Error bars on graphs represent the variability or spread of data by showing the standard deviation or range. A normal distribution is a bell-shaped curve where the most frequent data points fall in the center. The standard deviation measures how spread out data points are from the mean and 68% of values fall within one standard deviation of the mean. A t-test statistically determines if two data sets are significantly different from each other. While a correlation means two variables vary together, it does not prove that one variable causes the other.