The Nature and Origins of Mass Opinion (Cambridge Studies in Public Opinion and Political Psychology) Kindle Edition

What would a comprehensive graduate exam in political science or political behavior be without a review of Zaller's now classic text? Elite discourse, Zaller informs us, shapes public opinion. That is, powerholders have the ability to mold both the agenda of the public -- the political issues that citizens feel to be most important -- as well as their beliefs about these issues. The topic raises important normative and ethical issues and the book offers a number of starting points for follow-up research. A decidedly good, important buy, this one.

Zaller's argues public opinion on issues is unreliable, primarily because elite sources of information provide competing or multiple considerations causing public opinon polls to measure whatever recent elite message an individual has stored in thier short term memory. A classic and important text. Zaller is not a straightforeward read, but lays out an important theory. The theory lacks focus on the use of core values and partisanship on public opinion; two important variables. Other than that it has held up to criticism for over a decade

Zaller turns our perception of polling data on its ear. People don't think about most issues until they are asked, so opinion surveying might actually do more to shape opinions than actually report on them.

Zaller's book presents an interesting elite-driven model of public opinion, which essentially concludes that opinion polls are less than reliable and we should be skeptical of attempts to measure the public's attitudes. However, it is very dense, is chock-full of advanced statistics above the head of most political science graduate students, and therefore at the end of the day is somewhat unsatisfying.
If you need a primer on public opinion research, this isn't it. But if you want to read cutting-edge work, and you have a good grasp of statistics, dive in.

Zaller offers an insight into how public opinion is created and shifted to individuals. His view is that opinion is created within the circles of the elite, where political issues are given the proper amount of thought and consideration. After all, many in the public will not have the same time to devote to deciding what should be done in a particular instances. Zaller argues that the fact the media discusses the opinions of the elite ensures that the public will gain some insight into what they should think. A great example would be Glenn Beck and his legion of listeners. These people listen to Beck and get an idea of what is important to think about, but not how to think. They become informed about the what, but not exactly the how, unless they like those who are telling them of the issue. What all this means for the public is that issues are created among the elite and then communicated to the public, which insures that issues that are relevant to the public might not get the coverage they deserve. Anyway, this is a good book, but not for the masses. Sorry, but still good.

This book is a must read for political scientists. Even as someone who is not too interested in the public opinion literature, I found this to be an enjoyable read.

You want to know about public opinion? Read this book. You want to study political science past the undergraduate level? Read this book.

An interesting, and occasionally vexing, topic of study in political science is public opinion. In his book The American Democracy Thomas Patterson touches upon myriad issues that surround this topic: the difficulties in accurately measuring public opinion, the inconsistencies and fluidity of public opinion, and political socialization are just a few of the factors of interest to political scientists that Patterson examines. However, no introductory text can examine such a multifaceted topic in great depth, and as a result on page two-hundred eleven of his book Patterson recommends John R. Zaller's The Nature and Origins of Mass Opinion as a good source for more in-depth information about public opinion formation and measurement. Indeed, Patterson characterizes Zaller's work as "[a] superb analysis of the nature of public opinion" (211). Zaller examines many, if not all, of the factors that influence public opinion, spending a great deal of time examining the affect of information on political opinions and election choices, and the impact of "elite domination of public opinion" (310). But Zaller's work is perhaps best known for its thoughtful examination of public opinion instability, an examination that challenges traditional thinking on the topic. And, while not perfect, it is easily one of the most important and influential works written on this subject.

As Zaller highlights, variances in survey results has traditionally been attributed to "measurement error" which is built into a survey or the presence of "nonattitudes"

(i.e. respondents answer questions about which they have no strong opinion) (31). In the broadest sense, The Nature and Origins of Mass Opinion strives to refute these explanations; and instead posits that a combination of factors, including the degree of ambivalence an individual has related to a specific topic, the general level of interest a person has in a topic, and what information has been encountered and how recently, more accurately explain response instability. Zaller does an excellent job of building his case for this perspective; in particular the use of literature from disciplines other than political science is especially compelling. Supported by this literature, and compelling in its own right, are the axioms that comprise his "Receive-Accept-Sample (RAS) model" (58). The author painstakingly examines each of his axioms and examines each in relationship to real world data which serves to further strengthen his arguments. As a result it seems clear that his contentions that individuals who are more aware of the political process are also more likely to perceive political messages, while at the same time resisting messages that run counter to their own political biases, have great validity. Further, Zaller's argument that there is a high degree of ambivalence on many issues is well supported by the analysis of the findings from the 1987 pilot study that is cited (63). Finally, even though the author acknowledges the inherent challenges in proving the validity of axiom four, he makes a compelling case by utilizing literature from the field of psychology (62). The result is an intriguing model that, rather than assuming that "response error is simply so much noise . . . [or] signifies nothing of interest" instead argues that "response variation is rooted in an important substantive phenomena, namely the common existence of ambivalence in people's reactions to issues" (75). Of further importance and interest are the series of deductions that the author forms through the application of his model to real world circumstances.

While it is impractical to examine each of Zaller's deductions there are two that are especially significant. The author's third deduction is representative of the very essence of his RAS model and its implications: If people form conflicting considerations on most issues, and if they base their survey

responses on whichever of these considerations happen to be at the top of the head atthe moment of response, one should expect a fair amount of variability in people's responses to survey questions. (64).

This "top of the head" argument is fundamental to the author's perspective, and Zaller offers numerous citations in support of this deduction. The result is a sea change in the way that response instability is perceived. A second deduction that is quite compelling examines the relationship between political attitudes and the political messages of the powerful and elite:

[T]he population as a whole should be able to develop more stable attitudes for issues on which partisan elites divide sharply and clearly, thereby providing clearer message cues for everyone. Conversely, attitude stability should be weaker for issues on which partisan divisions are hazy or nonexistent, because in such cases the public gets few message cues. (67).

The validity of this deduction has been proven out over the past thirty years. On a whole series of issues, including reproductive health and immigration, the messages espoused by the major political parties have become increasingly polarized and that polarization has increasingly been reflected in attitudes amongst the electorate. The RAS model, and the deductions that flow from it, represent a significant addition to the body of literature in political science. However, though Nature and Origins of Mass Opinion is an invaluable work, it is not without its faults.

One small criticism that can be made is the relatively sparse discussion of the survey instrument from which much of the data used to test the RAS model. More to the point, it would be helpful to have the instrument included as an appendix to the book. Zaller does identify the pilot study used, and it is currently available online (after registering with the online host, and assuming that one has access to the statistical software necessary to open the documents), however, in order to test, and adequately assess, Zaller's conclusions the raw data must be readily accessible. A more significant criticism of the RAS model itself is an important assumption upon which it is built - that of the type of information individuals receive and process:

The Receive-Accept-Sample Model is . . . a set of claims about how citizens acquire "information" and convert it into attitude statements . . . [O]ne cannot test the model without making definite assumptions about the information environment that sustains citizens' attitudes on a given issue. For purposes of this chapter, I make the following simple assumption about this environment: that it consists of moderately intense,

temporally stable information flows favoring both the liberal and

conservative side of each issue. By moderately intense I refer to information flows that involve neither dominating headline stories . . . nor obscure or esoteric stories. (58)

The definition suffers from vagueness that is difficult to overstate, and ignores the fact that what a researcher might define as "esoteric" could be perceived as most significant to a member of the public. Given that the RAS model is what Zaller defines as an "information processing model" (58), the imprecise manner in which information environment and information flow are defied is a serious flaw.

For many decades there has been an assumption amongst political scientists that variations in survey responses stemmed from flaws inherent to the survey instrument itself. In The Nature and Origins of Mass Opinion, John Zaller exposes the flaws inherent in such thinking. Despite some non-fatal shortcomings the author builds a persuasive case that, where survey instruments are free of bias, response instability is the expression of the uncertainty that many people feel on a wide array of issues.