The replicability crisis in psychology

Repeating research while obtaining the same results as other scientists is a challenge.

In recent years, since the early 2010s, the scientific community has been calling attention to the existence of a replicability crisis in science, especially in psychology and medicine: the results of much research are impossible to replicate.The results of much research are either impossible to replicate or attempts to replicate them are simply not being made.

However, problems related to the confirmation of hypotheses are not the only ones that fall under the replication crisis, which has a broader character. In this regard, the relevance of falsification of results, particularly in the field of social psychology, and other very significant methodological factors should be highlighted.

The replicability crisis in science

One of the foundations of the scientific method is the replication of results.. Although many people have a strong tendency to take the conclusions of a single study as credible and definitive, the truth is that a hypothesis only becomes truly solid when it is confirmed by several valid studies from different research teams.

In the same vein, negative results, i.e., the refutation of hypotheses, are just as important as their verification. However, the proportion of studies that refute hypotheses seems to have declined in science in general; consequently, there is a clear of publications that corroborate experimental hypotheses, i.e., the refutation of hypotheses, as well as their verification..

Many of the publications that have been carried out on the replication crisis highlight the magnitude that it has taken on in psychology. However, it should be made clear that this crisis affects science as a whole and has a particular intensity in the case of medicine. and also has a particular intensity in the case of medicine. This is due to a series of interrelated factors.

The main causes of this phenomenon

A meta-analysis carried out by Daniele Fanelli (2009) concludes that publication fraud is more common in medical and pharmaceutical research than in other fields. than in other fields. The author suggests that this may be due to the greater economic incentives for publications or to a higher degree of awareness in these fields.

There are, however, several factors that influence the replicability crisis beyond the explicit falsification of data. One of the most significant is the selectivity of publications: in general, positive and eye-catching results have a greater potential to appear in journals and to provide recognition and money to researchers.

This is why the "crate effect" frequently occurs, whereby studies that do not support the expected hypotheses are studies that do not support the expected hypotheses are discarded, while those that do are while those that do are selected by the authors and published more commonly. In addition, the non-replication of positive studies reduces the risk of the hypotheses being disproved.

Other common practices that have similar objectives are to select a large number of variables and subsequently focus only on those that correlate, to modify sample sizes (e.g., include subjects until the results are positive), or to conduct multiple statistical analyses and report only those that support the hypotheses.

Why is it so serious in psychology?

The replication crisis in psychology is considered to date back to the early 2010s. During this period numerous cases of fraud involving relevant authors emerged; for instance, the social psychologistFor example, the 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 early 20th century are replications of previous studies. This is a very low figure since it strongly suggests that many of the conclusions obtained by isolated studies cannot be taken as definitive.

The number of successful independent replications is also low, at around 65%.The number of successful independent replications is also low, at around 65%; on the other hand, more than 90% of those carried out by the original research team corroborate the hypotheses. On the other hand, papers with negative results are also particularly infrequent in psychology; the same is true of psychiatry.

Solutions to the research crisis

The replicability crisis in psychology and in science in general not only compromises the results of a large number of studies, but may also lead to the legitimization of hypotheses that have not been confirmed with the necessary rigor. with the necessary rigor. This could lead to the widespread use of incorrect hypotheses, altering the development of the sciences.

At present, there are many economic interests (and others related to prestige) that favor the continuation of the replication crisis. As long as the criteria followed for the publication of studies and the dissemination of their results in the mass media continue to have this monetarist character, the situation is unlikely to change.

Most of the proposals that have been made to help solve this crisis are associated with rigorous methodology in all its aspects. rigorous methodology in all its phases, as well as theThe majority of proposals that have been made to help solve this crisis are associated with rigorous methodology in all its phases, as well as with the participation of other members of the scientific community; in this way, the peer-review process should be strengthened and efforts to encourage replication efforts should be encouraged.

Concluding

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. This makes it very easy for elements that are not taken into account in the research to "contaminate" the results.

On the other hand, the limitations of the ways in which it is decided whether there are real phenomena or only statistical phenomena means that false positives sometimes appear: the simple fact that the p-value is significant need not be sufficient to indicate that it reflects a real psychological phenomenon.

Bibliographical references:

• Fanelli, D. (2009). How many scientists fabricate and falsify research? A systematic review and meta-analysis of survey data. PLoS ONE 4(5).

• Makel, M. C., Plucker, J. A. & Hegarty, B. (2012). Replications in psychology research: how often do they really occur? Perspectives on Psychological Science, 7(6): 537-542.

• Nosek, B. A., Spies, J. R. & Motyl, M. (2012). Scientific Utopia: II. Restructuring incentives and practices to promote truth over publishability. Perspectives on Psychological Science, 7(6): 615-631.