by Back in the high school chemistry, I remember that I was waiting with my bank partner to form crystals on our stick in the cup of blue solution. Other groups around us jumped with joy when their crystals formed, but my group was just waiting. When the bell rang, everyone went out of me. My teacher came over, collected an unopened bag on the counter and told me: “Crystals cannot grow if the salt is not in the solution.”
Science worked for me: what you expect is clear and concrete. And if it doesn’t happen, they did something wrong.
If it were just so easy.
It took me many years to realize that science is not just a few activities where you know what will happen in the end. Instead, science is about discovering and generating new knowledge.
Now I am a psychologist who examines how scientists make science. How are new methods and tools used? How do changes happen in scientific areas and what disabled changes in the way we do science?
A practice that has fascinated me for many years is replication research, in which a research group tries to repeat an earlier study. As with the crystals, it is not always that the same result from different teams does not always happen, and if they are in a team whose crystals do not grow, they do not know whether the study did not work because the theory is wrong or whether they have forgotten to put the salt into the solution.
The replication crisis
An executive regulation in May 2025 by President Donald Trump emphasized the “reproducibility crisis” in science. While reproducibility and reproducibility may sound similar, they are different.
Reproducibility is the ability to use the same data and methods from a study and to reproduce the result. In my editorial role in the Journal Psychological Science, I carry out computing reproducibility tests in which we carry out the registered data and check whether all results can be reproduced independently at work.
But we do not carry out the study again or collect new data. While reproducibility is important, research that is incorrect, false and sometimes harmful can continue to be reproducible.
In contrast, replication repeats when an independent team repeats the same process, including collecting new data, to determine whether it achieves the same results. If research work is reproduced, the team can be safer that the results are no coincidence or a mistake.
The “replication crisis”, a term in the early 2010s, has expanded to many areas, including biology, economy, medicine and computer science. Failures in replication of top -class studies affect many scientists in these areas.
Why replicate?
Replicability is a central scientific value: researchers want to be able to find the same result again and again. Many important insights are only published when they are replicated independently.
Random results can occur in research. Imagine that one person has turned a coin 10 times and got two heads, then told the world that “coins have a 20% chance of finding their heads.” Although this is an unlikely result – about 4% – it is possible.
Replications can correct these random results and scientific errors to ensure that science is corrected itself.
In search of the Higgs Boson, two research centers in CERN, the European Council for Mern Research, Atlas and CMS, he reproduced the detection of a particle with a large unique mass, which led to the Nobel Prize 2013 in physics.
The initial measurements from the two centers actually estimated the mass of the particle as slightly different. While the two centers did not find identical results, the teams rated them and found that they were close enough. This variability is a natural part of the scientific process. Just because the results are not identical does not mean that they are not reliable.
Research centers such as CERN have installed replication in their process, but this is not possible for all research results. For projects that are relatively inexpensive, the original team often reproduces their work before publication – however, this does not guarantee that an independent team could achieve the same results.
If projects are costly, urgent or time -specific, it is often not possible to replicate them independently before spreading results. Do you remember as a person all over the country for a Covid 19 vaccine?
The initial Pfizer-Biontech Covid 19 vaccine lasted 13 months from the beginning of the study to the approval of the Food and Drug Administration. The results of the first study were so clear and convincing that replication would have been unnecessarily delayed to bring the vaccine to the public and to slow down the spread of diseases.
Since not every study can be replicated before the publication, it is important to carry out replications after the publication of studies. Replications help scientists to understand how well research processes work, identify errors and self -correction. What is the process of carrying out replication?
The replication process
Researchers were able to repeat the work of other teams, such as CERN. And that happens. But if there are only two studies – the original and replication – it is difficult to know what to do if you do not agree. For this reason, large multigroup teams often carry out replications in which all the same study replica.
Alternatively, each team can replicate a different study when it comes to appreciating the reproducibility of research – for example Cancer biology – and the focus is on the percentage of studies that replicate across many studies.
These large -scale replication projects have been created worldwide and include manylabs, many babies, psychological accelerators and others.
Replicators start learning as much as possible about how the original study was carried out. You can collect details about the study from reading the published paper, discuss work with your original authors and advise online materials.
The replicators want to know how the participants were recruited, how the data was collected and which tools were analyzed.
Sometimes studies can skip important details, such as the questions asked to the participants, or the equipment used. Replicators must make these difficult decisions themselves, which can affect the result.
Replicators also often explicitly change the details of the study. For example, many replication studies with larger samples – more participants – are carried out than the original study to ensure that the results are reliable.
Registration and publication
Unfortunately, replication research is difficult to publish: only 3% of papers in psychology, less than 1% in education and 1.2% in marketing are replications.
If the original study repeats, magazines can reject the paper because there is no “new insight”. If it does not replicate, magazines can reject the paper because they assume that the replicators have made a mistake – think of the salt crystals.
Due to these problems, replicators often use registration to strengthen their claims. Pre -registration is a public document that describes the plan for the study. It is stamped before the study is carried out.
This type of document improves transparency by recognizing changes in the plan for experts. Registered reports go one step further, the research plan is subject to the Peer check before the study was carried out.
If the journal approves registration, publish the results of the study regardless of the results. Registered reports are ideal for replication research, since the experts do not know the results if the journal is committed to publishing the paper and whether the study is repeated or not whether it is published.
About 58% of registered reports in psychology are replication studies.
Replication research often uses the highest standards in research practice: large samples and registration. Although not all replication research have to use these practices, those who contribute strongly to our trust in scientific results.
Replication research is a useful thermometer to understand whether scientific processes as intended. The active discussion of the replicability crisis in both scientific and political areas suggests many researchers that there is space for growth. While no field would expect a replication rate of 100%, new processes among scientists aim to improve the rates of the past.
This article will be released from the conversation, a non -profit, independent news organization that brings you facts and trustworthy analyzes to help you understand our complex world. It was written by: Amanda Kay Montoya, University of California, Los Angeles
Read more:
Amanda Kay Montoya is Associate Professor at the University of California in Los Angeles. She is a member of the Board of Directors of the Center for Open Science. It receives funds from the US National Science Foundation.
