Psychometric Properties of WISC-IV Verbal Scales: A Study of Students in China Who Are Blind

Abstract

Introduction:

Numerous studies have investigated the use of Wechsler tests on individuals with visual impairments. However, few studies have examined the utilization of the fourth edition of the Wechsler Intelligence Scale for Children (WISC-IV) for such individuals. This study investigated the reliability and validity of the WISC-IV Verbal Scales’ (i.e., the Verbal Comprehension Index [VCI] and the Working Memory Index [WMI]) scores for children with blindness in China.

Methods:

One hundred children with blindness were assessed using the Chinese version of the WISC-IV Verbal and Nonverbal Cognitive Development Scales.

Results:

The results of this study generally support the internal consistency reliability and validity of WISC-IV Verbal Scale scores for Chinese children who are blind. The split-half reliability coefficients ranged from .79 to .97, while Cronbach’s alpha or KR-20 (KR-20 is a measure of internal consistency reliability for measures with dichotomous choices) reliability coefficients ranged from .83 to .97. In terms of construct validity through exploratory factor analysis, the subtests of the WISC-IV Verbal Scale were confirmed to belong to VCI or the WMI as intended by the original structure of the WISC-IV. The convergent validity showed that the VCI and WMI scores were moderately and positively correlated with the Nonverbal Cognitive Development Scale.

Discussion:

The WISC-IV VCI and WMI were found to be valuable instruments for assessing the cognitive abilities of students in China with blindness. Finally, the limitations and significance of the study were also discussed.

Implications for practitioners:

The WISC-IV Verbal Scales may be used to assess the cognitive abilities of the numerous students in China who are blind and may be used as one of many tests to provide reference information for educational placement, educational planning, employment, and project evaluation.

Keywords

WISC-IV Verbal Scales verbal comprehension working memory children with blindness reliability and validity

The Wechsler Intelligence Scale for Children (WISC) is widely used in intelligence and cognitive ability evaluation and educational placement decision making. Furthermore, it is used in the placement of differently abled children, in clinical intervention, and in the neuropsychological assessment of children aged 6–16 years (Wechsler, 2003). Although the WISC-V test is widely used around the world, the WISC-IV is still the latest version of the Wechsler test in use in China (Ding et al., 2016), and the Chinese WISC-IV is widely used in schools, hospitals, workplaces, and other arenas (Cao & Zhu, 2011; Chen, 2016). Therefore, this study presents an investigation of the WISC-IV test. Adjustments to this version of the test have been made based on previous versions, particularly in terms of internal structure. Its structure includes four major components, namely, verbal comprehension (VC), perceptual reasoning (PR), working memory (WM), and processing speed (PS), and these four indexes together synthesize the Full-scale Intelligence Quotient (FSIQ). The WISC-IV was revised and supplemented by Zhang (2009) at the end of 2007 and has been used on a large scale in China since March 2008. The Chinese version of WISC-IV retains its original structure but also considers cultural factors, and several items have been revised to improve its suitability for children (Zhang, 2009).

Traditional cognitive tests for people with visual impairments mainly include verbal (e.g., the Wechsler Verbal Scales) and nonverbal tests, which are mainly tactile. Two main types of tactile tests exist: The first was designed specifically for people with visual impairments (e.g., Cognitive Test for the Blind; Nelson et al., 2002), and the second were adapted from tests designed for children, such as the Intelligence Test for Visually Impaired Children, nonvisual performance subtests, or the Tactile Test of Nonverbal Intelligence (Atkins et al., 2011; Duncan et al., 1989). Although scholars have developed numerous intelligence tests for individuals with visual impairments, the Wechsler tests are still among the most widely used (Atkins et al., 2011; Dial & Dial, 2010; Groenveld & Jan, 1992; Hannan, 2007; Lund et al., 2014; J. C. Miller & Skillman, 2003; Reid, 1997). J. C. Miller and Skillman (2003) argued that the Wechsler scale is still widely used for children with visual impairments because of its positive reputation, familiarity, flexibility, and accessibility. Moreover, Lund et al. (2014) also noted that the Wechsler tests are authoritative compared with other tests, and many of the other tests that are specifically designed for people with visual impairments are outdated.

When applied to people with visual impairments, the use of the whole Wechsler test is controversial; however, it is generally considered appropriate to use the verbal (excluding nonverbal) aspects of the test (Dekker, 1993; Lund et al., 2014; Nelson et al., 2002). Researchers have discussed the application of the Wechsler Verbal Scales for students who are visually impaired and have investigated the psychometric properties of the tests (Atkins et al., 2011; Baker, 1989; Dauterman & Suinn, 1966; Morash & McKerracher, 2016). Many studies have supported the satisfactory validity of the Wechsler verbal test, particularly the concurrent validity. Specifically, WISC-R IQ scores were positively correlated with Blind Linguistic Aptitude Test (BLAT) scores, with a correlation coefficient of .60 (Baker, 1989). Suinn et al. (1966) also indicated that the Wechsler verbal tests have acceptable validity coefficients: In particular, the verbal IQ could substantially predict the academic achievement, professional reputation, and work income of individuals with visual impairments. However, research on the reliability of the Wechsler test for visually impaired students is relatively limited. Lund et al. (2014) studied 15 participants with visual impairments to explore the reliability of the Wechsler verbal test scores, and the reliability was found to be low. More attention has been paid to the verbal tests of WISC-R than to WISC-IV in the past, and few scholars have discussed the use of the WISC-IV for visually impaired students.

Furthermore, previous studies on the use of the Wechsler test on children with visual impairments have been concentrated in Western Europe and the United States. Few studies have been conducted in China, which may be related to the values that are held by teachers of children with visual impairments in China. Many teachers in China do not realize the importance of ability assessment but instead carry out uniform teaching for all students according to the national unified curriculum often without assessing children’s cognitive ability or deeply considering its influence on learning (Zou, 2018). In addition, for a long time, China’s education system lacked psychoeducational assessment tools for visually impaired students, and schools for blind students were generally not familiar with such tests, thereby leading to very low, or even no, use of the Wechsler test (Xu & Mou, 2018; Zou, 2018). Thus, this study aimed to explore the psychometric properties of the WISC-IV verbal test for children with visual impairments in China. The study may help teachers understand the usability and importance of this test.

Previous studies have found several controversial issues regarding the use of the nonverbal aspects of the Wechsler tests for children with low vision (Atkins et al., 2011; Lund et al., 2014; J. C. Miller & Skillman, 2003). Thus, this study tests the reliability and validity of only the verbal aspects of the WISC-IV for children in China with visual impairments. Specifically, the study addresses the following research questions.

What are the internal consistency estimates for the WISC-IV Verbal Scales’ scores among Chinese students with visual impairments?

What are the concurrent validity estimates for the scores on the WISC-IV Verbal Scales among Chinese students with visual impairments? Is the structure of the WISC-IV Verbal Scales similar to that of the original Wechsler test for Chinese students with visual impairments?

Method

Participants

The participants attended the largest school for blind students in Southern China, which is located in Guangdong Province. We chose participants from classes for students with visual impairments only because of concerns that students with multiple disabilities would not be able to perform the test. We originally identified more than 100 students with visual impairments, but, ultimately, only 100 students between the ages of 7 and 18 years were able to take part in the test. Among this group, some were excluded from participation for two reasons. First, the students’ teachers identified some of the students as possibly having multiple disabilities, although they were not in classes with other multiply disabled students. Second, some students were unwilling to take part in the test after understanding the process. Of these 100 students, 23 were in lower grades (Grades 1–3), 34 were in middle grades (Grades 4–6), and 43 were in higher grades (Grades 7–9). A total of 73 participants were male, and 27 were female. In terms of the onset of blindness, 75 students were congenitally blind (blind at birth or before the age of 1 year), and 25 cases were adventitiously blind (blind at the age of 1 year or older) or lost their vision more than 2 years prior to the study. Regarding access to early education, 67 participants had received early education, whereas 33 had not received early education and began school in first grade. All of the students learned in braille and had not previously taken the Wechsler or any other haptic tests. In addition, all participants were native Chinese speakers. The study was approved by the university’s institutional review board. The informed consent of each child and their families was obtained prior to data collection. Table 1 shows in detail the distribution of students by the type of onset of blindness, gender, grade, and access to early childhood education.

Table 1.

Qualitative demographic information.

Demographic categories	Sample size (n)	Percentage
Type of onset of blindness
Congenital blindness	75	75
Adventitious blindness	25	25
Gender
Male	73	73
Female	27	27
Grades
Lower grades	23	23
Middle grades	34	34
Higher grades	43	43
Access to early education
Yes	67	67
No	33	33

Procedure

The study was conducted one-on-one in a quiet room at the school. Some authors of this study took part in administering the test, for which they had received thorough training: They had obtained the national certification of the Wechsler test master test certificate. Before the formal test, they also conducted some pretests with the participants, discussed the problems in the test process, and made efforts to standardize the procedure among themselves. The test was carried out in strict accordance with the standard procedures of the Wechsler test. The standard basal and ceiling rules were applied, noting the concern raised by Atkins et al. (2011) that once the test process or method is changed, the results lose verifiability and are no longer comparable to those of tests conducted under standard conditions.

At the time of the formal test, the examiners obtained the participants’ personal information, explained the purpose of the test, and reensured the participants that the results would be kept confidential, which helped them feel comfortable and relaxed. Moreover, the examiners clearly stated the test instructions before each subtest and conducted the testing in strict accordance with the test instructions. The arithmetic subtest of the Wechsler test requires a 30-second time limit for each item; thus, the examiner needed to observe the time accurately. Meanwhile, the other subtests and the Nonverbal Cognitive Development Scale had no time limits. The examiner made a note of the time at which the participants requested a break. The duration of each test was approximately 30–60 minutes. All participants completed the WISC-IV VC Index (VCI), WM Index (WMI), and the Nonverbal Cognitive Development Scale. The order of presentation of the WISC-IV and the Nonverbal Cognitive Development Scale was counterbalanced. Therefore, half of the participants completed the WISC-IV tests first, whereas the other half completed the Nonverbal Cognitive Development Scale first. The participants were encouraged to complete the tests in full. Small gifts were given to the participants after the tests were completed.

Measures

Chinese version of WISC-IV

The overall structure of the Chinese version of the WISC-IV test is similar to that of the American version. It includes four indexes, namely, VCI, PR Index (PRI), WMI, and PS Index (PSI; see Table 2; Zhang, 2009). VCI consists of four subtests, namely, similarities, vocabulary, comprehension, and information. Information is a supplementary subtest that is mainly used to measure language acquisition ability, language concept formation and assimilation, and abstract thinking related to language, analysis, and generalization. PRI likewise consists of four subtests, namely, block design, picture concepts, matrix reasoning, and picture completion. Picture completion is a supplementary subtest that measures participants’ spatial perception, visual organization, and fluid reasoning ability. Meanwhile, WMI consists of three subtests: digit span, letter–number sequencing, and arithmetic. Arithmetic is a supplementary subtest that mainly measures mental manipulation, attention, short-term and long-term memory, numerical reasoning ability and mental alertness. PSI consists of three subtests, namely, coding, symbol search, and cancellation. Cancellation is a supplementary subtest that investigates the speed at which participants process simple and regular information and the speed and accuracy of recording, attention, and writing ability. FSIQ is the most important indicator in intelligence testing as a general assessment of cognitive ability. The Chinese version of WISC-IV includes adjusted test items or subtests based on cultural and linguistic considerations. For example, the word reasoning subtest in VCI has been removed, and several pictures that are not easily recognizable by Chinese children have been adjusted in the other subtests. In addition, this version increases the difficulty of the memorization tests. The Chinese version of WISC-IV has satisfactory psychometric properties (Zhang, 2009). In addition, this test also sets norms for children with several types of disabilities, such as intellectual and learning disabilities, but not including children with visual impairments (Zhang, 2009).

Table 2.

The framework of the Chinese version of the WISC-IV test.

VCI	PRI	WMI	PSI
Similarities	Block design	Digit span	Coding
Vocabulary	Matrix reasoning	Letter–number sequencing	Symbol search
Comprehension	Picture concepts
Information*	Picture completion*	Arithmetic*	Cancellation*

Note. *Supplementary subtests.

VCI = Verbal Comprehension Index; PRI = perceptual reasoning index; WMI = working memory index; PSI = processing speed design.

Nonverbal Cognitive Development Scale

The Nonverbal Cognitive Development Scale was previously utilized to investigate the cognition ability of students with visual impairments in Taiwan (Chao-An, 1999). This tool has a total of 75 questions, all of which are multiple-choice, and the test format is similar to the Raven matrices tests. It is a nonverbal test and consists of pictures. The pictures are raised materials that can be touched and perceived. Students with visual impairments due to any visual impairment can perform the test as can sighted students (Chao-An, 1999). This scale mainly assesses nonverbal abilities and is used here as a tool to verify the concurrent validity of the Wechsler Verbal Scales’ scores. The test contains normative data not only for sighted children but also for those who are visually impaired. Chao-An71 (1999) reported a study of 418 students with visual impairments from Grades 1 to 9 who were selected to participate in creating normative samples. The total test–retest reliability coefficient was .83 for these students. In terms of validity, the correlation coefficients between the Nonverbal Cognitive Development Scale total score and each subtest of the Learning Aptitude Test for the Blind ranged from .31 to .71. The correlation coefficient of the entire test was .85 at a statistical significance level of .001 (Chao-An, 1999).

Data analysis

The participants’ descriptive statistics were calculated. The mean scores of the students who are blind were compared with the normative scores of the WISC-IV test, and the one-sample t-test was adopted to investigate whether the scores of the students who are blind differed from the norm group. Subsequently, the internal consistency reliability of the scales was calculated by using split-half reliability coefficients and the Cronbach’s alpha coefficient (or the KR-20 reliability coefficient where appropriate). Moreover, exploratory factor analysis was applied to analyze the construct validity. Regarding the justification of the exploratory factor analysis sample sizes, MacCallum et al. (1999) had pointed out that even if the sample was small (sample size less than 100), the value of commonality was high and there were fewer factors, and it was also suitable for exploratory factor analysis. The value of commonality in this study is relatively high (the mean level of commonality exceeded .7 in this study), and the factors are relatively few (only two factors in this study); therefore, exploratory factor analysis in this study is also acceptable. Finally, the Pearson product–moment correlation coefficients were used to calculate the concurrent validity. Statistical analyses were conducted using SPSS Version 18.0.

Results

Descriptive statistics

Table 3 shows that the VCI and the WMI mean composite scores of the students with blindness were lower than the norm groups. The mean scores of the students with blindness for the similarities, vocabulary, comprehension, information, letter–number sequencing, and arithmetic subtests were significantly lower than the norm group. The one-sample t-test revealed that significant differences existed between the VCI and the WMI mean composite scores of the students with blindness and the norm group mean of 100 (t = −13.15 and −2.60, p < .05). Among the differences, performance in the VCI was worse than that in the WMI. However, no significant difference existed between the mean score of the digit span subtest of students with blindness and the norm group mean of 10 (t = .13, p = .90).

Table 3.

Differences between WISC-IV Verbal Scale scores for students with blindness and norm group mean scores.

Subtests and indexes	Students with blindness		Norm group		t
Subtests and indexes	M	SD	M	SD	t
VCI	79.96	15.24	100	15	−13.15***
WMI	95.30	18.08	100	15	−2.60*
Similarities	6.66	3.15	10	3	−10.60***
Vocabulary	6.14	2.85	10	3	−13.56***
Comprehension	6.75	2.90	10	3	−11.22***
Information	6.50	2.09	10	3	−16.77***
Digit span	10.04	3.09	10	3	0.13
Letter–number sequencing	8.35	3.95	10	3	−4.18***
Arithmetic	7.91	2.83	10	3	−7.40***

Note. N = 100, WISC-IV = fourth edition of the Wechsler Intelligence Scale for Children; VCI = Verbal Comprehension Index; WMI = Working Memory Index

*p < .05. **p < .01. ***p < .001.

Reliability

The internal consistency reliability of each subtest and index score was calculated by using the split-half method (i.e., odd and even terms) and alpha coefficient (or the KR-20 reliability coefficient where appropriate). In terms of the split-half method, we applied Spearman–Brown correction to analyze the entire (equal length) test reliability coefficients (Krouse & Braden, 2011; Wechsler, 2003). Table 4 shows that the split-half reliability coefficients of each subtest and composite index were acceptable except for the comprehension subtest, which was lower than .80.

Table 4.

Internal consistency reliability of WISC-IV Verbal Scale scores.

Subtests and indexes	Split-half reliability coefficients	Cronbach’s alpha or KR-20 reliability coefficients
VCI	.94	.94
WMI	.97	.96 ^a
Similarities	.88	.83
Vocabulary	.90	.88
Comprehension	.79	.83
Information	.81	.88 ^a
Digit span	.89	.88 ^a
Letter–number sequencing	.95	.97 ^a
Arithmetic	.93	.91 ^a

Note. N = 100, WISC-IV = fourth edition of the Wechsler Intelligence Scale for Children; VCI = Verbal Comprehension Index; WMI = Working Memory Index.

^a Indicates KR-20 reliability coefficients and all others are Cronbach’s alpha coefficients.

Meanwhile, the alpha coefficient and KR-20 reliability coefficient were used to analyze the internal consistency. Dichotomously scored items were evaluated using KR-20; thus, the index “WMI” and the subtests “information,” “digit span,” “letter–number sequencing,” and “arithmetic” used KR-20 reliability coefficients, while the others, namely, the index “VCI” and the subtests “similarities,” “vocabulary,” and “comprehension,” used alpha coefficients. Table 4 shows that the alpha and KR-20 reliability coefficients of each subtest and composite index were also acceptable, the reliability coefficients ranging from .83 to .96.

Construct validity

We used the data from the exploratory factor analysis of the seven WISC-IV verbal subtests to identify the possible latent variables. This procedure is consistent with the analysis of standardized sample data reported in the WISC-IV technical manual (see Wechsler, 2003). The principal axis factoring method was used to extract factors, and the oblique rotation method was used to consider the correlation between factors. This study also implemented two different factor analysis. The first exploratory factor analysis uses both core and the supplementary subtests. The second one only includes the core subtests. For the first, as shown in Table 5, the Kaiser–Meyer–Olkin and Bartlett tests passed the significance test. The value for Kaiser–Meyer–Olkin was .83, and Bartlett’s sphericity test was significant (p < .001). We extracted two factors; specifically, the eigenvalue of each factor was greater than 1, whereas the factor load of each item was greater than .65. Factor 1 included four items (similarities, vocabulary, comprehension, and information) and explained 54.11% of the variance. Meanwhile, Factor 2 included three items (digit span, letter–number sequencing, and arithmetic) and explained 17.54% of the variance. The two factors combined explained 71.65% of the total variance. For the second, the results are similar to the first one, and the results are shown in Table 6. The Kaiser–Meyer–Olkin and Bartlett tests passed the significance test: The value for Kaiser–Meyer–Olkin was .71, and Bartlett’s sphericity test was significant (p < .001). Factor 1 included three items (similarities, vocabulary, and comprehension) and explained 56.81% of the variance. Meanwhile, Factor 2 included two items (digit span and letter–number sequencing) and explained 23.50% of the variance. The two factors combined explained 80.31% of the total variance. The results confirmed that the subtests belonged to the VCI or the WMI scores as intended by the original WISC-IV structure.

Table 5.

Exploratory factor analysis results of WISC-IV core and supplemental verbal subtests.

WISC-IV subtests	Factor loadings
WISC-IV subtests	Factor 1	Factor 2
Similarities	0.87
Vocabulary	0.90
Comprehension	0.80
Information	0.74
Digit span		0.92
Letter–number sequencing		0.84
Arithmetic		0.68
% of variance accounted for	54.11	17.54
Cumulative % of variance	54.11	71.65
Kaiser–Meyer–Olkin test	0.827
Bartlett’s test of sphericity	Chi-square = 322.792, df = 21, significance < .001

Note. N = 100, WISC-IV = fourth edition of the Wechsler Intelligence Scale for Children.

Table 6.

Exploratory factor analysis results of WISC-IV core verbal subtests.

WISC-IV subtests	Factor loadings
WISC-IV subtests	Factor 1	Factor 2
Similarities	0.88
Vocabulary	0.91
Comprehension	0.84
Digit span		0.93
Letter–number sequencing		0.86
% of variance accounted for	56.81	23.50
Cumulative % of variance	56.81	80.31
Kaiser–Meyer–Olkin test	0.707
Bartlett’s test of sphericity	Chi-square = 219.204, df = 10, significance < .001

Note. N = 100, WISC-IV = fourth edition of the Wechsler Intelligence Scale for Children.

Concurrent validity

We correlated the WISC-IV Verbal Scale scores with the Nonverbal Cognitive Development Scale scores to investigate the concurrent validity of the former. This analysis was performed using the Pearson product–moment correlation coefficients. Table 7 shows that there was a significant positive relationship between the Wechsler VCI and Nonverbal Cognitive Development Scale scores (r = 0.64, p < .01). Similarly, a significant positive correlation was observed between the Wechsler WMI and Nonverbal Cognitive Development Scale scores (r = 0.68, p < .01). Therefore, the Wechsler test scores for VCI, WMI, and the Nonverbal Cognitive Development Scale were correlated with one another. The correlation between VCI and WMI was also highly significant (r = 0.69, p < .01), which indicated that the WISC-IV Verbal Scales had an acceptable concurrent validity.

Table 7.

Correlation among Nonverbal Cognitive Development Scale, WMI, and VCI.

Subtests indexes	VCI	WMI	Nonverbal Cognitive Development Scale
VCI	1.00
WMI	.69^**	1.00
Nonverbal Cognitive Development Scale	.64^**	.68^**	1.00

Note: N = 100, WISC-IV = fourth edition of the Wechsler Intelligence Scale for Children; VCI = Verbal Comprehension Index; WMI = Working Memory Index.

*p < .05. **p < .01. ***p < .001.

Discussion

This study assessed the reliability and validity of the Chinese version of the WISC-IV Verbal Scale scores in 100 students who are blind. The VCI and WMI mean composite scores for the students we studied were 79.95 and 95.30, respectively. The students’ scores on the WISC-IV Verbal Scales were significantly different from those of the norm. Students who are blind are likely to exhibit a series of unique verbal cognitive abilities that differ from those of sighted individuals (see Bruce et al., 2018; J. C. Miller et al., 2007; L. R. Miller, 1977).

There are some differences between the results of this study and those of earlier studies. Previous studies have shown that people with visual impairments perform similarly to sighted individuals on the Wechsler Verbal Scales (MacCluskie et al., 1998; Vander Kolk, 1977, 1982). The poor performance of students with blindness in VCI may be due to the visual experience involved in conceptual comprehension. For example, a question in the vocabulary subtest asks students to explain the concept of “transparency (透明).” Students with blindness may have never seen transparent objects and, thus, have difficulty understanding these types of questions. Morash and McKerracher (2016) discussed this phenomenon and stated that because these tests are mainly aimed at sighted children, some questions do not consider the life experiences of visually impaired students, making the tests unfair. In addition, these differences in scores may also be attributable to the differences among the participants who were evaluated. Although the participants in this study had no other severe disabilities, they came from a school for blind students (students with visual impairments with higher cognitive abilities attend mainstream schools in China). Dekker (1993) also pointed out that the cognitive ability of students with visual impairments who attend mainstream schools is better than that of those in special schools. Moreover, most participants assessed with the Wechsler Verbal Scales in the past were adults, such as in the studies of Tillman and Osborne (1969), Vander Kolk (1977, 1982), and MacCluskie et al. (1998). Adults with visual impairments have acquired more life experiences than children, which may compensate for their cognitive difficulties.

Meanwhile, the results of this study were consistent with previous studies in that the WM ability of children with visual impairments was better than their VC. Previous studies have shown that people with visual impairments scored lower on Wechsler’s VC compared with those with typical vision but higher in WM than those who are sighted (Dekker, 1993; Smits & Mommers, 1976; Tillman & Osborne, 1969). Students who are blind often perform well in the WMI, which may be influenced by their daily listening practices. Listening and memorizing are crucial for students who are blind since they receive a large amount of verbal information every day and often need to memorize the information efficiently (Atkins et al., 2011). Classroom learning also entails the auditory and tactile reception of information; thus, students who are blind may be comparable to sighted students in terms of memorization. Atkins et al. (2011) noted that individuals with visual impairments rely more heavily on WM to process information compared with sighted people, who can rely on their visual experiences to understand and process information differently.

Further, the WISC-IV VCI and WMI scores showed acceptable concurrent validity, which was consistent with previous studies. The results of the current study were similar to those of Baker (1989), who suggested that the Wechsler Verbal Scales scores were significantly correlated with BLAT scores. Furthermore, Reid (2002) also noted that the partial correlation between the adapted Kohs block design test and the WAIS-R score was significant in all participant groups. Meanwhile, the present study revealed acceptable construct validity of the Wechsler Verbal Scales scores. Through exploratory factor analysis and confirmatory factor analysis, the seven subtests of the Wechsler Verbal Scales were revealed to exhibit the same structure as the original scale. Four subtests (i.e., similarities, vocabulary, comprehension, and information) belonged to VCI, and three (i.e., digit span, letter–number sequencing, and arithmetic) belonged to WMI.

In addition, the reliability coefficients of this study were acceptable. The split-half reliability coefficient ranged from .79 to .97, whereas the alpha coefficient (or KR-20 reliability coefficient where appropriate) ranged from .83 to .97. However, the study results were inconsistent with those of Morash and McKerracher (2016), who found the reliability of the vocabulary subtest of the Wechsler Abbreviated Scale of Intelligence (WASI) scores was low and that the alpha and split-half reliability coefficients of the Verbal subtest scores were only .85 and .72, respectively. Two reasons may explain this discrepancy. First, Morash and McKerracher utilized the WASI test, whereas this study used the WISC-IV Verbal Scales. Therefore, the different tests and questions may have led to different results. Second, the sample of that study was small (15 individuals), and the participants were adolescents. In addition, the present study likewise found differences in reliability within the test. The reliability of the VCI scores was not as high as those of WMI, especially in the subtests. For example, the split-half reliability coefficient of the comprehension subtest did not exceed .8. This subtest may be partly dependent on visual factors, making it unfair for students who are blind and reducing the stability of the test results. Morash and McKerracher (2016) likewise concluded that the Wechsler VCI was unfair to visually impaired students because of the visual factors, which may reduce the reliability of the test scores.

Some important limitations of the present study must be acknowledged. The study did not analyze the stability of the data at different time periods; that is, it did not analyze the test–retest reliability coefficient. Future studies could explore this issue. Second, the sampling had some limitations. At present, children with visual impairments in China mainly study in schools for blind students. Thus, the participants were all students with visual impairments who attended schools exclusively for blind people. Students with visual impairments who are studying at mainstream schools are widely scattered and, thus, difficult to find. Therefore, students with visual impairments in mainstream schools were not selected for this study. Moreover, the sample did not include all types of visually impaired groups, such as children with low vision and those with multiple disabilities. Further, there was an imbalance in the proportion of sample distribution, which may limit the situation in which the research conclusion can be applied (the proportion of children who are congenitally blind in this study was 75% and that of children who are adventitiously blind was 25%). The cognitive characteristics of children who are congenitally blind and those who are adventitiously blind may be different; because the sample size of this study was not large enough, however, the psychometric data of these two groups were not analyzed. Compared with adventitious blindness, the conclusion of this study may be more convincing when applied to children with congenital blindness. Therefore, we hope that future research can select a representative sample of students with visual impairments to take the Wechsler test. Finally, when using WISC-IV to test children with visual impairments, at present, only the verbal test can be used, and the application of the whole test to children with visual impairments cannot be explored. It is unfortunate that the entire test cannot be used, and we hope that future research can explore the application of the whole test to children with low vision.

Nevertheless, the study’s findings are of considerable importance. Previous studies on the WISC-IV tests have reported the intelligence score profiles of people with various types of disabilities (Krouse & Braden, 2011; Wechsler, 2003); however, no research has yet investigated the population with visual impairments. Thus, this study provided information with psychometric validity. WISC-IV Verbal Scales could be used to assess the cognitive ability of a vast number of students in China who are blind. Furthermore, although there were other factors that affected students’ educational placement and life outcomes, the results of WISC-IV Verbal Scales may be used as an important reference information. According to the Second National Sample Survey of Persons with Disabilities in China, the total number of people with visual impairments was 12.33 million, and the total number of persons with visual impairments and multiple disabilities was 4.69 million (Yang & Wang, 2011). Therefore, this study is highly relevant to the education and the employment of people in China who are visually impaired.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by grants from The Institute for Advanced Study of educational development in Guangdong HongKong Macao Great Bay Area, SCNU (Grant no. GD17YJY02), Research on the Development Status of Integrated Education in Guangdong Province(grant no. GDJY-2019-T-a03), and Research Project on Curriculum Reform of Basic Education in Guangdong Province (grant no. 2015JJKGYJ045).

References

Atkins

Cobb

Keil

Home

Wilkins

S. M.

(2011). Assessing the ability of blind and partially sighted people: Are psychometric tests fair. RNIB Centre for Accessible Information.

Baker

C. P

. (1989). The relationship between blind learning aptitude test scores and braille reading speed and comprehension of children who are blind [PhD thesis]. Texas Tech University.

Bruce

S. M.

Luckner

J. L.

Ferrell

K. A.

(2018). Assessment of students with sensory disabilities: Evidence-based practices. Assessment for Effective Intervention, 43(2), 79–89.

Cao

H. J.

Zhu

(2011). The WISC administrated to children with special needs: Challenges, reforms and future development. Chinese Journal of Special Education, 6, 17–24.

Chao-An

. (1999). A study on the pattern cognitive abilities of 6- to 15-year-old students with visual impairment. Bulletin of Special Education, 17, 139–162.

Chen

(2016). Application of WISC-IV Chinese version in special need children, questions and suggestions. Journal of Modern Special Education, 6, 32–35.

Dauterman

W. L.

Suinn

R. M

. (1966). Stanford-Ohwaki-Kohs tactile block design intelligence test for the blind part one: Final report.

Dekker

(1993). Visually impaired children and haptic intelligence test scores: Intelligence test for visually impaired children (ITVIC). Developmental Medicine & Child Neurology, 35, 478–489.

Dial

J. G.

Dial

C. L.

(2010). Assessing and intervening with visually impaired children and adolescents. In Miller

D. C.

(Ed.), Best practices in school neuropsychology: Guidelines for effective practice, assessment, and evidence-based intervention (pp. 465–479). Wiley.

10.

Ding

Xiao

Fan

C. H.

Zhang

(2016). New advances in intelligence testing: An introduction to the Wechsler Intelligence Scale for Children (fifth edition). Chinese Journal of Special Education, 7, 18–25.

11.

Duncan

Weidel

Prickett

Vernon

Hollingsworth-Hodges

. (1989). The tactile TONI: A possible new performance test for blind adults. Journal of Visual Impairment & Blindness, 83, 510–511.

12.

Groenveld

Jan

J. E.

(1992). Intelligence profiles of low vision and blind children. Journal of Visual Impairment & Blindness, 86(1), 68–71.

13.

Hannan

C. K.

(2007). Exploring assessment processes in specialized schools for students who are visually impaired. Journal of Visual Impairment & Blindness, 101(2), 69–79.

14.

Krouse

H. E.

Braden

J. P.

(2011). The reliability and validity of WISC-IV scores with deaf and hard-of-hearing children. Journal of Psychoeducational Assessment, 29(3), 238–248.

15.

Lund

E. M.

Miller

K. B.

Ganz

J. B.

(2014). Access to assessment? Legal and practical issues regarding psychoeducational assessment in children with sensory disabilities. Journal of Disability Policy Studies, 25(3), 135–145.

16.

MacCallum

R. C.

Widaman

K. F.

Zhang

Hong

(1999). Sample size in factor analysis. Psychological Methods, 4, 84–99.

17.

Maccluskie

K. C.

Tunick

R. H.

Dial

J. G.

Paul

D. S.

(1998). The role of vision in the development of abstraction ability. Journal of Visual Impairment & Blindness, 92(3), 189–199.

18.

Miller

J. C.

Skillman

G. D.

(2003). Assessors’ satisfaction with measures of cognitive ability applied to persons with visual impairments. Journal of Visual Impairment & Blindness, 97, 769–774.

19.

Miller

J. C.

Skillman

G. D.

Benedetto

J. M.

Holtz

A. M.

Nassif

C. L.

Weber

A. D.

(2007). A three-dimensional haptic matrix test of nonverbal reasoning. Journal of Visual Impairment & Blindness, 101(9), 557–570.

20.

Miller

L. R

. (1977). Abilities structure of congenitally blind persons: A factor analysis. Journal of Visual Impairment & Blindness, 71,145–153.

21.

Morash

V. S.

Mckerracher

(2016). Low reliability of sighted-normed verbal assessment scores when administered to children with visual impairments. Psychological Assessment, 29(3), 343–348.

22.

Nelson

P. A.

Dial

J. G.

Joyce

(2002). Validation of the cognitive test for the blind as an assessment of intellectual functioning. Rehabilitation Psychology, 47, 184–193.

23.

Reid

J. M. V.

(1997). Standardized ability testing for vocational rehabilitation. Journal of Visual Impairment & Blindness, 91(6), 546–554.

24.

Reid

J. M. V.

(2002). Testing the nonverbal intelligence of working-age visually impaired adults: Evaluation of the adapted Kohs block design test. Journal of Visual Impairment& Blindness, 96, 585–595.

25.

Smits

B. W. G. M.

Mommers

M. J. C

. (1976). Differences between blind and sighted children on WISC verbal subtests. New Outlook for the Blind, 70(6), 240–246.

26.

Suinn

Dauterman

Shapiro

(1966). The standard Ohwaki-Kohs tactile block design intelligence test for the blind. New Outlook for the Blind, 60, 77–79.

27.

Tillman

H. M.

Osborne

R. T.

(1969). The performance of blind and sighted children on the Wechsler Intelligence Scale for Children: Interaction effects. Educate Visually Handicapped, 1, 1–4.

28.

Vander Kolk

C. J.

(1977). Intelligence testing for visually impaired persons. Journal of Visual Impairment& Blindness, 71(4), 158–163.

29.

Vander Kolk

C. J.

(1982). A comparison of intelligence test score patterns between visually impaired subgroups and the sighted. Rehabilitation Psychology, 27(2), 115–120.

30.

Wechsler

(2003). WISC-IV technical and interpretive manual. Psychological Corporation.

31.

Mou

. (2018). Intelligence tests in visually-impaired persons: A review and prospects. Chinese Journal of Special Education, 2, 22–26.

32.

Yang

Wang

. (2011). Present status of persons with visual impairment in China. Disability Studies, 1, 29–32.

33.

Zhang

(2009). The revision of WISC-IV Chinese version. Psychological Science, 32(5), 1177–1179.

34.

Zou

(2018). An action research based on an individualized education mode for visually-impaired students. Chinese Journal of Special Education, 5, 29–32.