Market Research.

Market research is used to figure out how things work. This is often by systematically gathering data about people or companies using surveys.

Here, we write about the main market research techniques. These techniques include correspondence analysis, MaxDiff, and choice modeling. We also share tips and tricks for significance testing. And how to best present market research findings using PowerPoint and online dashboards (often considered a better alternative to PowerPoint) are explored.

Statistical testing of net promoter score
Statistical Testing of the Net Promoter Score
20 March 2018 | by Justin Yap

  A previous blog post described a smart way of computing the Net Promoter Score by recoding. The idea was to recode NPS ratings from 0 to…

Continue reading

Checking Convergence When Using Hierarchical Bayes for MaxDiff
16 January 2018 | by Justin Yap

Hierarchical Bayes is an approach to modelling MaxDiff data which can provide better predictive accuracy than other approaches (like latent class analysis). However, it is…

Continue reading

Using Q
Using Hierarchical Bayes for MaxDiff in Q
16 January 2018 | by Justin Yap

In this blog post I describe how to run Hierarchical Bayes for MaxDiff in Q, and explain the options and outputs available. Getting started Your…

Continue reading

Using Displayr
Using Hierarchical Bayes for MaxDiff in Displayr
16 January 2018 | by Justin Yap

In this blog post I describe how to run Hierarchical Bayes for MaxDiff in Displayr, and explain the options and outputs available. Getting started Your…

Continue reading

Using Displayr
Case Study: Updating Reporting
04 December 2017 | by Tim Bock

This is the sixth in a series of blog posts designed as a DIY training course for using Displayr. It illustrates how to update analyses and reporting…

Continue reading

NPS column chart with sparklines
Using Sparklines to Show Trends in Bar and Column Charts
09 November 2017 | by Tim Bock

Adding sparklines to a line chart that has multiple series of data can vastly improve readability and interpretation. Line charts are one of the standard…

Continue reading

Comparing Tricked Logit and Rank-Ordered Logit with Ties for MaxDiff
17 October 2017 | by Justin Yap

In this blog post, I compare the two ways in which MaxDiff data is treated in analyses such as latent class analysis and Hierarchical Bayes….

Continue reading

Image of dirty laundry in a basket
Automatic Data Checking: Let Your Analytics Package Do Your Dirty Laundry for You
16 October 2017 | by Matt Steele

Data checking is a bore (and a step we skip at our peril!). The laundry-list of data checking tasks can be seemingly endless: identifying bogus respondents,…

Continue reading

Great looking dashboard
Create and Update PowerPoint Reports using R
31 August 2017 | by Tim Bock

In my sordid past, I was a data science consultant. One thing about data science that they don’t teach you at school is that senior managers…

Continue reading

Choice Simulator
How to Create an Online Choice Simulator
22 August 2017 | by Tim Bock

The seven steps to create an online choice simulator are detailed in the post. In addition, an interactive example of a choice simulator is provided…

Continue reading

Adding Supplementary Points to a Correspondence Analysis
17 August 2017 | by Jake Hoare

Retrospectively adding supplementary points to a correspondence analysis can greatly assist in the interpretation of results. In other words, including supplementary row or column points…

Continue reading

Moonplot to show brand mapping from a correspondence analysis
Moonplots: A Better Visualization for Brand Maps
15 August 2017 | by Tim Bock

Moonplots are a better way to visualize brand maps than standard correspondence analysis outputs, which are often difficult to read correctly. The Moonplot resolves the…

Continue reading

MaxDiff analysis example
MaxDiff: 11 Tips to Do Your Own Analysis
03 August 2017 | by Tim Bock

MaxDiff analysis is one of those advanced techniques that can be run by any good quant researcher. In this post, I share 11 tips to…

Continue reading

experimental design
How to Check an Experimental Design (MaxDiff, Choice Modeling)
25 July 2017 | by Tim Bock

In this post, I explain the basic process that I tend to follow when doing a rough-and-ready check of an experimental design. The last step,…

Continue reading

Correspondence Analysis for brand switching
Correspondence Analysis of Square Tables
25 July 2017 | by Jake Hoare

Square tables are data tables where the rows and columns have the same labels, commonly seen as a crosstab of brand switching or brand repertoire data….

Continue reading

Put PowerPoint into Cruise Control: How to Automatically Update Your Reports
14 July 2017 | by Matt Steele

The ability to automatically update PowerPoint slides with new data can save time, money, error, and your sanity. Some analysis software packages allow your reporting to go into…

Continue reading

Using Q
Customization of Bubble Charts for Correspondence Analysis in Q
08 July 2017 | by Tim Bock

When you insert a bubble chart in Q (Create > Charts > Visualization > Labeled Bubbleplot), you can customize some aspects of its appearance from the…

Continue reading

correspondence analysis
Using Bubble Charts to Show Significant Relationships and Residuals in Correspondence Analysis
08 July 2017 | by Tim Bock

While correspondence analysis does a great job at highlighting relationships in large tables, a practical problem is that correspondence analysis only shows the strongest relationships, and sometimes…

Continue reading

Keep updated with the latest in data science.