Social Priming, Failed Replications, and Egos

2 02 2013

The Chronicle of Higher Education has a good article on social priming—the effects on our behavior from subtle cues in our social environment. It reviews some of the key studies such as the famous elderly prime makes people walk more slowly finding. There is an appropriately strong focus on John Bargh, whose studies made the effect well known, as well as some of the individuals who have had trouble replicating his results. As it remains today, there is serious doubt about whether the effect is very robust. Instead, there might be some moderators that enhance or weaken the effect that are unknown. If there is anything good coming out of this, it may be that we start begin respecting the publication of failed replications. That will be good for the entire discipline.

Link to the article at the Chronicle.

Image courtesy Flickr user sunnydelishgirl. Licensed under Creative Commons




David Brooks on using psychological research for public policy

11 01 2013

Today’s NY Times has an op-ed by David Brooks that points out that much of public policy is derived from “common sense” or “folk psychology” that has no scientific basis. Instead, he argues for public policy created by using research findings to enact behavioral changes in the population. I agree. There are some interesting comments to the article as well, but many of them amount to “David Brooks is on his agenda” and do not offer much of a criticism that is useful to the discussion.

Link to the article at the NY Times.




SPSP comments on the final Levelt report on the Deiderik Stapel debacle

10 01 2013

The Society for Personality and Social Psychology just released a statement on the Levelt report. It is not (yet) on their web site, so I will reproduce it here. It arrived on the SPSP listserv. I am glad we are undertaking an assessment of where we have gone wrong and how our research practices can be improved to help avoid something like this in the future.

Society for Personality and Social Psychology Statement on the Levelt Report

The recent Levelt report from the Netherlands details the breadth of Deiderik Stapel’s fraudulent activities and offers reflections on the scientific culture that enabled this magnitude of deception to go (nearly) undiscovered for many years. It is a sobering read. Both the European Association of Social Psychology and the Society of Experimental Social Psychology have issued statements on the report. These statements expressed appreciation to the Levelt committee for its thorough investigation of the Stapel case and indicated reservations about the report’s indictment of the field at large for what the Levelt report considered “sloppy science” practices. In large measure, we support these previous statements, and our statement is meant to complement rather than reiterate the points made by our allied societies.

In the aftermath of the Stapel case and other recently discovered cases of fraud, it would behoove us to reflect on the steps we can take as we move forward to protect our science against incidents of fraud in the future and to repair the image of our science. The core foundation of any field of scientific endeavor is trust and integrity.

The Society for Personality and Social Psychology has consistently maintained the stance that we must work together as a professional organization and as individual scientists to promote a context in which good scientific practices are celebrated and are embedded into the training we provide to young scientists joining the field. Indeed, this was the theme of a recent letter I wrote to the Society regarding the Stapel case. It is a theme that merits repeating. These recent events provide an opportunity not only for constructive reflection but also specific action. Upholding sound scientific practices will insure that our science has integrity. We should not assume, however, that because we all believe in the principles of ethical conduct that this is sufficient. In this regard, we can all profitably discuss ways to accomplish these goals. These goals should include, but not be limited to:

  • Identifying effective ways to build discussion of ethics and good scientific practices into our course work and everyday discussions in our laboratories;
  • Developing safe venues for trainees and others to report concerns about breaches of ethics within universities and within the journal review process;
  • Establishing clear standards for what personality and social psychology papers should present in methods and results sections of articles;
  • Providing formal training in how to review articles;
  • Clarifying within our formal training acceptable practices for addressing, for example, missing data, eliminating cases from analysis, and providing clear detail on methods and measures;
  • Increasing opportunities and incentives for conducting and reporting direct replications of important findings; and
  • Evaluating the pressures that can lead to a careerist focus as opposed to a focus on true discovery among scientists.

These are but a few of the issues we, and all sciences, need to consider. Recent months have borne witness to a number of activities designed to address these issues. For example: replication issues have been the subject of a recent special issue of Perspectives on Psychological Science (http://pps.sagepub.com/content/7/6.toc), a forthcoming special target article in the European Journal of Personality, and recent issues of our own Dialogue (see http://tinyurl.com/auzkr73). Brian Nosek and Daniel Lakens are co‐editing a special issue on replication of important findings in social psychological research in Social Psychology (see http://tinyurl.com/asujp7s). Finally, SPSP commissioned a Task Force for Responsible Conduct, which outlined a variety of ways we could take positive steps to ensure the integrity of our science (https://www.spsp.org/?ResponsibleConduct) and the Task Force continues to work on these issues.

Our upcoming meeting in New Orleans provides a number of immediate opportunities to explore and discuss these issues with our community. Two formal symposia address issues related to good scientific practices.

The first is titled “Openness in Scientific Reporting: Potential and Reaction” and is scheduled for Friday, January 18 from 11:15 am to 12:30 pm (Rooms R03‐R05).

The second symposium is titled “False Positive Findings: Effect Sizes Too Large, Too Small, or Just Right” and is scheduled for Friday, January 18 from 3:30 pm to 4:45 pm (Rooms R03‐R05).

Finally, I suggested to you in a recent letter that the leadership of SPSP was likely to hold a Special session for the membership to come together to discuss these issues or any issues of interest to the membership. We have scheduled this session for Saturday, January 19 from 3:30 pm to 4:45 pm Room 203‐205. In attendance will be David Funder who is the 2013 President of the Society, Jenny Crocker who chaired the SPSP Task Force on Responsible Conduct, Jack Dovidio the current Executive Officer for the Society, Jamie Pennebaker the new President‐Elect of the Society, and me. We invite you to come to this session with questions and we will do our best to address these and other issues relevant to the Society and engage the membership in a productive discussion.

Yours Sincerely,

Patricia G. Devine for the SPSP Executive Committee

Past-­‐President, Society for Personality and Social Psychology




Scrivener overview tutorial

18 11 2010

I am using a writing program called Scrivener for my academic work lately. It is a very compelling program for academic writers, because it allows for the construction of documents in a very non-linear, non-top-down approach. As I research for an article or chapter or paper for class, I can add parts of the document and not worry about how those parts are connected to the rest of the parts. After I have all the parts laid out, I then start to see connections and can begin to assemble the piece in a way that provides a sensible narrative.

A fellow academic has produced a tutorial outlining some of the features of Scrivener and why it makes sense for academic writibng over a more linear approach such as a word processor. Word processors are great for 2 things I think: relatively short documents such as letters or short papers where there aren’t lots of pieces that have to be interconnected. They are also great for the final assembly and polishing of a manuscript. I like Scrivener for the ability it has to produce the manuscript from bits and pieces of thoughts.

Here’s the tutorial: Five Essentials Tutorial