2.3 Looking for Links: Descriptive/Correlational Research

The Concept of Correlation

• Positive: High scores on variable X are associated with high scores of variable Y; low scores on X are associated with low scores on Y
• Negative: High scores on variable X are associated with low scores of variable Y; low scores on X are associated with high scores on Y
• Correlation Coefficient: can very b/w 0 and +1.00 or 0 and -1.00. Coefficient near 0 means no relationship b/w variables. +1.00 or -1.00 = perfect relationship b/w two variables. Closer to +1.00 or -1.00, stronger the relationship.
• Does NOT tell us causation; only that X and Y are related in some way. We can't know if X causes Y, Y causes X, or if they're both caused by a 3rd variable.

Naturalistic Observation

• *Particularly useful in studying animals in their natural habitats
• Major strength: allows researchers to study behavior under conditions less artificial than in experiments. Good to use as a starting point when little is known about the behavior under study. Can be used to study animal behavior
• Major weakness: Researchers have trouble making observations unobtrusively so as to not affect participants' behavior. Difficult to translate naturalistic observations into numerical data that permit precise statistical analyses

Case Studies

• Variety of data collection techniques can be used: interviewing subject/people close to subject, direct observation of participants, examination of records, psychological testing
• Used in clinical and research settings. When used in research, investigators look for threads of continuity across series of case studies/ analyze collection of case studies to look for patterns that permit general conclusions
• Advantage: well suited for investigating certain phenomena, esp., roots of psych. disorders & efficacy of selected therapeutic practices. Provide real-life illustrations that support hypothesis
• Disadvantage: highly subjective. Info needs to be pulled together in an impressionistic way, which lends to researchers' theoretical slants. Easy to see what you expect to see in the research.

Surveys

• Used to obtain info on aspects of behavior difficult to observe directly.
• Pros:
• easy to collect data on attitudes and opinions from large sample of participants
• Cons:
• Potential participation has declined in recent decades due to growing resentment of telemarketers and privacy/ID theft concerns
• Depends on self-report data which can be distorted by intentional deception, wishful thinking, memory lapses, and poorly worded questions

Pros & Cons of Correlational Research

• Advantage:
• Give researchers a way to explore questions unable to explore w/experiment
• Disadvantage:
• Can't prove causation