Bias can be defined as an act "to present or hold a partial perspective at the expense of (possibly equally valid) alternatives." Bias can come in many forms and could impact a scientific research paper in a number of different ways, none positive. The purpose of scientific research is to present evidence supporting possible new conclusions regarding some question in the scientific field. Bias would invalidate the information being presented.
Bias could take the form of omitting the relationship or results created by a variable in an experiment. For example, if a new type of metal alloy is being presented for use in building airplane bodies, it would be important to verify the durability of that metal under the extreme stresses that would be repeatedly exerted on it during takeoff and landing conditions. If a paper's authors were intent on having this metal widely adopted by the manufacturers, information regarding performance of the alloy during tests might be minimized or omitted for some conditions that could be critical.