ABSTRACT
Correlation assesses the strength of association between variables, and linear regression allows us to use one variable to predict another. The shape of the dots on the scatterplot shows an imperfect negative correlation between age and bone mineral density (BMD). BMD does indeed appear to generally decrease with age. This alone, however, is not enough to demonstrate a correlation and test significance — for this we need to calculate Pearson’s product moment correlation coefficient. We can therefore conclude that there is a significant negative correlation between age and BMD in women, and can accept the consultant rheumatologist’s hypothesis that BMD decreases as age increases. We have a small sample size in our age and BMD study, so in this case it is also appropriate to use the Spearman’s rank correlation coefficient. We can also use linear regression to predict the value of BMD for any specific age. This is achieved by calculating a straight line that best fits the association.
