Industries: Pharma / Biotech

Introduction to Statistical Analysis of Laboratory Data

Course Director: Al Bartolucci, Ph.D.

Course Fee: $2150.00 Regular Registration / $1950.00 Early Bird (30 Days in Advance)

Course Description - Course runs 9:00 to 5:00 both days (Breakfast & Lunch Included)

Basic Methods (Day One). This section of the course will detail the basic and intermediate statistical concepts that are essential for professionals in the field. The first day emphasizes the principles of descriptive and inferential statistical applications and focuses on actual study examples, problem solving and interpretation of results. Throughout the course the participants are encouraged to ask questions and discuss examples relevant to their own work. Topic areas to be discussed include, but are not limited to:

  • Basic statistical terminology including simple statistics as well as geometric ( e.g. means, standard deviations) transformations needed to effectively communicate and understand your data results
  • The statistical testing (one sided, two sided, non parametric, sample size, and power considerations) essentials required to initiate a research investigation (i.e., research questions in statistical terms)
  • Concepts of accuracy and precision in measurement analysis to ensure appropriate conclusions in experimental results including between and within laboratory variation results
  • Discussion of statistical techniques to compare experimental approaches with respect to specificity, sensitivity and linearity
  • The instructor gives a detailed description of topics discussed in the his latest publication, "Introduction to Statistical Analysis of Laboratory Data" by  Alfred  A. Bartolucci, Karan Singh and Sejong Bae (2015).

Advanced Topics (Day Two). This section of the course will go beyond the basics and cover more complex issues in laboratory investigations with examples. Topics will include:

  • Association studies including correlation and regression analysis with laboratory applications
  • Analysis of robustness and ruggedness
  • Method comparison using more accurate alternatives to correlate analysis and other pair-wise comparisons
  • Outliers, limit of detection and limit of quantitation
  • Statistical quality control for process stability and capability

Who Should Attend

This course is designed as an introduction to the statistical principles of laboratory data analysis and quality control that form the basis for the design and analysis of laboratory investigations. The course curriculum will benefit R&D managers, analytical laboratory supervisors and staff, manufacturing and production professionals, scientists, technicians and others who wish to comprehend and interpret methods of data analysis relevant to laboratory experimentation. Where applicable, topics are presented with relevant regulatory requirements.

This training will concentrate on the philosophy and understanding of the statistical principles required in conducting sound scientific investigations of laboratory processes and validation, including design and sample size issues. It will not simply present statistical formulae and the lectures are oriented toward professionals having minimal formal training in statistics or mathematics beyond basic algebra. However, for those with more formal training in statistics wishing to actually apply the techniques, appropriate time and references will be given for the procedures involved.

Course Agenda

First Day

Statistical Measures and Descriptive Statistics: Central tendency (average or mean, median, mode), dispersion measures such as range, variance, standard deviation, coefficient of variation, unbiased estimates, measurement summary and precision.

Graphical Techniques: Histograms, bar charts, scatter plots. Graphical representation of lab results.

Distributions and Formal Statistical Laboratory Tests: Normal, t-distribution (one sample, two sample, paired), one way ANOVA to assess effect and necessity of replication, skewed distributions with applications to experimental results with alternative statistical comparison methodologies.

Estimation Statistics: Point and interval estimates, accuracy, precision. Further concepts of method validation such as sensitivity, specificity, selectivity, linearity.

Second Day

Defining Robustness and Ruggedness: Design selection criteria, calculations, interpretation, effects of repeated experimentation, multiple lab results.

Defining Linearity Further: Applications to method comparison and interpretation. Examination of outliers in exploratory analysis of assay results.

Alternative Strategy to Linearity: Alternative advanced method for assessing agreement between two methods of laboratory measurements.

Limit Strategies: Limit of detection, limit of quantitation.

Calibration problem: Techniques involving crude and precise methodologies and measurement of bias.

Validation Using Statistical Process Control: Use of quality control charts to determine laboratory process stability and capability.

Learning Objectives

Those completing the course will have an understanding of the concepts of statistical design, analysis and graphing methods required in laboratory data analysis and reporting. Attendees will be able to interpret and report results related to design and analysis issues as presented in the scientific literature concerning laboratory data analysis, as well as, quality control methods.


"Excellent presentation in statistical measurements techniques that can be applied to many disciplines and projects." Keith B., Associate Director, Reliant Pharma
"I've taken undergrad & graduate level statistics courses (long ago) and the Course Director presented the material so I could understand it much better. He is great at explaining the concepts." Lori B., Senior Chemist, Barr