Koreliacija

Koreliacija is a statistical concept describing the degree to which two variables move together. The most common measure is the Pearson correlation coefficient, denoted r, which quantifies the strength and direction of a linear relationship. Its values range from -1 to 1: 1 indicates a perfect positive linear relationship, -1 a perfect negative linear relationship, and 0 effectively no linear relationship. If the relationship is non-linear, r may be small even when the variables are related.

In addition to Pearson r, non-parametric measures such as Spearman's rho and Kendall's tau assess monotonic relationships using ranks rather than raw values. These are more robust to outliers and do not assume a linear form, making them suitable for ordinal data or skewed distributions.

Interpreting koreliacija requires attention to several caveats. Correlation does not imply causation; two variables may move together due to a common cause, indirect linkage, or chance in small samples. Outliers can disproportionately affect Pearson r, and strict linearity is assumed for Pearson yet not for rank-based measures. Sample size influences the reliability of the estimate and its statistical significance, typically assessed with a p-value or confidence interval.

Calculation is conceptually straightforward: r equals the covariance of X and Y divided by the product of their standard deviations. For Spearman or Kendall methods, the calculation operates on ranked data. Koreliacija is widely used across disciplines—from finance to psychology—to gauge relationships, inform models, and study co-movements, while acknowledging its limitations and the possibility of spurious findings.