Long-Term Stability of Process Scores in the French WISC-IV

Poster

Reference

Long-Term Stability of Process Scores in the French WISC-IV

KIENG, Sotta, et al.

Abstract

In order to provide clinicians with a more thorough understanding of child's performance on Block Design, Digit Span and Cancellation, seven process scores are included in the Wechsler Intelligence Scale for Children-Four edition (WISC-IV): Block Design No Time Bonus (BDN), Digit Span Forward (DSF), Digit Span Backward (DSB), Longest Digit Span Forward (LDSF), and Longest Digit Span Backward (LDSB). For relevance of prognostic statements, it is essential to rely on test scores that are stable across time. The goal of this study was to explore the long-term stability of the WISC-IV process scores (BDN, DSF, DSB, LDSF, and LDSB). The sample consisted of 277 nonclinical French-speaking Swiss children aged between 7 and 12 years (at first testing: mean age = 8.87 and SD = 0.82; at second testing: mean age=10.64 and SD = 1.11). The WISC-IV were administered twice in an average test-retest interval of 1.77 years (SD = 0.56).

KIENG, Sotta, et al. Long-Term Stability of Process Scores in the French WISC-IV. In: The 19th Meeting of the European Society for Cognitive Psychology (ESCOP), Paphos (Cyprus), 17th to 20th September 2015, 2015

Available at:

http://archive-ouverte.unige.ch/unige:76718

Disclaimer: layout of this document may differ from the published version.

1 / 1

Long-Term St abilit y of Process Scores in t he French WISC-IV ^*

S. Kieng

, N. Favez

, J. Rossier

, S. Geist lich

, & T. Lecerf

1Facult y of Psychology and Educat ional Sciences, Universit y of Geneva, ²Inst it ut e of Psychology, Universit y of Lausanne, and ³Dist ance Learning Universit y, Swit zerland

The 1 9^{t h} Meet ing of t he European Societ y f or Cognit ive Psychology Paphos, Cyprus, Sept ember 2 0 1 5 ; Cont act : sot t a.kieng@unige.ch

INTRODUCTION

 The Wechsler Int elligence Scale f or Children-f ourt h edit ion (WISC-IV) is f requent ly used t o assess a child’s overall int ellect ual abilit y. The clinical int erpret at ion based on a Full Scale Int elligence Quot ient (FSIQ) and 4 indices ( Verbal Comprehension Index: VCI; Percept ual Reasoning Index: PRI; Processing Speed Index: PSI; and Working Memory Index: WMI) allows t he pract it ioners t o develop hypot heses regarding normat ive and personal st rengt hs and weaknesses.

 In addit ion t o FSIQ and t he 4 index scores, 7 process scores are also yielded: Block Design No Time Bonus (BDN) , Digit Span Forward (DSF) , Digit Span Backward (DSB) , Longest Digit Span Forward (LDSF) , and Longest Digit Span Backward (LDSB) . The LDSF score is t he number of digit s correct ly repeat ed on t he last f orward t rial, and t he LDSB is t he number of digit s correct ly repeat ed on t he last backward t rial.

 The process scores provide clinicians wit h a more t horough underst anding of child’s performance on visual-spat ial processing, working memory, and processing speed.

*This work was support ed by Grant 1 3 5 4 0 6 awarded by t he Swiss Nat ional Science Foundat ion ( Long -t erm st abilit y of t he WISC- IV: St andard and CHC composit e scores, Lecerf , Favez & Rossier)

OBJECTIVE

The goal of t his st udy was t o explore t he long-t erm st abilit y of t he WISC-IV process scores ( BDN, DSF, DSB, LDSF, and LDSB) wit h an average Test (T1) – Ret est (T2) int erval of 1 .7 7 years ( SD = 0 .5 6 years; range f rom 1 t o 3 .2 5 years) .

 Mean Full Scale IQ was 1 0 0 .8 1 (SD = 1 3 .9 4 ) at T1 and 1 0 3 .3 4 (SD = 1 2 .7 8 ) at T2 .

 At an int erindividual level, st abilit y coef f icient s of process scores ranged f rom .2 9 on LDSB t o .6 6 on BDN.

RESULTS

 2 7 7 non clinical French-speaking Swiss children;

 Aged bet ween 7 and 1 2 years;

 All of t hem had prof iciency in speaking, underst anding, and reading French;

 None of t hem has doubled or skipped class;

 Relat ively represent at ive of t he Geneva children’s populat ion (sex, SES) .

 The 1 0 core subt est s and t he subt est Pict ure Complet ion of t he WISC-IV were administ ered t wice t o each child.

 The 7 process component s of BDN, DSF, DSB, LDSF, and LDSB were comput ed according t o t he inst ruct ions report ed in t he WISC-IV Administ rat ion and Scoring Manual.

 For BDN, DSF, and DSB, t he individual changes in scores across t he ret est int erval were explored wit hin st andard error of measurement (SEM) . More precisely, a score was considered st able across t he ret est int erval when perf ormances remained wit hin ±2 SEM. Thus, approximat ely 9 5 % of t he ret est scores should be f oreseen wit hin ± 2 .9 4 , ± 2 .5 4 , and ± 2 .9 6 point s of t he init ial score f or BDN, DSF, and DSB, respect ively.

 For LDSF and LDSB, individual changes were present ed wit h use of number of children which earned ident ical scores at bot h assessment s.

METHOD

T1 T2

M ( SD) M ( SD) r p d

BDN 1 1 .0 3 ( 3 .0 9 ) 1 0 .8 3 ( 3 .0 6 ) .6 6 0 .1 9 0 .0 7 DSF 9 .3 7 ( 2 .5 9 ) 9 .9 1 ( 2 .8 3 ) .5 2 <0 .0 1 -0 .2 0 DSB 9 .8 6 ( 2 .6 0 ) 9 .7 1 ( 2 .9 2 ) .3 2 0 .4 3 0 .0 5 LDSF 4 .9 3 ( 0 .9 3 ) 5 .4 8 ( 1 .0 6 ) .4 7 <0 .0 1 -0 .5 5 LDSB 3 .5 1 ( 0 .8 3 ) 3 .9 2 ( 1 .0 4 ) .2 9 <0 .1 -0 .4 4

% of “stable” individuals within +/ -2 SEM

BDN 6 6 .4

DSF 6 2 .5

DSB 5 5 .2

CONCLUSION

 In line wit h Ryan, Umf leet , and Kane ( 2 0 1 3 ) , our f indings on a sample of 2 7 7 nonclinical children suggest ed t hat t he process scores were not st able in long-t erm.

 At an int erindividual level, result s showed low st abilit y coef f icient s f or all process scores (r < .7 0 ) .

 At an int ra-individual level, more t han 3 0 % of t he children earned BDN, DSF, and DSB scores t hat dif f ered by more t han ±2 SEM point s bet ween t he t wo t est ings.

 Caut ion should be exercised in t he int erpret at ion of t hese process scores. It is not recommended t o use t hem f or individual predict ions, but f or t he int erpret at ions in t he here and now.

 At an int ra-individual level, less t han 7 0 % of children had t heir process scores

wit hin a conf ident int erval of ±2 SEM bet ween t he t wo assessment s. Thus, t he predict ions will be correct f or less t han seven children of t en.

Sample N Mean Age T1 ( SD) Mean Age T2 ( SD)

Girls 1 4 5 8 .9 3 ( 0 .8 1 ) 1 0 .6 9 ( 1 .0 6 )

Boys 1 3 2 8 .7 9 ( 0 .8 6 ) 1 0 .5 8 ( 1 .1 7 )

Tot al sample 2 7 7 8 .8 7 ( 0 .8 2 ) 1 0 .6 4 ( 1 .1 1 )

 For LDSF and LDSB, only 3 4 .7 % and 3 8 .6 % of children earned ident ical scores at bot h assessment s.