Is the Wechsler Flynn-inflated?

 

Between WAIS-R and WAIS III this is not known, as the scores are not comparable. Between WAIS III and WAIS IV intelligence probably did rise.

 

http://www.baylor.edu/soe/faculty/index.php?id=925418WAIS

EXAMINING THE FLYNN EFFECT IN THE WECHSLER ADULT INTELLIGENCE SCALE Nicholas F. Benson1 , A. Alexander Beaujean1 and Gordon E. Taub2 1 Baylor University, Nicholas_Benson@baylor.edu. 2 University of Central Florida.

The Flynn effect (FE: i.e., increase in mean IQ scores over time) is commonly viewed as reflecting population shifts in intelligence, despite the fact that most FE studies have not investigated the assumption of score comparability. Consequently, the extent to which these mean differences in IQ scores reflect population shifts in cognitive abilities versus changes in the instruments used to measure these abilities is unclear.

This study used participants from the Wechsler Adult Intelligence Scale’s revised (WAIS-R; n = 1,800), third (WAIS-III; n = 2,450), and fourth (WAIS-IV; n = 2,200) editions’ standardization samples. First, WAIS subtest scores were equated using data obtained from participants who were administered two editions of the WAIS, either the WAIS-R and WAIS-III (n = 192) or the WAIS-III and WAIS-IV (n = 284). Each WAIS-R and WAIS-IV subtest was equated to the corresponding subtest raw score on the WAIS-III. Score equating enabled the combination of scores from all three instruments into 1 of 13 age groups before converting raw scores into Z scores.

The second part of the study involved examining invariance of the WAIS structure across standardization samples via multi-group latent variable models. Some factor loadings and all of the subtest intercepts were found to be non-invariant when comparing the WAIS-R and WAIS-III samples on the equated scores. Thus, score changes reflect, at least in part, a recalibration of the instrument’s metric or scale. Conversely, when comparing the WAIS-III and WAISIV samples on the equated scores, strict invariance was tenable. Thus, score changes between the WAISIII and WAIS-IV most likely reflect changes in psychometric g rather than changes to the instrument.

These results suggest that there is some evidence for an increase in intelligence, but also call into question many published FE findings as presuming the instruments’ scores are invariant when this assumption is not warranted. Even though score comparability across instruments depends on a minimum level of invariance, FE studies do not typically examine this assumption. Thus, any difference reported in manifest scores from these instruments (e.g., FSIQ) could just as easily be due to changes in the instrument as due to changes in the examinees. Our use of score equating and assessment of invariance in the present study allows for a more direct test of whether the FE arises from genuine secular changes in intelligence or simply reflects changes to the test.

It appears that there is evidence for a rise in the level of psychometric g from between the time the WAIS-III and WAIS-IV were normed, but little is known about the period between the WAIS-R and WAIS-III as the scores are not comparable.

Correlated vectors

 

Jensen developed the “method of correlated vectors”  in which a positive correlation between group differences and g-loadings strongly supported the hypothesis that the group differences were largely due to general mental ability. I first met Jelte at the Amsterdam ISIR conference in 2007, when he had just finished his PhD and was discussing the need for confirmatory factor analyses to establish measurement invariance. This is his critique of the method of correlated vectors.

http://wicherts.socsci.uva.nl/CVJMW.pdf

MORE PSYCHOMETRIC PROBLEMS WITH THE METHOD OF CORRELATED VECTORS Jelte M Wicherts1 1 Tilburg University, j.m.wicherts@uvt.nl.

The method of correlated vectors (MCV) was developed by Jensen (1980, 1998) to study the hypothesis that the relation between cognitive measures and an extraneous variable (e.g., ethnicity) is fully explained by g. Despite criticism, MCV continuous to be widely used in the study of group differences in intelligence test performance, most often to study Spearman’s hypothesis stating that ethnic group differences in cognitive performance are most pronounced on the most highly g loaded tests or items. In addition, recent studies have submitted MCV results to psychometric meta-analytic techniques in which MCV results are corrected for particular psychometric artifacts.

In this talk, I critically evaluate the psychometric assumptions underlying MCV. I focus on meta-analytic corrections applied to vectors of g loadings and on the application of MCV to item-level data. I use both formal arguments and empirical data to illustrate the drawbacks of MCV. I particularly address studies that have applied MCV to study group differences on items of Raven’s Standard Progressive Matrices (SPM). SPM data have yielded strong MCV correlations (i.e., Jensen effects) that have been interpreted as showing that the SPM measures g similarly across ethnic groups and is not subject to item bias. Using formal arguments, I show that failures of metric measurement invariance (group differences in g loadings) actually lead to higher MCV estimates in meta-analyses despite the fact that such violations clearly contradict Spearman’s hypothesis that group differences are due to g. Moreover, I show that MCV applied to item-level data does not provide accurate information about the comparative psychometric properties of a cognitive test and the role of g.

The empirical results show that MCV applied to SPM items in one group yields substantial Jensen Effects even when the items in the second group (N=252) are not from the SPM but rather from a test composed of items from the State-Trait Anxiety Inventory and the State-Trait Anger Scale. Combined, these results highlight problems with meta-analytic corrections to MCV results and show that MCV applied to item level data does not accurately refect the degree to which item bias or g plays a role in the ethnic group differences. Although MCV may be useful in some circumstances, it is better to use model-based approaches like multi-group confirmatory factor analysis or Different Item Functioning (DIF) analyses whenever these are feasible.

Schizophrenia and cognitive decline

Yes, that Stuart Ritchie, author of “Intelligence: All that matters”,

http://www.amazon.co.uk/Intelligence-That-Matters-Stuart-Ritchie/dp/1444791877

is attending ISIR15 in person, “en vivo y directo” as us Latins would say. Well, delegates aren’t going to fly all that way just for nothing, are they? Anyway, he is part of the Edinburgh gang, and here is a finding to give us pause: genetic risk for schizophrenia is associated with cognitive decline.

Only last Wednesday I met up with a group of work friends, and the psychiatrist among us, whom I hadn’t seen for quite a while, and who knows nothing of my blogging, replied to my query about her research by spontaneously saying: “I think that the main problem with my schizophrenic patients is low intelligence. Those with higher intelligence seem to deal with it better”.  I gave her advance warning of this paper:

http://www.ppls.ed.ac.uk/people/stuart-ritchie

POLYGENIC RISK FOR SCHIZOPHRENIA IS ASSOCIATED WITH STEEPER GENERAL COGNITIVE DECLINE

Stuart J Ritchie, Andrew M McIntosh, Alexandra Bannach-Brown, Toni-Kim Clarke, and Ian J Deary The University of Edinburgh, stuart.ritchie@ed.ac.uk.

Few predictors of people‘s differences in ageing-related cognitive decline have been discovered to date. One potential candidate is genetic risk for schizophrenia, which has previously been linked to cognitive change between childhood and old age in individuals with no schizophrenia diagnosis. To date, no studies have investigated the link between polygenic risk for schizophrenia and cognitive decline within older age.

We used data from longitudinal cognitive testing of subjects from the Lothian Birth Cohorts of 1921 (LBC1921; tests performed at mean ages 79, 83, 87, 90, and 92 years; initial n = 550) and 1936 (LBC1936; tests performed at ages 70, 73, and 76, initial n = 1,091). We examined the association between individual differences in polygenic risk for schizophrenia and differences in the trajectory of cognitive ageing. Polygenic risk scores were derived from the Psychiatric Genomics Consortium‘s most recent GenomeWide Association Study (GWAS) for schizophrenia.

Meta-analysis across both cohorts showed that individuals with higher polygenic risk scores for schizophrenia declined more steeply in their later-life general cognitive ability. These results were significant in the larger LBC1936 study alone. Although the results were in the same direction in LBC1921, they did not reach statistical significance. This work describes a new predictor of greater age-associated cognitive decline. It shows how GWAS results can be useful in predicting ostensibly unrelated phenotypes, and raises important theoretical questions about the links between genetic risk for schizophrenia and lowered intelligence.

Comment: Schizophrenia, in clinical practice, is usually described according to the nature of the disordered thoughts. Yes, disordered thinking is recognised as a great part of the problem, but there is a meta-cognitive dimension: odd thoughts have to be given a reality check, and it seems to be at that stage that many sufferers fail.

Chris Frith studied the condition for many years. Written with Eve Johnson, this is a very good short introduction

http://www.amazon.com/Schizophrenia-A-Very-Short-Introduction/dp/0192802216

Willpower makes you brighter?

 

There is no such thing as a daft hypothesis, merely one that requires testing. With great generosity Timothy Bates has tested three hypotheses that I would have left in a hedge. However, having left them there I would not have been able to refute them, and would be able to say, out of wisdom or abject lethargy, only that they did not seem likely to be true. That is always a weak argument, because many unlikely things are true, like visual receptors facing backward in the retina. Still can’t get over that one, but recent work shows that by doing so they pick up other interference cues which are helpful. Anyway, Tim has given three mental-state-IQ hypotheses a chance to shine, which is what we should all do, and finds they do not refute the concept of trait-IQ. Null results should be as interesting as positive results and are often truer to reality.

 

http://www.psy.ed.ac.uk/people/view.php?name=timothy-bates

TESTING ALTERNATIVES TO TRAIT-IQ: DWECK’S MINDSET, WOOLLEY’S EMOTIONAL COLLECTIVE, AND BAUMEISTER’S DEPLETED WILL MODELS Timothy C. Bates University of Edinburgh, tim.bates@ed.ac.uk.

Numerous theories seek to account for differences in reasoning without recourse to trait-IQ. Among these are Dweck’s (Mueller & Dweck, 1998) Incremental Mindset, Baumeister’s Resource Depletion Theory (Vohs, Baumeister, & Schmeichel, 2012) and, for working in groups, Woolley’s Collective IQ (C: Woolley, et al.., 2010).

In this presentation, we test the extent to which these models predict performance independent of g (if at all). The Baumeister and Dweck models of will power are tested in a repeated measures design involving 80 students. The Woolley Collective IQ model is tested in three experiments with 28 to 80 groups of individuals. In addition manipulations of empathy-equality are contrasted with manipulations of authority-obedience to test non-cognitive origins of group performance. The Dweck incremental versus fixed mindset model of IQ test performance was tested in three experiments of between 80 and 400 subjects, testing the predicted link of beliefs about performance to actualized performance both observationally and via belief priming.

We found no significant support for willpower depletion as a cause of cognitive decrements during testing. Moreover we find no support for incremental beliefs about will-power on measured cognitive test scores. Group-IQ performance showed a strong g-factor, but this was almost completely explained by individual differences in IQ. We found no support for empathizing, or for the role of women as factors raising group IQ scores. Study three showed, instead, support for authority/group morality manipulations in raising collective performance. We found no support for incremental versus fixed mindset on grades. We further found no significant effect of mindset priming on IQ scores post a performance setback challenge in either of two replications. Across multiple studies, we were unable to support empathizing, will-power conservation, or mindset (typical or primed) as factors affecting IQ or cognitive control.

Diurnal gloom

I have never bothered to record the daily variation in my moods, for fear that the effort would damage the frailty of my good humour and cast me into even deeper despair. At a gut level, or perhaps that should be a gut-wrenching level, we all seem to have good days and bad days. Some days proceed from defeat to defeat like ink spreading across a damask table cloth. Others (less frequent) soar above earthly concerns to deliver joyous triumphs and accolades. Raise high the roof-beams, carpenters!

Does any of this emotional tittle tattle and swooning effluvia have any effect on serious things like intellectual performance? It seems so. You may remember Sophie von Stumm, who showed that the Dunning-Kruger effect affects people’s self-assessments of their own intellectual abilities:

http://drjamesthompson.blogspot.co.uk/2014/05/so-you-think-youre-intelligent.html

 

Here is what she has found when she subjects volunteers to up to 5 days of intellectual testing.

 

DAY-TO-DAY VARIABILITY IN IQ AND MOOD Sophie von Stumm

Goldsmiths University of London,

s.vonstumm@gold.ac.uk.

Intelligence, also known as IQ, has been shown to be subject to systematic developmental changes in childhood and in late life and to be relatively stable from adolescence through most of adulthood. However, little is known about the stability – or in fact variability – of cognitive functioning and intelligence test performance across days. Here, the day-to-day variability across 6 IQ tests was studied and its associations with a) day-to-day variability in mood and b) personality traits.

Overall 98 participants (age range 18 to 75, mean 23 years) were assessed 5 times on 5 consecutive days in the lab, where they completed each day different versions/ items of 6 cognitive ability tests (short-term memory, logical reasoning, image rotation, pattern comparison, working memory, processing speed) and the Positive and Negative Affect Scale (PANAS). On day 1, they also completed the NEO-FFI to assess the Big Five.

All participants completed at least 2 study days, and 77 participants contributed on all 5 study days. Day-to-day variance in mood and cognitive ability tests were adjusted for trial-to-trial variance (i.e. within-test variance/ internal consistency). Day-to-day variability in cognitive ability tests ranged in IQ points (mean 100, SD = 15) from 0 to 20 with an average of approximately 6 IQ points across tests (SD = 3), demonstrating a) considerable day-to-day variability in cognitive function and b) individual differences in day-to-day variability. Associations between individual differences in the extent of day to-day variability in cognitive function and mood were weaker than their respective correlations with personality traits.

This study demonstrates the extent of day-to-day variability in IQ and mood, thereby highlighting the importance of the individual differences dimension of variability per se. In addition, the findings help understanding the relative importance of more (i.e. mood) and less (i.e. personality) fluctuating factors for variability in cognitive ability test performance.

http://www.gold.ac.uk/psychology/staff/stumm/

What do college admission tests predict?

 

Put the brightest people onto the hardest problems and everyone benefits.

The first step in that process is to use very well validated tests in order to find the brightest people, and then give the best students the best education so that they then go on to solve the hardest problems. The rest of us, speaking for myself, then try to understand and implement their findings. When calculating my location, my satnav implements a time correction based on Einstein’s theory of relativity (satellites move at very fast speeds, such that their atomic clocks seem slightly wrong when their time signals are received by the cheap quartz clocks in the satnav).

Paul Sackett identifies an issue which can prevent the brightest people solving the hardest problems. He says: One recurring theme in my work is the tension between designing selection systems to maximize job performance vs. to maximize ethnic, racial, and gender diversity. The current controversies over the future of affirmative action attest to the prominence of this concern.

You know what I think. I think it is wrong to confuse competence with demographic representativeness. Competence is needed so as to obtain the best outcomes of skilled behaviour. Representativeness is needed to establish that a sample represents a population. Confusing the two will lead to bad decisions and sub-standard performance.

 

 

I am sure Paul Sackett knows all this, because he has been instrumental in long-term, high quality work on admissions tests, and is particularly hot on the methodological issues which bedevil the field (and most of psychology). His work shows the continuing relevance and power of testing. Sackett and Kuncel have a sample a million students which, as they modestly say, “delivers robust answers”.

Sackett and Kuncel found that SAT and high school grades contributed to predicting academic performance in college. Taking parents’ education and family income into account had little effect on the relationship between SAT scores and college performance. The SES of students actually enrolled in college was very similar to the SES of students who were applying to college. When they examined the data more closely and looked at the entire applicant pool, they found that fewer low-SES students were entering the college admissions process.

Based on these findings, it seems that low-SES students are not underrepresented in colleges because low SAT scores prevent them from gaining admission, but rather because fewer low-SES students apply to college in the first place.

“We view this as broadly relevant,” says Sackett. “Entrance tests such as the SAT receive a great deal of public scrutiny and it is important for all involved—students, parents, college officials—that accurate information about how the test functions be available.”

This duo should receive a medal for doing educational research in a minefield. In fact, that is what ISIR is offering them, in the form of the President’s Invited Symposium. We will listen carefully to these scholars.

 

What do college admissions tests predict, and for whom? Insights from a large scale research program. Paul Sackett & Nathan Kuncel (Introduced by Michael A. McDaniel)

Student success in higher education is a critical and fascinating topic. Student success is multifaceted; stakeholders emphasize different outcomes such as degree completion, work skills and critical thinking. Multiple individual characteristics are associated with different facets of success. College admissions use numerous measures ranging from interviews to standardized tests. Here we will discuss decades of our research on these issues including many studies that have come out of our joint research laboratory. We make use of a still-growing primary database now following over 1,000,000 students through college, combined with meta-analyses on hundreds of thousands of students. Higher education policymakers need to know about test validity, the role of SES on student admissions and performance, linearity of test-performance relationship, and predictive bias by race and gender. Our very large N delivers robust answers to these and other key questions.

http://www.psych.umn.edu/people/profile.php?UID=psackett

Poster session

http://www.isironline.org/wp-content/uploads/2015/09/ISIR-2016-program1.pdf

In the above link, just find “Posters” and have a look through. You will be spotting the rising stars, as well as catching up on briefer point by established researchers. The interview with Robert Plomin will be on the ISIR website in due course.

Coffee time now, and see you tomorrow.

EXECUTIVE FUNCTIONS AND CONSCIENTIOUSNESS

By special dispensation of the authors, there is a Powerpoint version of Laura’s talk which, though it was changed somewhat for the conference, gives you far more of the general argument and the results. Saves you a ticket to Albuquerque. If you are grateful, hit the Donate button and send me a coffee.

https://drive.google.com/file/d/0B3c4TxciNeJZaGExNDVuaXJYMEk/view?usp=sharing

GENETIC AND ENVIRONMENTAL STRUCTURE OF SELF-REGULATION:

Daniel A Briley, 1 Laura E Engelhardt, Frank D Mann, K Paige Harden, Elliot M Tucker-Drob

1 University of Texas at Austin, daniel.briley@utexas.edu.

Self-regulation refers to the ability and general tendency to maintain alignment between desired and actual psychological states. This core psychological task is thought to guide behavior and depend on the joint input of several cognitive systems. For example, executive functions refer to high-level cognitive control mechanisms that direct lower-level processes necessary for attention, learning, and decision making. Similarly, conscientiousness refers to general tendencies toward disciplined, self-controlled, responsible, and achievement-oriented behavior. Although these constructs appear similar on the surface, the psychometric relation between self-regulatory ability (i.e., executive functioning) and general selfregulatory behavior (i.e., conscientiousness) is not well known.

The current project uses data from the Texas Twin Project to test the structure of self-regulation in a genetically informative sample. Participants (N = 509 individuals, 82 MZ pairs and 195 DZ pairs, mean age = 11.0 years, 46.2% male) completed an extensive battery of twelve executive functioning tasks, and a hierarchical factor model was constructed. Participants self-reported on their levels of conscientiousness, and one of the participants’ parents provided an informant-report of conscientiousness.

We tested the association between executive functioning and self- and informant-report of conscientiousness, as well as two facets. We used behavior genetic methodology to decompose this association into genetic and environmental pathways. A hierarchical factor model of executive functioning fit the data well. We specified a general executive functioning factor that was indicated by four sub-factors, working memory, updating, switching, and inhibition. Phenotypically, conscientiousness was weakly associated with executive functioning (r’s approximately.15). Convergent validity was somewhat larger for informant-report variables and for facets related to self-discipline (as compared to order). These associations were primarily genetically mediated. Higher levels of executive functioning and conscientiousness are both associated with a host of beneficial life outcomes, such as academic achievement, health, and occupational success.

One possible explanation for this common result is that individuals who are able to better self-regulate can use cognitive resources to accomplish goal-directed tasks. The current results indicate that executive functioning and conscientiousness primarily index different self-regulatory mechanisms, despite the high similarity of the motivating theoretical background for each construct. Identifying the causal processes that link self-regulation and intelligence with beneficial life outcomes needs to consider both levels of ability and typical patterns of behavior.

Do women find bright men sexy?

Although I don’t do trigger warnings, you may want to skip this item. One of the few consolations to shy men with intellectual aspirations was the thesis that women would be tempted to mate with them on the basis of intellect alone. Of course, men were still required reveal their intellects in some way, but telling jokes was judged sufficient. Personally, I cannot recall this ever working. “A friend, seeing Franz Kafka sitting alone at a cafe table, walked across the hall to him and said “Franz, I am sitting with some friends. Would you like to join us?” “No” replied Kafka.”

The young women on whom I attempted this approach would smile appreciatively, and then dance with someone else, usually of taciturn demeanour and singularly lacking in social skills. Of course I would not dream of suggesting that this was a shallow and heartless response. Perhaps the music was too loud, and the joke could not be heard.

So, what happens when you test the hypothesis?

MALE GENERAL INTELLIGENCE (G) DOES NOT INCREASE FEMALE SEXUAL ATTRACTION

Lars Penke, Ruben C. Arslan, and Juliane Stopfer1 1 Georg August University Göttingen, lpenke@uni-goettingen.de.

Human general intelligence (g) has been hypothesized to be an indicator of genomic mutation load and under sexual selection for indirect genetic benefits (‘good genes’ for the offspring), implying that high g should be sexually attractive. People clearly report preferences and assortatively mate for intelligence, but these effects can be due to direct phenotypic benefts of g and social homogamy. Only one study (Prokosch et al., 2009) with methodological limitations has directly tested if higher male intelligence increases female initial sexual attraction.

We tested 88 young men (age 19 to 31 years) on six psychometric intelligence subtests and two measures of information processing speed, from which a g factor was extracted, and on the Big 5 personality dimensions. Standardized photos, voice recordings and videotapes of three behavioral tasks (reading headlines, charade, tell-a-joke) were also taken. Sixteen women and 14 men judged the intelligence and personality of the target men based on the videos. A second group of 16 women rated the attractiveness of the men as long-term and short-term partners. A third group of 25 women received information about each men in five steps, with intelligence cues being increasingly present over and above physical attractiveness information, and rated long- and short-term attraction after each step. Both men and women could accurately judge intelligence and extraversion, but not the other Big 5, from the videos. Measured male g had no effect on female short-term attraction, but a small positive effect on long-term attraction, though only after extraversion and independently rated physical attractiveness were controlled. When information on male intelligence was presented incrementally, measured g did not predict changes in female long-term or short-term attraction ratings formed based on physical attractiveness. Overall we found no support for intelligence being sexually attractive to women on first encounters, and limited support that it increases initial impression of the potential as a long-term romantic partner. This is only the second study on the attractiveness of measured intelligence at zero acquaintance, and the first one that assessed a true g factor, had a sufficiently large sample of target men, and tested whether increasing availability of intelligence information alters women’s reported attraction. Taken together with very limited support for an association between g and mutation load in the currently available genomic data, these results cast doubt on the hypothesis that g is an indicator of genetic fitness under ‘good genes’ sexual selection.

Future US intelligence

 

FUTURE US INTELLIGENCE: IQ PREDICTION UNTIL 2060 BASED ON NAEP Heiner Rindermann and Stefan Pichelmann1

1 Technische Universität Chemnitz, heiner.rindermann@psychologie.tu-chemnitz.de.

US National Assessment of Educational Progress (NAEP) measures cognitive competences in reading and mathematics of US students (last 2012 survey N=50,000). The long-term development based on results from 1971 to 2012 allows a prediction of future cognitive trends. For predicting US averages, also demographic trends have to be considered. We want to answer the following questions: (1) Will the Flynn effect be continued? (2) Will there be a decrease or increase in gaps between ethnic and racial groups? (3) What effect has the rising share of minorities? (4) What effect has gap reduction on society’s average ability level? (5) What effect has national ability development on GDP?

The White average 1978/80 set at M=100 and SD=15 was used as a benchmark. Based on two past NAEP development periods for 17-year-old students, 1978/80 to 2012 (more optimistic) and 1992 to 2012 (more pessimistic), and demographic projections from the US Census Bureau, cognitive trends until 2060 for Whites, Blacks, Hispanics, Asians and the entire age cohort were estimated.

Estimated population averages for 2060 are 103 (optimistic) or 102 (pessimistic). White-Black gaps from currently 11 IQ decrease to 6 IQ (op.) or 7 IQ (pe.), White-Hispanic from 9 IQ to 4 IQ (op.) or 3 IQ (pe.), Asian-White gaps increase from currently 3 IQ to 9 IQ (op.) or 12 IQ (pe.) resulting in a distinctive top Asian group at around 114 IQ.

The catch-up of minorities (their faster ability growth) contributes around 2 IQ to the general rise of 3 IQ; however, their larger demographic increase reduces the general rise at about the similar amount (-1.4 IQ). Because minorities with faster ability growth also rise in their population proportion the interactive term is positive (around 1 IQ). Consequences for economic and societal development are discussed. Based on past NAEP trends and population estimations US future IQ is predicted.

For the US in 2060 an average IQ of 102-103 points is predicted. General FLynn effects contribute positively to IQ development. Minority catch-ups contribute positively to IQ development. Non-Asian minority population increases contribute negatively to IQ development.