How to Select Statistics for a Single Variable
How to Select Statistics for a Single Variable
Published on
11 January 2020

What are the different statistics available for a single variable? Selecting statistics for a single variable depends of the type of variable: nominal, ordinal, or interval.


OK, you have collected your data and want to learn more about a single variable in it. How can you do that? What kinds of statistical metrics (statistics) can you use?

The answer depends on the scale of measurement of that variable. The table below provides a summary of that:

Type of variableCentral TendencyFrequenciesDispersionSymmetryPeakenessNormality
NominalModeRelative frequencies (e.g. percentages) or absolute frequenciesThe relative frequency of modal value
OrdinalMedianRelative frequencies (e.g. percentages) or absolute frequencies, or N-tilesInter-quartile deviation
IntervalMean , Median, and other measures (see below)Relative frequencies (e.g. percentages) or absolute frequencies, or N-tilesStandard deviation or coefficient of variation, or rangeSkewnessKurtosisKolmogorov-Smirnov one sample test or Lilliefors extension of the Kolmogorov-Smirnov test, or the Chi-square goodness of fit test

Selecting an appropriate metric for central tendency of interval variables

This can be tricky. We suggest you should consider the following:

  1. Are there significant outliers (for which you may need to do a plot of the data)?
  2. Is the distribution skewed?

If there are outliers, we recommend doing one of the following transformations to your variable:

  1. Winsorizing the dataset
  2. Trimming it (i.e. removing extreme values)

If your distribution is not symmetric (i.e. is skewed), you should use both mean and median.


This post is based on A Guide for selecting statistical techniques for analyzing social science data.


Read these articles next:

Creating a perception map

Can you make perceptual maps from fake data? (A replication study)

This study investigates the claim that synthetic LLM-generated data could be used effectively for creating perceptual maps in market research contexts.

View article
Comparing the new and old WH Smith Logo

WH Smith Trial Rebrand — New vs. Old

To assess the impact of WH Smith's recent trial logo update, this Logo Test compares the newly introduced WHS logo with the WH Smith logo, which has been in use since 2004.

View article

Mini-lecture: Conjointly's guide to sample selection

This mini-lecture discusses the importance of sampling in research, and will walk you through the steps of finding the right sample for your experiment.

View article