Changing the way you learn

Other Experimental Designs

Small N Design
1. A design where each participant is treated as a separate experiment
2. N is refering to the number of participants
3. Can make a causal statement from a study with a single person as long as the study is designed appropriately
  1. Examples of Small N Designs
    1. ABA Designs
      1. Also known as Reversal Design
        Procedures include if represented in a clinical setting:
        Observe a baseline of behaviour before treatment
        Observe behaviour during treatment
        Remove treatment to see if behaviour reverts back or not
    2. Multiple Baseline Designs
      1. Multiple measures taken before and after the manipulation
4. Pros and Cons
  1. Cons
    1. A single reversal is not extremely power evidence
    2. Ethical concerns removing a beneficial treatment
    3. Cannot test hypotheses using statistical procedures
  2. Pros
    1. It is possible to replicate procedures
    2. Present individual rather than group data
    3. ABA designs can be improved by extending them to ABAB or ABABAB designs
      1. Increases statistical power
      2. More ethical ending with the treatment
Quasi-Experimental Design
1. Used when the control (e.g. Random Assignment) of true experiments cannot be achieved
2. Lower internal validity than true experiments
3. Six Different Designs
  1. One-group posttest only
    1. Between groups design
    2. The responses of a number of individuals exposed to the same event
  2. One-group pretest/posttest
    1. Adding a baseline measure to the one-group to provide a basis for comparison
  3. Non-equivalent control group design
    1. Between-groups design
    2. Separate control group introduced
    3. Creates Selection Differences
    4. Not equivalent due to lack of random assignment
  4. Non-equivalent control group pretest/posttest design
    1. An improvement on the non-equivalent control group design
    2. Addition of a pretest improves internal validity
    3. Can assess for difference prior to testing
  5. Interrupted time series design
    1. Examination of the dependent variable for a long period of time
      1. Before and after the IV is implemented
    2. Repeated measures design
  6. Control series design
    1. A improvement to the interrupted time series design
      1. Finding an appropriate "control group"
    2. Involves finding a similar population that did not receive a particular manipulation
4. Threats to internal validity
  1. History threats
    Annotations:
    - Some external or "historical" event affects most members of the treatment group at the same time as the treatment
  2. Selection Effects
    Annotations:
    - Unaccounted-for differences between groups
  3. Maturation Threat
    Annotations:
    - Spontaneous change not caused by the independent variable
  4. Testing effect
    Annotations:
    - An order effect in which there is a change in participants as a results of experiencing the Dependent Variable more than once
  5. Instrument decay
    Annotations:
    - Occurs when a measuring instrument changes over time
  6. Regression toward the mean
    Annotations:
    - Extremely low or extremely high performance at Time 1 is likely to be less extreme at Time 2 (i.e., closer to average
  7. Attrition threat
    Annotations:
    - People who completed the study vary systematically from those who did not.
Developmental Research Designs
1. Longitudinal
  1. Same group is observed at different times (e.g., as they age)
2. Sequential Method
  1. Combination of both Developmental Research Designs
3. Cross-Sectional Method
  1. People of different ages are measured at the same point in time

Next up

Other Experimental Designs

Description

Resource summary

Media attachments

Similar

	Created by Mahir Mohiuddin about 10 years ago