Table of Contents:
Background: Early childhood development (e.g. motor, linguistic or intellectual development),
defined as a classified, directed and long-term change in experience and behavior, has been
a broad field of research in psychology for many decades. The empirical research of
developmental psychopathological processes as well as the diagnosis and treatment of mental
illnesses in young children is still a comparatively new topic. Epidemiological studies have
been available since the beginning of the turn of the century and point to a point prevalence of
approximately 10% - 20% over all mental illnesses in preschool-age children. It has been
shown that the common classifications according to DSM-5 and ICD-10 (with disorder-specific
adaptations) can also be used for the early childhood. Issues in the classification, that
necessitate adjustments, are for example the rapid intrapersonal development of the stage of
life as well as the differentiation between conspicuous and development-adequate behavior.
A normative comparison by means of validated measuring procedures of behavior and
development for the identification of mental illness therefore appears to be useful. This
dissertation presents the validation study of the Conners Early Childhood Scales, which
combines both relevant areas in one procedure.
Method: In the validation study of the German version of the Conners EC, a total of 720
parental judgements and 599 caregiver judgements were surveyed throughout Germany, in
respect to an equal gender and age distribution. In addition, the Strengths & Difficulties
Questionnaires (SDQ) and the Behavioral Assessment Sheet for Preschool Children (VBV 3-
6) were also surveyed. A total of 43 children were additionally examined with the observation
sheet for pre-school children (BBK 3-6). The aim of the present study is to examine the quality
criteria (objectivity, reliability, validity) of the German adaptation of the Conners EC. The
collected data was tested for reliability by determining internal consistency and correlative
analyses, and for validity by correlative analyses, confirmatory and exploratory factor analyses
and covariance analyses.
Result: Satisfactory objectivity of execution, evaluation and interpretation was assumed.
Regarding reliability, acceptable to very good values were found across almost all scales of
internal consistency. The retest reliability was in an acceptable range. When assessing validity,
the postulated factor structure could be replicated with confirmatory analyses for the behavior
scales of the original version. In explorative analyses, further factors were found for the parent
version of the Conners EC. Within the developmental milestone scales the global factor
"general development" for both versions was found in explorative analyses. The analyses of
variance demonstrated an influence of age, sex, parent educational level as well as the
attending in any clinical institution on parents and caregivers judgements.
Discussion: Overall, the validation of the German adaptation is satisfactory. The reported
characteristic values of the original version of the Conners’ EC were replicated on the basis of
the surveyed sample (with minor changes). Critical aspects of the conception of the Conners’
EC, such as inaccurate information on the creation of the itempool and the structuring and
verification of the factor structure, are discussed. The largest limitation of the study is the
limited representativeness of the norm sample and the inaccurate coverage of the clinical subsample.
Outstanding research questions with respect to the Conners’ EC include the collection
of data for samples with specific clinical disorders to determine sensitivity and specificity as
well as the investigation of a more comprehensive factor structure with respect to behavioral
scales and the developmental milestones.