Normed for ages 3:0 to 17:11 (Usual age range is 3:6 to 17:11; Out of level age range is 3:0 to 3:5).

This subtest assesses visual-spatial ability; perception of spatial orientation; analysis of visual data; and nonverbal reasoning. The presentation booklet is two-dimensional while blocks are three-dimensional. The task requires the child to make a two-dimensional construction while ignoring the third dimension.

Young children create designs using foam squares with sides of black or yellow. Older children use three-dimensional blocks with sides that are black, yellow, black and yellow divided diagonally, and black and yellow divided vertically. Items begin as two-block patterns and increase to 9 block patterns. Daniel (1986) found that flat squares and cubic blocks measured the same ability with a group of sixth grade children. Elliott (1990b, p. 48) determined that, for younger children, the flat squares were less confusing.

This subtest can be scored in one of two ways, Standard and Alternative (Unspeeded). Standard scoring is dependent upon both speed and accuracy, while Alternative (Unspeeded) is dependent solely on accuracy, although time limits are still enforced.

The Pattern Construction subtest contains a total of 26 items. Items are generally scored from a minimum (failure) of 0 points to a maximum (correct reconstruction with bonus points for speed) in the Standard administration. In the Alternative (Unspeeded) administration, scoring is pass-fail (2 points for first trial, 1 for second trial, when available, and 0 for failure or exceeding the time limit). Three different starting points are available (age 3:0 to 6:11 start at sample A and item 1; 7:0 to 12:11 start at sample C and item 8; 13:0 to 17:11 start at sample D and item 14). There is a typographical error on p. 221 in at least some printings of the Manual. The starting-point samples are wrong for two ages and should be the same as above and the same as those shown on p. 210 and in the protocol. The last three items are provided for those cases in which the examiner chooses to score the subtest using the untimed, alternative procedure.

The Pattern Construction subtest is considered a fair measure of g across all ages (overall r = .65). For ages 3 to 11, it has fair g loadings, while from ages 12 to 17, it is considered to have good g loadings. This subtest contributes substantially to the Spatial factor (loading = .82). Specificity is ample for all age groups 6:0 to 17:11.

Pattern Construction is considered to possess high overall reliability (r = .91), with reliability coefficients ranging from .80 to .93 across all of the age groups It correlates best with Recall of Designs (r =.57) and least with Recall of Objects (r = .15). It has a moderate correlation with the GCA (r = .77).

Gf-Gc classification

In the Broad stratum definition of abilities, Pattern Construction is considered to be a logical measure of Visual Processing (Gv). In the Narrow stratum of abilities, it is considered to be a probable measure of Spatial relations (SR) and a possible measure of Visualization (VZ) (McGrew & Flanagan, 1998, p. 106).

The Pattern Construction subtest is described on pages 210 to 221 in the DAS Administration and Scoring Manual and discussed on pages 48 to 50 and 57 in the DAS Introductory and Technical Handbook. The record form has helpful notations for all items -- M, P, or D -- referring to the method for presenting the items to the child: Model (M) refers to the examiner building the pattern in front of the child and then leaving the completed model in place while the child builds his or her pattern. Picture (P) refers to showing the child a picture of the pattern from either Booklet 2 (items 1 through 7) or booklet 1 (sample c through item 26) and leaving the picture in full view while the child completes the pattern. Finally, Demonstrate (D) refers to those cases in which the examiner builds the pattern using the child’s own blocks, and then mixes the pattern up and has the child try again. In five specific cases, there are multiple notations, so that, for example, "M, P" for item 1 means that the examiner creates the model as well as shows the picture to the child.

Some children try to complete the patterns by building their designs directly on top of the model or the picture. On early items, this strategy may be helpful, but on later items, where the picture is much smaller than the blocks themselves, the use of this strategy results in their actually covering up the pictures they are trying to copy. If this is the case, the child should be encouraged to make the patterns on the table directly in front of the child.

Although rotations of 30 degrees or more are scored as 0, in all cases that rotations occur the examiner should show the child the rotation and indicate how the pattern should have been made.

Timing is important on this subtest since, since in the timed administration bonus points are given for successful completion within certain time frames. Timing of the subtest begins when the examiner finishes with the instructions and continues until the child has completed the item. Because many children often do not tell the examiner when they are done, examiners should watch the construction carefully and note the time at which the pattern is successfully completed. Stop the watch when the child indicates by word or gesture, that he or she is complete. If the child has not changed the design successfully completed earlier, give credit for the earlier time.

The DAS Pattern Construction subtest allows for an Alternative (Unspeeded) administration. If the examiner feels that the imposition of the strict time limits is not a fair or valid procedure (e.g., a motor impairment prevents speedy manipulation of the blocks or the child is a slow, thoughtful worker) the Alternative scoring procedure can be utilized. In this case, examiners should refer to the Alternative starting, decision, and stopping points on p. 221 of the Manual and in the record form. In general, if one chooses to score with the Alternative procedure, more items must be administered. Items 24-26 are administered only as part of Alternative scoring. According to the Manual (Elliott, 1990a. pp. 210, 220) The choice between the Standard and Alternative administrations need not be made in advance. However, because the examiner must administer any needed additional items the decision must at least be contemplated before completing the subtest.

The Alternative (Unspeeded) procedure is a valuable option. Like the Stanford-Binet, 4^th ed. (SB: FE), and unlike the Wechsler scales with their heavy emphasis on speed (see Chapters XX and XX ), the Alternative procedure offers more valid assessment of the abilities of students who, for any reason work slowly.

The decision points for the Pattern Construction subtest may at first appear a bit confusing: <3 with less than maximum score: Continue; <3 First-trial Passes: Go Back. This is the 3 by 3 rule with the slight twist that failure is not a requirement. In order to discontinue at the decision point, a child needs to pass (with any amount of credit) 3 items and additionally must obtain less than perfect (not necessarily zero) scores on 3 items. These authors have found that placing some mark in the margin next to the scoring table on the record form helps to keep track of any less than perfect (e.g., 3 points on an item for which 4 points are possible) scores. If it is found that there are only 1 or 2 marks (<3 with less than maximum) this signals us to continue testing through the next block.