Module 6/7

Types of Studies

• Two Main Types: Observational Studies and Experiments

Observational Study

• Definition: A study where variables are simply observed without manipulation.

• Approach: Hands-off; data results occur naturally.

• Conclusions: Allow for associations, links, or correlations between variables. Cannot infer cause and effect.

Experiment

• Definition: Involves intentionally manipulating a variable to observe effects.

• Purpose: To show cause and effect relationships.

• Approach: Hands-on; control and manipulate variables such as whether participants watch TV while snacking.

• Conclusions: Can determine cause and effect based on manipulation of independent variables.

Experiment Designs:

• Design A: Assign participants to different environments (TV on vs. TV off) to see snacking habits (Experiment).

• Design B: Participants log eating times after watching TV (Observational).

• Design C: Participants record previous day’s eating times while watching TV (Observational).

• Design D: Conduct a sample survey on opinions about snacking and TV (Sample Survey).

Research Questions

• Well-Posed Research Question Characteristics:

  • Clear population identification.

  • Clear variables identified.

  • A population value for comparison.

• Example Question: "Do you snack more while watching TV?" (Not well-posed due to unclear population).

Vocabulary and Key Concepts

• Prospective Data: Collecting data moving forward in time.

• Retrospective Data: Collecting data backward in time.

• Random Selection: Ensures all individuals have an equal opportunity to participate.

• Random Assignment: Ensures fairness in assigning participants to treatment groups.

• Control Group: A baseline for comparison that receives no treatment.

• Blinding: Participants or researchers unaware of group assignments to prevent bias.

• Placebo Effect: Improvement in condition due to the belief in treatment, not the treatment itself.

Research Pitfalls

• Hawthorne Effect: Changes in behavior when individuals know they are being observed.

• Noncompliance: Failure to adhere to study protocols (e.g., smoking a cigarette when in a non-smoking group).

• Lack of Realism: Study settings may not accurately replicate real-life situations, affecting results.

• Dishonesty: Participants may not truthfully disclose sensitive information, impacting data quality. Solutions include anonymity or confidentiality.

Data Collection: Sample Surveys

• Types of Surveys: In-person, online, telephone, or email.

• Cautions in Surveys:

  • Avoid trigger words that elicit strong emotions.

  • Use clear, unbiased questions to obtain valid responses.

• Question Types:

  • Open Questions: Unlimited response options.

  • Leading Questions: Involve suggested answers that guide responses.

  • Sensitive Questions: Requires confidentiality for honesty.

Additional Topics for Review

• Types of Variables: Categorical and Quantitative.

• Types of Sampling Plans: Volunteer, Convenience, Cluster, Stratified, Systematic, Census.

• Recognizing Conclusions: Distinguishing valid conclusions from observational studies (only correlations) versus experiments (cause and effect).

Important Considerations for Exam Preparation

• Be able to identify observational studies vs. experiments from scenarios.

• Understand the implications of study designs and potential biases.

• Familiarize with vocabulary relevant to experiments and surveys.