Features of correlational studies

Correlational studies examine the relationship between two variables without manipulating them. This approach is useful in psychology for exploring connections that might be difficult to test through experiments.

Key elements of correlations

• Co-variables - These are the two variables being studied in a correlation; they are measured to see if they relate to each other, rather than one being changed to affect the other.
• Data collection - For each participant, measurements are taken on both co-variables. These can come from various sources, such as questionnaires, observations, or tests.
• Scatter graphs - The data points are plotted on a graph with one co-variable on the x-axis and the other on the y-axis. This visual tool helps reveal any patterns in the relationship.
• Purpose and applications - Correlations are often chosen when it's not practical or ethical to control variables, allowing researchers to investigate real-world links without direct intervention.

Examples of correlations in psychological research

Psychological studies frequently use correlations to explore relationships in areas like sleep, mindfulness, and social cognition. Below are key examples, broken down for clarity.

Dement and Kleitman: Sleep and dreams

• Aim - This study looked at the link between rapid eye movement (REM) sleep – a stage of sleep where eyes move quickly under closed lids – and dream content.
• Method - Participants were woken during REM sleep and asked to describe their dreams; time in REM and word count in narratives were measured.
• Results - A positive relationship was found: longer REM periods corresponded to more detailed dream descriptions.
• Conclusions - This suggests REM sleep is associated with vivid dreaming, prompting further research into sleep stages.

Hölzel et al.: Mindfulness and brain scans

• Aim - This research examined mindfulness – a practice of focused attention and awareness – and its effects on brain structure.
• Method - Participants in a mindfulness-based stress reduction (MBSR) programme had brain scans before and after; time spent on exercises and changes in grey matter – the brain tissue involved in processing information – were measured.
• Results - No consistent relationship was found between extra mindfulness time outside sessions and grey matter increases.
• Conclusions - Core programme activities may drive brain changes, but additional practice time did not strengthen the effect.

Baron-Cohen et al.: Revised Eyes Test

• Aim - This study investigated autism spectrum quotient (AQ) – a score measuring traits associated with autism spectrum conditions – and performance on a test of emotion recognition.
• Method - Participants completed the AQ questionnaire and the Revised Eyes Test, which involves identifying emotions from eye images; scores on both were compared.
• Results - A negative relationship emerged: higher AQ scores were linked to lower accuracy on the Eyes Test.
• Conclusions - This highlights potential difficulties in social cognition for those with higher autism traits, informing theories of empathy.

Operationalising co-variables for hypotheses

To make correlations testable, researchers must clearly define the co-variables. This process, called operationalisation, involves specifying exactly what each variable is and how it will be measured.

Example of operationalised co-variables:

• Co-variable 1 - Scores on a memory test, measured out of 50 possible points.
• Co-variable 2 - Hours spent revising in one week, recorded via self-report logs.

Clear operationalisation enables the formation of hypotheses, such as predicting a positive link between revision time and test scores, and supports reliable replication.

Types of correlations: positive, negative, and none

Correlations describe how co-variables relate: they can move together, in opposite directions, or show no pattern. Understanding these types helps interpret psychological data.

Positive correlation

This occurs when both co-variables change in the same direction. As one increases, so does the other; as one decreases, the other follows.

For example, in Dement and Kleitman's study, more time in REM sleep was associated with longer dream narratives, showing both measures rising together.

Negative correlation

Here, co-variables move in opposite directions. When one increases, the other decreases.

For example, Baron-Cohen et al. found that higher AQ scores corresponded to lower Eyes Test performance, illustrating an inverse relationship.

No correlation

This is when there is no consistent pattern between the co-variables; changes in one do not predict changes in the other.

For example, Hölzel et al. observed no link between additional mindfulness time and grey matter changes, with data points scattered randomly.

Recognising these types is essential for analysing whether variables are associated and in what way.

Understanding correlation coefficients

A correlation coefficient is a numerical value that quantifies the strength and direction of the relationship between co-variables. It provides a precise measure beyond visual graphs.

Key features of correlation coefficients

• Range - Coefficients range from -1 to +1.
  • A positive value (e.g., +0.8) indicates a positive correlation.
  • A negative value (e.g., -0.8) indicates a negative correlation.
  • Zero means no correlation.
• Strength - The closer to +1 or -1, the stronger the relationship. For example, +0.75 or -0.75 suggests a strong link, while values near 0 show a weak or absent connection.
• Interpretation - Coefficients help compare correlations objectively; a value of +0.9 is stronger than +0.5, even if both are positive.

This tool allows psychologists to assess how reliably co-variables are linked.

Strengths and weaknesses of correlational research

Like all methods, correlations have advantages and limitations. Evaluating them helps understand their role in psychology.