The figure at left shows three different formulas for calculating the heritability coefficient. The top formula is used when the data available consists of correlations between pairs of monozygotic twins reared apart (MZA). 
The second formula can be used when the data consists of both correlations between pairs of monozygotic (MZ) twins, and pairs of dizygotic (DZ) twins. It doesn't matter whether the twins have been reared together or apart, so long as both groups of twins have been reared in the same way (together or apart).
The third formula us used for calculating the heritability coefficient when when the data consists of DZ twins reared apart (DZA).

The heritability coefficient (h²) is not a magic formula that instantly tells us how much of some phenotypic characteristic is due to our genes. Instead, the heritability quotient merely tells how how much of the total variability in some phenotypic characteristic can be attributed to genetic variability within our sample.
There are other limits to our interpretation of the heritability coefficient, some of which which are shown in the figure at left. Two of these are particularly important:
1. h² tells us nothing about how much of an individual's phenotype can be attributed to his or her genes: h² is about the source of variation between individuals.
2. h² is not a constant. The relative contribution of genetic variability to phenotypic variability may change over time, and may be different in different populations, depending on their genetic homogeneity, and on the range of environmental variables that contribute to whatever phenotypic characteristic we're interested in.

The study by Thomas, Chess, and Birch was one of the very first attempts to define early measures of temperament. The variables that they looked at are listed at left.

Thomas et al (1970) described three different categories of infants, based on the pattern of responses on their assessments. The two main ones, together with their characteristics, are shown at left.