2012-2013 Fall Colloquia

ABSTRACT: With uncertainty (unknown probabilities) as opposed to risk (known probabilities), it is impossible in a straightforward way to assign costs and benefits to a set of alternative strategies. Indeed, it is questionable whether objective knowledge, in the sense of knowledge derivable from an algorithm independent of the participants involved, is possible in such a situation. The absence of potentially crucial information means that judgments about likely outcomes are inevitably subjective to some extent.

This uncertainty and subjectivity raise ethical issues for scientists and for policy-makers. Scientists need to decide how to communicate knowledge effectively in a way that non-scientists can understand and use, while not pretending to know what they are incapable of knowing. And they need to decide what role their own subjectivity should play in their communications on scientific issues with policy implications. Policy-makers need to understand their own ethical choices in using scientific information to make decisions and to present themselves to their own audiences.

We propose to focus this issue by examining a case study from the Intergovernmental Panel on Climate Change (IPCC) with respect to possible sea-level rise from melting of the Arctic ice sheet (see O’Reilly, Oreskes and Oppenheimer 2012). Scientists studying this issue in the IPCC decided that uncertainties prevented them from making an objective estimate of ice sheet melting, so omitted any estimate of ice sheet melting from their report on sea level rise [details to follow]. This decision raises ethical issues for scientists and for the policy-makers who need to use such estimates in making and justifying policy decisions.

We are using this actual situation to focus our discussion of the ethics of communicating about climate change for scientists and policy-makers. We seek to integrate philosophical approaches to questions of the ethics of communication under uncertainty with an understanding of scientists’ perspectives and an analysis of the ethical issues that such questions raise for policy-makers.

ABSTRACT: When writing questions for web or pen-and-paper surveys, surveyors can choose from a variety of formats, including check all that apply or a forced-choice format (e.g., yes-no). Previous research shows that asking questions in a forced-choice format consistently yields higher endorsement rates than a check-all format, and provides some evidence that respondents process forced-choice answers at a deeper cognitive level. We conduct a meta-analysis of the available (published and unpublished) studies and found reported support levels increase an average of 48% when items are asked in a forced-choice rather than check-all format. We also conduct two business-to-business (B2B) web survey experiments. The results from the first survey confirm the previous findings and extend them to a B2B population, in a variety of countries and languages. In the second survey, we assess the validity of these two formats by linking each response to internal behavior information. The results from this validation proved however inconclusive. We discuss possible reasons and highlight a research agenda for the future.

ABSTRACT: In this presentation, I’ll try to present some of the major principles and findings of sociolinguistics, or the study of language and society, the 60-year old branch of linguistics (the “scientific study of language”) in which I specialize. Rather than sending a single, long, relatively technical paper (like my 60-page paper on “Language Change and Stylistic Variation” that’s forthcoming in the J. of Sociolinguistics, I’ll attach three short, relatively non-technical papers (and a brief letter) that are be more accessible to non-linguists and might relate to one of your group’s interests, attitudes.

One key theoretical assumption and empirical finding of sociolinguistics is that the FORMS or FEATURES of spoken language vary and change constantly, much more so than non-linguists realize, according to a variety of factors, including the region, social class, ethnicity, gender, age and network of speakers and their interlocutors, the topics they’re speaking about, the relative formality of the recording context, and other aspects of “style.” Paradoxically, the style that is most regular and most reflective of ongoing change in language is the most vernacular or least monitored style, and to elicit more of the vernacular, sociolinguists employ methodologies like recording speakers in peer groups or asking about topics like danger of death or childhood games in which speakers tend to get more involved and pay less attention to “correct” speech. My introductory article in the 2011 American Heritage Dictionary of the English Language (itself the subject of some controversy in the May and June 2012 issues of The New Yorker) addresses to some extent the nature of variation and change in our living language, but I’ll also bring in additional data from my research on African American English in East Palo Alto, and from Guyanese Creole English in Guyana, South America.

People’s expressed or implied attitudes to speech can also vary quite significantly, depending on who is doing the asking, and the contexts or genres in which questions about language are being asked. This I’ve tried to illustrate through a (1985) 9-page paper I did on “standard” and “non-standard” language attitudes in creole English speaking communities, and a more recent (2004) 6-page paper on “Ebonics: The beloved, belittled language of Black America.” In both cases, apparently paradoxical language attitudes co-exist, and unraveling the paradox requires attending to the context in which these attitudes are expressed, and recognizing their complex, multi-faceted nature.

ABSTRACT: According to numerous studies, the election-year economy influences presidential election results far more than cumulative growth throughout the term. Here we describe a series of surveys and experiments that point to an intriguing explanation for voter behavior that runs contrary to the standard explanations political science has offered, but one that accords with a large psychological literature. Voters, we find, actually intend to judge presidents on cumulative growth. However, since that characteristic is not readily available to them, voters inadvertently substitute election-year performance because it is more easily accessible. This “end-heuristic” explanation for voters’ election-year emphasis reflects a general tendency for people to simplify retrospective assessments by substituting conditions at the end for the whole. The end heuristic explanation also suggests a remedy, a way to align voters’ actions with their intentions. Providing people with the attribute they are seeking—cumulative growth—eliminates the election-year emphasis.

ABSTRACT: Many commentators view the Tea Party as an organization whose rise has been fueled in part by the racial antipathy. This view is based primarily on indirect evidence – e.g., the Tea Party’s racially homogeneous membership, positions on issues related to race, and vitriol towards President Obama – or anecdotes – various racist incidents associated with the organization. However, there has been little systematic study of the link, and what research does exist offers limited insight on the causal role racial prejudice might play in fostering Tea Party support. In a survey study (N = 501) we found that the extent of respondents’ agreement with a series of blatantly racist statements was highly correlated with Tea Party support, moreso than a variety of political positions linked with the Tea Party in prior research (economic concerns, opposition to the Affordable Care Act, opposition to Obama). A second study sought to demonstrate a causal link by manipulating the salience of the president’s African-American racial heritage. Participants were presented with a picture of Obam in which his skin color was either artificially darkened or lightened. White participants presented with a darkened picture of Obama were more likely to report supporting the Tea Party. Together these data suggest a significant causal link between racial prejudice and Tea Party support.

ABSTRACT: Political psychologists have found that political attitudes are often grounded in moral convictions and that differences in moral convictions help explain the diverging attitudes of liberals and conservatives. An implication of this research is that persuasive appeals based on moral considerations that are not strongly endorsed by the target audience will fail, but moral and political arguments can be successful if presented as consistent with the moral convictions of the target. Applying a Moral Foundations Theory (Haidt & Graham, 2007) framework, here we test whether framing policy stances in terms of moral foundations endorsed by the target of the message will lead that target to be more supportive of these stances, offering a mechanism of morally based political persuasion. Results of Studies 1 and 2 showed that conservatives were more accepting of same-sex marriage and the Affordable Care Act when these more liberal policies were presented in terms of moral convictions endorsed at higher levels among conservatives (ingroup-loyalty and purity-sanctity, respectively). In Study 3, liberals were more supportive of high levels of military spending when exposed to arguments framed in terms of fairness and equality, values endorsed more among liberals. Finally, in Study 4, we presented participants with messages supporting President Barack Obama’s reelection that were framed in terms of one of the moral foundations endorsed more by conservatives (ingroup-loyalty, respect for authority, purity and sanctity). Relative to a control condition with no persuasive appeal, conservative participants exposed to one of these messages reported greater support and willingness to vote for Obama. Mediation analyses suggested the reframed persuasive appeals were effective because they led individuals to consider the stance (Study 3) and the political candidate (Study 4) as more consistent with their moral principles.

ABSTRACT: Despite the fact that youths in contact with the juvenile justice system are significantly more likely than other youths to have mental disabilities, the justice system has not put enough emphasis on the overall need for mental health treatment to prevent recidivism. The juvenile justice system has in some ways become a “‘dumping ground’ for mentally ill, learning disabled, [and] behaviorally disordered juveniles. Many juvenile offenders have a history of involvement with the mental health system, but migrate to the juvenile justice system because the mental health system has failed to serve their needs.” (Geary 2005) Using the National longitudinal study of adolescent health, this research aims to answer the following questions: What is the relationship between the trajectory of mental health problems during adolescence and aggressive behavior in early adulthood? How do age, gender, race, socioeconomic status mediate the association between earlier mental health wellbeing and later criminal behavior? In this study, we have confirmed that adolescents’ depressive symptoms can significantly predict their aggressive behavior later on in adulthood. Trajectory analysis shows that chronicle depression is dangerous. It also shows that the deterioration of mental health well being is alarming in terms of predicting aggressive behavior. A policy implication of this finding is that instead of only relying on the immediate diagnoses / evaluation of adolescents’ mental health, it is worthwhile for the criminal justice system to look at the record of adolescents’ mental health wellbeing and pay special attention to those whose situation have been deteriorating.

ABSTRACT: In states with direct democracy, political parties, interest groups, and others spend vast sums to publicize their endorsements of initiatives and disseminate policy information about them. How does this affect public opinion? To address this question, we conduct survey experiments where citizens express their opinions about pending initiatives. We manipulate whether they receive party cues, policy information, both, or neither type of information. Contrary to much existing research, we find that policy information affects citizens¹ opinions even when party cues are present. The nature of the effects depends upon whether the policy information reinforces, undermines, or is neutral with respect to citizens¹ own political party¹s positions on the initiatives. In some instances, the presence of a party cue can alter how citizens respond to policy information. When the policy information is neutral, citizens respond more positively to it when party cues are present versus absent. However, citizens do not discount information that undermines their party¹s positions when party cues are present. These results suggest lessons for political scientists who study the effects of information and for political parties, interest groups, and others who seek to influence citizens¹ opinions.