experimental design

Experimental design is crucial for determining how different conditions affect participants' performance on dependent variables (DVs).
Examples: Looking at sleep duration (eight hours or more versus less than eight hours) and the impact of watching TV while doing homework (TV on vs. TV off).

Repeated Measures Design
- Definition: All participants receive all levels of the independent variable (IV).
- Example: Each participant engages in a task with the TV on, and potentially a week later, they engage in a similar task without the TV.
- Data Comparison: The performance (DV) of the same participant is compared across the two scenarios.
- Key Considerations:
  - High control of variables since the same participants are used.
  - Example: Performance on a memory test is compared for each individual across conditions.
Independent Groups Design
- Definition: Participants are divided into separate (independent) groups with each group being assigned to one level of the IV.
- Example: Group A performs the task with the TV on, while Group B performs the task with no TV.
- Data Comparison: The performance (DV) is compared between the two distinct groups.
Matched Pairs Design
- Definition: Combines aspects of independent groups and repeated measures by matching participants on key characteristics before allocating them into groups.
- Example: One member of each matched pair is assigned to Group A, and the other to Group B.
- Matching Variables: Must align with the purpose of the study, such as matching participants based on characteristics that impact performance (e.g., time spent watching TV), but not irrelevant characteristics (e.g., gender unless evidence suggests a relevant difference in memory).

Advantages:
- Control over participant variables since all participants undergo both conditions.
Disadvantages:
1. Potential for order effects, affecting performance due to:
- Practice Effects: Improved performance in the second test due to prior exposure.
- Boredom Effects: Diminished performance from repetition.
1. Participants may guess the experiment's purpose during the second test, possibly altering their behavior.

Advantages:
- Eliminates order effects since different participants are used for each condition.
- Less likely to be affected by participants guessing the experiment's purpose.
Disadvantages:
1. Cannot control differences in participant characteristics between groups, leading to confounding variables (e.g., memory ability differences between Group A and Group B).
2. Requires more participants to achieve equivalent data due to the grouping structure.

Advantages:
- Reduces participant variable differences since participants are matched prior to assignment in experimental groups.
Disadvantages:
1. Time-consuming and challenging to find suitable matched pairs based on important variables.
2. May not control all participant variables, as only known relevant variables can be matched; unknown variables could still influence results.

To reduce practice effects, researchers can administer different tests that are equivalent but designed to avoid the exact same questions.
Counterbalancing: A method used to mitigate order effects by varying the order in which conditions are presented to participants.

Random allocation techniques (e.g., drawing names from a hat) can help distribute participant variables evenly across groups.
Consider conducting a pilot study to identify key variables to match on, thus simplifying the matching process.

When evaluating the strengths and weaknesses of each design, consider how control over participant variables, order effects, and demand characteristics may impact the study objectives.
Each design's advantages can often be assessed by reflecting on the disadvantages of the others.