JMP Training

Data Access and Management

Data access and management

Getting data into JMP from wherever it actually lives — Excel, CSV and text files, SQL databases over ODBC, SAS data sets — and then shaping it into a form fit for analysis. This covers joining and concatenating tables, stacking and splitting between long and wide layouts, recoding inconsistent values, handling missing data, building formula columns, subsetting and filtering, and saving the steps so the same import and clean-up runs again unattended next month.

Most analysis time is spent here, and it is where most analysis errors originate.

Data Analysis and Modelling

Correlation

Measuring the strength and direction of the relationship between variables, using scatterplot matrices and multivariate correlation — and knowing where correlation stops being evidence of cause.

Regression and ANOVA

Fitting models that relate a response to one or more predictors: simple and multiple linear regression, comparing group means with one-way and multi-factor analysis of variance, interaction effects, residual diagnostics, and reading p-values and effect sizes without over-reading them.

Predictive modelling

Building models that predict outcomes not yet observed — partition and decision trees, neural networks and regression-based models — and validating them against held-back data so they hold up outside the sample they were fitted on.

Quality and Reliability

Control charts

Statistical process control: plotting a process over time against statistically derived control limits to separate ordinary common-cause variation from special-cause signals worth investigating. Covers the Shewhart family — individuals and moving range, X-bar and R, and the p, np, c and u attribute charts — along with the run rules that decide when a process is out of control.

Process screening

Reviewing many process variables at once to find the few that are unstable, drifting or out of specification, so attention lands where it matters instead of on one chart at a time.

Process capability

Whether a stable process can consistently hold specification — Cp, Cpk, Pp and Ppk, capability indices compared across many characteristics at once, and how to read a goal plot.

Measurement System Analysis

How much of the variation you observe comes from the measuring system rather than the thing being measured — gauge repeatability and reproducibility (Gauge R&R), bias, linearity and stability. If the measurement system is not capable, nothing downstream of it is trustworthy.

Quality Analysis and Quality Control in the Mining Industry

QA/QC basic principles

Why analytical QA/QC matters when resource estimates and public reporting depend on assay data: the control sample types — certified reference materials, blanks, and field, coarse and pulp duplicates — insertion rates, sample chain of custody, and the documented QA/QC that resource reporting codes expect.

Pass/Fail criteria and case studies

Conventional rule-based acceptance limits, how failures are actioned and batches re-assayed, and — through worked case studies on real assay data — where a pure pass/fail view stops telling you what you need to know.

Facilities in JMP that support QA/QC analysis

The platforms already built into JMP that do this work: control and variability charts, distributions, bivariate fits, Graph Builder, data filters, and scripting to automate per-batch QA/QC reporting.

Using Basic3QAQC for analytical QA/QC

Working with our own JMP-based Basic3QAQC add-in — paired data for precision, blanks and standards for accuracy, and performance-based method precision limits. See where Basic3QAQC fits alongside rule-based systems, or our QA/QC consulting.