Structural Factor Analysis Experiments with Incomplete Data

McArdle, John J.

doi:10.1207/s15327906mbr2904_5

Cited by 189 publications

(183 citation statements)

References 83 publications

Supporting

Mentioning

178

Contrasting

Order By: Relevance

“…Prime among these, as common to general structural equation models, is the assumption of incomplete data as missing at random (Rubin, 1974). This assumption justified our estimation procedures based on raw maximum likelihood (Arbuckle, 1996;McArdle, 1994). We contend, though, that rather than basing our findings on the missing at random assumption, we rely on the assumption that the incompleteness mechanisms underlying the overall cognitive-sensory data structure weaken, rather than strengthen, the dynamic relations among variables Singer et al, 2003).…”

Section: Discussionmentioning

confidence: 99%

“…2 Instead, we used the Raw Maximum Likelihood (Arbuckle, 1996;McArdle, 1994) estimation algorithm, introduced by Lange, Westlake, and Spence (1976) in the context of pedigree analysis. This algorithm relies on the missing at random assumption (MAR; Rubin, 1974) and produces unbiased estimates if missing information is either completely at random or dependent on variables included in the statistical model (which in the longitudinal setting usually includes previous measurements of the same variable).…”

Section: Statistical Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Exploring structural dynamics within and between sensory and intellectual functioning in old and very old age: Longitudinal evidence from the Berlin Aging Study

Ghisletta

Lindenberger

2005

Intelligence

View full text Add to dashboard Cite

Cross-sectional and longitudinal analyses of age-heterogeneous samples have revealed correlational links between and within intellectual, sensory, and sensorimotor domains. Due to basic limitations of cross-sectional designs and a reluctance to disentangle antecedent-consequent relations in longitudinal designs, the functional significance and dynamics of these links have remained unclear. Advanced structural equation models allow representing multivariate longitudinal changes as a function of time-based and directed relations. We applied this methodology to longitudinal data from the Berlin Aging Study (at inception total N = 516; age range = 70-103 years) to explore the structural dynamics among perceptual speed, verbal knowledge, close visual acuity, and distance visual acuity. We found reliable, occasion-based, age-partialed latent regression paths that influence longitudinal changes within and across intellectual and sensory domains. We conclude that intellectual and sensory domains are dynamically linked in old and very old age, and discuss implications of this finding for theories of cognitive aging. D

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Statistical Modelsmentioning

confidence: 99%

Exploring structural dynamics within and between sensory and intellectual functioning in old and very old age: Longitudinal evidence from the Berlin Aging Study

Ghisletta

Lindenberger

2005

Intelligence

View full text Add to dashboard Cite

show abstract

“…We may guess that this will be the common case in applications (for further discussion, see Heckman, 1979;Heckman & Robb, 1986;McArdle, 1994;Rubin, 1987Rubin, , 1991.…”

Section: Sampling and Missing Datamentioning

confidence: 99%

Principles and practice in reporting structural equation analyses.

McDonald

2002

Psychological Methods

3,643

2,447

View full text Add to dashboard Cite

Principles for reporting analyses using structural equation modeling are reviewed, with the goal of supplying readers with complete and accurate information. It is recommended that every report give a detailed justification of the model used, along with plausible alternatives and an account of identifiability. Nonnormality and missing data problems should also be addressed. A complete set of parameters and their standard errors is desirable, and it will often be convenient to supply the correlation matrix and discrepancies, as well as goodness-of-fit indices, so that readers can exercise independent critical judgment. A survey of fairly representative studies compares recent practice with the principles of reporting recommended here.Structural equation modeling (SEM), also known as path analysis with latent variables, is now a regularly used method for representing dependency (arguably "causal") relations in multivariate data in the behavioral and social sciences. Following the seminal work of Jöreskog (1973), a number of models for linear structural relations have been developed (Bentler & Weeks, 1980;Lohmoller, 1981;McDonald, 1978), and work continues on these. Commercial statistical packages include LISREL (Jöreskog & Sör-bom, 1989, 1996, EQS (Bentler, 1985(Bentler, , 1995, CALIS (Hartmann, 1992), MPLUS (Muthén & Muthén, 1998), RAMONA (Browne, Mels, & Cowan, 1994), SEPATH (Steiger, 1995), and AMOS (Arbuckle, 1997). Available freeware includes COSAN (Fraser & McDonald, 1988) and Mx (Neale, 1997). McArdle and McDonald (1984) proved that different matrix formulations of a path model with latent variables are essentially equivalent. Programs such as those listed supply essentially the same basic information, with minor variations in the details supplied. Thus, the eight parameter LISREL model, which arose out of the work of Keesling and Wiley (see Wiley, 1973) and was subsequently developed to its current state by Jöreskog (see Jöreskog and Sör-bom, 1996), the four-matrix model of Lohmoller (1981), the three-matrix EQS model of Bentler and Weeks (1980), and the two-matrix RAM model (see McArdle & McDonald, 1984) rearrange the same set of parameters. Not surprisingly-and perhaps not regrettably-user guides and texts on this topic are not in agreement in their recommendations about the style of presentation of results (e.g., see Bollen, 1989;Loehlin, 1992;Long, 1983aLong, , 1983b. There is even less agreement in the form of the results actually reported in articles on applications.It would be immodest for any journal article to offer a code of practice for the presentation of SEM results. It could also be counterproductive. (We note that for a long time there was a uniformly accepted convention for the publication of analysis of variance, or ANOVA results: the standard ANOVA table and the table of cell means. The near-disappearance of this from journals is regrettable.) Sound guidelines for the reporting of SEM results have been offered previously

show abstract

“…This is accomplished by specifying models that are identical to Model 3 except that, in turn, a common implicit method factor is not specified (Model 4) and a common explicit method factor is not specified (Model 5). By comparing the fits of each with that of Model 3, we may discern to what extent accounting for common method variance is important to understanding the structure of relations among these measures.We used the full information maximum likelihood (FIML) estimation approach in our analyses (Enders & Bandalos, 2001;McArdle, 1994) be based on all available data. That is, even though our four samples varied in terms of the collection of attitude domains on which they were measured (see Table 2), the FIML technique allows variable interrelations and parameter standard errors to be estimated from the combined data of all samples.…”

mentioning

confidence: 99%

A Multitrait-Multimethod Validation of the Implicit Association Test

Nosek

Smyth

2007

Experimental Psychology

310

View full text Add to dashboard Cite

Abstract. Recent theoretical and methodological innovations suggest a distinction between implicit and explicit evaluations. We applied Campbell and Fiske's (1959) classic multitrait-multimethod design precepts to test the construct validity of implicit attitudes as measured by the Implicit Association Test (IAT). Participants (N = 287) were measured on both self-report and IAT for up to seven attitude domains. Through a sequence of latent-variable structural models, systematic method variance was distinguished from attitude variance, and a correlated two-factors-per-attitude model (implicit and explicit factors) was superior to a single-factor-per-attitude specification. That is, despite sometimes strong relations between implicit and explicit attitude factors, collapsing their indicators into a single attitude factor resulted in relatively inferior model fit. We conclude that these implicit and explicit measures assess related but distinct attitude constructs. This provides a basis for, but does not distinguish between, dual-process and dual-representation theories that account for the distinctions between constructs. Keywords: implicit social cognition, attitudes, individual differences, construct validity, structural equation modelingRealizing that the human mind is more than the sum of its conscious processes, a number of theorists have proposed a conceptual distinction between evaluations that are the products of introspection, called explicit attitudes, and those that occur automatically and may exist outside of conscious awareness, called implicit attitudes (Greenwald & Banaji, 1995;Wilson, Lindsey, & Schooler, 2000). Greenwald and Banaji (1995, p. 8), for example, defined implicit attitudes as "introspectively unidentified (or inaccurately identified) traces of past experience that mediate favorable or unfavorable feelings toward an attitude object." This theory has developed in conjunction with the invention of measurement tools that assess automatic evaluative associations without introspection (e.g., Fazio, Sanbonmatsu, Powell, & Kardes, 1986;Greenwald, McGhee, & Schwartz, 1998;Nosek & Banaji, 2001;Wittenbrink, Judd, & Park, 1997).Some experiences with these new measurement tools have spawned doubts about whether they measure attitudes at all (Karpinski & Hilton, 2001), and whether a conceptual distinction between implicit and explicit attitudes is worthwhile (Fazio & Olson, 2003). Fazio and Olson contend "it is more appropriate to view the measures as implicit or explicit, not the attitude (or whatever other construct)" (2003, p. 303). The purpose of the research we report was to test whether the structure of attitude variance derived from an implicit measure (the Implicit Association Test [IAT]; Greenwald et al., 1998) and from an explicit one (semantic differentials) is best represented by one latent factor or by two correlated latent factors, when stripped of confounding method variance. If our hypothesis that the latter specification will fit the data better is sustained, it will support a view ...

show abstract

Structural Factor Analysis Experiments with Incomplete Data

Cited by 189 publications

References 83 publications

Exploring structural dynamics within and between sensory and intellectual functioning in old and very old age: Longitudinal evidence from the Berlin Aging Study

Exploring structural dynamics within and between sensory and intellectual functioning in old and very old age: Longitudinal evidence from the Berlin Aging Study

Principles and practice in reporting structural equation analyses.

A Multitrait-Multimethod Validation of the Implicit Association Test

Contact Info

Product

Resources

About