Three ways to perceive reliability
When measuring a phenomenon, you need to select the right tool. To measure temperature, you use a thermometer; alcohol content in breath is measured with an alcometer; and weight,...
Is gambling addictive? What if the respondent has no opinion?
The standard approach to investigating a single categorical variable involves a few elements.
Which car to buy: Marginal gains from chart margins
When it comes to charts ‘less’ is not necessarily better and ‘more’ can often lead to more functional data visualisation.
Trees that grow from tables
When stepping outside the domain of distributive and descriptive statistics for individual variables, we usually take interest in correlations between variables.
Weights and Tables
When creating a table, you can present weighted or unweighted data. As usual, the answer depends on the situation. You will get my meaning in a minute, but first, let me explain...
Multiple choice: which percentage base to use?
Multiple choice questions where the respondent may select more than one answer are commonly used in surveys. How should the responses to such questions be presented in a table? In...
Cross-selling and tables
What combination of products frequently end up in a shopper's basket? What services are usually bought together by consumers?
The Marimekko chart or analytical patchwork
What data can we show on a regular bar chart? How many variables can we use? For those who answered ‘not enough’ the Marimekko chart may be the answer.
Crosstabs as charts: introducing the Contingency Map
Tabular presentation is one of the most universal reporting techniques.