Skip to main content
SAGE
Search form
  • 00:06

    SPEAKER 1: Chapter 12 focused on covariations,which is when there are relationshipsbetween the independent variables themselves.Covariations can have consequences for the effectsizes that we measure between the independent variablesand the dependent variables.When we use multiple independent variables in psychology,there is a high chance that there will be some covariation.

  • 00:29

    SPEAKER 2: Let's look at variables and routes.The variables are like containersthat hold information.We can think of the links that connect variablesas being like pathways along which information flows.

  • 00:42

    SPEAKER 1: The presence of covariation leadsto there being multiple routes betweeneach independent variable and the dependent variable.Direct routes are links that connect to variableswithout any intervening variables.Indirect routes are sequences of direct linksfrom one independent variable to another and onwardsto the dependent variable.

  • 01:03

    SPEAKER 2: Now, let's look at howcovariation changes the effect sizes that we measure.There are three different ways of describingthe sizes of effects.Total effect size, the effect sizeyou would see if you just used a single independent variable.This is the sum of all the different routes,direct and indirect, from a single independent variableto a dependent variable.

  • 01:29

    SPEAKER 2 [continued]: Unique effect size, the effect sizeseen when all covariation between independent variablesis removed.The unique effect size is that partof the effect of an independent variablethat it doesn't share with any other independent variables.Direct effect size, the effect sizethat determines the flow of informationthrough a direct link.

  • 01:56

    SPEAKER 1: For effect sizes for a single independent variableor multiple independent variablesthat are all independent of each other, the total, unique,and direct effect sizes are all the same.When multiple independent variables covary,the different types of effect sizesare all different and may even have quite different meanings.

  • 02:17

    SPEAKER 2: How to untangle all of this?There are two rules for combining effect sizes.First, create a sequence of links.Multiply the direct effect sizes along the sequence together.Second, add effect sizes of parallel routes startingand ending with the same two variables.

  • 02:40

    SPEAKER 1: The presence of covariationoften requires us to change our understanding of the connectionbetween the independent variable and the dependent variable.And it points to the underlying psychological meaningof independent variables.For more information, see study.sagepub.com/statisticsforpsychology.

Video Info

Publisher: SAGE Publications Ltd.

Publication Year: 2019

Video Type:Tutorial

Methods: Covariance, Independent variables, Dependent variables

Keywords: covariance; effect size; independent variables; research methods

Segment Info

Segment Num.: 1

Persons Discussed:

Events Discussed:

Keywords:

Abstract

Covariations, the relationships between independent variables, are summarized, including their influence on effect size.

Looks like you do not have access to this content.

Covariations

Covariations, the relationships between independent variables, are summarized, including their influence on effect size.

Copy and paste the following HTML into your website