Factorial Designs (In depth)

Factorial Design

• Examines the effect of more than one independent variable at the same time to see how they interact.

  • Ex) In a study examining the effect of therapy type, (EMST vs. LSVT vs. EMST & LSVT) and session frequency (two times a week vs. four times a week) on vocal loudness, researchers compare how the two independent variables influence vocal outcomes and whether there is an interaction between them. 

2 by 2 independent design

• Factorial design with two independent variables, each with two levels, where participants are assigned to each condition. This allows researchers to examine main effects of each IV and their interaction, while keeping groups independent.

  • Ex) In a study investigating the effects of therapy type (EMST vs. LSVT) and diagnosis (MS vs. PD) on vocal loudness, participants would be assigned to one of the four groups MS with EMST, MS with LSVT, PD with EMST or PD with LSVT and vocal loudness would be measured after the intervention. This allows researchers to determine whether therapy type and diagnosis independently influence outcomes (vocal loudness) and whether the combination of factors produces a different effect than either factor alone.

2 by 2 repeated measures 

• Factorial design where two independent variables, each with two levels, are tested within the same participants. Every participant experiences all four combinations of IV levels. This allows researchers to examine main effects and interaction effects while controlling for individual differences. 

  • Ex) In a study investigating the effects of therapy type (EMST vs. LSVT) and task (reading task vs. conversation) on vocal loudness in Parkinsons Disease, all participants would be assigned to all four conditions: EMST + Reading task, EMST + Conversational task, LSVT + Reading task, and LSVT + conversational task. This allows researchers to determine whether therapy type and task independently influence outcomes (vocal loudness) and whether the effect of therapy type differs depending on the task. 

2 by 2 mixed 

• Factorial design with two independent variables, where one is a between-subjects factor and one is a within-subjects factor. This design allows researchers to examine the main effects of each variable and the interaction between them.

  • Ex) In a study investigating the effects of therapy type (EMST vs. LSVT) on vocal loudness in indivduals with Parkinson’s Disease, particpants are randomly assigned to receive either EMST or LSVT therapy. Vocal loudness is measured for all participants before and after the intervention. This mixed design allows researchers to examine whether therapy type and time each affect vocal loudness on their own, and whether the amount of improvement from pre- to post-treatment differs depending on therapy type.