The effects of third variables in psychological research

  • Manuel Ato Universidad de Murcia
  • Guillermo Vallejo Universidad de Oviedo
Keywords: mediation, moderation, covariation, spuriousness, confounding, suppression, third variables


Given a causal relationship between an independent variable (X) and a response variable (Y), the interest of some applied researchers is sometimes focused on knowing the role playing for alternative intervening variables (third variables or Z-variables), and particularly the role of mediating and moderating variables. Some influential works have repeatedly denounced that in applied psychology some confusion exists with respect to the meaning and appropriate use of third variables effects as causal hypotheses. This work is intended to reinforce the comprehension and using of five of more known effects of third variables and concretely the effects of covariation, spuriousness or confounding, suppression, and particularly of mediation and moderation, with special attention towards the application of statistical tests and the report of most common mistakes that usually appear when we tackle a causal analysis. An accessible language for applied researcher with path diagrams is used. A flow diagram is finally proposed in order to help applied researchers to distinguish between third variable effects and to make easier the application of the appropriate statistical procedures.


Download data is not yet available.

Author Biography

Manuel Ato, Universidad de Murcia
Departamento de Psicología Básica y Metodología
How to Cite
Ato, M., & Vallejo, G. (1). The effects of third variables in psychological research. Anales De Psicología / Annals of Psychology, 27(2), 550-561. Retrieved from