Consumer Preference Research

Conjoint research involves conditional logistics regression, or conditional logit model, developed by McFadden and the bias-corrected maximum likelihood estimator described by Firth. In this research, JMP was used for analysis of preference of tactile varnishes on consumer cosmetic packaging. 

The results of this study confirmed up to 5% price utility for specific applications of decorative coatings used in packaging of cosmetics valued around $50.00.
Marginal utility of coating by price

This research provided evidence that decorative coatings, and particularly "Hi Rise" coating, were preferred over no coating by consumers.

Utility for coatings at $52.15 price

This conjoint study provided evidence for 5% increased price utility.

View Study

Market Research Study

The 2022 PI Sales Compensation Study was published by Americas Printing Association Network. This sample (redacted sections) shows the work involved in surveying more than 100 printing companies and creating visuals that provide quick and insightful information for benchmarking sales professionals.

PI Sales image


View Sample

Experimental Design

In this DOE study, our team explored the relationships of input variables in the printability of fine silver lines by screen printing to determine the optimal settings considering the interactions of these variables in the process. Optimal settings are summarized below in Table III.
Input variables
Anova
LS Means 1
LS Means 2
Optimized settings
View Abstract

Analysis of Variance

This research used Analysis of Variance (Anova) and post-hoc analysis using Tukey’s HSD to determine if differences in color measurement devices were statistically significant or merely a result of chance. Tukey’s HSD, a T-test that corrects for family-wise error, conducts significance tests of all combinations of pairs..

This research focused on differences in measurement of several low-cost and professional colorimeters and spectrophotometers and confirmed that color measurements among most devices were statistically different.
Tukey
View Abstract
Next