AI Brief: The trilogy-of-the-mind individual difference construct (cognitive, conative, affective) “band is getting back together” as CAMML

I am currently working to expand my skill set by incorporating AI tools. Although adapting to new technologies can be challenging, leveraging these resources offers significant benefits for professional growth.

This AI Brief was produced by requesting Google NotebookLM—recommended by Dr. Adam Lockwood—to generate a narrative summary of my 2022 PDF article describing the Cognitive-Affective-Motivation-Model of Learning (CAMML). While I found the first results promising, I made more edits to enhance its accuracy and informativeness. My next goal is to use AI to summarize multiple articles, find similarities and differences, and potentially create comparative tables (again following guidance graciously provided by Dr. Lockwood).

These incremental steps mark my transition toward utilizing AI to support one of my primary professional interests: producing informative blog and social media posts aimed at professionals such as school psychologists and special education teachers working with students who often are marginalized in educational contexts. The goal is to help bridge the gap between theory, technology, research, and practical application.

 

Feedback is encouraged and may be directed to iqmcgrew@gmail.com or via the social media platform (LinkedIn, Twitter/X, BlueSky) comment feature where this blog post was discovered. I’m hoping to add AI Briefs as a regular feature of IQs Corner Blog and associated social media platforms.

 

 

AI Brief: The trilogy-of-the-mind individual difference construct (cognitive, conative, affective) “band is getting back together” as CAMML

 

Dr. Kevin McGrew with assist from Google NotebookLM

 

The Cognitive-Affective-Motivation Model of Learning (CAMML; McGrew, 2002)[1] is a proposed theoretical framework designed to integrate contemporary motivational, affective, and cognitive constructs into a unified model for the practice of school psychology. The central thesis of the framework is that school psychologists must move beyond a narrow focus on intelligence (general intelligence or psychometric g in particular) to embrace an updated "trilogy-of-the-mind" model, which views intellectual functioning as the inseparable interaction of cognition, conation (motivation/volition), and affect.

Theoretical Foundations and the Rebirth of Conation

 

The CAMML framework is heavily rooted in the seminal work of Richard Snow, specifically his research on aptitude trait complexes. McGrew argues that the field of school psychology has historically neglected Snow’s broader definition of aptitude—which includes personality and motivational differences alongside cognitive abilities—and instead, has favored a restricted view of aptitude as synonymous with IQ or psychometric g.

 

CAMML seeks to resurrect conation (the proactive part of motivation connecting cognition and affect to behavior) as a core pillar of intellectual functioning. By "standing on the shoulders of giants" like Snow, Spearman, and Wechsler, the model asserts that cognitive processes cannot be understood in isolation from the "nonintellectual" (conative) factors that drive and direct them. For example, David Wechsler defined intelligence as "the aggregate or global capacity of the individual to act purposefully, to think rationally, and to deal effectively with his environment." While this is the core of Wechsler’s definition, he also strongly emphasized this capacity is influenced by non-intellective (conative) variables such as drive, persistence, interest, emotional states, and personality traits.

 

Structural Components of the CAMML Framework

 

The model organizes individual differences characteristics into three functional categories, which could be called the 3-D model.

 

●      Affective “Dispositions”: These are distal-to-learning traits, primarily represented by the Big 5 personality traits (specifically Openness and Conscientiousness) and their associated social-emotional facets (e.g, curiosity, creativity, persistence, focus, determination). These personality traits act as dispositions that indirectly influence learning through more proximal mechanisms.

 

●      Motivational "Drivers": Motivation is conceptualized as the initiation of behavior, formed by achievement orientations (e.g., goals, interests) and self-beliefs (e.g., self-efficacy, self-concept). These constructs are typically domain-specific (e.g., math) and work in synergistic "complexes" to energize a student's readiness to act.[2]

 

●      Volitional "Directors": Volition, or self-regulated learning (SRL), represents the post-decisional phase of action. These are the mechanisms that direct, control, and regulate behavior toward goals once a commitment to learn has been made.

 

A more detailed list of the 3-D CAMML domain constructs and definitions is available here. See figure below for a visual representation of the major affective and conative MACM constructs (click on image to enlarge for easy viewing and reading).

 

The "Crossing the Rubicon" Investment Model

 

The functional heart of CAMML is the "Crossing the Rubicon" model, which illustrates the pathway from initial desire to engaged motivated learning. In this model:

 

1. Pre-decisional Phase: Achievement orientations and self-beliefs drive or prepare the learner to start a wish—>want—> intention sequence, that eventually eventuates in motivated action.

2. Commitment: When the learner "crosses the Rubicon," they are making a firm commitment to motivated action through cognitive engagement.

3. Action Phase: Volitional (SRL) strategies steer the cyclical process via action results feedback while the learner invests cognitive abilities (such as those defined by CHC theory) to acquire knowledge.

4. Outcomes: This personal investment of fluid cognitive processes (Cattell's general gf that subsumes broad Gf, Gv, Ga, Gwm, Gl, Gr, and Gs abilities) during learning results in the development of crystallized knowledge systems (Cattell’s general gc that subsumes broad Gc, Grw, Gq, and Gkn abilities).

 

See figure below for visual representation of “the CAMML crossing the Rubicon model of motivated learning” (click on image to enlarge for easy viewing and reading).

 

Implications for School Psychology Practice

 

CAMML advocates for a paradigm shift in assessment and intervention. It suggests that school psychologists should transition from routine, comprehensive cognitive testing toward more time-efficient selective, referral-focused cognitive assessments combined with the assessment of key conative (non-cognitive) characteristics that contribute to learning aptitude complexes. This approach prioritizes identifying manipulable instructional levers, such as a student's motivational orientation (e.g., intrinsic motivation, interests, goal orientation), self-beliefs (e.g., locus of control, self-efficacy, growth or competence mindset), rather than relying exclusively on cognitive ability scores (especially full-scale IQ or g) that have proven hard to modify.

 

The framework provides a "whole-child" perspective to better address the nuances of individual differences, particularly as students move from the traditional “industrial” model of education (i.e., regularly scheduled, structured, in-class teacher-directed learning) to more of an “information-age” paradigm of education—a paradigm that requires a fuller expression of independent motivated self-regulated learning (SRL).

PS - other CAMML related posts on this blog can be found by clicking here.

[1] All relevant references can be found in McGrew (2022).

[2] The motivation constructs included in the CAMML framework are drawn from earlier efforts to develop the McGrew Model of Achievement Competence Motivation (MACM). A detailed explanation of the evolution and development of the MACM model is available elsewhere (McGrew et al., 2004). A series of recent MACM PowerPoint® modules is available here.

Meta-analysis supports cognitive ability differentiation hypotheses (SLODOR)

A B S T R A C T

The cognitive ability differentiation hypothesis, which is also termed Spearman's Law of Diminishing Returns, proposes that cognitive ability tests are less correlated and less g loaded in higher ability populations. In ad-dition, the age differentiation hypothesis proposes that the structure of cognitive ability varies across respondent age. To clarify the literature regarding these expectations, 106 articles containing 408 studies, which were published over a 100-year time span, were analyzed to evaluate the empirical basis for ability as well as age differentiation hypotheses. Meta-analyses provide support for both hypotheses and related expectations. Results demonstrate that the mean correlation and g loadings of cognitive ability tests decrease with increasing ability, yet increase with respondent age. Moreover, these effects have been nearly constant throughout the century of analyzed data. These results are important because we cannot assume an invariant cognitive structure for dif-ferent ability and age levels. Implications for practice as well as drawbacks are further discussed.

Article link.

Psychometric taxonomic models of intelligence: A visual tour from g to CHC

Click on images to enlarge:

 

Posted with Blogsy

Did Spearman abandon g? Guest post by Ruben Lopez

The following is a guest post by Ruben Lopez, school psychologist with the Moreno Valley Unified School District, CA and member of the IQs Corner Virtual Community of Scholars.

Ruben reviewed the following article and has provided his comments below.

• Deary, I. J., Lawn, M., & Bartholomew, D. J. (2008). A conversation between Charles Spearman, Godfrey Thomson, and Edward L. Thorndike: The international examinations inquiry meetings 1931-1938. History of Psychology, 11(2), 122-142 (click here to view)

At the end of his life, did Spearman abandon g?

After his retirement and late in life at age 68, what did Charles E. Spearman, discoverer of the general factor of human intelligence (symbolized as g), really think about general factor g? A recent article in the journal History of Psychology (complete citation above) addresses this question and concludes that Spearman’s long-standing adversaries Godfrey Thomson and Edward Thorndike proposed an incipient version of the CHC broad factors known as fluid (Gf) and crystallized (Gc). If Thomson and Thorndike indeed proposed an incipient non-hierarchical Gf-Gc versus Spearman’s hierarchical g, they began a debate which would continue more than 70 years later between CHC luminaries John Carroll (in support of hierarchical g) against John Horn (in support of non-hierarchical Gf-Gc)—a debate they waged until fairly recently to the end of both of their lives.

In their article, Deary, Lawn, and Bartholomew analyze word for word transcriptions of the contributions Spearman, Thomson, and Thorndike made to the theory and practice of testing intelligence at three meetings in 1931, 1935, and 1938 to which Spearman had been invited by Thorndike. Yet Deary, et al. report that Thorndike and Thomson found their interaction with Spearman to be negative and Spearman felt the same. Moreover, it seems to me that Deary, et al. clearly favored Thomson and Thorndike over Spearman’s contributions and personal style. Deary, et al., for example, describe Thomson as practical and winning, while Spearman in contrast is described as mechanically theoretical and a “contrarian”, waiting only to challenge and correct the theoretical views of the other participants.

Deary, et al. cite statements by Spearman that appear to indicate that he was uncertain or even disbelieved that g exists beyond being nothing more than a statistical phenomenon, as many anti-g proponents wish to communicate when they say that Spearman believed that g was just a “positive manifold.” Deary, et al., for example, suggest that supporters of g should know that at these meetings Spearman said about g, “There is no such thing, but only a general factor in intelligence.” (p. 126)

Yet at the same meeting Spearman said, “This than is what the G term means, a score-factor and nothing more. But this meaning is sufficient to render the term well defined so that the underlying thing is susceptible to scientific investigation; we can proceed to find out facts about this score-factor, or G. We can ascertain the kind of mental operations in which it plays a dominant part as compared with the other or specific factor.” (p. 126) This says to me that Spearman as an objective researcher admitted that g was at least a psychometric fact but at the time was a phenomenon that was not understood in psychological or biological terms.

Therefore, this article convinced me that even as an old man, likely battle worn by younger adversaries like Thorndike and Thomson, Spearman did not appear to have abandoned g. In fact, one year after the last meeting with Spearmen at a symposium of the British Psychological Society in 1939, Godfrey Thomson is quoted in the article as saying, “I myself lean at the moment more toward Spearman’s g and his later group factors than I do to Thurstone’s….” (p. 129) Also, 69 years ago, Thomson conceded at the symposium, “Surely the real defense of g is simply that it has proved useful.” (p. 129)

Technorati Tags: psychology, educational psychology, school psychology, neuropsychology, intelligence, IQ, g, general intelligence, Charles Spearman, CHC, Cattell-Horn-Carroll, History of Psychology, IQs Corner

g (general IQ): Historical converstations - stay tunned

Yesterday on the NASP listserv NASP Historian Tom Fagan drew attention to the following article. One of IQ's Corners occasional contributors (Virtual Community of Scholars) indicated an interest in reading this article and providing a guest blog post. Below is the reference and abstract. Stay tunned for the forthcoming blog post.

Deary, I.J., Lawn, M., & Bartholomew, D. J. (2008). A conversation between Charles Spearman, Godfrey Thompson, and Edward L. Thornkdike: The International Examinations Inquiry Meetings 1931-1938. History of Psychology, 11(2), 122-142.

Abstract

• Even within “an appreciation of the fundamentally social nature of scientific activity” (K. Danziger, 1990, p. 3), it is unusual to read what key scientists actually said to each other, directly or in audience. Here the authors describe, structure, illustrate, and interpret the verbatim statements made by, and a detailed conversation that took place between, Charles Spearman, Godfrey Thomson, and Edward Thorndike within the Carnegie-funded International Examinations Inquiry meetings in 1931, 1935, and 1938. Unusually, there were transcriptions of all comments at these meetings, even of the smallest verbal utterance. The transcriptions offer a novel look at these researchers’ theoretical and practical approaches to intelligence testing and its place in education. Aspects of Thomson’s and Spearman’s personalities are in evidence too, from this unique source. One particular conversation among the three leads to an important new insight about intelligence and intelligence testing. These conversations provide new and complementary information on a trio of leading intelligence researchers whose individual contributions and interactions with each other were seminal in the scientific study of human cognitive abilities.

Technorati Tags: psychology, educational psychology, school psychology, neuropsychology, cognition, intelligence, IQ, g, general intelligence, Spearman

Quantoids corner-bifactor and second-order FA comparisons-Guest post by Matthew Reynolds

The following is a guest blog post by Matthew Reynolds, one of Tim Keith's Doctoral Student in Educational Psychology (School Psychology & Quantitative Methods) at the University of Texas at Austin, Department of Educational Psychology.

This is an excellent post by a future quantoid to be reckoned with in the field of school/educational psychology research. Kudos to Dr. Tim Keith for suggesting that one of his doctoral students make a quest blog post. This is the first such doctoral student virtual scholar post. If there are other professors who would like to entertain the idea of doctoral students being assigned articles to review and prepare for guest posts on IQ's Corner, then drop me an email.... iap@earthlink.net

• Chen, F. F., West, S. G., & Sousa, K. H. (2006). A comparison of bifactor and second-order models of quality of life. Multivariate Behavioral Research, 41, 189-225. (click to view)
  

Although not directly related to intelligence, this article compares two confirmatory factor analytic (CFA) models frequently used in psychometric research of intelligence: bifactor and second-order models. Chen et al. (2006) describe the bifactor model as having a general factor that accounts for the communality in all items and domain specific factors that account for influences above and beyond the general factor. The second-order model is described as having interrelated first-order factors with a general factor that accounts for those relations.

Study 1 compared the two models by applying the factor structure to a quality of life measurement from the AIDS Time-Oriented Health Outcome Study. Study 2 was a Monte Carlo study investigating whether there was enough power to detect differences in the bifactor and second-order model. Previous research had suggested that it was empirically impossible to distinguish between the two in typical samples used in social science research (i.e. Mulaik & Quartetti, 1997).

Results from Study 1:

• Bifactor and second-order factor models were imposed on a 17 item health-care related quality of life survey. The models had a general overall quality of life factor and four domain specific factors. The four domain-specific factors included cognition, vitality, mental health, and disease worry.

• The results from the bifactor model suggested that the mental health factor did not provide unique information above and beyond the general factor. Therefore, the model was re-specified without a mental health factor.

• The second-order factor model was specified with four first-order factors and a general quality of life factor that accounted for the relations among the first-order factors. The residual variance for the mental health factor, however, was statistically significant suggesting that there was some unique contribution of this factor (although the general factor accounted for 91.4% of the variance in that factor). Note this finding was different from the bifactor model. In the bifactor model the mental health factor did not provide unique information. Therefore, to be consistent with the bifactor model the authors also re-specified the second-order model so that only three factors, and the subtests related to mental health factor loaded directly on the second-order factor.

• The results comparing the two different models showed that both the bifactor and second-order factor models provided adequate fit. Because the second-order model is a more constrained version of the bi-factor model, the likelihood ratio test (i.e., chi-squared difference test) was used to compare the fit of the models. The second-order model fit worse than did the bifactor suggesting that the constraints applied to the bifactor model to get to the second-order model were too restrictive. Also, a power analysis suggested that there was adequate power to detect the difference.

• Next, the authors used these models to predict social functioning. Both models resulted in almost identical standardized estimates. This finding was rather reassuring in regards to the interpretability of the ability factors.

Study 2:

• The findings suggested that even with a sample size of 200 there appears to be enough power to detect differences between the bifactor and second-order models.

Discussion:

• The authors concluded that the bifactor model offers several advantages over the second-order model. One advantage was that it identified three factors instead of four. I am not quite convinced that this is necessarily an advantage. Two, they noted that researchers may miss potential non-significant first-order factor variances when looking at their results. I thought this was a good point by the authors; however, I also have had the same concern about using bifactor models. For example, a not-so-careful researcher may not consider the non-significant domain specific factor loadings as well as a non-significant domain specific factor variance.

• The second advantage was that the bifactor model fit better. That is, the relations between the general factor and the items could not be fully mediated by the first-order factors.

• Third, they stated that the bi-actor model is easier to interpret when predicting external criteria because the domain factors are represented as common factors in bifactor models whereas they are residualized factors in the higher-order model. Although true, I think the point is rather minor.

• Last, and perhaps most importantly, they conclude that BOTH models are useful in research. I agree completely with this point as CFA models should be consistent with theoretical models.

• In general, the article provides great information for those interested in hierarchical factor analysis, and it is provided in a straightforward manner. I think that the advantages of the bifactor model were a bit overstated. I do agree that it is useful to examine both models in research, especially since the second-order model can be derived from the bifactor model.

• In my own research, one weakness of the bifactor model has been related to empirical under-identification. I believe that perhaps it runs into some of the same difficulties as the multi-method multi-trait models in that they are over-parameterized. A recent study that used the bifactor model to test for method effects also found that the bifactor model may fit well even when it is an incorrect model (Maydeu-Olivares & Coffman, 2006).

• In terms of research in psychometric intelligence, the interpretation of the two models is slightly different as well. For example, in a bifactor model all of the effects of the general factor are direct. In intelligence research it seems to me that the contemporary theories are more consistent with the higher order model in which the general factor explains the interrelations of the broad abilities and its relation to test performance is mediated through the broad abilities.

• To make this more germane to intelligence researchers I have included some output of analyses that I performed using the Holzinger & Swineford correlation matrix reported in their 1937 study. Shown are the specifications, the models with standardized loadings, and the unstandardized loadings, variances, and the total effects shown separately. Just as a warning, the models are not in publication form, but suffice for a demonstration. I hope these models help to clarify how the second-order model is in fact a more constrained version of the bifactor model. See Yung, Thissen, and McLeod (1999) for a more technical account.

• Last, as an aside, I thought I would share the last two sentences from the Holzinger & Swineford 1937 article in Psychometrika. In this article the authors introduced the bifactor model:

• “The Bi-factor analysis illustrated above is not only very simple, but the calculation is relatively easy as compared with other methods. The total time for computation, done by one person, was less than ten hours for the present example.”

• I just ran a bifactor model in Amos 5, and other than setting the model up, the actual computational time took 0.29 seconds. You have to appreciate all of the time and patience that researchers have put in over the years to get us where we are today!

Technorati Tags: psychology, educational psychology, neuropsychology, statistics, factor analysis, confirmatory factor analysis, quality of life, QOL, g, general intelligence

powered by performancing firefox

CHC cognitive abilities and expertise - Guest post by Ruben Lopez

Over on the CHC listserv there has been some good discussion of CHC cognitive abilities, SLODR, and the development of expertise. Ruben Lopez, a regular and thoughtful contributor to the IAP list, made the following post today. I liked it so much I asked if I could post it as a guest blog comment here at IQs Corner. Ruben agreed...thanks Ruben.

"I just ran across a discussion of CHC and SLODR in The Cambridge Handbook of Expertise and Expert Performance by Ericsson, Charness, Feltovich, and Hoffman (2006). (It seems that the findings on expertise and expert performance have important implications for teaching kids to become experts-well, at least proficient-in school subjects.)

About CHC and SLODR, Earl Hunt says, "Gf and Gc are correlated, which makes it possible to speak reasonably about g. However, correlations between measures of different types of cognitive abilities are highest toward the low end of the general intelligence scale, and markedly lower at the high end (Detterman & Daniel, 1989, Deary et al., 1996). This is important, as expertise is generally associated with high levels of performance.

Measures of Gf have substantial correlations with measures of the performance of working memory. A high-Gf person is probably good at keeping track of several things at once and concentrating his or her attention in the face of distractions (Engle, Kane & Tulhoski, 1999; Kyllonen & Christal, 1990). These talents are good to have during the learning phase of most psychomotor activities (e.g., skiing, riding a bicycle, playing tennis). However, they are much less needed once an activity has been learned. Laboratory studies of how people learn to do psychomotor tasks have shown that intelligence is a reasonably good predictor of performance early in learning but does not predict asymptotic levels of learning very well (Ackerman 1996; Fleishman, 1972).

... Some aspects of expertise, such as swinging a golf club, require learning a constant relationship between stimulus and response. Other aspects, such as the analogical reasoning typical of the law, involve varied mappings, the development of mental models of a situation, and extensive knowledge. Demands on both Gf and Gc never cease." (pp. 32-33)

CHC applied to the study of expert performance-cool."

Technorati Tags: psychology, educational psychology, school psychology, IQ, intelligence, Spearman, CHC, CHC theory, cognitive, cognition, expertise, Gf, Gc, SLODR, working memory

powered by performancing firefox

CHC-based analysis of Spearman's law of diminishing returns in KABC-II

Kudos to one of the best quantoids in school/educational psychology (Dr. Tim Keith), and a serious quantoid in the making (Keith's doctoral student, Mathew Reynolds) re: their "in press" article in Intelligence.

• Reynolds, M. & Keith, T. (2006, in press). Spearman's law of diminishing returns in hierarchical models of intelligence for children and adolescents, Intelligence (click here to view)

Using CFA methods, Reynolds and Keith investigated Spearman's law of diminishing returns (SLODR) in the norm sample for the CHC-based KABC-II. I'm excited about this paper due to the elegant use of CFA to evaluate whether SLODR exists, and, if it does, at what level of the CHC taxonomy (stratum I, II, or III - with five broad CHC abilities being represented--Gf,Gc,Gv,Glr,Gsm). The abstract provides a sufficient summary...so I won't waste any bandwidth.

Abstract

• Spearman's “law of diminishing returns” or SLODR refers to a decrease in g saturation as ability level increases. SLODR has been demonstrated in a number of intellectual batteries but several important aspects of the phenomenon are not yet well understood. We investigated the presence of SLODR in the Kaufman Assessment Battery for Children—Second Edition (KABCII), a popular measure of intelligence for children. We used confirmatory factor analysis to investigate the invariance of two hierarchical factor structures across ability groups; the subtest variance explained by the ability factors across groups; and whether SLODR was produced only by subtests with low loadings on the general ability factor. We found that SLODR was present in the KABC-II, and its presence was not dependent on the hierarchical model of intelligence. Moreover, our findings suggest that SLODR acts on g and not on the broad abilities, although the contribution of g to various broad abilities is lower in the high ability group. Finally, SLODR was not produced by the subtests with the lowest g loadings on the general factor.

Technorati Tags: psychology, educational psychology, neuropsychology, school psychology, special educaton, IQ, IQ testing, intelligence, CHC, Cattell-Horn-Carroll, g, general intelligence, Spearman, SLODR, CFA, Gf, Gc, Gv, Glr, Gsm, KABCII, KABC-II

powered by performancing firefox

IQ PREPLOG: CHC abilities and effects on reading skills--beyond g--some specific abilities important

This is an IQ PREPLOG post. The following manuscript is now "in press" and can be viewed by clicking here.

Floyd, R., Keith, T., Taub, G. McGrew, K. (2006). Cattell–Horn–Carroll Cognitive Abilities and Their Effects on Reading Decoding Skills: g Has Indirect Effects, More Specific Abilities Have Direct Effects. School Psychology Quarterly (in press 2-28-06)

Abstract

• Previous research has indicated the importance of several cognitive abilities to the acquisition and development of reading decoding skills. The findings from such research may be limited because some abilities, such as general intelligence (g), frequently have been omitted from analyses. Drawing on the ability taxonomy of the Cattell–Horn–Carroll (CHC) theory, this study employed structural equation modeling to examine the effects of CHC abilities on reading decoding skills using 5 age-differentiated subsamples from the standardization sample of the Woodcock–Johnson III (Woodcock, McGrew, Mather, 2001). Using the Spearman Model including only g, strong direct effects of g on reading decoding skills were demonstrated at all ages. Using the Two-Stratum Model including g and broad abilities, direct effects of the broad abilities Long-Term Storage and Retrieval, Processing Speed, Crystallized Intelligence, Short- Term Memory, and Auditory Processing on reading decoding skills were demonstrated at select ages. Using the Three-Stratum Model including g, broad abilities, and narrow abilities, direct effects of the broad ability Processing Speed and the narrow abilities Associative Memory, Listening Ability, General Information, Memory Span, and Phonetic Coding were demonstrated at select ages. Across both the Two-Stratum Model and the Three-Stratum Model at all ages, g had very large but indirect effects. The findings suggest that school psychologists should interpret measures of some specific cognitive abilities when conducting psychoeducational assessments designed to explain reading decoding skills.

Technorati Tags: psychology, cognition, psychological assessment, iq testing, CHC theory, Cattell-Horn-Cattell, Gf-Gc theory, reading, decoding, special education, education, achievement, SEM, structural equation modeling, g, general intellignece, Spearman