For those of you who like to read German… The Sueddeutsche Zeitung (SZ) is a major newspaper in Germany. The article reports on psychology’s recent popular problems (Bem, Stapel, Smeesters) and it characterizes them as a consequence of a “systematic problem that presumably affects all empirical sciences”: The preference for new and counter-intuitive findings and the lack of replication studies. Finally, it describes the Reproducibility Project and its goals – in a quite favorable way.
> For those of you who like to read German… The Sueddeutsche Zeitung (SZ) is a major newspaper in Germany. The article reports on psychology’s recent popular problems (Bem, Stapel, Smeesters) and it characterizes them as a consequence of a “systematic problem that presumably affects all empirical sciences”: The preference for new and counter-intuitive findings and the lack of replication studies. Finally, it describes the Reproducibility Project and its goals – in a quite favorable way.
I could´t find a link version. But here is a google translation of the newspaper article:
"The psychologist Daryl Bem is inclined to the supernatural. A few months ago, the professor from Cornell University, a study which he values as th reference to the existence of precognition - as psychologists call it, what laymen refer to as clairvoyance. Bem-te his leg with subjects including Lis-th word before and asked subsequently from which terms had noticed the subscriber. Only then he handed the list of subjects but some-times made, that they were able to characterize this one-consuming. And just the subjects who were cramming for the actual test had been trimmed a little better, Bem reports in his study. Thus events could have been a-nis in the immediate future influence on one in the past.
Worldwide, psychologists were wondering about the experiments and showed Un-rhymed units out in the study design and statistical analysis. Before al-lem, but they shook their heads, that this kind of work in one of the most reputable journals in their discipline, had been published: How could it happen that the reviewers of the Jour-nal of Personality and Social Psychology, had rubber-stamped the work? "Even in top journals can be found again and again studies that have somehow slipped through," the psychologist Stefan Schulz-Hardt says of the University of Goettingen. Sol-che things happen, all right.
However, the processes to take the work of Daryl Bem, a spotlight on a system fault of science, which not only psychology, but rather probably affects all empirical disciplines. The British psychologist Ri-chard Wiseman, Stuart Richie and Christopher French tried to study by Bem namely as closely as possible to repeat - and it could not
Note to explore the effect described. Nor do they succeeded, however, the repetition of the study blizieren pu - and this is the real prob-lem. The Journal of Personality and So-cial Psychology denied a publication, Science, and Psychological Science Brevia waved off as well. Only the British Journal of Psychology was to replicate the study to evalu-tion by experts. One reviewer said, not to publish from - it was Daryl Note finally pu-blizierten the British psychologist to Wiseman in the online journal Plos One.
It is published only what is surprising and counterintuitive
"The replication of studies is the foundation of science," said Christopher French from the University of London recently at the World Skeptics Congress in Berlin. For an-other researchers achieved using the same methods, the same results as the authors of the first study, this increases the informative value of a study. And only after many tests achieve comparable results can be se to secure knowledge.
But hardly anyone replicated studies, such as French, Wiseman and Richie, he goes to scientists around the world: "It is almost impossible to publish onsstudien replication," says Schulz-Hardt. Studies are usually observed only in prestigious international journals, and publishes deliver results if they new it, the best and are surprisingly counterintuitive. When researchers repeated the study but a colleague to check over-the results, then the work is usually for a ton: For the second rope team
has conquered the Mount Everest, in-terested finally hardly a man. Public attention and speed just enjoy the Pioneers.
But the result of a single study that may have come about by chance, despite all care. "Sometimes you actually feel like you could toss a coin, if you have an effect on experimentally-verified - sometimes you can find him, not once," says social psychologist Dieter Frey of the University of Munich. This is prob-lem here is that found in the literature, no evidence of failed replica-tion, because these results are not published, but disappear in the drawers in institutions.
What significance have bioinor-te studies then? Do they offer a reliable indication tot epidemiologist John Ioannidis of, or has at Stanford University, right, in 2005, a much-publicized on-set with the heretical entitled "Why most published research findings are false" published? Wants psychologists to Brian Nosek of the University of Virginia have launched a project to which this question at least for parts of the psychology be answer-: The Working Group will systematically replicate studies to an idea of the rate of false-of positive results will be in their discipline. The psychologists at Nosek NEH-men is so, each of 30 studies, the beginning of 2008 in the journals Psychological Science, Journal of Personal ity and Social Psychology and Journal of Experimental Psychology: Learning, Me-mory and Cognition are published.
These works are fresh enough, as with the authors of the work still to-grip on their material or the raw data should have the same long history nug and published, that they already
other researchers have been quoted as saying. This is relevant, Nosek says, because "we want to know what serves as an indicator of whether a study can be successfully re-takes." Can often-cited, so much attention in the scientific world as little more replicate studies relevant-relevant results? That would be an indication that the quality will prevail and that the science itself successfully purified. But the result could also provide the opposite: that pub-kumswirksamer nonsense gets a lot of attention and is quoted often, but not to replicate is reproducibly.
What is left at the end of science?
With the intention of the psychologists he give-up new questions: When is a re-plication is considered successful and when they failed? Where this limit is Ver-locates? And what does it mean when a repeat study provides a different result-nis? The study is thus refuted? "Results vary randomly," says Frank of Renkewitz der Universität Erfurt, who along with Stephanie Miller and other colleagues to Brian Noseks reproduction project cooperated tet. "After a failed replication-on is far from clear that the corresponding result is false, it could also be the result of the replication study to be defective." Is Still clear in this case that "we have some knowledge then with a little more caution to enjoy, "adds Schulz-Hardt. Two wi-scientific controversies results also mean that further attempts ben-GE must be interested in the supposedly no-one will.
Ahead were the reactions of most positive psychology, says Nosek, even among scientists, whose studies are now being reviewed. "That was long overdue," says Peter about Fischer, lead author of a study, which was also Frey and Schulz-Hardt Ren wa-involved, which will now be examined by a working group. The psychologist from the University of Regensburg and is believed firmly that his study will be reproduced. "I have great confidence in my data."
However, not all representatives of the energy-psychologists are happy about the project. Science magazine cited the social psycho-login Nalini Ambady from Stanford University, whose argument that fears-tions of many in their field sammenfasst to-: argues that the project has a very low rate of successful replications could be the reputation of psychology energy-sustainable be damaged. Once set up-to-get out of the box in recent months some of the negative penetrated. The Dutch social psychologists acknowledge Diederik Sun-stack had to accept that he has had numerous studies GE invented and falsified all records. On 21 June was now back Dirk Smees-age of the Erasmus University Rotterdam. An anonymous tipster had made the university point out to sam that data is in some high-level NER-published studies "too good to be true" were.
Is there a project like that of Bri-an inconvenient not Nosek? Could it not discredit an entire discipline, and confidence in science, he vibration of-? On the contrary, it could also be an example for the integrity of a compartment. "The school should ma chen," Fischer says, "it might even be a starting point for other disciplines."
> For those of you who like to read German… The Sueddeutsche Zeitung (SZ) is a major newspaper in Germany. The article reports on psychology’s recent popular problems (Bem, Stapel, Smeesters) and it characterizes them as a consequence of a “systematic problem that presumably affects all empirical sciences”: The preference for new and counter-intuitive findings and the lack of replication studies. Finally, it describes the Reproducibility Project and its goals – in a quite favorable way.
