Abstract
This special issue highlights new methodological approaches to advance developments in social psychology and expand the methodological toolbox for quantitative, qualitative, and mixed methods researchers. We showcase articles that offer a range of innovations, including new methodological approaches, applications, and procedures. The authors link their methodological innovations to the new theoretical insights produced and provide empirical examples that show how their methods advance social psychological theory.
Two of the articles propose new statistical approaches leading to theoretical insights. First, Jonathan H. Morgan, Kimberly B. Rogers, and Mao Hu highlight how methodological specification can affect theoretical conclusions in affect control theory. The authors compare four different models in estimations of impressions of social events: stepwise regression, ANOVA, Bayesian Model Averaging, and Bayesian Model Sampling. While discussing the theoretical implications, the authors recommend the use of Bayesian methods for model specification.
Second, in a research note, Bruno Arpino proposes a two-step approach for estimating double standards in a split-ballot survey. The first step is based on a matching protocol which facilitates the second step, statistical techniques applied to the new data. Arpino demonstrates the method with an application concerning gender double standards on attitudes toward the age at leaving home using data from the third round of the European Social Survey.
Philip S. Brenner and John DeLamater investigate how desirability bias can occur even in self-administered modes like mail and web surveys. They offer an explanation rooted in identity theory and test this by focusing on measurement directiveness as a cause for bias. They randomly assign participants to conditions that either mention the focus of thestudy, physical exercise, or do not mention the focus. Comparing survey responses, text updates, and records from recreation facilities, they find that direct measures generated bias while the nondirective text condition did not.
Jason Radford, Andrew Pilny, Ashley Reichelmann, Brian Keegan, Brooke Foucault Welles, Jefferson Hoye, Katherine Ognyanova, Waleed Meleis, and David Lazer provide an overview of Volunteer Science, an online laboratory for conducting experiments. To demonstrate the use of the online laboratory, they report the results of six canonical social psychological studies. Their results show that the online laboratory is capable of performing a variety of studies with large numbers of varied volunteers. The researchers intend Volunteer Science to grant researchers, regardless of their access to traditional laboratories, the ability to design and conduct experiments.
Two articles present novel applications of established methods with archival data. First, Matthew Hollander and Douglas W. Maynard use conversation analysis (CA) to study original audio recordings from the classic Milgram Experiment. They demonstrate that conversational analysis produces substantive advances in knowledge about the “obedience to authority” paradigm. CA treats the experimental encounters as three-party interactional scenes and explicates the interactional dilemma for each “Teacher” subject. Hollander and Maynard focus on two kinds of resistance to directives enacted by both obedient and defiant participants and also find that defiant participants adopt two other-attentive practices almost never used by obedient participants. The authors suggest that CA can illuminate the social organization of other social scientific methods such as experiments and survey interviews. They also propose that using CA holds promise for increased understanding of authority-subordinate relations in other settings and situations, such as police-citizen encounters.
Second, in an upcoming article for the March 2017 issue of SPQ, stef m. shuster and Celeste Campos-Castillo examine archival data concerning framing strategies from the failed 1980 Iowa Equal Rights Amendment (ERA). The distinctive ideologies of the proponents for and against the ERA made it a critical case to study. The authors’ analysis indicates that pro-ERA groups used “frame resonance,” a widely discussed strategy in the social movement literature in which activists align issues with ideologies. shuster and Campos-Castillo find, however, that anti-ERA groups used another strategy, what they call “frame dissonance,” by depicting how passing the ERA clashed with these groups’ ideologies. The authors used a novel application of Interact, the computer program for affect control theory, to simulate constituents’ likely responses to frames. The simulations confirmed the authors’ categorization of framing strategies as either resonance or dissonance. shuster and Campos-Castillo’s use of Interact not only provides a new technique for triangulating archival analyses but also introduces a new framing strategy to the social movement literature.
Finally, in another article to appear in the March 2017 issue of SPQ, Jessica Santana, Paolo Parigi, and Karen Cook introduce a methodology they term an “online field experiment” and differentiate it from more traditional experiments. The three components of this design are collaboration with online platforms, recruitment of participants involved with the online community, and retention of participants regardless of their probability of compliance with the treatment. These online experiments exploit the properties of Big Data. Further, the authors argue that this approach enables examination of treatment complexity, a treatment that depends on interaction within a particular context.
Some contributions entail new revelations about “old data,” such as the reexamination of the Milgram data. Some entail gathering new data from “old designs,” via Volunteer Science. Others provide us with new models, new techniques, and ways to include technology into research design. All provide promise and assure us that our field is advancing in exciting ways.