Ahh ... effect. You would think that the use of the word "effect" or the phrase "has an effect on" or "X affects Y" would be/should be a clear signal of a purported causal relationship. However, in my research with consumers reading headlines about scientific research in the popular press I have found that readers confuse "effect/affect" with correlation more than just about any other causal verb. In a content analysis we conducted on 16 weeks of Google Health News headlines we found over 300 different causal verbs used. Not surprisngly, consumers sometimes have difficulty recognizing these as causal verbs even though that is there intended meaning in the headline. Yet, when we have asked readers to interpret scientific headlines as causal, correlational, or descriptive, the causal headlines using some form of "effect" give readers more trouble than the headlines with other causal verbs we have used. One possibility for such confusion is that I often see the phrase "direct effect" used when someone is talking about a positive correlation. "There is a direct effect between X and Y."
Yes, in review of these actual headlines, "risk factors" is commonly used in correlational headlines. That is misleading use of terminology, but an accepted phrase in medical research. Jon =============== Jon Mueller Professor of Psychology North Central College 30 N. Brainard St. Naperville, IL 60540 voice: (630)-637-5329 fax: (630)-637-5121 [email protected] http://jonathan.mueller.faculty.noctrl.edu >>> Annette Taylor <[email protected]> 7/19/2017 9:07 AM >>> Back in the good old days....when I was in graduate school...I specifically being told by my advisor that "effect" could not be used in a title unless it was a clearly causal effect. So this does err on the side of emphasizing causal. Nevertheless, I also heard somewhere from someone (???) that the reason that the APA guidelines reduced the maximum number of words for a title in APA style was to focus on the actual variables in the title and eliminate any suggestion of "effect" in the title to reduce the abuse of the term "effect" Now, it makes for splashier headlines when your study gets published and people can talk about something BY INFERENCE "causing" something else simply because it is systematically linked with it. Finally, on a similar topic, I woke up this morning to a news story about "risk factors" for Alzheimer's and my immediate thought was, how are these things "risk factors?" Specifically it mentioned hearing loss and sleep apnea. My understanding of a "risk factor" when talking about health research is that these are things that are either set: a family history of ....xyz; or something we can manage such as obesity or smoking. So hearing loss may be associated with Alzheimer's, might predict that some amount of the variance in developing Alzheimer's is accounted for by something like hearing loss. But is the use of the phrase "risk factor" correct in this instance. Again, it seems to be a phrase that is being abused, much like "effect" is being abused. Early morning musings--so they might be mushy. Annette Annette Kujawski Taylor, Ph.D. Professor, Psychological Sciences University of San Diego 5998 Alcala Park San Diego, CA 921210 [email protected] On Tue, Jul 18, 2017 at 10:00 PM, Teaching in the Psychological Sciences (TIPS) digest <[email protected]> wrote: Subject: Opinions needed From: Dap Louw <[email protected]> Date: Tue, 18 Jul 2017 18:27:55 +0000 Tipsters I am well aware that (and often frustrated by) all sorts of buzz words, concepts, theories, etc become the flavour of the month/year in organizations, including universities. I would therefore appreciate your viewpoint on the following, especially as research methodology is not my field of specialization: To what extent can we measure 'effect'? In the last 40 years in Psychology I've been involved in hundreds of studies on "The effect of ......... (television on ...; poverty on ....., etc, etc)". BTW, when I used ' "the effect of" psychology' in Google Scholar search I got 2 460 000 results. However: According to the latest recommendations of our University's Research Committee we cannot measure effect unless you make use of especially the longitudinal design. Therefore any title such as "The effect of ......... (television on ...; poverty on ....., etc, etc)" is unacceptable and should be replaced by "the perceived effect of ....." or something similar. Is this a case of methodology or semantics? I look forward to hearing from you. It's high time to get the TIPS ball rolling again! Regards from this side of the ocean. Dap --- You are currently subscribed to tips as: [email protected]. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13269.01f6211e00cc8f00a7b68e8e24b1b4d6&n=T&l=tips&o=51065 (It may be necessary to cut and paste the above URL if the line is broken) or send a blank email to leave-51065-13269.01f6211e00cc8f00a7b68e8e24b1b...@fsulist.frostburg.edu --- You are currently subscribed to tips as: [email protected]. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=51067 or send a blank email to leave-51067-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
