Reconsidering the Psychometrics of the GRS-S: Evidence for Parsimony in Measurement

Abstract

The authors evaluated measurement-level, factor-level, item-level, and scale-level revisions to the Gifted Rating Scales–School Form (GRS-S). Measurement-level considerations tested the extent to which treating the Likert-type scale rating as categorical or continuous produced different fit across unidimensional, correlated trait, and bifactor latent factor structures. Item- and scale-level analyses demonstrated that the GRS-S could be reduced from a 72-item assessment on a 9-point rating scale down to a 30-item assessment on a 3-point rating scale. Reliability from the reduced assessment was high (ω > .95). Receiver operating characteristic (ROC) curve comparisons between the original and reduced versions of the GRS-S showed that diagnostic accuracy (i.e., area under the curve) of the scales was comparable when considering cut scores of 120, 125, and 130 on the WISC-IV Full Scale ( Wechsler Intelligence Scale for Child–Fourth Edition) and verbal IQ and the WIAT-III (Wechsler Individual Achievement Test–Third Edition) composite score. The findings suggest that a brief form of the GRS-S can be used as a universal or selective screener for giftedness without sacrificing key psychometric considerations.

The Gifted Rating Scales-Preschool/Kindergarten Form (GRS-P): A Preliminary Examination of Their Psychometric Properties in Two Greek Samples

“…Artistic Talent Scale: χ 2 (54, N = 98) = 167.65, ρ < 0.000, CFI = 0.93, SRMR = 0.03, RMSEA = 0.15 (CI 90% 0.12–0.17). Motivation Scale : χ 2 (54, N = 99) = 191.37, ρ < 0.000, CFI = 0.93, SRMR = 0.03, RMSEA = 0.16 (CI 90% 0.14–0.18) [ 44 , 63 ]. Although all parameters of the aforementioned models were found to be statistically significant ( p < 0.05) and standardized root mean squared residual (SRMR) values were below 0.05, also indicating a good fit for the models tested, the chi-square goodness-of-fit test was statistically significant for all the initial models resulting in a rejection of the null hypothesis of good fit.…”

“…Artistic Talent Scale : χ 2 (42, N = 98) = 46.03, ρ < 0.309, CFI = 1.00, SRMR = 0.01, RMSEA = 0.03 (CI 90% 0.00–0.08). Motivation Scale : χ 2 (46, N = 99) = 87.32, ρ < 0 0.000, CFI = 0.98, SRMR = 0.02, RMSEA = 0.10 (CI 90% 0.06–0.12) [ 44 , 63 ].…”

“…However, the information, which is presented in the manual [ 33 ], does not support the two GRS forms’ structural validity, and taking into account that no overall score is available on both of them, the proposed (in an item-level analysis) theoretical structure of the GRS-P and the GRS-S consists of 5 and 6 first-order constructs (latent variables), respectively [ 42 , 43 ]. Although Pfeiffer et al [ 33 ] stated that they conducted factor analyses to examine the internal structure of the GRS, they did not specify, in the manual [ 33 ], which kind of factor analyses were conducted [ 42 ], while they have presented this kind of results from item-level analyses supporting the proposed 6-factor structure as regards mainly the GRS-S [ 44 , 45 , 46 ].…”

“…To our knowledge, the proposed theoretical structure of the GRS-P consisting of 5 first-order constructs (latent variables) has been tested, with item-level analyses (CFA), by Karadag and Pfeiffer [ 41 ] in a Turkish sample. However, it is noteworthy that the indices of their proposed verified model ranged from not accepted (RMSEA = 0.070) to marginally accepted (CFI = 0.919) [ 44 , 45 , 46 ]. Furthermore, although the possibility that a general factor could also account for most of the variance captured by GRS ratings cannot be rejected, the GRS-P scoring model has been examined using a bifactor model approach only by Benson and Kranzler [ 47 ] in the standardization sample in the norming of the GRS-P.…”

Αποδομώντας Πέντε Διαδεδομένους Μύθους Για Τη Χαρισματικότητα Και Τους Χαρισματικούς Μαθητές

“…Πράγματι, η επικράτηση της εξίσωσης της χαρισματικότητας με την υψηλή γενική νοητική ικανότητα g είναι ιδιαίτερα διαδεδομένη (Robinson et al, 2000, με το ελληνικό τμήμα της MENSA να αναγνωρίζει ως χαρισματικά, άτομα με δείκτη ευφυίας στο υψηλότερο 2% του πληθυσμού. Αν και η ουδός που καθορίζει τη διάκριση χαρισματικών και μη χαρισματικών ατόμων αλλάζει ανάλογα με το εργαλείο, τους λειτουργικούς ορισμούς ή τους πληθυσμούς (Fernández et al, 2017‧ Heacox & Cash, 2013‧ NAGC, 2017‧ Reis & McCoach, 2000, οι προσεγγίσεις αυτές προωθούν από κοινού μια ψυχομετρική θεώρηση της χαρισματικότητας, η οποία τελεί όμως συχνά υπό αμφισβήτηση (Petscher & Pfeiffer, 2019‧ από διαφορετικούς ορισμούς και τρόπους μέτρησης. Για παράδειγμα, στο πλαίσιο της τριαρχικής θεωρίας του, ο Sternberg (1999Sternberg ( , 2000 ορίζει τη χαρισματικότητα ως την εξαιρετική επίδοση στην αναλυτική, την πρακτική και τη δημιουργική νοημοσύνη, υιοθετώντας μια τυπολογία της ευφυίας που υπερβαίνει τις παραδοσιακές προσεγγίσεις με εστίαση στον γενικό παράγοντα g. Έτσι, ακόμη και στις πιο απλουστευτικές ψυχομετρικές προσεγγίσεις, για να απαντηθεί τι είναι «χαρισματικότητα» προϋποτίθεται η απάντηση στο ερώτημα «τι είναι νοημοσύνη‧».…”