What’s the difference between dependent and independent variables?

/in /by

Screen Shot 2013-12-29 at 8.05.00 PMby Adar Ben-Eliyahu

  • Independent Variables

Independent variables are the characteristics that we think predict the outcomes of interest. In mentoring research, we might think that such factors as the mentors’ age, the youth gender, the quality of the relationship, and the programs’ level of supervision would predict better youth outcomes. In this case, those factors would be the predictors or independent variables.

  • Dependent Variables

Dependent variables are the main outcomes or characteristics of interest. In mentoring research, we might think that the program affects youth grades, self-esteem, and delinquency. If so, those factors would be our dependent variables in statistical analysis.

  • Other Variables (Confounding, Mediating and Moderating)

Confounding Variable – A confounding variable is a variable that co-occurs with the independent variable and offers a different explanation of the results. For example, we might find that match length (an independent variable) predicts youth well being (the dependent variable).  This could be the case, but it might also be the case that youth with better social skills are better able to develop and maintain strong relationship. In this sense, a confounding variable provides alternative explanation to the findings.

Mediating Variable – A mediating variable, or mediator, helps to explain the relationship between the independent and dependent variable. For example, Rhodes (2005) has proposed that mentoring (the independent variable) leads to various positive outcomes (dependent variables) via youth’s improved relationships with parents.

Moderating Variable – A moderating variable, or moderator, is a variable that heightens or diminishes the relationship between the independent and dependent variables. For example, the relationship between mentoring and outcomes may vary, depending on the gender of the mentee. In this case, gender is a moderator.