- The document discusses different types of sampling bias that can occur when collecting data samples to represent a population. - There are biases that can affect parameter estimates, standard errors, confidence intervals, test statistics, and p-values. These are all related - a bias in one area like confidence intervals will lead to biases in other areas like test statistics. - Five assumptions that can lead to bias if violated are: the presence of outliers, additivity and linearity of relationships, normality of data, homoscedasticity (equal variances), and independence of observations. Outliers, violations of linearity or additivity, and unequal variances can bias parameter estimates, standard errors, and test statistics.