Correlation Methods

Correlation is a statistical technique that can show whether and how strongly pairs of variables are related. Correlation is computed into what is known as the correlation coefficient and it normally ranges between -1 and +1. We talk of perfect positive correlation when there is movement of a security either up or down and the other security moving in the same direction. On the other hand, perfect negative correlation means that if one security moves in either direction the security that is perfectly negatively correlated will move by an equal amount in the opposite direction. If the correlation is zero, the movements of the securities are said to have no correlation i.e. they are completely random.

Perfectly correlated securities in real life are very rare but securities with some degree of correlation are likely to be found in abundance, Spiegel pg 312

Discussion

According to Spiegel pg 295, correlations allow an assessment of whether two variables are systematically related within a group of individuals. A single person or a few individuals may show behavior that differs from the rest but this won’t affect the variables so much. This means that exceptions exist for individuals, even if the rest of the group is predictable.

In the cases of study provided, there are some confounding factors that would have made it difficult to establish a causal connection between use of day care and the mental adjustment and cognitive functioning of a child.

One is stableness. It applies to the relationship between age and height. Though there is an increase in height as a child progresses in years, the growth is not even as in to form a straight line. There are some periods whereby the growth rate may be high and vice versa considering the health status of a child, diet, season and even the parents. The genetical make up of a child is important as there are those who may be coming from parents who are short and others from parents who are tall and every child will take after the parents.

Two is the gender of a child. This is evident from the experiment on brain size and intelligence. It was noted that for the males the correlation between the IQ and brain sizes is r = 0.51 and 0.33 for the female counterparts and a higher value of r = 0.65 for male and a corresponding value of r = 0.35 for the females. This is to be attributed to gender and Sherry Ortner supports this when she purports that the wiring in the brains of women is inferior to that one of men.

The third confounding factor may be that not all babies in the experiment may be independent. Others are dependent on others and this is evident in the case of twins who may be depending on one another and this may be extreme in case of identical twins. In cases of twins they have some ‘linking factor’ in that if one starts to crawl the other one will be induced very fast to crawl and if none of them is interested in crawling then they are likely to stay in that mode for a very long period of time. As one starts to crawl the other one will imitate and each will want to outdo the other hence the rate of crawling will be mastered very first. This is closely related to the environment where a child is brought up and how. If the person who takes care of that child fastens it to her back most of the time then the kid will take a longer period of time to know how to crawl but if given freedom and mingles freely with other children then the child is likely to start crawling at a very early stage.