Module 5 - Research in Communication Disorders: Group Designs pt. 2

Single-Group Research Design

Research design where you observe one group of participants in two or more conditions

Pre-test and post-test

When are single-subject group research designs typically observed?

Causality

The link between the implementation of the IV and the impact it has on the DV

Exploratory

The more _____ in nature the starting off point it is, the more promise the IV and if anything is worth exploring

Weaker

Is a single-group research design considered a weaker research design?

Randomization

A powerful thing as it is a strategy when you randomly assign someone that reduces research bias and more

Researcher bias

What does randomization reduce?

Validity

What adds to statistical tests?

Confounding / extraneous variables

What does randomization minimize?

Participants are treated equally

What does randomization ensure?

• Computer generated random assignment

• Random number tables

• Shuffled cards or tossed coins

What are some ways randomization achieved?

You can’t randomly assign

When do you use matching?

Group equivalency

What does matching attempt to establish?

How well the participants are matched

What does internal validity depend on?

Random assignment

What is matching inferior to?

Matching

• The step that is used when you can’t randomly assign

  • If you randomly assign and have a small sample size and you don’t have initial group equivalency

Blinding

This reduces participant and investigator biases as participants won’t alter their behaviors and investigators won’T consciously or unconsciously influence outcomes

Behaviors

With blinding, we don’t want participants to alter what?

Outcomes

With blinding, we don’t want investigators to consciously or unconsciously influence what?

Two Group Research Design

Where you have to come up with a plan for assigning people to groups through random assignment or matching

Group equivalence

What is important for two group research designs?

Significant outcome

Large differences between the experimental and control groups indicate what?

Independent Research Design

Require random assignment into the treatment and control groups where random assignment reduces researcher biases and ensures compatibility between the groups since they are assumed to be equivalent

Related Research Design

Similar to quasi-experimental counterpart that uses matching instead of random assignment

Complex Research Designs

This type of research design is either looking at comparing several treatments or trying to mimic these types of real-life situations where multiple factors are all playing together

Conditions

What does complex research designs add to the independent variable?

Solomon 4 Group Design

What an example of a complex research design?

Multivalent complex research design

Looking at multiple factors to try and understand the different things that may impact someone’s response to the intervention

Factorial complex research design

Tries to model itself after real life that can be hard to interpret with a combination of IVs that may affect the DV

• Time in therapy could have 3 levels: 1, 2, and 4 hours

• Service delivery could have 3 levels: pull out, classroom-based, and consultation

• Two factors with 3 levels

What are 3 levels of IV in factorial complex research designs?

If variables coexist and if they may affect behaviors in combination

What does factorial complex research design allow researchers to study?

Multiple-Group Designs

Two or more comparison / control groups