Logo
  Sign In / Register
 
 
 

  
Kids IQ Test
Test Manual

Take the FREE Premium IQ Test
 

<  Back to Test Description    

Description

The Kids IQ Test is a 201-item multiple choice measure of verbal intelligence designed for children ages 5 to 17.  It was designed specifically for administration over the Internet, with the assistance of an adult for younger children.  It has about 40 items for each of five of the content areas measured by the Wechsler Intelligence Scale for Children: Information, Similarities, Arithmetic, Comprehension and Vocabulary.  When a child misses 5 items in a row, that section of the test is discontinued and the next begun.

Design

The test items were written by the author to measure a range of ability from a mentally slow 6 year old to a bright 16 year old.  The content of the items was written to reflect the content of the corresponding areas of the Wechsler test.  A multiple-choice format was chosen to permit Internet administration and automatic, computer scoring.  Each item has 5 options, four which are possible answers and one for “I don’t know”.   The test is administered over the Internet at FunEducation.com and the author’s web site (www.testmasterinc.com).  There is only one correct answer for each item.

Statistical Properties

To obtain initial statistics, the test was administered over the Internet to 395 children ranging in age primarily from 5 to 17 with a few 3, 4 and 18-year-olds.   An invitation to participate in return for a few dollars worth of web site testing services at the site was sent to 2,000 adults who had previously visited a web site with which the author is affiliated (Funeducation.com).  Respondents were also told they would receive test scores when norming was complete.  These adults had visited the FunEd web site to take other tests offered by it. They were invited to test their children.  The invitation was also sent to 13,000 children under the age of 18 who previously had taken another intelligence test offered at the web site.  The children responded within a few week period in the summer of 2003.  Most were from the United States, with some from Canada, Great Britain, Australia and other countries.

Data

Data from this initial study were processed by the author using SPSS software.  Correlations were run between each test item and the total score of which it is a part for the total sample.  Every one of the items except the simplest Arithmetic item correlated significantly with its total score, usually at the .01 level or better, indicating that each item was contributing as expected.   Similar analysis by age group showed less consistent item characteristics.

Alpha reliability coefficients were computed for each age, section and total score, a sample of which is provided below.

Kids IQ Test Alpha Coefficient Reliabilities

Age

N

Info

Simil

Arith

Comp

Vocab

Total I.Q.

6

12

 

 

 

 

 

.97

7

9

 

 

 

 

 

.97

8

22

.84

.85

.94

.88

.91

.97

9

12

 

 

 

 

 

.93

10

18

 

 

 

 

 

.95

11

22

.94

.80

.86

.85

.92

.97

12

29

 

 

 

 

 

.95

13

42

 

 

 

 

 

.96

14

60

.77

.89

.94

.92

.93

.97

15

65

 

 

 

 

 

.89

16

62

.72

.72

.78

.66

.76

.91

17

30

 

 

 

 

 

.91

Thus, it appeared that the section and total scores had adequate reliability for accurately differentiating various levels of verbal intelligence overall, especially as reflected in the Total Verbal I.Q. score.

By way of comparison, consider the corresponding reliabilities for the WISC-III subtests and total Verbal WISC-III I.Q. for 14-year-olds (1).

Comparison of Reliability Data, Kids IQ Test versus WISC-III, 14-year-olds

 

N

Info

Simil

Arith

Comp

Vocab

Total V.I.Q.

Kids IQ Test

60

.77

.89

.94

.92

.93

.97

WISC

200

.87

.84

.77

.76

.91

.95

Means and standard deviations are proprietary information.  Means were plotted on graph paper across ages and smooth curves drawn to obtain estimated values for larger populations.  Standard deviations were roughly similar across ages, especially for sample groups of 20 or more.  The mean standard deviation across all ages was computed and is used as the standard deviation estimate for larger populations. 

The curves for the mean scores were compared to similar curves drawn for corresponding raw score data published in the WISC-III manual.  Kids IQ Test curves were similar to WISC-III curves, though the WISC-III curves tended to be straigher.  The Kids IQ Test curves tend to rise in a curvilineal arc which tends to level out at age 16 to 17.  This may reflect a leveling of brain neurological maturity at these ages.

While some older children obtained perfect scores on one or more section tests, none obtained a perfect total score.  The highest raw total score was 195 of 201 items, obtained by a 14-year-old British girl.

Validity Data

The test is designed to have content validity by mimicking the content of the WISC-III.  Readers familiar with both instruments can judge for themselves how well this mimicking was achieved.  Eventually a validity study testing a sample of children on both instruments can provide correlations between them for further validation.

Inter-test correlations 

The correlations between sections and the total score are provided below.  These are for the total initial sample, all age groups.

Inter-test Correlations, All Age Groups Combined, N = 393

(All significant at .000 level)

Test section

Total

QVQ

Info

Simil

Arith

Comp

Vocab

Info

.83

1.00

.70

.69

.70

.50

Similar.

.78

 

1.00

.62

.60

.44

Arith

.83

 

 

1.00

.73

.49

Compreh

.86

 

 

 

1.00

.56

Vocab

.80

 

 

 

 

1.00

Thus, as expected, each section correlates substantially with the total Verbal I. Q. to which it contributes.  The correlations between sections are less high, again as expected.  Each section score contributes to the measure of intelligence but adding something different from the other sections.

Similar results were obtained for each age group.  For example, for 14-year-olds: 

Inter-test Correlations, 14-Year-Olds, N= 60

(All significant at .000 level)

Test section

Total 

QVQ

Info

Simil

Arith

Comp

Vocab

Info

.81

1.00

.77

.59

.53

.59

Similar.

.77

 

1.00

.59

.45

.46

Arith

.80

 

 

1.00

.61

.57

Compreh

.87

 

 

 

1.00

.89

Vocab

.87

 

 

 

 

1.00

 

Again by way of comparison, consider the corresponding data for the WISC-III, for 14-year-olds (2).

Section and Total Score Intercorrelations, 200 14-year-olds, WISC-III (3)

Test section

Total V.

I.Q.

Info

Simil

Arith

Comp

Vocab

Info

.84

1.00

.66

.60

.52

.71

Similar.

.85

 

1.00

.55

.63

.68

Arith

.77

 

 

1.00

.50

.60

Compreh

.80

 

 

 

1.00

.65

Vocab

.88

 

 

 

 

1.00

Low values are desirable for the correlations between the section scores, as they indicate greater independence of content, adding more uniquely to the total verbal intelligence concept measured.  The mean of the inter-correlations for the Kids IQ Test is .61.  For the WISC-III it is also .61.

Updating the norms

The test was administered over the Internet to several thousand children.  By November of 2005 data were accumulated for 4672 children.  This data was analyzed in several ways, with reassuring results. 

Again, no significant differences between genders was found, as expected.

There was a tendency for persons who paid for a detailed report to get slightly higher scores with slightly smaller standard deviations.  However, using statistics for the entire sample generated I.Q. scores very close to corresponding Verbal I.Q. scores on the WISC-III for a few children tested on both instruments, so norms for the total sample, paid and unpaid are now used.

Analysis of this large sample reveals that all of the test sections and all of the total scores have very good reliabilities across all ages from 6 to 16.  Consider the data for a sample of age groups, below.  Note for comparison the inclusion of corresponding reliability data for the Wechsler Intelligence Scale for Children 10 to 11 years old (WISC10), based on a sample of 119 children and reported in the manual for the Wechsler test.(1)

Kuder-Richardson 21 Reliability Coefficients

Age

N

Info

Simil

Arith

Comp

Vocab

Total I.Q.

6

354

.89

.92

.95

.92

.90

.98

8

332

.85

.89

.90

.90

.87

.97

10

487

.79

.83

.89

.85

.83

.96

WISC 10

119

.87

.80

.75

.89

.76

.93

14

409

.70

.77

.88

.80

.80

.94

16

238

.79

.86

.90

.85

.82

.96

Calibrating to WISC-III

In prior research by the author with FunEducation, it has been found that adults taking an Internet-administered verbal I.Q. test seemed to be of above average intelligence compared to the general population, as might be expected.  For this reason, the Kids IQ Test based on the initial sample data of several hundred children was calibrated to the WISC-III by adjusting scores based on a sample of children administered both tests.

Scores for the Kids IQ Test were adjusted, by reducing normative  means by multiplying them by .93.  Using these adjusted means, the Kids IQ Test scores an 8-year-old boy were 91 on the WISC and 93 on the Kid’s I.Q. test.  For a 14-year-old girl the scores were 125 on the WISC and 123 on the Kid’s I.Q. test.  For some children scores were lower than expected on the Kid’s I.Q. test.  Given the excellent statistical properties of the Kid’s I.Q. test, this difference was considered to reflect less than optimal motivation or attention on the latter test.  Parents will be encouraged to make sure their children are well-rested and well-motivated when taking the Kid’s I.Q. test as a result. 

For the larger norm group of 4672 children, the .93 multiplier adjustment was found to be unnecessary and has been dropped.

While normative data is proprietary, means and standard deviations in raw score units for 10-year-olds are provided below as an example:

Means and Standard Deviations for 487 10-year-old Boys and Girls

 

Informat.

Similar.

Arith.

Compreh.

Vocab.

Total Score

Mean

25.06

26.08

31.67

28.15

26.63

137.60

S. D.

6.12

6.97

7.03

7.20

6.83

29.20

Report Format

Scores are provided in terms of both  I.Q. and percentile levels for all five sections and the total score.  The printed report provides a brief description of verbal intelligence, reliability data and recommended interpretation of scores.  It advises also that obtained scores can be cross-checked by seeking additional evaluations by local professionals.

Intended Use

The Kids IQ Test is intended as a test that parents of young children and teenagers can use over the Internet for measuring verbal intelligence.  Scores obtained can help persons with self-awareness, self-understanding and making plans for schooling and/or career choices. Reports can be used by professionals as well, as in assessing for career planning, school placement and problem diagnosis.  For example, when diagnosing ADHD and dyslexia, it is important to rule out mental retardation.  Low scores on the Kids IQ Test should be verified with individual I.Q. testing, such as via the WISC-III or a similar instrument.

References

1.  Wechsler, David, Wechsler Intelligence Scale for Children - Third Edition, Manual, The Psychological Corporation, Harcourt Brace Jovanovich, Inc, New York, 1991.  P 166.

2.  Ibid. p.171.

3.  Ibid.  p. 278.
 

     <  Back to Test Description


Take the FREE Premium IQ Test

 

 

Fun Tests & Quizzes  
 
 
Your Tests  
 
 
 
© 1999-2014 FunEducation, Inc. All Rights Reserved.   Give Feedback  |  Terms of Service  |  Privacy