the chronicle of higher education has an intriguing post in yesterday’s edition, about fraudulent images used in scientific publications. the images discussed in the article were altered in a way that made it look like the results of the research were different than the actual results, but the researcher commented that she was simply, “trying to present it even better.”
if a researcher does alter an image, those changes should be noted in the caption, a footnote, or in the methods section. it is easy to note that “the contrast in this image was increased to demonstrate more clearly the difference between a and b.” if you think of an image as data, a researcher should be able to understand that he/she should disclose all changes to images just as he would note any statistical weighting to data.
how do we go about altering images in a way that maintains their integrity? i started a draft of do’s/don’ts at this wiki: http://orgmonkey.wetpaint.com/page/images+as+evidence. will you check it out and make changes or add new things to think about? i’d like to be able to hand a copy of this to researchers when i meet with them to assist in image preparation for publications.
