In recent years, since the early 2010s, the scientific community has drawn attention to the existence of a reproducibility crisis in science, especially psychology and medicine: The results of much research are either impossible to reproduce or simply no attempt is made to do so.
However, problems with confirming hypotheses are not the only ones that are included in the replication crisis, but that it is of a broader nature. In this sense, the relevance of the falsification of the results, especially in the field of social psychology, and other very important methodological factors should be emphasized.
The replicability crisis in science
One of the foundations of the scientific method is the replication of results. While many people have a marked tendency to view the findings of a single study as credible and definitive, the truth is that a hypothesis only gains real strength when it is confirmed by several valid studies from different research teams. .
Likewise, negative results, that is, refuting hypotheses, are as important as their verification. However, the proportion of studies refuting the approaches seems to have decreased in science in general; therefore, there is a clear primacy of publications corroborating experimental hypotheses.
Numerous publications that have been produced around the replication crisis highlight the extent that it has taken on in psychology. However, it should be noted that this crisis affects science as a whole and that it also has a particular intensity in the case of medicine. This is due to a number of interrelated factors.
The main causes of this phenomenon
A meta-analysis by Daniele Fanelli (2009) concludes that publication fraud is more common in medical and pharmaceutical research than in the rest of the fields. The author suggests that this may be due to the large scale of economic incentives for publications or to greater awareness in these areas.
There are, however, several factors that influence the replicability crisis beyond explicit data tampering. One of the most important is the selectivity of publications: in general, positive and striking results have a greater potential to appear in journals and to bring recognition and money to researchers.
This is why the “drawer effect” often occurs studies that do not support the expected hypotheses are rejected while those that do are selected by the authors and published most often. In addition, the non-replication of positive studies decreases the risk that the hypotheses will be refuted.
Other common practices that have similar goals are to select a large number of variables, then focus only on those that are correlated, changing the sample size (e.g., including subjects until the results are positive) or to perform several analyzes. those who support the hypotheses.
Why is this so serious in psychology?
The replication crisis in psychology is considered to date back to the early years of the 2010s. numerous cases of fraud involving the perpetrators concerned have occurred; for example, social psychologist Diederik Stapel falsified the results of several publications
A meta-analysis by Makel, Plucker and Hegarty (2012) found that only about 1% of psychology studies published since the turn of the 20th century are replicas of earlier studies. This is a very low number because it strongly suggests that many of the conclusions drawn from isolated studies cannot be taken as definitive.
The number of successful independent replications is also low, Standing about 65%; in contrast, more than 90% of those conducted by the original research team corroborate the hypotheses. On the other hand, jobs with negative results are also particularly rare in psychology; the same can be said of psychiatry.
Solutions to the research crisis
The reproducibility crisis in psychology and science in general not only compromises the results of a large number of studies, but can lead to the legitimation of unconfirmed hypotheses with the necessary rigor. This could lead to the widespread use of incorrect assumptions, altering the development of science.
There are currently many economic interests (and others also linked to prestige) which favor the persistence of the replication crisis. As long as the criteria followed for the publication of studies and the dissemination of their results in the mainstream media continue to have this monetarist character, the situation will hardly change.
Most of the proposals that have been made to help resolve this crisis relate to the methodological rigor in all its phases, As well as with the participation of other members of the scientific community; in this way, it would be to strengthen the “peer review” process and try to encourage replication efforts.
It should be borne in mind that in the field of psychology we work with many variables, on the one hand, and it is difficult to establish a context in which the starting point is similar to that of another study, on the other hand. This allows elements that are not taken into account in the search to very easily “contaminate” the results.
On the other hand, the limits of the modes of decision concerning the existence of real phenomena or only statistical phenomena sometimes lead to false positives: the simple fact that the p-value is significant should not be sufficient to indicate that it reflects a real psychological phenomenon.
Fanelli, D. (2009). How many scientists fabricate and falsify research? A systematic review and meta-analysis of survey data. PLoS ONE 4 (5).
Makel, MC, Plucker, JA and Hegarty, B. (2012). Replications in Psychological Research: How Often Do They Really Occur? Perspectives on Psychological Science, 7 (6): 537-542.
Nosek, BA, Spies, JR and Motyl, M. (2012). Scientific utopia: II. Restructure incentives and practices to promote truth about advertising. Perspectives on Psychological Science, 7 (6): 615-631.