Experimental Methods in Research

Experimental Methods

• Aim: A general statement of the researcher's plan for investigation.
• Hypothesis: A precise statement of the relationship between variables.
• Independent Variable (IV): Manipulated variable, impacting the dependent variable.
• Dependent Variable (DV): Measured variable, responding to changes in the IV.

Types of Experiments

• Laboratory Experiment: Conductec in a controlled environment, allowing high control over variables for replicability; however, limited ecological validity and can be affected by demand characteristics.

• Field Experiment: Conducted in a natural environment with some control over variables, resulting in higher ecological validity and observation of natural behaviour. However, lower control over extraneous variables and ethical concerns like privacy could be present.

• Natural Experiment: An experiment where the IV occurs naturally (e.g., natural disaster). Allows studying events that cannot be manipulated but are hard to replicate and confounding variables are hard to control. High external validity.

Sampling Methods

• Opportunity Sampling: Participants recruited based on availability, quick and cost-effective but not representative and can be affected by researcher bias.

• Random Sampling: Every member has an equal chance of selection, minimising researcher bias and commonly representative but can be time-consuming and may not represent the population, due to refusals.

Observational Techniques

• Naturalistic Observation: Observing behavior in its natural setting; high ecological validity but uncontrolled extraneous variables and challenging to replicate.

• Covert Observation: Participants are unaware of being observed, reducing demand characteristics and capturing authentic behaviour but raises ethical concerns regarding informed consent.

Data Analysis

• Measures of Central Tendency: Ways to calculate an average (mean, median, mode), summarising large datasets and identifying central patterns. However, each measure has limitations, such as the mean influenced by outliers.

• Qualitative Data: Non-numerical data (interviews, text) rich in detail and captures participant insight, but difficult for analysis and subjective interpretations can bias conclusions.

• Quantitative Data: Numerical data (test scores, measurements), easy to analyze and allows comparison through statistical methods, but lacks depth and context.

Ethical Issues

• Informed Consent: Participants are fully aware of study aims and potential risks before participation for ethical conduct and to build trust.

• Deception: Misleading participants about the true purpose of the study to permit natural behaviour, but can cause distress if participants feel betrayed.