JMP analysis preparation

Lesson 1

Start with a question and synthetic data

The original JMP tutorial collection begins with Opening JMP and Getting Started during the first month in Lab. Use the official JMP Learning Library for current tutorial destinations and software-specific instructions. Start practice with synthetic or openly licensed example data.

Write the research question before selecting a statistical test. Identify the unit of analysis, outcome, exposure, groups, timing, and whether observations are independent or paired. A clear question makes software choices meaningful.

Put it into practice

1. Open the official getting-started lesson.
2. Create a small synthetic table or use an approved JMP sample.
3. Record the JMP edition and version.
4. Write one sentence describing the comparison you intend to explore.

Lesson 2

Understand the data table

Define each variable in a data dictionary with its meaning, units, permitted values, and missing-value convention. Check column types and modeling roles. Preserve raw input and perform transformations in a documented working copy.

Inspect row counts, duplicate records, impossible values, and missingness before creating figures. A numeric code for a category should receive a categorical modeling role. A date column needs a consistent format and a meaningful time origin.

Put it into practice

1. Import a practice CSV or Excel file.
2. Compare imported row and column counts with the source.
3. Set continuous, nominal, or ordinal roles according to the variable.
4. Create a documented formula column.
5. Save the table and cleaning notes.

Lesson 3

Explore distributions and graphics

Use distribution summaries to understand scale, spread, outliers, and missing observations. Match the graphic to the variables: bar charts for categories, histograms or box plots for a continuous variable, and scatter plots for two continuous variables.

Graph Builder supports exploratory composition. Label axes with units, identify the denominator, and include meaningful captions. Preserve the script or reproducible steps for each output. An exploratory association can generate a question for formal analysis.

Put it into practice

1. Create a distribution summary for the outcome.
2. Create one comparison graphic across groups.
3. Explain one unusual observation and your handling decision.
4. Save a publication-quality export and its analysis script.

Lesson 4

Select inference with a mentor

Choose the method from the design, estimand, distribution, and assumptions. The recovered collection includes proportions, confidence intervals, chi-square tests, paired and independent t-tests, ANOVA, nonparametric methods, correlations, linear regression, and logistic regression. Learn the concepts alongside the software.

Before interpreting an output, specify what the estimate means, its units, the comparison, and the uncertainty. Review confounding, independence, multiple testing, sample size, and missing data with the project mentor. Report effect estimates and confidence intervals with the relevant assumptions.

Put it into practice

1. State whether the comparison concerns a mean, proportion, or association.
2. Explain whether the observations are paired or independent.
3. List the assumptions you will evaluate.
4. Record the chosen method and reviewer feedback.
5. Interpret a synthetic result in plain language.

Lesson 5

Build reproducibility and model validation

Save scripts alongside the practice data, dictionary, software version, and interpretation. Document exclusions and transformations so another learner can repeat the work. Keep the practice package free of sensitive information.

For predictive modeling, separate training and evaluation data at the appropriate unit, document split rules, and consider leakage. The source collection includes validation columns, model comparison, trees, neural networks, and other advanced topics. Advance to these methods after establishing statistical foundations and a reviewed plan.

Put it into practice

1. Save a repeatable analysis script or journal.
2. Restart from the preserved practice input and reproduce the outputs.
3. Compare estimates, row counts, and graphics with the first run.
4. Write a limitations paragraph.
5. Request a methods review before applying the workflow to an assigned research dataset.

Resources and original materials

These lessons combine recovered Lab guidance with new practice activities. Original document permissions remain in effect. Historical institutional procedures require confirmation for the current project. Product interfaces depend on your software edition.

• Original JMP tutorial topic collection
• Official JMP Learning Library
• Opening JMP and Getting Started
• Getting Started for Students

Resource review: October 14, 2026.