Beth Loftus is not a fan of the replicability movement

Professor Elizabeth Loftus, legend in psychological science due to her influential work on memory, is not a fan of open scientific practices such as direct replications and making the data sets publicly available. And she defends the scientist’s right to peek at the data while it is being collected.

“The many strict, methodological precepts did not make scientific research any more fun. For instance, you would be disallowed from peeking at your data as it is being collected. But how can you be a curious researcher if you can’t check at an early stage whether your research is headed in the predicted direction? Should I really feel guilty when I do this? Nowadays there is much hostility among researchers, some feel harassed because they have to make their data publicly available. Sometimes researchers tell me that they wish to replicate my very first study on the misinformation effect. I just think Really, what’s the point? I don’t believe that single studies are important in psychology. Moreover, hundreds of studies have been done on that effect. Back in the day we used borrowed film material that we returned: Good luck retrieving that!”

Obviously we can’t know for sure what practices Loftus herself has used in her work; however, her statement kind of implies that at least some of her work indeed is based on questionable research practices (QRPs). Not very surprising given the recent literature showing that QRPs are norm rather than exception in psychological science. What disturbs me, thus, is not the fact that her work (likely) includes QRPs, but rather that she chooses to the defend them.

EJ Wagenmakers responded - in a too careful way? - in a blog post yesterday (Feb 21, 2019):

They both seem reasonable and the conversation was very productive. Fun and spontaneous creativity is very important and should not be discouraged.

To me it seemed as Loftus didn’t really understand what Wagenmakers said about the data peeking. The types of statistical analyses used in Loftus’ field are not Bayesian or sequential. Thus, the type of data peeking Loftus referred to is - most likely - to consider questionable. I believe Wagenmakers could and should have made this clearer in the post.