IQ as in "intelligence quotient"? Or is this some distance learning acronym now? Just a little confused... maybe because my iq isn't that high!
That all depends on what you mean by real, valid, and reasonable. If a test is designed properly and given to an unselected population, the scores will be distributed in a certain way. What does that mean? What can be determined from it? What does it validate? Jensen might be able to tell you. ------------------ Quinn
Indeed it is, Covey! IQ is an acronym for "Internet Quota," which relates to Degree Mill advertising. When a sales representative speaks of her/his monthly IQ, reference is being made to the number of people lured into enrolling in their Internet degree mill program. Russell
IQ is a valid measure of performance on an IQ test. But it doesn't correlate highly, I believe, with most real world measures of performance. The four major long-term longitudinal studies I know of, of people who tested high as children (Hollingworth's "Children Above 180 IQ"; Terman; Breen; and the Hunter College-Bellevue 1941 study) found that in general, these people did well, but not spectacularly well as might have been predicted. The 60 people in the Hunter-Bellevue 1941 study, who were together for 8 years (3 through 11), two foreign languages early on, algebra in 3rd grade, calculus in 5th, etc., came together for a reunion and scrutiny 39 years later, when all were 50. Significantly (but not wildly so) higher numbers of doctors, lawyers, and authors including a famous author of teenage romance novels. But there was nothing that might not have been found in any gathering of 50-year old college graduates, I suspect.
I used to work for an agency in which having a "diagnosable handicapping condition" would qualify a person for services. If I had a client who might be on the cusp of eligibility I would use a psychologist with a foreign accent. Most clinicians believed the unfamiliar accent would lower the IQ at least 6 points for most children. I would sometimes have a person whose new IQ scores on the adult version (WAIS) were much different than the scores on the childrens version (WISC). The theory of the test is that the IQ score shouldn't change significantly. But they seemed too. Many schools currently use the Woodcock Johnson system of tests to measure both achievement and IQ. Many psychologists feel that the WISC is more accurate than the Woodcock Johnson for IQ. In my work the Woodcock Johnson was more likely to put my clients in the average range, while the WISC showed they were below average, and hence eligible. There are many forms of IQ tests. In corrections we used to use the "revised Beta". It was fast, but I don't know how accurate. Finally, one client I was working with had a 21 point IQ increase between the 6th grade and 10th grade. When I talked with the psychologist the comment was, "depressed kids always do worse..." Sometimes the test is administered when it should not be. All the best!
OK, I don't think my daughter/coauthor Mariah will mind my telling this one. When she was 5, in kindergarten, and bored, the school gave her an IQ test to see if maybe she should be in first grade. The psychologist called us in for a conference. Seems she had gotten 19 of the 20 questions right on the IQ test she took. Should be promoted twice, he said. We asked, and he told us, which question she had "missed." Well that evening,we asked Mariah. The question was: "The Jones family was on a picnic. An animal came along. And then they had to burn all their clothes. What do you think happened?" Mariah told us she had no idea. What do you think it could have been, we asked. "I just don't know," she replied. "It couldn't have been a skunk, because then all they would have had to do was boil their clothes in tomato juice, so I can't imagine what happened." Five year old knows more than IQ test maker-upper, and suffers lower score as a result.
Dave, I’m not sure if you’re asking about the construct of intelligence per se or if you are asking about tools that measure intelligence. One way to view intelligence is the sum of psychological abilities: spatial, reasoning and memory among many others which we now recognize as having emotional and interpersonal- as well as neurocognitive components and also sub-components (eg, verbal reasoning, non-verbal reasoning). Any test is simply a sample of behavior the purpose of which is to discriminate performance between people and estimate an individual’s performance. Binet’s goal was not to measure intelligence but to stratify Parisian schoolchildren. A test has numerous psychometric properties with the most well known being reliability (does the test measure the same way each time) and validity (does the test measure what it purports to measure), which determine how well it does its job. Is any test perfect? Of course not! A test can only provide an ESTIMATE but some, those with good psychometric properties, do better than others. Allow me to make a point by way of a measurement every student encounters: the final exam. Imagine that the western civilization course you took last semester is made up of 1,000,000 data points. A four-hour exam can only estimate mastery of the material in that it is only possible to review, say, 1,000 of those 1,000,000 data points in four hours. Imagine how many data points are included under the category of intelligence! Technical aspects may account for peculiarities such as Irat describes. It is NOT a good idea to lock onto the overall score. The overall score is, ta da, just an ESTIMATE of the examinee’s true abilities. It is a better idea to account for a standard error of estimate by reporting a confidence interval such as there is a 95% probability that the examinee’s score is in the interval x to y. Flynn’s hypothesis is interesting. Flynn found that on a societal level mean IQ scores increase over time. I want to say that on standard measures where the mean is 100 and the standard deviation is 15 he found that the mean increases by six points every decade but I’m not positive about the exact figure. The interpretation is that IQ creep is due to improved health, nutrition, information transfer and so on. So, you would expect an individual examinee to do better on the Wechsler-Bellview than the WAIS than the WAIS-R and so forth. In the case of Irat’s example it could be that (factoring in Flynn’s effect and the error of estimate) the confidence intervals might overlap or much of that 21 point difference disappear if the person was first tested with an earlier version of the WISC then at age 16 tested with a current version of the WAIS. Intelligence testing is an important tool. It is a very complex endeavor and requires a significant degree of training. The layperson doesn’t realize that the pattern among the subtests tells you much more about the individual than the overall score. This is really where the rubber meets the road. Mistakes do happen that can have very adverse consequences. Imagine how you might feel if you discovered that your quite substantially average child was placed in a special ed program of if after suffering a stroke you were found to be legally incompetent to manage your affairs due to a computational error or an inaccurate interpretation made by an inadequately trained examiner? Taking an analogy from medicine, anyone can read the results of a chemistry panel and see there is an elevated BUN but it takes an experienced clinician to differentiate amongst the entire panel that the elevated BUN is due to dehydration v. renal insufficiency.
