correlations

Correlations

Overview

  • Correlation analysis assesses relationships between co-variables.
  • Important to differentiate correlations from experiments.
  • Proper terminology suggests correlation is a method of analysis, not a research method.
  • Despite this, "correlation" is commonly used to refer to studies using correlational analysis.

Key Terms

  • Correlation: A mathematical technique to investigate the association between two variables known as co-variables.
  • Co-variables: The variables that are investigated in a correlation (e.g., height and weight). They differ from independent and dependent variables as correlation seeks to explore associations rather than cause-and-effect relationships.

Types of Correlation

  1. Positive Correlation:

    • When one co-variable increases, the other also increases.
    • Example: Higher number of people in a room is associated with increased noise.

  2. Negative Correlation:

    • When one co-variable increases, the other decreases.
    • Example: Increased number of people in a room correlates with reduced personal space.

  3. Zero Correlation:

    • Indicates no relationship between co-variables.
    • Example: The number of people in a room in Manchester versus total daily rainfall in Peru is considered likely to show no correlation.

Scattergrams and Interpretation

  • Correlations are represented visually on scattergrams, where each point indicates the x and y positions of the respective co-variables.
    • Positive Correlation: Indicates direct association between variables (e.g., caffeine consumption and anxiety levels).
    • Negative Correlation: Indicates inverse association (e.g., caffeine consumption versus hours of sleep).
    • Zero Correlation: No noticeable trend exists (e.g., caffeine consumption vs number of dogs seen).

Differences Between Correlation and Experiments

  • Experiments: Researchers manipulate the independent variable (IV) to measure changes in the dependent variable (DV). This allows for cause-and-effect inferences.
  • Correlations: No manipulation of variables occurs, so establishing causation is impossible. For example, a strong positive correlation between caffeine and anxiety does not imply caffeine induces anxiety.

Evaluation of Correlations

Strengths

  • Useful as a preliminary research tool; helps identify potential relationships for further exploration.
  • Provide quantifiable measures showing how two variables are related.
  • Quick and economical to conduct; does not require controlled environments.
  • Can use secondary data, reducing time spent on data collection.

Limitations

  • Cannot ascertain causation; only indicates how variables relate.
  • Possible ambiguity regarding which co-variable influences the other. For instance, anxious individuals may consume more caffeine rather than caffeine causing anxiety.
  • Uncontrolled third variables may influence correlations (i.e., intervening variables).
    • Example: Job stress could lead to increased caffeine consumption and anxiety.
  • Correlations can be misinterpreted, leading to false conclusions about causality.
    • Example: Single-parent families and crime association does not imply causation; intervening factors may play a role.

Correlational Hypotheses

  • Hypotheses for correlational studies differ from those for experiments due to the absence of IVs and DVs. Must clearly state expected relationships between operationalised co-variables.
    • Directional Hypothesis: e.g., "There is a positive correlation between the price of a chocolate bar and its tastiness rating (out of 20)."
    • Non-Directional Hypothesis: e.g., "There is a correlation between the price of a chocolate bar and its tastiness rating (out of 20)."

Curvilinear Relationships

  • Some relationships are not strictly positive or negative.
  • Yerkes-Dodson Law: Indicates optimal performance at moderate arousal levels; performance declines with too low or high arousal levels.
    • Question: Suggest pairs of co-variables demonstrating curvilinear relationships.

Check Questions

  1. What is meant by a "correlation"?

    • A correlation describes the relationship between two co-variables, highlighting their associations without implying causation. (2 marks)

  2. Explain one strength and one limitation of correlations in psychological research.

    • Strength: Correlations can reveal potential relationships between variables efficiently. (3 marks)
    • Limitation: Correlations do not establish cause and effect, limiting interpretative power. (3 marks)

  3. Differences between positive and negative correlations:

    • Positive correlation example: More aggressive parents correlate with more aggressive children.
    • Negative correlation example: Higher temperature leads to fewer clothes worn by people. (4 marks)