Measurement Invariance of the Chinese Gifted Rating Scales

Abstract

The Gifted Rating Scales-School Form (GRS-S) has been validated in several countries; however, no study has examined the rater invariance of this measure. The present study built on previous validity studies and examined configural and metric invariance between parent and teacher raters using the Chinese version of the GRS-S Teacher and Parent Forms. Analyses were conducted from a Jöreskog tradition, with testing of hypotheses related to group invariance. Model fit indices (i.e., comparative fit index [CFI], root mean square error of approximation [RMSEA], χ 2 /df) supported equivalence in factor structure and standardized loadings between raters. Additionally, hypothesis testing revealed equivalence in loadings for raters. Implications of the results, as well as limitations and directions for future research, are discussed. KeywordsGifted Rating Scales; Chinese; rater invariance China has had a long history of seeking out and educating child prodigies (Tsui-Chen, 1961). The main criteria in identifying a gifted child in China until the early 1900s had been literary creation (ability to compose poems and essays), reading, and exceptional memory of literary works. Recently, intelligence scores obtained from the Chinese versions of the Stanford-Binet Intelligence Test, the Wechsler Preschool and Primary Scale of IntelligenceRevised, and Test for Identifying the Cognitive Ability of Supernormal Children (TICASC) have been used to identify gifted children (Shi & Zha, 2000;Stevenson, 1998). Shi and Zha (2000) have suggested a broader definition of giftedness to encompass multiple dimensions, such as high cognitive ability, academic skills, leadership, creative potential, and even positive personality traits. Many of the aspects of giftedness are consistent with the recent definition of giftedness in other countries such as the United States. For example, the U.S. Department of Education has conceptualized giftedness as extraordinary intellectual and specific academic or artistic abilities or high performance capability in creativity or leadership (U.S. Department of Education, Office of Educational Research and Improvement, 1993 To conduct a comprehensive evaluation of a potentially gifted child, teachers and parents are often invited to provide information on the child (Pfeiffer, 2002). Although early reports (e.g., Gear, 1978;Pegnato & Birch, 1959) indicated that teachers were relatively poor at identifying gifted students because they were too close to students to have a sound perspective, more researchers (e.g., Gagné, 1994;Hunsaker, Finley, & Frank, 1997;Renzulli & Delcourt, 1986), however, have argued that teacher nomination and teacher rating for gifted children identification were very useful in identifying gifted students. Furthermore, teacher accuracy can be improved through the use of psychometrically sound rating scales that have been formulated from a list of characteristics of gifted children. Therefore, welldesigned teacher rating scales provide teachers with a reliable and efficient way to s...

The Reliability and Validity of a Spanish Translated Version of the Gifted Rating Scales

“…The present estimates of reliability measured here (i.e., internal consistency and standard error of measures) were comparable to the U.S. standardization sample of the GRS-S , which resulted in internal consistency ranging from .97 to .99, and standard error of measures values ranging 1.00 to 1.73 (Marguilies & Floyd, 2004; Pfeiffer & Jarosewich, 2003, 2007; Ward, 2005). Moreover, these findings are consistent with those found in separate international studies with a Korean-translated (Lee & Pfeiffer, 2006) and Chinese-translated (Petscher & Li, 2007) versions of the GRS-S .…”

“…Though partly confounded by a relatively small sample (i.e., approximately half of what would be considered minimally acceptable), perhaps future analyses should involve multiple theoretically driven modeling. Recent research by Petscher & Li (2007) showed that in a Chinese sample of students, a six-factor solution fit better than a g model, but also better than 2-factor, 3-factor, and multiple second-order factor models. Though evidence was observed to support the retention of two models, these results could be culturally contextual, and should be studied in greater depth with a larger sample to warrant such testing.…”

Merske značilnosti slovenskega prevoda Lestvic za ocenjevanje nadarjenosti GRS-S

“…kitajskem (Li idr. 2008b;Petscher in Li, 2008;Siu, 2010), južnokorejskem (Lee in Pfeiffer, 2006), turškem (Karadağ, Karabey in Pfeiffer, 2016), portoriškem (Rosado idr., 2008;, omanskem (Hassan Hemdan Mohamed idr., 2017), brazilskem (de Cássia Nakano in Siqueira, 2012) in italijanskem (Beretta, Pfeiffer in Zanetti, 2014) ter ustrezne za spletno administracijo (Yarnell in Pfeiffer, 2015). Povsod so se izkazale kot uporaben pripomoček, merske značilnosti pa sovpadajo z značilnostmi izvirnika.…”

“…Lestvice so preko različnih kulturnih okolij konfiguralno in metrično invariantne (predlagana faktorska rešitev je ustrezna v vsakem od njih), med kulturami pa tudi ne prihaja do pomembnih odstopanj v testnih rezultatih (Li idr., 2009). Poleg ocen učiteljev so raziskovalci pri kitajskem vzorcu potrdili tudi uporabnost lestvic za pridobivanje ocen staršev (Li idr., 2008a;Petscher in Li, 2008).…”

Longitudinal Measurement Invariance of Likert-Type Learning Strategy Scales

“…Moreover, though rare in the student approaches to learning field, numerous applications of multi-sample invariance testing can be found in other social science domains (e.g., Petscher & Huijun, 2008). A large number of these studies rely on data gathered using self-report Likert-type questionnaires.…”