Blog

Violin plot

Before starting a more complex data analysis, it is worth taking a closer look at variable distributions which are of interest to us.

Correspondence analysis: solution quality assessment

Correspondence analysis is a technique often thought of as an alternative to a crosstab (contingency table).

Basic steps when design tables

Humans presented data in tabular format already in the ancient times. Thus, sometimes we can be deluded that structuring and reading tables is intuitive and does not require expert knowledge

Perceptual maps: choosing the correct normalization method

In this article you can find perceptual maps prepared using different normalization methods.

Dartboards and dashboards

The dartboard visualization is another useful medium for presenting the results of analyses on a dashboard.

Speedometers in reporting

Speedometers, or gauges, are very intuitive hence their popularity in dashboarding. But they evoke mixed feelings among analysts.

Deciding to remove an item from the scale

Sometimes instead of asking a single question in a survey, we ask a whole set of questions, which are intended to be used later to build a scale.

What do waterfalls have to do with data visualization?

A waterfall graph, otherwise known as a cascade chart, may mean very little to those of us from outside the financial services industry.

How to prepare for reliability analysis?

The area of reliability analysis is of interest first of all to those who plan to use scales and indices in their studies, namely more complex indicators than a single survey question.

Thermometers on dashboards

What are the markers of a good dashboard? Why use a thermometer in it?

From crosstab to correspondence analysis: how exactly are the rows and columns of a crosstab transformed into coordinates used in a chart?

Reduction of multidimensionality

Is a simple two-dimensional coordinate system always enough for us?

Bullet graph

How can we improve a bar chart? Of course, it depends on what we want to achieve.

When letters count in a table

A common task analysts are often faced with is to establish whether there are any relationships between variables and, if there is a, what is the nature of the relationship.

A table or a chart? Why not both? Hollywood film production costs and the use of a combined table chart

Welcome to the world of optimal scaling, I.E. The conversion of a qualitative variable into a quantitative one

Dashboarding – reporting for decision-making

The dashboard is a type of panel board, a control panel, or instrument board.

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

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 briefly what weighting adjustment is.

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 this post we will present two methods and discuss the differences between them and when to use them.

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.

How to change how we think about visualization?

When presenting results, we often employ various types of charts that best suit the data. We want the presentation to be captivating.