This is a common belief. However, it is not true in most cases. To help dispel this myth, let's look at the Oklahoma law regarding the licensing of engineers (O.S. Title 59, Section 475). This statute defines the "practice of engineering" as follows:
"...any service or creative work, the adequate performance of which requires engineering education, training and experience in the application of special knowledge of the mathematical, physical and engineering sciences to such services or creative work as consultation, investigation, evaluation, planning and design of engineering works and systems, planning the engineering use of land and water, teaching of advanced engineering subjects or courses related thereto, engineering research, engineering surveys, engineering studies, and the inspection or review of construction for the purposes of assuring compliance with drawings and specifications; any of which embraces such services or work, either public or private, in connection with any utilities, structures, buildings, machines, equipment, processes, work systems, projects, and industrial or consumer products or equipment of a mechanical, electrical, chemical, environmental, hydraulic, pneumatic or thermal nature, insofar as they involve safeguarding life, health or property, and including such other professional services as may be necessary to the design review and integration of a multidiscipline work, planning, progress and completion of any engineering services."
Granted, your state may have a different definition, but my guess is that the above is fairly representative for the country.
I underlined some tasks to emphasize that research and other studies are not exempt if they eventually lead to structures, processes, systems or products (i.e. final designs). It makes perfect sense that the engineering decisions made during the final design process are dependent upon the engineering studies and research that was used as reference data and recommendations prior to the design. I know this might be a stretch, but I could make a case that even studies that have "negative" results might be considered as part of a design basis. For example, if a design option was excluded due to an experiment with a negative outcome, that research and outcome had an impact on the design. If the research was conducted improperly, or the results were interpreted improperly, the best design option might have been excluded. That in turn might have an adverse effect on safety. Bottom line: final design includes everything that came before. If your work was involved in the "everything that came before", you should have a PE license.