Do not try to hide negative or confounding results because they do not support your argument. If you have “unexpected results”, you should still accept them and discuss them in your work. Transparent reporting is the universal requirement.
It is scientifically important to report negative results. Some negative results are of great interest precisely because they contradict previously published results. Reporting negative results may be valuable to scientists! It might save them from repeating experiments that have already been performed. Therefore, it is important that all results are published, regardless of the outcome.
Publication bias can arise if the literature contains only positive results. Because negative results naturally attract fewer citations, many scientists simply do not attempt to publish negative results. But this has a negative impact for science. Theories that are false or incomplete might never be corrected. Instead we must serve the advancement of science.
Negative results also have an important impact on meta-analyses, which pool and analyze all available data. Meta-analyses draw conclusions from a series of related studies, so they are often important for validating clinical outcomes. Not having negative results may distort meta-analyses. The studies included in the analyses will be biased toward showing the preferred effect.
To address this problem, open access and broad-scope journals (e.g., PLOS ONE, Frontiers, and Biomed Central journals) are increasingly publishing papers with negative findings. Other journals have special sections to publish negative results. You can also save your negative results in data repositories such as Dryad and FigShare. Dryad will archive only data associated with a published study, but almost anything can be placed with FigShare, including poster presentations and slides. These all receive a permanent number or digital object identifier (DOI). This number allows them to be discoverable and citable.
Negative results are not “bad”. Bad results only come from poor study design or inappropriate methods. Your readers have a right to critically evaluate your work. They should be allowed to review all of the findings related to your study, both positive and negative. Don’t hide any results from your readers. Doing so is dishonest.