Contact Us

Advanced Analytics


MACRO Consulting, Inc. provides advanced analytic consulting to clients that don’t need full-service support.  We provide service using most commercially available statistical techniques:

  • Choice-based conjoint
  • Menu-based conjoint
  • Max/Diff Scaling
  • Hierarchical Bayes regression modeling
  • Latent Class regression modeling
  • Latent Class Cluster modeling
  • Structural Equation Modeling
  • Cluster Ensemble Analysis
  • Other commonly available techniques, such as factor analysis, discriminant function analysis, correspondence analysis, etc.

In addition to the above commercially available techniques, Paul Richard McCullough, president and founder of MACRO Consulting, Inc., has developed some innovative analytic techniques that have proven useful in the commercial sector:

  • Cake Method©: a method for handling a large number of attributes in conjoint studies
  • An SEM/Conjoint approach to understanding attitudinal drivers of consumer choices and preferences
  • An SEM approach to controlling brand halo in brand driver analysis
  • A new method for scaling Max/Diff scores that is more intuitive than current methods
  • MACROModel©: a holistic approach to evaluating large numbers of new product concepts, estimating price sensitivity and volumetrics
  • ImageMax ©: A  new method of brand measurement that would improve inter-item and inter-brand discrimination, eliminate brand halo and improve predictive validity

MACRO Consulting, Inc. strives to stay at the cutting edge of analytic techniques.  In a constantly developing field, utilizing the newest tools is vital.  Our current areas of inquiry include:

  • Ways to reduce the number of choice tasks necessary to estimate disaggregate choice models
  • Ways to eliminate the Number of Levels Effect bias

Recommended Reading

Abbreviated Task Sets: Estimating Disaggregate Choice Models With Extremely Few Tasks Per Respondent

Brand Halo: The Elephant in the Room – The good news is that it isn’t that hard to get him to move, Column “Beg To Differ”

Comparing Heirarchical Bayes and Latent Class Choice: Practical Issues for Sparse Data Sets

Is Max/Diff Really All That? As long as people make mistakes, the answer is yes, Column “Beg To Differ”

A Method For Handling a Large Number of Attributes in Full-Profile Trade-Off Studies

Putting the Why Into Conjoint: A Proposed Enhancement to Conjoint Analysis

Single Task Choice: Pipe Dream or Possibility?, Column “Beg To Differ”

Trade-off Study Sample Size: How Low Can We Go?

A User’s Guide to Conjoint Analysis