Hypothesis (Inferential statistics): A practical approach

Understand the normal distribution, a symmetric gaussian model with mean and standard deviation. Apply the empirical rule: about 68%, 95%, and 99% lie within 1, 2, and 3 standard deviations.

Explore the standard normal distribution and z scores, transform x to z using the formula (x - mu)/sigma, and read z-tables to estimate probabilities such as P(X < 30).

Differentiate data types into quantitative and qualitative, identify discrete versus continuous quantitative data, and classify qualitative data as nominal, ordinal, or binary.

Discover how data type and distribution shape the selection of hypothesis tests, from correlation and regression to t tests, chi square, and proportion tests, using Minitab.

Explore mini tab’s interface, enter data in the worksheet, copy headers into the gray bar to ensure correct data types, and stack data into a new worksheet.

Learn to test data normality with Minitab, formulate null and alternative hypotheses, and conclude normality from a p-value of 0.452 in a 40-value dataset.

Perform a one-sample z test to compare the new delivery method against the six-day standard, test null and alternative hypotheses, and evaluate the p-value with alpha 0.05.

Apply the one-sample t test to test if the mean differs from 1080, using 15 weeks of data, alpha 0.05, p-value 0.002, rejecting the null.

Explore one-sample percent defective tests to assess if a population proportion exceeds a target value, interpreting p-values, confidence intervals, and power.

Apply the two-sample t test to compare mean processing times of two independent plants with small samples and unknown population variance. Verify normality and equal variances before testing hypotheses.

Identify paired samples and perform a paired t-test to analyze before-and-after data, interpret p values, and draw conclusions from practical case studies.

explains how one-way ANOVA compares means across three or more groups, checks normality and equal variances, and tests whether any mean differs.

Compare defect rates with a two-sample proportion test, using independent processes, set null and alternative hypotheses, and note p = 0.07 implies no difference.

Use the chi-square test for association to determine if age and hiring are independent, by comparing observed and expected counts and using p-values and alpha to decide.

Explore how correlation measures the strength of linear relationships between X and Y, noting that correlation does not imply causation, using Pearson's r from -1 to 1.

Master regression analysis as a predictive model linking a continuous dependent variable Y to predictor variables X. Learn simple and multiple linear regression, the regression equation, and best fit line.

Develop a regression model in Minitab with project cost and time data to reveal a linear relationship, interpreting p-value 0.004 and equation y = 0.9 + 4.1x, adjusted r-squared 63.01%.

Develop a solid understanding of hypothesis concepts and hypothesis tests through self-practice with Minitab, building on this course’s inferential statistics foundation.