Experimental error: Under "Analytical resources," review "error analysis and significant figures" and "error representation and curvefiting."
Statistical tests: [Unpaired (independent) t test] [paired t test] [Chi-squared test] [sample problems] [solutions to sample problems]
Tables: ['Student's' t distribution] [Chi-squared critical values]

Selected Critical Values of the t-Distribution A test is 2-tailed if you ask the question, 'does population 1 differ from population 2'. Then, if the mean for population 1 is significantly greater or smaller than that for population 2, you reject the null hypothesis. If you ask simply, is the true mean for population 1 greater than that for population 2, then you reject the null hypothesis only if the experimental mean for population 1 is found to be significantly greater. It can be significantly smaller, but you weren't asking that question. That would be a one-tailed test. For a two-tailed test if the calculated value of t exceeds the tabled value, then report the p value in the table. For a one-tailed test, the p value is divided by two. So 'p < 0.05' becomes 'p < 0.025." The table should include values for p=0.1 so that a one-tailed test can be conducted at the p=0.05 level, but we never do such tests in my class, so why clutter up the table?