through election campaign data. To take an analogy, Olympics judges may not be able to maintain focus on every athlete during every second of an event, so how do they use what bits of each athlete they do see to decide which is the best athlete in the event? (Of course, it's somewhat different because each voter can have his own opinion of what qualities make the best candidate.) Lau also believes that one factor is a limited capacity (as well as inclination to put in the effort) to acquire all the data which could be relevant to comparing the candidates. So, an important factor in the book is the use of indicators (such as the candidate's party or endorsements he has received) which are suggestive of what a candidate might do, but which don't actually give specifics.
You'll already be past page 150 when the author says he knows up until now they've been taling about the "tools" used by the voter in the campaign, but now he's going to talk about how voters evaluate the relative preference between candidates.
The book gives statistical analysis of findings for those who can take advantage of this, but the kind of reader who is comfortable with popular science books should be able to gather the important facts being discussed.
They carried out experiments with simulated election campaigns in which each "voter" worked at a separate PC and could choose which of the information accessible from the screen they wanted to see (while the available choices scrolled down the screen so that items might disapppear before the voter got to it, as information may come and go in real campaigns.) Each "voter" participated in both a primary and general election (most of the time, the researchers intentionally had the voter's preferred candidate lose the primary.) There are aspects which are certainly distinct from real world campaigns which might make a difference, but this review will be long enough as is.
Lau tells us that past voter research worked under the premise that if a voter does not accidentally mark his ballot differently than he intended, he voted "correctly" (and that presumably reflected the voter's beliefs and policy preferences.) Lau tells us that sometimes there is a conflict between what the voter would like elected officials to do, and what one could rationally expect the candidate he voted for to do. In that sense, Lau believes a voter can vote "incorrectly." His point is not comparing which candidate researchers think would make sense for a voter to pick and the candidate that is actually picked. It's whether, for instance, an elderly person who says he wants Social Security maintained, but votes for a candidate who wants to defund Social Security.
It seems to me, that determining this is more difficult than is generally assumed. There's often a considerable difference between campaign promises and elected officials' action. Campaign promises are very available to voters, but more reliable indicators can be less available. Voting records only tell us so much - politicians sometimes vote the way their constituents want when they know the other side will win anyway, but vote against their constituents if their vote is needed by the other side. And voting records only tell us what happened to bills which reached the floor. Information on which committee members prevented bills from coming to a vote are even harder for most voters to get... Such information could be condensed into tables, but generally isn't (perhaps, because it would be disillusioning.)
Lau concludes that voters tend to maintain a sort of mental score for each candidate. Each time a voter comes across information on a candidate, he increases or decreases how favorably he feels about the candidate, but he does not necessary maintain specific memories of the particular items which contributed to these calculations. This can play a significant role in choosing candidates, but we are told that the number of specific memories affecting the feeling score can be decisive especially in some conditions.
Why would a person vote "incorrectly?" One factor noted in the book is that candidates who have more money to spend on influencing voters tend to win more "incorrect" votes.
- - - - -
For the most part, the book discusses elections in terms of the US "two party system." It's assumed that there are usually only two choices which could be considered "correct." There are rare exceptions - they do mention a few elections in which a third candidate (such as Ross Perot) was able to get above a certain percentage of the vote. (I wasn't entirely clear whether they considered such candidates who had little or no chance of winning as potential "correct" votes.) As best as I can tell, any party which can't get over 10% was viewed as never a "correct" vote. So, the researchers have at least restricted their definition of "incorrect" voting to assume a voter's objective must be to have a candidate win in this particular election who is less distant from the voter than the other candidate (that is, immediate gratification rather than delayed gratification.)
The authors say that about 70% of voters vote "correctly." That's better than a 50-50 coin flip, but in the context of an essentially two party system, where many people use a candidate's party as a major part of their decision method, it doesn't seem that much better than random to me.
- - - - -
The 2016 Democratic primary gives us an example of trying to determine what a candidate will do if elected. Clinton claimed to support campaign finance reform, but used big money against a primary candidate who did not (she wasn't "fighting fire with fire".) For her to champion campaign reform in office would be like a drinker working to outlaw alcohol - not likely. But at best, a voter may see this reasoning. Clinton switches to saying she was anti-TPP - which may have made some people believe and some people skeptical because she switched. But it was only after the primary voting that we learned she picked a historically pro-TPP VP and her people blocked an anti-TPP plank from the party platform. This is very suggestive - but not "proof" and after the voting ended.
- - - - -
It's true, third parties play a limited role in US politics. But over the last few decades, the number of registered voters who don't designate a party has grown, and "independents" are now a plurality of registered voters. Vast numbers of Americans don't vote in presidential elections (and even fewer in other elections.) If the intent of the study is solely to determine individual behavior (and not the resulting collective action), then the question of not voting may not be as signficant. But if we want to understand how particular candidates get elected, then we need to understand factors in whether people go to the polls (or at least establish this affects all candidates equally.) If we want to understand what Americans would like to see happen, we need to understand not only those who vote for major parties every election and those who vote for major parties some years, we also have to understand those who generally do not vote and those who vote for third parties. Many who don't vote in general may do that for reasons which aren't relevant to such studies, but even if 1/4 of non-voters have feelings relevant to these studies, that is still a considerable part of the population. Third party voters are a small group in the US. However, if studies of voters were done on a larger scale (say, 10,000 subjects) it would not be an entirely accurate picture if it excluded third party voters. Such a larger study would generally add statistical confidence to the results. The US electoral system makes it quite hard for third parties to win, and that affects voter willingness to vote for third party candidates they prefer. However, there's an intermediary category, such as the Liberal and Conservative columns on ballots which generally list the same candidate as the Democrats or Republicans. When it's the same person, what makes voters vote for that candidate under the "third party" column? That may also tell us about US citizen's political behavior. Another question: Is there a significant number of voters who say they "don't know" where a candidate stands because they've heard the campaign promises and negative ads, but have doubts what to really expect?
Why does anyone vote for a third party in the US? Consciously or unconsciously, it would seem to be one of two things: (1) Hopes that the third party will grow and be able to shape the political scene, or (2) To punish one of the major parties for failing in some way - hopefully encouraging it to do better in the future.
You'll already be past page 150 when the author says he knows up until now they've been taling about the "tools" used by the voter in the campaign, but now he's going to talk about how voters evaluate the relative preference between candidates.
The book gives statistical analysis of findings for those who can take advantage of this, but the kind of reader who is comfortable with popular science books should be able to gather the important facts being discussed.
They carried out experiments with simulated election campaigns in which each "voter" worked at a separate PC and could choose which of the information accessible from the screen they wanted to see (while the available choices scrolled down the screen so that items might disapppear before the voter got to it, as information may come and go in real campaigns.) Each "voter" participated in both a primary and general election (most of the time, the researchers intentionally had the voter's preferred candidate lose the primary.) There are aspects which are certainly distinct from real world campaigns which might make a difference, but this review will be long enough as is.
Lau tells us that past voter research worked under the premise that if a voter does not accidentally mark his ballot differently than he intended, he voted "correctly" (and that presumably reflected the voter's beliefs and policy preferences.) Lau tells us that sometimes there is a conflict between what the voter would like elected officials to do, and what one could rationally expect the candidate he voted for to do. In that sense, Lau believes a voter can vote "incorrectly." His point is not comparing which candidate researchers think would make sense for a voter to pick and the candidate that is actually picked. It's whether, for instance, an elderly person who says he wants Social Security maintained, but votes for a candidate who wants to defund Social Security.
It seems to me, that determining this is more difficult than is generally assumed. There's often a considerable difference between campaign promises and elected officials' action. Campaign promises are very available to voters, but more reliable indicators can be less available. Voting records only tell us so much - politicians sometimes vote the way their constituents want when they know the other side will win anyway, but vote against their constituents if their vote is needed by the other side. And voting records only tell us what happened to bills which reached the floor. Information on which committee members prevented bills from coming to a vote are even harder for most voters to get... Such information could be condensed into tables, but generally isn't (perhaps, because it would be disillusioning.)
Lau concludes that voters tend to maintain a sort of mental score for each candidate. Each time a voter comes across information on a candidate, he increases or decreases how favorably he feels about the candidate, but he does not necessary maintain specific memories of the particular items which contributed to these calculations. This can play a significant role in choosing candidates, but we are told that the number of specific memories affecting the feeling score can be decisive especially in some conditions.
Why would a person vote "incorrectly?" One factor noted in the book is that candidates who have more money to spend on influencing voters tend to win more "incorrect" votes.
- - - - -
For the most part, the book discusses elections in terms of the US "two party system." It's assumed that there are usually only two choices which could be considered "correct." There are rare exceptions - they do mention a few elections in which a third candidate (such as Ross Perot) was able to get above a certain percentage of the vote. (I wasn't entirely clear whether they considered such candidates who had little or no chance of winning as potential "correct" votes.) As best as I can tell, any party which can't get over 10% was viewed as never a "correct" vote. So, the researchers have at least restricted their definition of "incorrect" voting to assume a voter's objective must be to have a candidate win in this particular election who is less distant from the voter than the other candidate (that is, immediate gratification rather than delayed gratification.)
The authors say that about 70% of voters vote "correctly." That's better than a 50-50 coin flip, but in the context of an essentially two party system, where many people use a candidate's party as a major part of their decision method, it doesn't seem that much better than random to me.
- - - - -
The 2016 Democratic primary gives us an example of trying to determine what a candidate will do if elected. Clinton claimed to support campaign finance reform, but used big money against a primary candidate who did not (she wasn't "fighting fire with fire".) For her to champion campaign reform in office would be like a drinker working to outlaw alcohol - not likely. But at best, a voter may see this reasoning. Clinton switches to saying she was anti-TPP - which may have made some people believe and some people skeptical because she switched. But it was only after the primary voting that we learned she picked a historically pro-TPP VP and her people blocked an anti-TPP plank from the party platform. This is very suggestive - but not "proof" and after the voting ended.
- - - - -
It's true, third parties play a limited role in US politics. But over the last few decades, the number of registered voters who don't designate a party has grown, and "independents" are now a plurality of registered voters. Vast numbers of Americans don't vote in presidential elections (and even fewer in other elections.) If the intent of the study is solely to determine individual behavior (and not the resulting collective action), then the question of not voting may not be as signficant. But if we want to understand how particular candidates get elected, then we need to understand factors in whether people go to the polls (or at least establish this affects all candidates equally.) If we want to understand what Americans would like to see happen, we need to understand not only those who vote for major parties every election and those who vote for major parties some years, we also have to understand those who generally do not vote and those who vote for third parties. Many who don't vote in general may do that for reasons which aren't relevant to such studies, but even if 1/4 of non-voters have feelings relevant to these studies, that is still a considerable part of the population. Third party voters are a small group in the US. However, if studies of voters were done on a larger scale (say, 10,000 subjects) it would not be an entirely accurate picture if it excluded third party voters. Such a larger study would generally add statistical confidence to the results. The US electoral system makes it quite hard for third parties to win, and that affects voter willingness to vote for third party candidates they prefer. However, there's an intermediary category, such as the Liberal and Conservative columns on ballots which generally list the same candidate as the Democrats or Republicans. When it's the same person, what makes voters vote for that candidate under the "third party" column? That may also tell us about US citizen's political behavior. Another question: Is there a significant number of voters who say they "don't know" where a candidate stands because they've heard the campaign promises and negative ads, but have doubts what to really expect?
Why does anyone vote for a third party in the US? Consciously or unconsciously, it would seem to be one of two things: (1) Hopes that the third party will grow and be able to shape the political scene, or (2) To punish one of the major parties for failing in some way - hopefully encouraging it to do better in the future.
Displaying 1 of 1 review