d2 Test of Attention - Revised (d2-R)

Abbreviations: d2-R, Test d2-Revision

At the end of the 1950s, Brickenkamp constructed with the Test d2 an attention test that, since its first publication in 1962¹⁾, has gone through 10 editions (counting the d2-R) and is among the most successful German test instruments. The first seven editions appeared at intervals of two to five years without changes to the test, but with an extension of the age range of the norms and with an expanded literature review, in particular validation data. The eighth edition (1994)²⁾ brought with the new CP score (concentration performance) a further and better operationalization of the parameter for error-corrected working speed. For the ninth edition (2002)³⁾, new norms were collected. The tenth version presented here appeared in 2010 under extended authorship and with the new title d2 Test of Attention - Revised (d2-R) or Test d2 – Revision.⁴⁾ For the d2-R, the test sheet itself was changed for the first time. In addition, there are new parameters for scoring.

In TDB2Online, test results obtained with the ninth edition of the d2 or with the d2-R can be scored. Although the ninth edition of the d2 is currently (August 2021) still available from the test distributors in Germany and Switzerland, the improved d2-R is recommended for the initial assessment of patients or examinees.

For administration of the d2-R, one needs the complete test set, which in addition to an initial supply of test sheets, brief instructions and scoring sheets also contains the manual. The test sheets contain 14 lines with 57 characters each. Each character consists of the letter d or p, with 1 to 4 dashes drawn above and/or below it (at most 2 above and 2 below). The target character is a d with two dashes. The two dashes can both be above, both below, or one dash each above and below. All other combinations of letters and dashes are distractors. The task consists of marking all target characters in sequence with a cross stroke, working as quickly and as carefully as possible.

The new test sheets of the d2-R are carbonless. On the carbon copy, the positions of the correct characters are marked, which facilitates counting.

The material is protected by copyright and is, moreover, supplied only to certain professional groups. It is available from the test distributors in Germany or Switzerland or from international shops of Hogrefe.

The d2-R can be administered both individually and in a group. The entire administration is described very well and in great detail in the manual. There are also brief instructions, which serve to standardize the test instructions.

Administration of the d2-R as an individual test takes about 6 minutes including instructions, and scoring up to the determination of the raw scores takes a further 4 minutes.

The first and the last line of the test sheet are not scored.

TDB2Online scores the three most important raw scores, CP (Concentration performance), PT (Processed targets) and E% (Percentage of errors). PT is the number of processed targets, that is, a pure index of the speed of processing. E% quantifies the number of errors relative to the number of processed targets. E% is thus an index of the accuracy of the working style. CP is a complex performance index in which the number of correctly marked targets is calculated and reduced by the number of confusion errors. All three parameters are newly defined in the d2-R and differ from the d2. However, there are closely related parameters in each case in the old d2 (for details see the section Comparison with earlier editions).

The other parameters of the test, concentration curve and working style, are not suitable for representation in the profile.

A new norming study was necessary for the d2-R because the outer form of the test form and the scoring of the parameters were changed compared with the d2. The new manual norms of the d2-R are based on studies conducted between 2007 and 2008 primarily in Berlin, but also in five other German federal states. Norms are available in the form of percentile ranks and standard scores (100;10) for the three parameters E%, PT and CP, and additionally also for the old GZ score used in the d2. The norming range extends from standard score 70 to standard score 130, that is, over a total of 6 standard deviations. For children and adolescents, the age grading is fine-meshed (9–10, 11–12, 13–14, 15–16 and 17–19), for adults considerably wider (20–39, 40–60).

No further data that could be used for norming could be found. There are, to be sure, very extensive validation data for the d2-R, as previously for the d2, from the most diverse samples. However, none of these samples is even approximately population-based. Mostly they are either student samples, job applicants or clinical samples. Thus the only norm sources remaining are the data reported in the manual itself, even if their representativeness is similarly uncertain (see below).

In the age range of interest for TDB2Online, 2204 persons were examined for the norming, namely 500 in the age range 15–16 years, 728 in the age range 17–19, 708 in the age range 20–39 and 268 in the age range 40 to 60 years. In the range of children, adolescents and young adults the sample is thus large, in the older adult range from 40 years on it is small. At least this probably correlates with the use of the test, which is rather rarely used at higher ages.

In the young adult range, where the influence of age is not so large, the chosen span of 20 years (20–39) is probably still acceptable, because the changes over the course of age are relatively small here. In the higher age range of 40 to 60 years, however, it is certainly too wide. With only 268 persons, the group was probably too small to define narrower age groups. For persons over 60 years there are no d2-R norms.

The manual gives no further information on the representativeness of the sample. As unfortunately is often the case with German ability tests – and as in the earlier editions of the d2 – it is not apparent that the sampling was based on any quota and/or random plan.

see the detailed test documentation

see the detailed test documentation

see the detailed test documentation

It is a pity that there is no representative normative basis for a test that is used so frequently. This was already the case with the last two editions of the d2, and it is also the case with the d2-R. A lack of representativeness cannot be compensated for by large numbers. At least there are no major norm differences between the 9th edition of the d2, which is still frequently used, and the d2-R. The description of the norm construction in the d2-R is even more incomplete than in the last two editions of the d2. One does not learn what was calculated and how.

Both in terms of design and norms, the d2-R is a test that can be used profitably only with younger persons. For older persons, the characters to be processed are too small; norms are available only up to age 60, and in adequate group size only up to age 40.

The relatively short administration time is always a plus. However, this also applies to other attention tests: The Digit Symbol Coding subtest in the WAIS-IV, too, takes only two minutes plus one minute for the instructions.