Advanced Neuropsychological Diagnostics Infrastructure (ANDI): A Normative Database Created from Control Datasets

Abstract

In the Advanced Neuropsychological Diagnostics Infrastructure (ANDI), datasets of several research groups are combined into a single database, containing scores on neuropsychological tests from healthy participants. For most popular neuropsychological tests the quantity, and range of these data surpasses that of traditional normative data, thereby enabling more accurate neuropsychological assessment. Because of the unique structure of the database, it facilitates normative comparison methods that were not feasible before, in particular those in which entire profiles of scores are evaluated. In this article, we describe the steps that were necessary to combine the separate datasets into a single database. These steps involve matching variables from multiple datasets, removing outlying values, determining the influence of demographic variables, and finding appropriate transformations to normality. Also, a brief description of the current contents of the ANDI database is given.

Effects of low visual acuity on neuropsychological test scores: A simulation study

“…The inconsistent results among these studies are likely to be explained by the variety of both participant characteristics and LVA values. This study indicates that for the TMT, the currently available norm data (de Vent et al, 2016) should not be applied in people with LVA. In clinical practice, this means that for individuals with LVA, the current norm data of the TMT can only be used to state that a test score is in the normal range, but they cannot be applied to conclude that the cognitive abilities measured by the test are impaired.…”

The Mini Mental State Examination does not accurately screen for objective cognitive impairment in Fabry Disease

“…Raw test scores were converted to normative T‐scores (mean = 50, SD = 10, corrected for age, education, and sex where possible) using extensive normative data (median sample size = 471, range 121‐1000) . Language skills were assessed using the 30‐item short form of the Boston Naming Test (BNT) and the Wechsler Adult Intelligence Scale IV: Similarities (WAIS‐IV: Sim) . Memory was assessed with the Rey Auditory Verbal Learning Test (RAVLT) and the Rivermead Behavioural Memory Test (RBMT): Storytelling (van), both assessing immediate recall (IR) and delayed recall (DR).…”

Univariate comparisons given aggregated normative data

“…As such, on the second (skewed) test, the assessed individual requires a score excessively extreme compared to the corresponding test's norm distribution to be found significant, thus lowering the power. Should norm group test score distributions be found to substantially differ, transforming the data to normality is recommended (de Vent et al, 2016).…”

“…The norm group data were simulated as if test scores had already been corrected for demographic influences, meaning they had a mean of zero (de Vent et al, 2016). Thus, the scores of the norm group data only consisted of a within-study term epsilon (ε) and a between-study term denoted by nu (ν).…”

“…Such a normative sample will rarely be available. In order to provide a solution, we propose to merge already available datasets -the 'healthy' control groups of previously conducted studies -to create one dataset that meets these demands (Agelink van Rentergem, de Vent, Schmand, Murre, & Huizenga, in press;Agelink van Rentergem, Murre, & Huizenga, 2017;de Vent et al, 2016). With data-sharing increasing in popularity in the social sciences (Asendorpf et al, 2013;King, 2011;Poline et al, 2012;Vines et al, 2014), this seems like an opportune solution to the appropriate norm group problem.…”