@mhhfive that's true. I remember learning about the epicycles that were needed to make those orbits work out. The mathematics involved were quite comlex. 

Remember we are talking about design research here. Not people getting angry when IT asks them to verify that their machines are plugged in. 

I've moderated over 1,000 usability study sessions, and we very consistently see people blaming themselves when they are unable to complete a task, especially after they find out the way the designers intended them to complete the task. I've either managed to recruit more than 1,000 people with incredibly low self-esteem, or there is something else at play here.

And this isn't necessarily employees looking at their colleagues' or management's designs. Sometimes it's employees of a 50,000 person company looking at internal tools. Sometimes they're looking at third-party tools. Sometimes people are looking at the work of third-party consultants. Sometimes it's potential customers looking at stuff that hasn't even been released yet. Most of these people have zero vested interest in the outcome. In fact, usability study moderators go to great lengths to assure participants that they are not the designers, no one's feelings will be hurt regardless of the outcome, and that we are interested solely in the truth, good or bad. 

I'd suggest that IT people have their own bias, that the user is always at fault, because they aren't using the system the way a designer intended. That's what IT typically considers the "right" way. I know many IT people who, had they received a help call about the knife issue I described in my other comment, would absolutely turn around and tell their colleagues, 'Idiot was looking under "kitchen utensils,'" without a single thought as to whether that was reasonable or not. It just wasn't where the information was, therefore, the user was looking in the wrong place - i.e., the source of the error.

Almost everything that gets released passes QA testing - it does what it's designed to do. If there is a "save document" requirement, the software is technically capable of saving a document. But if the user can't figure out how to do it, it's still a poor design. 

Perhaps an example from the physical world will help:

http://baddesigns.com/doors.html

Basically, this is your standard "user-trying-to-pull-a-push-door" problem. The designers intended for people inside the vestibules to push on the doors to exit either side. But they put pull handles on *both* the pull and push sides of both sets of doors. People actually, literally, didn't know how to get out. 

Imagine a call to a help desk:

"Help! I"m trapped in a corridor! Both sets of doors are locked from the inside!"

The helpdesk person would help, then mute the phone and laugh and laugh at the stupid user who didn't push the door. From their perspective, the user made a mistake - they pulled when they should have pushed. 

And the user might also feel stupid. ("Damn. Why didn't I think to push on the door?!") Or they might (correctly) blame the bad design. (Though it would certainly be wrong to blame the corridor door help desk.)

Either way, the fact remains that it's a bad design and vastly increases the likelihood of such easily-preventable errors.

(For more great examples, check http://baddesigns.com and http://thisisbroken.com - both old, but the examples are pretty timeless.)

It's interesting, actually - some of the best ROI for this kind of work comes in the form of reduced calls to the help desk. 

Some of these biases seem easier to correct for than others. For example, how would you correct for the Reverse Confirmation Bias?

all research has inherent biases.. when the early astronomers looked up at the stars, they were biased towards thinking everything revolved around the Earth.. and in perfect circles. And incredibly complex models of the universe were devised based on these assumptions!

It's really amazing how far research can go when it's based on an underlying assumption that turns out to be wrong... 

@dgcooley but do they have a vested interest in saying the design is not at fault? Perhaps they feel their jobs are more secure if they affirm the design, rather like no one wants to tell the emperor the truth about his new clothes.  In regular tech use, as just confirmed my resident IT guy, the users always blame the program when really they have no clue. 

It's true that the FAE is better-documented than the RFAE and may be more prevalent in real-world scenarios. Certainly, you hear a lot of "Stupid computer!" coming from people.

But that's not what we usually see in formative design research. It may well be that that's one of the only places the Reverse FAE applies. It's incredibly rare to see someone blame the design in that setting.

I was aware of a couple of biases (which I learned when I was still in college), but I had no idea that they were this many! Most people I know pride themselves for being fair and unbiased, but to find someone who is truly unbiased is difficult. A truly unbiased person is hard to come by these days.

Generally, references to the Hawthorne effect all concern effects on an experiment's results of the awareness of participants that they are the subject of an intervention. However there are many different possible mechanisms, and all may be important in particular cases. What is not disputed is that there is an important issue here, and it is clear that there is a need for a term to refer to these issues: the term "Hawthorne effect" tends to get re-appropriated for any issue in the general area.

Learned link (long discussion)

@dgcooley That is what I inferred, but from what I find, the reverse attribution is not considered the norm. In fact on http://allpsych.com/psychology101/attribution_attraction.html, the explanation is as follows: " Self-Serving Bias.  We tend to equate successes to internal and failures to external attributes (Miller & Ross, 1975).  Imagine getting a promotion.  Most of us will feel that this success is due to hard work, intelligence, dedication, and similar internal factors.  But if you are fired, well obviously your boss wouldn't know a good thing if it were staring her in the face. This bias is true for most people, but for those who are depressed, have low self-esteem, or view themselves negatively, the bias is typically opposite.   [emphasis mine] For these people, a success may mean that a multitude of negatives have been overlooked or that luck was the primary reason.  For failures, the depressed individual will likely see their own negative qualities, such as stupidity, as being the primary factor."

In other words, most people don't say, "oh, I should have figured out that knives would be there." Rather, they would say, "What kind of an idiot classifies knives under 'chef's supplies' rather than 'kitchen utensils?'" Only someone who is lacking a great deal of confidence would likely say, "Well, they know better than I do what the correct classification is."

While I haven't seen it classified as a bias, the situation that I would consider closest to the the case where a person defers to another in a hierarchy, despite the fact that lower-ranked person is correct. That situation has come up hospital settings as described in Speaking up improves anaesthesiological treatment "Errors in the operating theatre can have tragic consequences. Researchers from ETH Zurich and the University Hospital Zurich demonstrate that the efficiency of anaesthesia teams depends greatly on their ability to communicate openly and speak up constructively with regard to the performance of colleagues."

@lin :"I would personally lend a lot of credence to the Hawthorne Effect - mostly because it seems to be the Heisenberg Uncertainty Principle appled to social research." I was thinking the same thing.