| Most developmental assessment systems have one or more weaknesses. | The Lectical Assessment System (LAS) addresses these weaknesses. |
|---|---|
| First, many developmental scoring systems, despite claims that they focus on structure, require some degree of concept-matching. In other words, the rater compares the conceptual content of a performance with the conceptual content of examples in a manual, making it difficult or impossible to score atypical performances. | LAS analysts determine the level of a performance by "looking through" its particular conceptual content to examine what we call its logical structure and hierarchical order of abstraction. After determining the level of the performance, they examine its level of elaboration to provide a phase score. |
| Second, many of these systems are based on an analysis of the development of a small sample of individuals over a limited time period. Sample sizes smaller than 60 have been common. This means that the scoring rules are based on the performances of individuals in the sample, which may represent only a narrow range of possible conceptual content. | The LAS was (and continues to be) validated on a growing database containing thousands of interviews and assessments on a variety of topics. Respondents have been from 5 to 86 years old and represent a wide demographic range. Using these data, we have shown the LAS to be a domain general assessment system that can be used to score verbal performances in any domain of knowledge. |
| Third, some popular developmental assessment systems are multidimensional. They either (1) confound the developmental dimension with other psychological constructs, such as personality, preferences, and biases, or (2) they provide a score that averages performances across multiple knowledge domains. | Multidimensionality is a problem, because, in the first case, it can bias developmental assessments toward particular content, and in both cases it makes it impossible to say exactly what has been measured. The LAS is a unidimensional measure, like a ruler. It measures only hierarchical complexity. This is a powerful property. |
| Fourth, the users of some developmental assessment systems measure performance in one domain and call the result a general measure of an individual's cognitive complexity. There is no such thing as an accurate general measure of cognitive complexity, because cognitive development generally occurs at different rates in different knowledge domains. | All of our assessments are domain specific. We never claim that a single assessment can determine an individual's overall cognitive complexity. Instead, since cognitive development usually progresses at different rates in different domains, we produce assessments for particular skills. This makes it possible to produce psychographs, showing relative rates of development for different skills. |
| Fifth, the statistical reliability of most developmental assessments is inadequate for the assessment of individuals. Further, scores lack precision because the confidence interval (margin of error) around a score is often greater than one developmental level. | We maintain a level of statistical reliability that is consistently above .90-the gold standard in individual assessment. And our Certified Analysts must minimally maintain an agreement rate of 85% within 1/5 of a level with a Certified Master Analyst. The margin of error around Lectical scores is about 1/5 of a level. |