World's IQ

Mankind’s IQ is 84 to 88: Some (provisional) reports from THE NIQ-DATASET V1.3 David Becker Department of Psychology, Chemnitz University of Technology, Germany Wilhelm-Raabe-Str. 43, D-09107 Chemnitz, Germany E-mail: david.becker@s2009.tu-chemnitz.de Abstract: Since the latest report at the London Conference on Intelligence in 2017, the project about a new edition of Lynn and Vanhanen’s 2002 and 2012 estimates of the IQs of nations ran through many steps of improvement plus extension and is now just before reaching a significant state. A report of provisional findings should be given to the audience. This will include a collection of psychometric measured IQs for 125 nations from 325 sources, 566 samples and a total number of 627,098 individuals, gives a global mean IQ of 84.74 (88.17 if weighted for sample-size). After the extension and restructuring of the 2017th version, it is now possible to make deeper insights into the data structure, for instance: comparing measurements from Raven’s Progressive Matrices, Wechsler Intelligence Scales and Cattell’s Cultural Fair Tests as well as looking for contingent effects of sampling characteristics (e.g. year of measurement, sample size etc.) on calculated IQs. Correlation analyzes show that such possible effects are limited to only a few variables (year of measurement, year of standardization of used test) and weightings for sample size and/or data-quality did not do any remarkable changes. Some important correlates from education, economics, demographics, geography, health, psychology, human development, politics and biology will be additionally shown. Keywords: cross-national; IQ; FLynn-Effect; sampling bias; correlates Sources: Lynn, R., & Vanhanen, T. (2002). IQ and the Wealth of Nations. Westport, CT: Praeger. Lynn, R., & Vanhanen, T. (2012). Intelligence: A Unifying Construct for the Social Sciences. London: Ulster Institute for Social Research.

The dataset of national IQs presented by Richard Lynn and Tatu Vanhanen first in 2002 and most recently in 2012 was both frequently criticized and used in cross-national research. It has also become very popular in public discussions, for example in web forums, and thus well known to a broader audience. Because of this popularity, we consider it important that the dataset behind this list is maintained, continued and improved for future research. Furthermore, an improved transparency and traceability of the methods can be useful in communication of the results. For about a year a group of researchers decided to carry out this task, beginning with the collection, as much as was possible, of sources for national IQs, along with restructuring of the original work notes, which were shared willingly by Richard Lynn. In this talk we want to present the first results of this preliminary revision. All available sources which give raw scores for Raven’s Matrices were considered and all data were picked out which are necessary to understand the estimations and calculations from the raw scores to the final national IQ. This includes also a new and standardized estimation of the required Flynn Effect corrections, based on a meta-analysis conducted by Pietschnig and Voracek in 2015. Final and intermediate results were correlated with the original results as well as with common associated variables (PISA, TIMSS, GDP/C, latitude, skin color, etc.). The talk will include discussions about discrepancies between the original data and the revisit. Problems with our methods will be discussed, and their validity will be assessed. In addition, future plans –– further work on national IQ, and an open access online data set -- will be presented.