♦ The EDI statistical study on IQ

STATISTICAL ABSTRACT
METHODOLOGY OF THE STATISTICAL STUDY  

The title of each graph of the of the statistical study indicates the parents variables (R or M & F) to which the correlations are related. These correlations are represented by each point of the coloured lines corresponding to each examined C variable (children).

Likewise, the variables of unknown order, formed by the different groups of 1 to 10 values from the 70 IQ values of each parent and children variables are placed on the left hand side of the graph. The groups of 1 to 10 values located on the right hand side have been previously ordered with the variable mentioned at the bottom of the graph.

Statistical data
 

Each graph condenses more than 5,000 different points of information for the interrelations between:

• 70 values of each CI variable of the fathers, mothers, and children.
• 8 variables of the CI fathers, mothers, and children.
• 3 variables of individual averages of the previous variables.
• 10 criteria of value arrangement.
• 10 sizes of groupings of individuals.
• 20 values of evolution’s parameters in a sensitivity analysis.
• Countless random variables created in the simulation model.

The set of graphs collects all of these interrelations, that is, more than 1,000,000 values. Note that the average of any two values has its own dynamic and is more or less independent of the other two values.

As an example of the information’s validity, one could put forth the case of having a historical sample of 70 packets of cigarettes. The sample can be considered consisting of 70 elements, or many more if we consider that for each packet the following could be investigated:

• The number of cigarettes per packet.
• The size of the cigarettes.
• The type of cardboard the packet is made of.
• The color.
• If it has any images.
• If it has any health warnings.
• The type or severity of these warnings.
• Information about the level of nicotine and tar.
• etc.

Indeed, an almost instantaneous perception of the exactitude of the particular specification is obtained; sixty coefficients of determination (r2) are shown in a way that highlights the global and underlying relations of the involved data.

In order to facilitate the comparative analysis, a multidimensional correlation index has been defined (hereinafter MCI) to represent the global precision of the adjustments shown in any graph with one number. It will be made up of the sum of the determination coefficients of the ten rearranged variables.

There will be an MCI for each variable and a global MCI for the three variables studied in each graph. The maximum G-MCI will be 30, since 3 different variables and ten groups are always used.

On the right hand side and below the variable, the r² and the GMCI are shown to help understanding the correlations involved.

The results are surprising, which can be observed both in the graphs of the statistical annex and in the following tables. An aspect that will especially allow us to reach some important conclusions is the model sensitivity of the arrangement criterion.

* * *

 

 

 

•