4.1 KiB
| title | chunk | source | category | tags | date_saved | instance |
|---|---|---|---|---|---|---|
| Invalid science | 3/3 | https://en.wikipedia.org/wiki/Invalid_science | reference | science, encyclopedia | 2026-05-05T07:02:34.957128+00:00 | kb-cron |
=== Lack of access to data and software === Clinical trials are generally too costly to rerun. Access to trial data is the only practical approach to reassessment. A campaign to persuade pharmaceutical firms to make all trial data available won its first convert in February 2013 when GlaxoSmithKline became the first to agree. Software used in a trial is generally considered to be proprietary intellectual property and is not available to replicators, further complicating matters. Journals that insist on data-sharing tend not to do the same for software. Even well-written papers may not include sufficient detail and/or tacit knowledge (subtle skills and extemporisations not considered notable) for the replication to succeed. One cause of replication failure is insufficient control of the protocol, which can cause disputes between the original and replicating researchers.
== Reform ==
=== Statistics training === Geneticists have begun more careful reviews, particularly of the use of statistical techniques. The effect was to stop a flood of specious results from genome sequencing.
=== Protocol registration === Registering research protocols in advance and monitoring them over the course of a study can prevent researchers from modifying the protocol midstream to highlight preferred results. Providing raw data for other researchers to inspect and test can also better hold researchers to account.
=== Post-publication review === Replacing peer review with post-publication evaluations can encourage researchers to think more about the long-term consequences of excessive or unsubstantiated claims. That system was adopted in physics and mathematics with good results.
=== Replication === Few researchers, especially junior workers, seek opportunities to replicate others' work, partly to protect relationships with senior researchers. Reproduction benefits from access to the original study's methods and data. More than half of 238 biomedical papers published in 84 journals failed to identify all the resources (such as chemical reagents) necessary to reproduce the results. In 2008 some 60% of researchers said they would share raw data; in 2013 just 45% do. Journals have begun to demand that at least some raw data be made available, although only 143 of 351 randomly selected papers covered by some data-sharing policy actually complied. The Reproducibility Initiative is a service allowing life scientists to pay to have their work validated by an independent lab. In October 2013 the initiative received funding to review 50 of the highest-impact cancer findings published between 2010 and 2012. Blog Syn is a website run by graduate students that is dedicated to reproducing chemical reactions reported in papers. In 2013 replication efforts received greater attention. Nature and related publications introduced an 18-point checklist for life science authors in May, in its effort to ensure that its published research can be reproduced. Expanded "methods" sections and all data were to be available online. The Centre for Open Science opened as an independent laboratory focused on replication. The journal Perspectives on Psychological Science announced a section devoted to replications. Another project announced plans to replicate 100 studies published in the first three months of 2008 in three leading psychology journals. Major funders, including the European Research Council, the US National Science Foundation and Research Councils UK have not changed their preference for new work over replications.
== See also == Metascience (research) Replication crisis Reproducibility Project Retraction Watch Séralini affair Statistical correlation
== References ==
== External links == "Has science gone wrong?". The Economist. 2013-10-19. Retrieved 2013-10-22. O'Grady, Cathleen (2020-12-09). "Psychology's replication crisis inspires ecologists to push for more reliable research". Science | AAAS. Retrieved 2020-12-16.