Random variables vs sample of mesurement i would like to know why a sample of measurement in Statistics is being represented through random variables even though it's about only one quality of measurement, say the weight of individuals of a population ? 
In the case of determining the mean value of the sample one then calls it determining the mean value of the sum of the random variables.
Can somebody explain me why it should be the same thing ?
THanks for your comment.
 A: One way to look at this is to assume you have a very, very large population you are sampling from to get height or whatever. Then assuming independent uniform random sampling of individuals from the population, to a very good approximation, each sample $j$ can be treated as an independent identically distributed (i.i.d.) random variable $X_j$ sample value where the distribution of each random variable $X_j$ is the same as the "true" distribution that generated the population. Then, given the $X_j$, you can define new random variables such as the mean of the $X_j$ and treat the observed computed value (e.g. empirical mean) as a sample from the random variable distribution defined in terms of the random variables $X_j$. The only last detail to consider is that you don't know the true population mean or standard deviation, and so these are typically estimated from the samples. This can introduce bias e.g. in the estimate of standard deviation because the empirical mean of the samples is not exactly the same as the true mean of the population distribution. There are often ways to correct for this bias, e.g. look up unbiased variance estimation vs. plain empirical variance.
