TY - JOUR
T1 - Examining the Latent Structure of the Delis-Kaplan Executive Function System
AU - Karr, Justin E.
AU - Hofer, Scott M.
AU - Iverson, Grant L.
AU - Garcia-Barrera, Mauricio A.
N1 - Publisher Copyright:
© The Author(s) 2018. Published by Oxford University Press. All rights reserved.
PY - 2019/1/7
Y1 - 2019/1/7
N2 - Objective The current study aimed to determine whether the Delis-Kaplan Executive Function System (D-KEFS) taps into three executive function factors (inhibition, shifting, fluency) and to assess the relationship between these factors and tests of executive-related constructs less often measured in latent variable research: reasoning, abstraction, and problem solving. Method Participants included 425 adults from the D-KEFS standardization sample (20-49 years old; 50.1% female; 70.1% White). Eight alternative measurement models were compared based on model fit, with test scores assigned a priori to three factors: inhibition (Color-Word Interference, Tower), shifting (Trail Making, Sorting, Design Fluency), and fluency (Verbal/Design Fluency). The Twenty Questions, Word Context, and Proverb Tests were predicted in separate structural models. Results The three-factor model fit the data well (CFI = 0.938; RMSEA = 0.047), although a two-factor model, with shifting and fluency merged, fit similarly well (CFI = 0.929; RMSEA = 0.048). A bifactor model fit best (CFI = 0.977; RMSEA = 0.032) and explained the most variance in shifting indicators, but rarely converged among 5,000 bootstrapped samples. When the three first-order factors simultaneously predicted the criterion variables, only shifting was uniquely predictive (p <.05; R2 = 0.246-0.408). The bifactor significantly predicted all three criterion variables (p <.001; R2 = 0.141-242). Conclusions Results supported a three-factor D-KEFS model (i.e., inhibition, shifting, and fluency), although shifting and fluency were highly related (r = 0.696). The bifactor showed superior fit, but converged less often than other models. Shifting best predicted tests of reasoning, abstraction, and problem solving. These findings support the validity of D-KEFS scores for measuring executive-related constructs and provide a framework through which clinicians can interpret D-KEFS results.
AB - Objective The current study aimed to determine whether the Delis-Kaplan Executive Function System (D-KEFS) taps into three executive function factors (inhibition, shifting, fluency) and to assess the relationship between these factors and tests of executive-related constructs less often measured in latent variable research: reasoning, abstraction, and problem solving. Method Participants included 425 adults from the D-KEFS standardization sample (20-49 years old; 50.1% female; 70.1% White). Eight alternative measurement models were compared based on model fit, with test scores assigned a priori to three factors: inhibition (Color-Word Interference, Tower), shifting (Trail Making, Sorting, Design Fluency), and fluency (Verbal/Design Fluency). The Twenty Questions, Word Context, and Proverb Tests were predicted in separate structural models. Results The three-factor model fit the data well (CFI = 0.938; RMSEA = 0.047), although a two-factor model, with shifting and fluency merged, fit similarly well (CFI = 0.929; RMSEA = 0.048). A bifactor model fit best (CFI = 0.977; RMSEA = 0.032) and explained the most variance in shifting indicators, but rarely converged among 5,000 bootstrapped samples. When the three first-order factors simultaneously predicted the criterion variables, only shifting was uniquely predictive (p <.05; R2 = 0.246-0.408). The bifactor significantly predicted all three criterion variables (p <.001; R2 = 0.141-242). Conclusions Results supported a three-factor D-KEFS model (i.e., inhibition, shifting, and fluency), although shifting and fluency were highly related (r = 0.696). The bifactor showed superior fit, but converged less often than other models. Shifting best predicted tests of reasoning, abstraction, and problem solving. These findings support the validity of D-KEFS scores for measuring executive-related constructs and provide a framework through which clinicians can interpret D-KEFS results.
KW - Assessment
KW - Executive functions
KW - Test construction
UR - http://www.scopus.com/inward/record.url?scp=85058362577&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85058362577&partnerID=8YFLogxK
U2 - 10.1093/arclin/acy043
DO - 10.1093/arclin/acy043
M3 - Article
C2 - 29733343
AN - SCOPUS:85058362577
SN - 0887-6177
VL - 34
SP - 381
EP - 394
JO - Archives of Clinical Neuropsychology
JF - Archives of Clinical Neuropsychology
IS - 3
ER -