Chapter 2 Development and Standardization (sic)


Development and Standardization (sic)


The DWEEEB was developed with several goals in mind.

The primary goal in the development of the DWEEEB was, plain and simple: to make money.

A second, much less important goal was to maintain the integrity of the original test by providing a well standardized, reliable and valid measure of the enduring theoretical underpinnings of the scale – the general factor, better known as Gee This goal was not reached.

The third goal of the DWEEEB was the destruction of factor structure. Because unpublished research has shown that examiners (experts) don’t know or care about factor structure, and couldn’t tell an oblique rotation from an orthogonal one if their wives depended on it, the test author simply ignored this and made up the whole chapter on Factors. (Don’t tell anyone).

The fourth and final goal of the development of the DWEEEB was the improvement of subtest content. Extensive files of user inquiries, resulting from an unstructured questionnaire/ telephone survey of a few close friends who think very much as the test authors, detailed lists of improvements suggested by numerous unknown people at local bars, and comments from consultants provided a wealth of information not used in the revision process.

Having reached 3 out of 4 goals (75th percentile), this project was deemed a complete success.


Description of the Sample

The DWEEEB norms presented in this manual are derived from the standardization sample that was not representative of the U.S. population of children. A stratified random sampling plan was not used to ensure that the representative proportions of children from each demographic group would be included in the standardization sample. The following sections presents the characteristics of the DWEEEB standardization sample.

Age: The standardization sample of 22 cases did not include 200 children in each of 11 age groups.

Gender: The sample included almost equal numbers of male, female, and neuter in each age group.

Race: For each age group in the standardization group, a three-legged, 200 yard dash was run. A great time was had by all.

Geographic region: The United States was divided into forty major geographic regions specified by no known source. (see Figure 2.1)

Figure 2.1 40 Geographic regions for Standardization Sample

Parent Education: The sample was stratified (whatever that means) according to the following parental education categories:

  • Yes
  • No

Locating and Testing the Sample

Invitations to participate in the DWEEEB standardization were mailed to every school district in the United States. Unfortunately, we forgot to include any return address and thus had a 100% non-return rate. Oops!! See Figure 2.2:

Figure 2.2 DWEEEB Invitation Sent to School Districts

Representativeness of the Sample

Figure 2.3 says it all.

Figure 2.3 Representativeness of the Sample

Tables 2.2-2.5 should present in great and extensive detail the demographic characteristics of the standardization sample. This data would have been helpful but the charts were too difficult to create so they were faked. A picture of an outside consultant administering a DWEEEB at a table is presented instead.

Table 2.2-2.5 Consultant Administering the DWEEEB on a Table

Sample by Community Size

Community Size DWEEEB Sample U.S. Population
Metropolis 0.0 ?
Statistical Area (MSA) 0.0 ?
Over a billion 0.0 ?
A little less than a billion 100.0 100.0
Total 100.0 110.0

Table 2.6 Percentages of the Standardization Sample by Community Size

Qualitative Descriptions of the DWEEEB FSIQ Scores

Many examiners utilize a qualitative system as well as a quantitative system to describe a child’s performance. Table 2.7 presents specific IQ score ranges and their corresponding categories and diagnostic classifications. Don’t, however, let this stop you from developing your own labels. Everyone knows that if a test has a mean of 100, then any score (99 to 0) should be considered BELOW AVERAGE. The same holds true for the individual scaled scores with a mean of 10. All numbers, 9 to 0, are definitely below average and indicate some measure of impairment. If a score does fall at the mean or above, don’t let that stop you from reporting weaknesses and impairments. Attach statements to your report that indicate how easy it was for the child to get the score, or for that matter how ‘difficult’ it was for the child to get the score. Always remember regression to the mean, and since no one other than yourself will be able to explain its meaning, tell everyone it means that a high score is really a low score because it has to be regressed to the mean. Keep saying it and someone will believe they understand what you mean.

Theoretical Actual
FSIQ Classification Normal Curve Sample
130 Very Good 2.2 2.222
120-129 Almost Very Good 6.7 6.7859
110-119 Pretty Good 16.1 83.0
90-109 Okay 50.0 50.123
80-89 Oops 16.1 12.03
70-79 Below Oops 6.7 21
69 and below Come on!! Try harder! 2.2 2.2

Table 2.7. Qualitative Descriptions of the DWEEEB FSIQ Scores