help

Comparison of the WAIS-III and WISC-IV in 16-Year-Old Special Education Students Shirley Gordon*, Simon Duff�, Terry Davidson� and Simon Whitaker§

*Glen Acre House Child and Family Service, Lindley, Huddersfield, UK; �Division of Clinical Psychology, University of Liverpool,

Quadrangle, Liverpool, UK; �CASPD, Salomons, David Salomans Estate, Southborough, Tunbridge Wells, Kent, UK; §The Learning Disability Research Unit, University of Huddersfield, Queensgate, Huddersfield, UK

Accepted for publication 20 November 2008

Background Previous research with earlier versions of

the WISC and WAIS has demonstrated that when

administered to people who have intellectual disabili-

ties, the WAIS produced higher IQ scores than the

WISC. The aim of this study was to examine whether

these differences still exist. A comparison of the

Wechsler Adult Intelligence Scale – Third Edition

(WAIS-III) with the Wechsler Intelligence Scale for

Children – Fourth Edition (WISC-IV) was conducted

with individuals who were 16 years old and receiving

special education.

Materials and Methods All participants completed the

WAIS-III (UK) and WISC-IV (UK). The order of admin-

istration was counterbalanced; the mean Full Scale IQ

and Index scores on the WAIS-III and WISC-IV were

compared.

Results The WAIS-III mean Full Scale IQ was 11.82

points higher than the mean Full Scale IQ score on the

WISC-IV. Significant differences were also found

between the Verbal Comprehension Index, Perceptual

Reasoning ⁄ Organization Index and Processing Speed

Index on the WAIS-III and WISC-IV, all with the WAIS-

III scoring higher.

Conclusions The findings suggest that the WAIS-III pro-

duces higher scores than the WISC-IV in people with

intellectual disabilities. This has implications for defini-

tions of intellectual disability and suggests that Psychol-

ogists should be cautious when interpreting and

reporting IQ scores on the WAIS-III and WISC-IV.

Keywords: intellectual disability diagnosis, intelligence

test, WAIS-III, WISC-IV

Introduction

Definitions of intellectual disabilities tend to include

three core criteria: significant impairment of intellectual

functioning, significant impairment of adaptive ⁄ social

functioning and age of onset before adulthood (British

Psychological Society, 2000). A significant impairment in

intellectual functioning is generally regarded as an IQ

below 70, assessed on a well-standardized and reliable

assessment on intellectual functioning. The Wechsler

Adult Intelligence Scale – Third Edition (WAIS-III,

Wechsler 1997; UK edition, Wechsler 1998a,b) and the

Wechsler Intelligence Scale for Children – Fourth Edi-

tion (WISC-IV, Wechsler 2003; UK edition, Wechsler

2004) are probably the most widely used tests of intelli-

gence in children and adults. Although these tests have

been found to be highly reliable and valid when used

with individuals functioning within the normal range of

intelligence (Wechsler 1997, 2003) there have been some

concerns about how accurately these tests measure intel-

lectual functioning at the extreme ends of the range

(Whitaker 2008; Whitaker & Wood 2008). Previous

research (Nagle & Lazarus 1979; Spitz 1988) has sug-

gested that earlier versions of the WAIS produced

higher scores than earlier versions of the WISC, in indi-

viduals who were functioning in the low IQ range.

Nagle & Lazarus (1979) compared the WAIS with the

WISC-R in a group of 30 participants who received spe-

cial education classes for children with an IQ between

55 and 75. The WAIS yielded significantly higher IQ

scores, scoring 13 points higher on the Full Scale IQ

than the WISC-R. Significant correlations between the

corresponding subtests and IQ scales were found, sug-

gesting that the content was related, despite the non-

equivalence of scores. Spitz (1988) investigated whether

the difference between the WAIS-R and WISC-R was

Journal of Applied Research in Intellectual Disabilities 2010, 23, 197–200

� 2010 Blackwell Publishing Ltd 10.1111/j.1468-3148.2009.00538.x

Published for the British Institute of Learning Disabilities

consistent across different levels of IQ. He looked at

existing data from seven previously published studies

and combined this with the data from two new studies,

giving a total of 236 participants overall. He found that

‘there was an inverse relationship between the size of

the WISC-R ⁄ WAIS-R score disparity and the level of the

WISC-R IQ, such that the lower the WISC-R IQ, the

higher the increment in IQ derived from the WAIS-R’

(Spitz 1988, p. 377). However, there has been a lack of

recent research to replicate these findings using the most

recent versions of the tests and also a lack of research

using a sample of participants from the UK. Whitaker &

Wood (2008) and Whitaker (2008) examined the criteria

for gaining a scaled score of two, as outlined in the

manuals for the WISC-IV (UK), the WISC-III (UK) and

the WAIS-III (UK). On the basis of this they suggested

that the WAIS-III (UK) would produce significantly

higher IQs in the low ability range than either the

WISC-IIII or WISC-IV. An unpublished dissertation

(Bresnahan 2008), reported an American study where

the WISC-IV and WAIS-III were given to 20 school chil-

dren with IQs less than 75. The results showed that the

mean Full Scale IQ score on the WISC-IV was 14 points

lower than the mean Full Scale IQ score on the WAIS-

III. Therefore, there is a need to look at the relationship

between the UK versions of these tests.

Methods

Participants

Participants were selected from four special education

schools within one Local Authority area. The inclusion

criteria for participants were that they had to be able

to speak English, be 16 years of age (the age at which

both the WISC-IV and the WAIS-III can be given) and

attend a special school. Adolescents who had severe

or profound intellectual disabilities and would not be

able to complete the IQ assessments were excluded

from the study. A total of 17 participants were

recruited to the study, which included eight female

and nine male students. The mean age of participants

at the time of the first testing was 16 years 2 months

with a range between 16 years 0 months and 16 years

6 months.

Design and procedure

The research proposal was approved by Division of

Clinical Psychology Research Committee at Liverpool

University, who also considered the ethical issues. The

head teachers of each of the schools involved gave writ-

ten informed consent to take part in the study.

The WAIS-III UK Edition (Wechsler 1998a,b) and The

WISC-IV UK Edition (Wechsler 2004) were administered

to all participants. The standardized instructions were fol-

lowed for both tests. All the compulsory subtests were

administered alongside the optional subtests if they

appeared on both the WAIS-III and the WISC-IV. A total

of 13 WAIS-III subtests and 13 WISC-IV subtests were

administered. There were four participants who did not

complete the optional subtests on the WISC-IV (Picture

Completion, Information and Arithmetic); however, anal-

ysis using these subtests is not reported in this paper.

A repeated measures design was used, with the

administration order of the tests counterbalanced to

reduce practice effects. Participants were randomly

assigned to groups. Group 1 consisted of nine partici-

pants who completed the WAIS-III first followed by the

WISC-IV. Group 2 consisted of eight participants who

completed the WISC-IV first and then the WAIS-III. The

testing sessions took place during school hours and in a

suitable quiet room within each school. The number of

sessions needed to complete the two tests ranged

between two and three. The mean test–retest interval

was 2 weeks and ranged between 1 and 10 weeks. A

total of 12 participants had their tests administered and

scored by the principal researcher, with an assistant psy-

chologist administering and scoring the tests for five

participants. There were no systematic differences

between the two examiners. Paired samples t-tests were

used to compare the mean results of Group 1 and

Group 2.

Results

Table 1 provides a summary of the means and standard

deviations of the Full Scale IQ and Index scores. The

mean Full Scale IQ score on the WAIS-III was 64.82,

(SD = 8.51) and 53.00 (SD = 10.08) on the WISC-IV,

11.82 points lower. A paired samples t-test was used to

determine the significance of difference between the IQ

and Index scores on the WAIS-III and the WISC-IV. The

results were that the WAIS-III yielded significantly

higher scores on the Full Scale IQ [t(16) = 13.02,

P < 0.001], Verbal Comprehension Index [t(16) = 7.98,

P < 0.001], Perceptual Organization ⁄ Perceptual Reason-

ing Index [t(16) = 3.68, P < 0.002] and Processing Speed

Index [t(16) = 4.86, P < 0.001] than on the WISC-IV.

There was no significant difference between the WAIS-

III and WISC-IV scores on the Working Memory Index

[t(16) = 1.93, P < 0.071].

198 Journal of Applied Research in Intellectual Disabilities

� 2010 Blackwell Publishing Ltd, 23, 197–200

Significant correlations were found between the

WAIS-III and WISC-IV Full Scale IQ scores (r = 0.93,

P < 0.001), Verbal Comprehension Index scores (r = 0.77,

P < 0.001), Perceptual Organization ⁄ Reasoning Index

scores (r = 0.76, P < 0.001), Processing Speed

Index scores (r = 0.88, P < 0.001) and Working Memory

Index scores (r = 0.67, P < 0.005), which suggests that

the content of the IQ and Index scores on the two tests

was strongly related.

The differences between participant’s Full Scale IQ

scores on the WAIS-III and WISC-IV ranged between 5

and 20 points and all participants had a higher score on

the WAIS-III. There was a negative correlation

(r = )0.57, P < 0.05) between these differences and the

WISC-IV Full Scale IQ, suggesting a greater disparity in

Full Scale IQ between the two tests at lower IQ levels. It

is also notable that four participants’ Full Scale IQ scores

on the WISC-IV were 40, which was also the modal

score for this test. This is also the lowest Full Scale IQ

obtainable on the WISC-IV, which suggests the possibil-

ity of a floor effect. The lowest Full Scale IQ score

obtained on the WAIS-III was 49, which was well above

the test’s lowest obtainable score of 45. With regard to

the diagnosis of intellectual disabilities, using the WISC-

IV only one participant had a Full Scale IQ of 70 or

greater; however, on the WAIS-III six participants had

Full Scale IQ scores of 70 or more.

Discussion

The WAIS-III produced a significantly higher mean Full

Scale IQ score than the WISC-IV, in a sample of 16 year

olds who attended special school. The difference

between the mean Full Scale IQ score on the WAIS-III

and the WISC-IV was 11.82 points, with the WAIS-III

scoring higher. All the participants scored lower on the

WISC-IV than the WAIS-III, the smallest difference

between any participant’s scores was 5 points. The

Index scores, with the exception of Working Memory,

were also significant higher on the WAIS-III, with the

differences ranging between 9.50 and 12.58 points.

Although this result was obtained with a small sample,

it is consistent with the predictions made by Whitaker

(2008), a similar unpublished study by Bresnahan (2008)

in the US and studies comparing earlier versions of the

WISC and WAIS.

Part of this almost 12 point difference is likely to be

due the Flynn Effect, which is the trend for the intellec-

tual ability of the population as a whole, including peo-

ple with low intellectual ability, to increase from one

generation to the next (Flynn 2006). Flynn has estimated

this growth in ability to be about 0.3 of an IQ point per

year. Since the WISC-IV was standardized 6 years after

the WAIS-III, it would be expected to score two points

less than the WAIS-III. Therefore, there is likely to be a

10 points difference between the two assessments in the

low IQ range that is due to factors other than the Flynn

Effect. It is not clear why the remaining 10 points differ-

ence in scores occurs and further research is clearly

called for.

The large difference in scores obtained by special edu-

cation students on the WAIS-III and WISC-IV, which

has been found in this study, clearly has implications

for professionals working with people who have

intellectual disabilities. Currently the WAIS-III and

WISC-IV are both used to measure an individual’s intel-

lectual ability, to determine whether they meet the diag-

nostic criteria of having an IQ less that 70 for a

diagnosis of intellectual disability. Having a diagnosis of

intellectual disability can influence many decisions

Table 1 Comparison of WAIS-III and

WISC-IV means, standard deviations (SD),

t scores and correlation coefficients (r) for

the Full Scale IQ and Index scores

IQ scale ⁄ Index

score

WAIS-III WISC-IV

Diff t RM SD M SD

FS IQ 64.82 8.51 53 10.08 11.82 13.02*** 0.93***

VCI 67.59 7.62 55.76 9.50 11.83 7.97*** 0.77***

PRI ⁄ POI 70.29 10.59 62.88 12.58 7.41 3.68** 0.76***

WMI 64.35 9.21 60.71 10.09 3.65 1.93 0.68**

PSI 71.41 10.38 64.82 11.89 6.59 4.86*** 0.88***

*P < 0.05, **P < 0.005, ***P < 0.001.

WAIS-III, Wechsler Adult Intelligence Scale – Third Edition; WISC-IV, Wechsler Intel-

ligence Scale for Children – Fourth Edition; FS, Full Scale; VCI, Verbal Comprehension

Index; PRI, Perceptual Reasoning Index; POI, Perceptual Organization Index; WMI,

Working Memory Index; PSI, Processing Speed Index.

Journal of Applied Research in Intellectual Disabilities 199

� 2010 Blackwell Publishing Ltd, 23, 197–200

about an individual’s life, including parenting assess-

ments, capacity to consent, access to services and access

to educational courses. If the tests are producing differ-

ent scores, and it is unclear which test is producing the

most accurate results, then important decisions are

being made on the basis of inaccurate information.

Therefore, professionals working with people who have

intellectual disabilities need to exercise caution when

interpreting IQ scores on the WAIS-III and WISC-IV,

particularly when the IQ score is in the mild or border-

line ability range.

This study has provided the first empirical compari-

son of the WAIS-III and WISC-IV in people with intel-

lectual disabilities in the UK. However, the study is

limited by the use of a small sample of adolescents, who

were all aged 16 and from a single area of the UK.

There was also a relatively a large range of test–retest

intervals. Therefore, further investigations are needed

using larger samples in other countries to add support

to these findings. Further research is also needed to

investigate the reasons for these differences. The results

of such research could help to improve the process of

standardization of future versions of the Wechsler tests.

Acknowledgments

Thank you to all the schools and young people who

gave their time to participate in the study.

Correspondence

Any correspondence should be directed to Shirley

Gordon, Clinical Psychologist, Glen Acre House Child

and Family Service, 21 Acre House Avenue, Lindley,

Huddersfield, HD3 3BB, 07901 503898, UK (e-mail:

[email protected]).

References

Bresnahan J. A. (2008) A preliminary study of WISC-iv and WAIS-

III IQ scores for students with extremely low cognitive functioning.

Dissertation Abstracts International: Section B: The Science

and Engineering. Fairleigh Dickinson University, Teaneck,

NJ, USA.

British Psychological Society (2000) Learning Disability: Defini-

tions and Contexts. British Psychological Society, Leicester.

Flynn J. R. (2006) Tethering the elephant: capital cases, IQ and

the Flynn Effect. Psychology, Public Policy and Law 12, 170–189.

Nagle R. J. & Lazarus S. C. (1979) The comparability of the

WISC-R and WAIS-R among 16 year-old EMR children. Jour-

nal of School Psychology 17, 362–367.

Spitz H. H. (1988) Inverse relationship between the WISC-

R ⁄ WAIS-R score disparity and IQ level in the lower range of

intelligence. American Journal on Mental Retardation 92, 376–

378.

Wechsler D. (1997) WAIS-III, WMS-III: Technical and Interpretive

Manual. The Psychological Corporation, San Antonio, TX.

Wechsler D. (1998a) Wechsler Adult Intelligence Scale – Third Edi-

tion, UK edition. The Psychological Corporation, London.

Wechsler D. (1998b) Wechsler Adult Intelligence Scale – Third UK

Edition: Administrative and Scoring Manual. The Psychological

Corporation, London.

Wechsler D. (2003) Wechsler Intelligence Scale for Children –

Fourth Edition: Technical and Interpretive Manual. The Psycho-

logical Corporation, San Antonio, TX.

Wechsler D. (2004) Wechsler Intelligence Scale for Children –

Fourth UK Edition: Administrative and Scoring Manual. The Psy-

chological Corporation, London.

Whitaker S. (2008) WISC-IV and low IQ: review and compari-

son with the WAIS-III. Educational Psychology in Practice 24,

129–137.

Whitaker S. & Wood C. (2008) The distribution of scaled scores

and possible floor effects in the WISC-III and WAIS-III. Jour-

nal of Applied Research in Intellectual Disabilities 21, 136–141.

200 Journal of Applied Research in Intellectual Disabilities

� 2010 Blackwell Publishing Ltd, 23, 197–200

Copyright of Journal of Applied Research in Intellectual Disabilities is the property of Blackwell Publishing

Limited and its content may not be copied or emailed to multiple sites or posted to a listserv without the

copyright holder's express written permission. However, users may print, download, or email articles for

individual use.

Comparison of the WAIS-III and WISC-IV in 16-Year-Old Special Education Students Shirley Gordon*, Simon Duff�, Terry Davidson� and Simon Whitaker§

*Glen Acre House Child and Family Service, Lindley, Huddersfield, UK; �Division of Clinical Psychology, University of Liverpool,

Quadrangle, Liverpool, UK; �CASPD, Salomons, David Salomans Estate, Southborough, Tunbridge Wells, Kent, UK; §The Learning Disability Research Unit, University of Huddersfield, Queensgate, Huddersfield, UK

Accepted for publication 20 November 2008

Background Previous research with earlier versions of

the WISC and WAIS has demonstrated that when

administered to people who have intellectual disabili-

ties, the WAIS produced higher IQ scores than the

WISC. The aim of this study was to examine whether

these differences still exist. A comparison of the

Wechsler Adult Intelligence Scale – Third Edition

(WAIS-III) with the Wechsler Intelligence Scale for

Children – Fourth Edition (WISC-IV) was conducted

with individuals who were 16 years old and receiving

special education.

Materials and Methods All participants completed the

WAIS-III (UK) and WISC-IV (UK). The order of admin-

istration was counterbalanced; the mean Full Scale IQ

and Index scores on the WAIS-III and WISC-IV were

compared.

Results The WAIS-III mean Full Scale IQ was 11.82

points higher than the mean Full Scale IQ score on the

WISC-IV. Significant differences were also found

between the Verbal Comprehension Index, Perceptual

Reasoning ⁄ Organization Index and Processing Speed

Index on the WAIS-III and WISC-IV, all with the WAIS-

III scoring higher.

Conclusions The findings suggest that the WAIS-III pro-

duces higher scores than the WISC-IV in people with

intellectual disabilities. This has implications for defini-

tions of intellectual disability and suggests that Psychol-

ogists should be cautious when interpreting and

reporting IQ scores on the WAIS-III and WISC-IV.

Keywords: intellectual disability diagnosis, intelligence

test, WAIS-III, WISC-IV

Introduction

Definitions of intellectual disabilities tend to include

three core criteria: significant impairment of intellectual

functioning, significant impairment of adaptive ⁄ social

functioning and age of onset before adulthood (British

Psychological Society, 2000). A significant impairment in

intellectual functioning is generally regarded as an IQ

below 70, assessed on a well-standardized and reliable

assessment on intellectual functioning. The Wechsler

Adult Intelligence Scale – Third Edition (WAIS-III,

Wechsler 1997; UK edition, Wechsler 1998a,b) and the

Wechsler Intelligence Scale for Children – Fourth Edi-

tion (WISC-IV, Wechsler 2003; UK edition, Wechsler

2004) are probably the most widely used tests of intelli-

gence in children and adults. Although these tests have

been found to be highly reliable and valid when used

with individuals functioning within the normal range of

intelligence (Wechsler 1997, 2003) there have been some

concerns about how accurately these tests measure intel-

lectual functioning at the extreme ends of the range

(Whitaker 2008; Whitaker & Wood 2008). Previous

research (Nagle & Lazarus 1979; Spitz 1988) has sug-

gested that earlier versions of the WAIS produced

higher scores than earlier versions of the WISC, in indi-

viduals who were functioning in the low IQ range.

Nagle & Lazarus (1979) compared the WAIS with the

WISC-R in a group of 30 participants who received spe-

cial education classes for children with an IQ between

55 and 75. The WAIS yielded significantly higher IQ

scores, scoring 13 points higher on the Full Scale IQ

than the WISC-R. Significant correlations between the

corresponding subtests and IQ scales were found, sug-

gesting that the content was related, despite the non-

equivalence of scores. Spitz (1988) investigated whether

the difference between the WAIS-R and WISC-R was

Journal of Applied Research in Intellectual Disabilities 2010, 23, 197–200

� 2010 Blackwell Publishing Ltd 10.1111/j.1468-3148.2009.00538.x

Published for the British Institute of Learning Disabilities

consistent across different levels of IQ. He looked at

existing data from seven previously published studies

and combined this with the data from two new studies,

giving a total of 236 participants overall. He found that

‘there was an inverse relationship between the size of

the WISC-R ⁄ WAIS-R score disparity and the level of the

WISC-R IQ, such that the lower the WISC-R IQ, the

higher the increment in IQ derived from the WAIS-R’

(Spitz 1988, p. 377). However, there has been a lack of

recent research to replicate these findings using the most

recent versions of the tests and also a lack of research

using a sample of participants from the UK. Whitaker &

Wood (2008) and Whitaker (2008) examined the criteria

for gaining a scaled score of two, as outlined in the

manuals for the WISC-IV (UK), the WISC-III (UK) and

the WAIS-III (UK). On the basis of this they suggested

that the WAIS-III (UK) would produce significantly

higher IQs in the low ability range than either the

WISC-IIII or WISC-IV. An unpublished dissertation

(Bresnahan 2008), reported an American study where

the WISC-IV and WAIS-III were given to 20 school chil-

dren with IQs less than 75. The results showed that the

mean Full Scale IQ score on the WISC-IV was 14 points

lower than the mean Full Scale IQ score on the WAIS-

III. Therefore, there is a need to look at the relationship

between the UK versions of these tests.

Methods

Participants

Participants were selected from four special education

schools within one Local Authority area. The inclusion

criteria for participants were that they had to be able

to speak English, be 16 years of age (the age at which

both the WISC-IV and the WAIS-III can be given) and

attend a special school. Adolescents who had severe

or profound intellectual disabilities and would not be

able to complete the IQ assessments were excluded

from the study. A total of 17 participants were

recruited to the study, which included eight female

and nine male students. The mean age of participants

at the time of the first testing was 16 years 2 months

with a range between 16 years 0 months and 16 years

6 months.

Design and procedure

The research proposal was approved by Division of

Clinical Psychology Research Committee at Liverpool

University, who also considered the ethical issues. The

head teachers of each of the schools involved gave writ-

ten informed consent to take part in the study.

The WAIS-III UK Edition (Wechsler 1998a,b) and The

WISC-IV UK Edition (Wechsler 2004) were administered

to all participants. The standardized instructions were fol-

lowed for both tests. All the compulsory subtests were

administered alongside the optional subtests if they

appeared on both the WAIS-III and the WISC-IV. A total

of 13 WAIS-III subtests and 13 WISC-IV subtests were

administered. There were four participants who did not

complete the optional subtests on the WISC-IV (Picture

Completion, Information and Arithmetic); however, anal-

ysis using these subtests is not reported in this paper.

A repeated measures design was used, with the

administration order of the tests counterbalanced to

reduce practice effects. Participants were randomly

assigned to groups. Group 1 consisted of nine partici-

pants who completed the WAIS-III first followed by the

WISC-IV. Group 2 consisted of eight participants who

completed the WISC-IV first and then the WAIS-III. The

testing sessions took place during school hours and in a

suitable quiet room within each school. The number of

sessions needed to complete the two tests ranged

between two and three. The mean test–retest interval

was 2 weeks and ranged between 1 and 10 weeks. A

total of 12 participants had their tests administered and

scored by the principal researcher, with an assistant psy-

chologist administering and scoring the tests for five

participants. There were no systematic differences

between the two examiners. Paired samples t-tests were

used to compare the mean results of Group 1 and

Group 2.

Results

Table 1 provides a summary of the means and standard

deviations of the Full Scale IQ and Index scores. The

mean Full Scale IQ score on the WAIS-III was 64.82,

(SD = 8.51) and 53.00 (SD = 10.08) on the WISC-IV,

11.82 points lower. A paired samples t-test was used to

determine the significance of difference between the IQ

and Index scores on the WAIS-III and the WISC-IV. The

results were that the WAIS-III yielded significantly

higher scores on the Full Scale IQ [t(16) = 13.02,

P < 0.001], Verbal Comprehension Index [t(16) = 7.98,

P < 0.001], Perceptual Organization ⁄ Perceptual Reason-

ing Index [t(16) = 3.68, P < 0.002] and Processing Speed

Index [t(16) = 4.86, P < 0.001] than on the WISC-IV.

There was no significant difference between the WAIS-

III and WISC-IV scores on the Working Memory Index

[t(16) = 1.93, P < 0.071].

198 Journal of Applied Research in Intellectual Disabilities

� 2010 Blackwell Publishing Ltd, 23, 197–200

Significant correlations were found between the

WAIS-III and WISC-IV Full Scale IQ scores (r = 0.93,

P < 0.001), Verbal Comprehension Index scores (r = 0.77,

P < 0.001), Perceptual Organization ⁄ Reasoning Index

scores (r = 0.76, P < 0.001), Processing Speed

Index scores (r = 0.88, P < 0.001) and Working Memory

Index scores (r = 0.67, P < 0.005), which suggests that

the content of the IQ and Index scores on the two tests

was strongly related.

The differences between participant’s Full Scale IQ

scores on the WAIS-III and WISC-IV ranged between 5

and 20 points and all participants had a higher score on

the WAIS-III. There was a negative correlation

(r = )0.57, P < 0.05) between these differences and the

WISC-IV Full Scale IQ, suggesting a greater disparity in

Full Scale IQ between the two tests at lower IQ levels. It

is also notable that four participants’ Full Scale IQ scores

on the WISC-IV were 40, which was also the modal

score for this test. This is also the lowest Full Scale IQ

obtainable on the WISC-IV, which suggests the possibil-

ity of a floor effect. The lowest Full Scale IQ score

obtained on the WAIS-III was 49, which was well above

the test’s lowest obtainable score of 45. With regard to

the diagnosis of intellectual disabilities, using the WISC-

IV only one participant had a Full Scale IQ of 70 or

greater; however, on the WAIS-III six participants had

Full Scale IQ scores of 70 or more.

Discussion

The WAIS-III produced a significantly higher mean Full

Scale IQ score than the WISC-IV, in a sample of 16 year

olds who attended special school. The difference

between the mean Full Scale IQ score on the WAIS-III

and the WISC-IV was 11.82 points, with the WAIS-III

scoring higher. All the participants scored lower on the

WISC-IV than the WAIS-III, the smallest difference

between any participant’s scores was 5 points. The

Index scores, with the exception of Working Memory,

were also significant higher on the WAIS-III, with the

differences ranging between 9.50 and 12.58 points.

Although this result was obtained with a small sample,

it is consistent with the predictions made by Whitaker

(2008), a similar unpublished study by Bresnahan (2008)

in the US and studies comparing earlier versions of the

WISC and WAIS.

Part of this almost 12 point difference is likely to be

due the Flynn Effect, which is the trend for the intellec-

tual ability of the population as a whole, including peo-

ple with low intellectual ability, to increase from one

generation to the next (Flynn 2006). Flynn has estimated

this growth in ability to be about 0.3 of an IQ point per

year. Since the WISC-IV was standardized 6 years after

the WAIS-III, it would be expected to score two points

less than the WAIS-III. Therefore, there is likely to be a

10 points difference between the two assessments in the

low IQ range that is due to factors other than the Flynn

Effect. It is not clear why the remaining 10 points differ-

ence in scores occurs and further research is clearly

called for.

The large difference in scores obtained by special edu-

cation students on the WAIS-III and WISC-IV, which

has been found in this study, clearly has implications

for professionals working with people who have

intellectual disabilities. Currently the WAIS-III and

WISC-IV are both used to measure an individual’s intel-

lectual ability, to determine whether they meet the diag-

nostic criteria of having an IQ less that 70 for a

diagnosis of intellectual disability. Having a diagnosis of

intellectual disability can influence many decisions

Table 1 Comparison of WAIS-III and

WISC-IV means, standard deviations (SD),

t scores and correlation coefficients (r) for

the Full Scale IQ and Index scores

IQ scale ⁄ Index

score

WAIS-III WISC-IV

Diff t RM SD M SD

FS IQ 64.82 8.51 53 10.08 11.82 13.02*** 0.93***

VCI 67.59 7.62 55.76 9.50 11.83 7.97*** 0.77***

PRI ⁄ POI 70.29 10.59 62.88 12.58 7.41 3.68** 0.76***

WMI 64.35 9.21 60.71 10.09 3.65 1.93 0.68**

PSI 71.41 10.38 64.82 11.89 6.59 4.86*** 0.88***

*P < 0.05, **P < 0.005, ***P < 0.001.

WAIS-III, Wechsler Adult Intelligence Scale – Third Edition; WISC-IV, Wechsler Intel-

ligence Scale for Children – Fourth Edition; FS, Full Scale; VCI, Verbal Comprehension

Index; PRI, Perceptual Reasoning Index; POI, Perceptual Organization Index; WMI,

Working Memory Index; PSI, Processing Speed Index.

Journal of Applied Research in Intellectual Disabilities 199

� 2010 Blackwell Publishing Ltd, 23, 197–200

about an individual’s life, including parenting assess-

ments, capacity to consent, access to services and access

to educational courses. If the tests are producing differ-

ent scores, and it is unclear which test is producing the

most accurate results, then important decisions are

being made on the basis of inaccurate information.

Therefore, professionals working with people who have

intellectual disabilities need to exercise caution when

interpreting IQ scores on the WAIS-III and WISC-IV,

particularly when the IQ score is in the mild or border-

line ability range.

This study has provided the first empirical compari-

son of the WAIS-III and WISC-IV in people with intel-

lectual disabilities in the UK. However, the study is

limited by the use of a small sample of adolescents, who

were all aged 16 and from a single area of the UK.

There was also a relatively a large range of test–retest

intervals. Therefore, further investigations are needed

using larger samples in other countries to add support

to these findings. Further research is also needed to

investigate the reasons for these differences. The results

of such research could help to improve the process of

standardization of future versions of the Wechsler tests.

Acknowledgments

Thank you to all the schools and young people who

gave their time to participate in the study.

Correspondence

Any correspondence should be directed to Shirley

Gordon, Clinical Psychologist, Glen Acre House Child

and Family Service, 21 Acre House Avenue, Lindley,

Huddersfield, HD3 3BB, 07901 503898, UK (e-mail:

[email protected]).

References

Bresnahan J. A. (2008) A preliminary study of WISC-iv and WAIS-

III IQ scores for students with extremely low cognitive functioning.

Dissertation Abstracts International: Section B: The Science

and Engineering. Fairleigh Dickinson University, Teaneck,

NJ, USA.

British Psychological Society (2000) Learning Disability: Defini-

tions and Contexts. British Psychological Society, Leicester.

Flynn J. R. (2006) Tethering the elephant: capital cases, IQ and

the Flynn Effect. Psychology, Public Policy and Law 12, 170–189.

Nagle R. J. & Lazarus S. C. (1979) The comparability of the

WISC-R and WAIS-R among 16 year-old EMR children. Jour-

nal of School Psychology 17, 362–367.

Spitz H. H. (1988) Inverse relationship between the WISC-

R ⁄ WAIS-R score disparity and IQ level in the lower range of

intelligence. American Journal on Mental Retardation 92, 376–

378.

Wechsler D. (1997) WAIS-III, WMS-III: Technical and Interpretive

Manual. The Psychological Corporation, San Antonio, TX.

Wechsler D. (1998a) Wechsler Adult Intelligence Scale – Third Edi-

tion, UK edition. The Psychological Corporation, London.

Wechsler D. (1998b) Wechsler Adult Intelligence Scale – Third UK

Edition: Administrative and Scoring Manual. The Psychological

Corporation, London.

Wechsler D. (2003) Wechsler Intelligence Scale for Children –

Fourth Edition: Technical and Interpretive Manual. The Psycho-

logical Corporation, San Antonio, TX.

Wechsler D. (2004) Wechsler Intelligence Scale for Children –

Fourth UK Edition: Administrative and Scoring Manual. The Psy-

chological Corporation, London.

Whitaker S. (2008) WISC-IV and low IQ: review and compari-

son with the WAIS-III. Educational Psychology in Practice 24,

129–137.

Whitaker S. & Wood C. (2008) The distribution of scaled scores

and possible floor effects in the WISC-III and WAIS-III. Jour-

nal of Applied Research in Intellectual Disabilities 21, 136–141.

200 Journal of Applied Research in Intellectual Disabilities

� 2010 Blackwell Publishing Ltd, 23, 197–200

Copyright of Journal of Applied Research in Intellectual Disabilities is the property of Blackwell Publishing

Limited and its content may not be copied or emailed to multiple sites or posted to a listserv without the

copyright holder's express written permission. However, users may print, download, or email articles for

individual use.