Standard Article

Causal Research

Part 2. Marketing Research

  1. Harmen Oppewal

Published Online: 15 DEC 2010

DOI: 10.1002/9781444316568.wiem02001

Wiley International Encyclopedia of Marketing

Wiley International Encyclopedia of Marketing

How to Cite

Oppewal, H. 2010. Causal Research. Wiley International Encyclopedia of Marketing. 2.

Author Information

  1. Monash University, Melbourne, Victoria, Australia

Publication History

  1. Published Online: 15 DEC 2010


Causal knowledge is one of the most useful types of knowledge. Causal research aims to investigate causal relationships and therefore always involves one or more independent variables (or hypothesized causes) and their relationships with one or multiple dependent variables. Causal relationships can be tested using statistical and econometric methods. However, in many cases conclusions about causality are stronger if they can be based on designed experiments. Experiments typically include one or more experimental conditions and a control condition. Random assignment of units to these conditions allows the application of statistical theory to infer the probability of effects being caused by other factors than the designed, manipulated factor. Designing an experiment involves making a series of decisions about, among others, the unit of observation, the setting in which the experiment is to be conducted, the creation of experimental stimuli and observation instruments, the number of conditions to create and how many test units to include, and how to assign units to the conditions. Design types vary based on whether they are quasi-experimental or true experiments, how many independent factors they include, whether they include all possible conditions or only a subset of conditions, and whether they expose units to single or multiple treatments.


  • causal relationships;
  • experimentation;
  • experimental design;
  • statistical modeling;
  • factorial designs;
  • blocking factors;
  • quasi-experiments;
  • experimental treatments;
  • experimental observations;
  • pretest effects;
  • within and between-participant designs