Hi Bob,
So far, I have made proposals to which you have responded, not always charitably I might add. I like to apply the principle of charity to all my communications. I would therefore like to offer my apologies if you think I was ignoring you but the error is yours. I have actually answered most of your points and even pre-empted many.
I construct - you demolish. Like others, I have found your comments amusing and I accept that this may be the way you prefer to work. I can accept that we are all different. I would actually prefer to state my case and let you state yours. Others can then read, compare and contrast the merits of each argument and reach their own decisions. I find your style abrasive at best and offensive at times.
I see demolition as a slight intellectual challenge compared with construction but if you insist: -
"And you think employees were treated better before testing? In the good old days the wrong skin color could not get hired or promoted, the wrong faith was a career killer, getting fired was easy. The good old days are old for a reason and one reason is better selection practices. Governments have finally learned and so have some employers that predicting job success requires more than competence, education, college degrees, certifications, age, color, faith, etc."
You assume because I am retired, I want to go back to the ‘good old days’. It seems not to have crossed your mind that an ‘old duffer’ like myself is capable of suggesting moving ‘forward’ to something better, not ‘backwards’ to something I simply imagine was better. Again, the error is entirely yours. As a supporter of psychometrics, stereotyping other people is an error you seem unable to avoid. Psychometrics is after all an institutionalised form of stereotyping.
As for “PREDICTING job success”, this is the very reason I quoted Miller. Prediction and control (the bit you think unnecessary if the ‘right’ people are chosen) may be desirable in chemical or manufacturing processes. However, the prediction and control (even by default) of people by a “rich and powerful elite” what made Miller tremble. The quote may be old but it is still relevant.
There are two mistakes people often make when evaluating ideas. The first is to assume that if something is old, it is good. The second is to assume that if something is new, it is better. You accuse me of the first yet you constantly fall foul of the second.
In my experience, far from eliminating racism or sexism, psychometrics offers a safe haven for bigots hide and continue in their prejudices by simply saying the rejection of candidate X or Y was due to a poor test scores. Since evidence rather than reason impresses you, I offer the following evidence.
The UK Civil Service uses batteries of the most up to date and sophisticated psychometric tests available at a dedicated assessment centre using specially trained staff. The selection process takes up to three days for higher grades. Despite these rigorous procedures, women in the Civil Service are proportionately less well represented as they move up the promotion ladder (70% of the workforce are women at lower grades – 15% at senior grades). If you are not white, proportionate representation at senior grades is even worse. Figures can be verified on the UK Civil Service website.
Since I am unable to accept that white European or American males are intellectually superior, more capable or have more ‘talent’ than women or other races, I can only conclude that bias is built into the selection process. The Civil Service use similar procedures to your own. In light of this evidence, I disagree absolutely with any suggestion that psychometrics may help to eliminate sexism and racism. The tests are biased and help perpetuate white male supremacy.
"Perhaps you don't understand the meaning of "all is not well with psychometric measurement"? That statement does not mean all assessments cannot work. I think you understand the written word well enough to know that but choose to ignore the meaning."
If a respected geologist publicly announces “all is not well with the San Andreas Fault”, I would be foolish to wait ten years before moving my family away from San Francisco. I think I understand the English language perfectly. I am deaf. You have chosen not to hear. The difference apparently eludes you.
I am not a believer in ‘evidence’ divorced from reason. The ‘evidence’ you offer is not factual. ROI is merely a mathematical representation of former customers’ opinions. It is certainly not a scientific measure. Presenting facts about opinions does not convert opinions into facts. You seem to have difficulty distinguishing between fact and opinion. I prefer to rely on absolute facts and deductive reasoning.
I have already made reference to Piaget’s demonstration which I will recount briefly for the benefit of those unfamiliar. Piaget posed two questions, identical in format but with very different types of methodology for providing the answer.
1. Are there more apples or more oranges in the world?
2. Are there more fathers or more sons in the world?
To answer the first question we must count and compare the figures. However, since results may vary from season to season or from year to year, this empirical method only ever yields provisional knowledge. Psychometrics therefore only provides provisional knowledge since psychometrics rest on a foundation of quantitative statistical evidence – an empirical method.
The second question can be answered using deductive reasoning alone. Since every father is also somebody’s son, and some sons are not fathers, there will always be more sons than fathers. There are no circumstances under which there could ever be more fathers than sons so there is not the remotest chance that this answer could possibly be wrong.
It is not that I doubt your figures. I doubt the value and your interpretation of your figures. I doubt them because I doubt ALL evidence that contradicts deductive reasoning no matter how strong the evidence may appear to some.
"You have yet to explain why non-ipastive assessments cannot work. You seem to focus on ipsative assessments which should not be used for selection."
I have actually explained my reasons before. You just failed to understand them. I use only deductive reasoning to show why non-ipsitive methods cannot work. This way I can be absolutely certain that there are no circumstances under which I can be mistaken. Arrogant if I am unable to back it up – confident if I can.
Freud provides a clue. He said that the personality has three distinct parts which he named, ego, superego and id. This is often dismissed as merely Freud’s opinion. It is not.
• Ego - aspects of the personality are universal so common to ALL people.
• Superego - group centred or socio-cultural aspects of the personality shared by SOME people.
• Id - self-centred aspects of the personality are unique to ONE person.
While Kant may use the terms unity, plurality and totality, the terms ONE, SOME and ALL are more accessible. These terms permit no possible exception since NONE is simply the negation of ALL. There are no other possible options.
This is something my three year old grand daughter understands perfectly as does any three year old. Given a packet of sweets, she can eat ONE of them, SOME of them or ALL of them. None left = All gone. Pearls of wisdom fall out of the mouth of babes but ‘sophisticated adults’ often miss them.
• It is clearly pointless exploring the universal aspects of personality that are common to ALL people since these will be the same for ALL people.
• Since psychometrics relies upon comparing people with one another, psychometrics MUST focus on those aspects of the personality that are common to SOME people. This is the point at which the bias in the UK Civil Service above manifests itself. SOME people (usually white European/American middle aged males) decide which factors are important in the selection process and which are not. They can keep their prejudices, blame the test and evade any ethical doubts about their own integrity.
• Most aspects of the personality are unique to the individual, and as you quite rightly observe, these cannot be used for direct comparison between individuals. This does not mean they cannot be used for selection. Kelly’s repertory grid allows us to explore the individual and I have seen it used very effectively in helping to make decisions for the most senior posts.
In ‘Inquiring Man’, (1985) Don Bannister and Fay Fransella say this.
“The castrating effect of separating personality off as a mini-psychology in its own right is perhaps best seen in the curiously named study of ‘individual differences’ which turns out to be a study of ‘group sameness’. Here we have focused on the establishment of some general dimensions, at some point along which all individuals can be placed rather than on a study of the dimensions which each individual develops in order to organise their world. ….
…The attempt to encompass the person within the study of personality is additionally bedevilled by the persistence of trait psychology. The habit of seeing others in a rather simple, rigid and typological manner has stunted the life of many individuals and its formalisation in psychology has had a similar effect upon the discipline.”
Time to eat your own words Bob - “I think you understand the written word well enough to know that but choose to ignore the meaning.”
Psychometrics uses dimensions of personality common to SOME people (the chosen ones) but has the arrogance to impose these dimensions upon ALL and protest if others disagree. If you are still having difficulty with the concept, the following analogy may help.
Psychometrics assumes people are all like chocolate cakes with varying amounts of the same ingredients - butter, sugar, eggs, flour and cocoa. But, psychometrics cannot account for a coffee and walnut cake because it cannot account for differences due to qualitative variations in ingredients.
People differ qualitatively as well as quantitatively. Psychometrics set the standard ingredients – the personality factors thought to be important. The greatest variation between people is found in the qualities unique to the individual – not in the qualities we share with SOME people and obviously not in the qualities we share with ALL of the rest of humanity.
When looking for the ideal employee, we must take into account the qualities that make the individual unique rather than concentrating on measuring group sameness (aka conformity).
Chandrasekhar quite rightly challenges your right to impose your conformities upon others since this is unjustifiable and unethical. Ethics are not a final step to scientific procedures. Ethics is an attitude that must be built into any procedure, theory or business from the very beginning. This lack of ethics is what Miller foresaw and feared. This lack of ethics is apparent in your attitude. Mankind does not exist to enhance business. All business should be aimed towards enhancing mans existence.
Your defence of psychometrics can be seen as inevitable because it perpetuates the self interests of an elite group of chosen ones - your group - as chosen by your group.
As for the rest of your comments, Chandrasekhar has already dealt with many of the ethical points quite admirably and questioned you definition of ‘talent’. Thank you Chandrasekhar.
My view is that psychometric tests are a biased, dehumanising and unethical form of institutional stereotyping. I am not alone. Whilst you may have every right to disagree, your inability to do so without humiliating others is a sign that you are not happy with yourself. In that at least I think you are right.
As for the ‘necessary but not sufficient’ mantra, ‘sufficient’ is an imposed, flexible and ill-defined construct that allows you to deceive the poorly educated. Kant’s categories of possibility, actuality and necessity are precise, unambiguous and permit no exception.
• Possibilities describe what might exist or might happen.
• Actualities describe what does exist or does happen.
• Necessities describe what must exist or must happen.
It seems to me you would benefit greatly by reading some philosophy. The opening two paragraphs of Kant’s ‘Critique of Pure Reason’ would be a good start. It is very old but it is enlightning. It is also very precise if you take the time to understand it.
Regards
Mark 51