The High Society: Dear I-O
Paul M. Muchinsky*
The University of North Carolina at Greensboro
*Unamused, indifferent, or entertained readers can contact the author at
At the SIOP conference in Los Angeles there was a session on TIP. Each columnist was asked to give some thought as to how
TIP might be made better. Although I believe it is difficult to improve upon perfection, I did give some serious thought to the request. The idea I came up with should be regarded as nothing more than a trial balloon.
The first advice columnist I can ever remember reading was Ann Landers. Everyday in my local newspaper were letters written by people seeking advice. They all began Dear Ann. Another one was written by her sister Abigail Van Buren, whose column appeared under the heading Dear Abby. More problems, more advice. These advice columns have exploded in popularity. One column gives advice on how to play a hand in bridge. Another one gives advice on how to play a hand in poker. Two guys on National Public Radio have a call-in show where they give advice on car problems.
I say it might be time for SIOP to get into this act. Nowadays just about everyone needs advice on something. I-Os are no better or no worse than the rest of humanity. I say if our members need advice, perhaps we should step up and provide it. I say maybe
TIP should start a regular advice column. Our members would submit questions on troubling issues, and a learned member of the
TIP staff would give the advice. To avoid even the appearance of conflict of interest, I asked a fellow SIOP member if he would serve as a guest advice columnist for this trial demonstration. He graciously agreed. Five letters were privately solicited. Here are the letters and the advice given. You be the judge.
For the longest time I have been telling people that validity coefficients are an underestimate of the true relationship among constructs. I advocated using various formulas to correct their magnitude. I even went so far as to refer to these corrected validity coefficients as an estimate of the truth. Lately however, I am beginning to question my own thinking and reasoning. I mean, in real life, things are what they are. For example, 7:30 is 7:30. 7:30 is not an estimate of 9:30, no matter how many corrections are applied to it. Im feeling so confused. Please help.
It sounds like you are going through the change of academic life. There is absolutely nothing wrong with applying hydraulics to correlations to inflate their magnitude. Size matters in life, and bigger is usually better. You are a statistical optimist. You see the glass as half full, not half empty. Even if the glass is not half full, after applying your correction formulas, you can make it appear that way. Imagine all the people youve made happy by convincing them that uncorrected correlations of .20 are simply corrected correlations of .60 in an earlier stage of development. Think of them as larva morphing into butterflies. Most people prefer to look at butterflies rather than larva, including journal editors. You fail to understand your overwhelming positive impact on our field. You single-handedly improved the mental health of an entire profession. For more than 75 years we believed our validity coefficients were serving permanent life sentences in the Punyville Penitentiary. Then you found a way for us to escape these feelings of inadequacy. I am personally saddened that you are going through this period of self-doubt in your career. I know that a particular
TIP columnist has, from time to time, poked fun at some of your ideas. He is simply envious of your professional acclaim, born of your superior quantitative ability. His quantitative ability is so low the only way he can count to 11 is if he puts both hands in his pockets. If his comments have in any way contributed to your current state of funk, the next time you go pheasant hunting, pretend he is one of those pheasants you are aiming at. I said pretend. Furthermore, quit calling corrected correlations an estimate of the truth. As you know, any numerical value can be an estimate; some are just more accurate than others. Call your corrected correlations the truth. If you get pushback, write an appendix to prove you are right. Besides, every hour on the half-hour it is 9:30 someplace in the world. And thats the truth.
Im in a pickle and I need a way out. I have been promoting 360-degree feedback as a way to impress clients. Ive been telling clients that information from multiple sources is always more accurate than feedback from one source. But here is my problem. Although information from multiple sources may be more accurate, the different sources are not equally important. In fact, the only source that matters at all is what your boss thinks of you. On a conceptual level Ive painted a pretty picture, but on an operational level its getting pretty ugly. Peers and subordinates cant fire you, and you cant fire yourself. What should I do?
I see your problem and its a big one. How long did it take you to get this blinding glimpse of the obvious? I say a geometric problem requires a geometric solution. Subordinates will never say anything bad for fear of retaliation. Peers will knife you in the back at the first chance to enhance their own careers. If you are still going to stick with this 360-degree albatross, tell your clients 360-degree feedback is nothing more than four subordinates being evaluated by their boss, because 4 times 90-degrees is 360-degrees. Negative feedback from your boss can be grounds to say, Youre fired. If your clients demand self-assessments as well, tell them a negative self-assessment can be grounds to say, I quit. In the future I suggest that as an I-O psychologist you stick with using multisyllabic words to impress clients, and leave the geometry to the mathematicians.
I work for a large corporation as an internal change agent. Last week the CEO pulled me aside and said I was to find a way to motivate the entire workforce. Furthermore, I was told my results must be immediately effective. I panicked. I went through every motivation book I could get my hands on, and now I am more confused than ever. Is motivation a trait I should look for in people, should I make people feel inequity to get them to work harder, do I impose a variable schedule of paychecks, do I determine valences, instrumentalities, and expectancies, or what?
None of the above. Given the urgency of your problem and its magnitude there is only one solution that is viable. The solution has an anatomical origin. Unbeknownst to the medical profession, there is a nerve that runs directly from the buttocks to the brain. A swift kick applied to the buttocks by a supervisor at the start of each workday will most likely achieve the results your CEO desires. However, I feel its advisable to bring two issues to your attention. First, although this particular technique may not motivate behavior for some recipients, it will nonetheless give the respective donors a great measure of satisfaction. Second, and this almost goes without saying, do not place one-legged people in supervisory roles.
Im head of organizational effectiveness for a large global corporation. Part of my job is to ensure uniform selection methods in all our locations around the world. It seems every location wants to use some pet selection method they are in love with. This one wants to use the interview, that one wants to use a personality inventory, another one wants to use graphology, and so on. Not only do they use different selection methods, but each one is absolutely committed to whatever method it uses. Im ready to pull my hair out. Any advice on how to solve this problem?
You actually have two problems, not one. The first problem is the differing selection methods across locations, and the second is the high degree of commitment each location has for the method it uses. There will be strong resistance to change. You said your company is global. This is the key to the solution. Because you are head of organizational effectiveness, it is your legitimate responsibility to effectuate change. Here is my advice. Get a high school cheerleaders uniform, complete with a megaphone. At each location tell them your job is to make the company great. Bellowing into the megaphone, ask them if they want to become great. Of course, they will all say, Yes! Now you have attained the critical buy-in you need. Then you yell into the megaphone, Give me a
g! They will all yell, g! You then yell, So you want a
g?! They will all yell, Yes! You then say, Because you want a
g, Ive got a g for you! Then you whip out of your briefcase a test of g, and tell them it is this test they will now use to hire all employees. Remind them that great always begins with
g. Repeat this fandango in every location of the company around the world. These folks wont know what hit them. You will become a legend in the corporation. Be sure to shave your legs before each performance.
I am a graduate student. I am just putting the finishing touches on my dissertation. It is a psychometric study. I derived the denominator term for the density function of the robust semiparametric Clafdapka distribution under conditions of nonnormality. I recently accepted a job with a HR consulting firm. My soon-to-be boss seems to be a wonderful man. He told me when presenting data to clients, they like it in the form of pie charts and bar graphs. They can understand percentages, and some can understand means, but I should never present standard deviations. He tells me measures of variability are too complex for clients to understand. He also told me clients like information in bullet form. I will have to learn about bullets because I have never fired a gun in my life. Given my job choice, do you think I wasted a lot of time studying item response theory, confirmatory factor analysis, structural equation modeling, and hierarchical linear modeling? Do you think I can convince every client that the world is best viewed through the lens of a variancecovariance matrix?
P.S. Congratulations on getting this job, your reward for many years of dedicated study! Would you kindly send me a postcard when you celebrate the first anniversary of your employment with this companythe first month anniversary?
So, what do you think? Should TIP have an advice column? Obviously this decision will have to be made by someone at a pay grade far above mine. Please send your thoughts to
Laura Koppes, editor of TIP. Just dont recommend me as the advice columnist. I already have the tough job of making you laugh once every 3 months.
July 2005 Table
of Contents | TIP Home
| SIOP Home