Correlational Research

Pearson’s R

• Correlation coefficient that quantifies the strength of a correlation between quantitative variables

  • Good for measuring linear relations, NOT good for nonlinear relationships

  • Between -1 & 1

correlational research

• Non-experimental research, no manipulation of independent variable or causal conclusions

• Can test many different research questions

  • Is conscientiousness associated with lifespan?

  • Is neuroticism linked to cardiovascular disease?

  • How are antagonism and romantic satisfaction related?

reasons for correlational research

• Researchers don’t think relationship is causal or are not interested in causality

• Cannot manipulate independent variable 

  • Unethical, not possible, or not practical

• Establish reliability and validity of measures

• Can be higher in external validity than controlled experiments

• Can provide additional, converging evidence following experiments

restriction of range

• It is important to ensure variables in your study are not limited/restricted in range

  • Is there full coverage of the entire variable?

• Restriction of range can bias correlation coefficients, resulting in misleading statistical

the bidirectionality problem

• Correlational research cannot tell us the causal directionality of a relationship; X could cause Y, but Y could also cause X

  • Ex: Stress is correlated with insomnia (stress could cause poor sleep, but poor sleep could also cause stress)

the third variable problem

• Two variables could be related because of a THIRD unmeasured variable (or confound)

  • Ex: Drownings are correlated with ice cream sales

    • Third variable: temperature (people are more likely to be in pools, people are more likely to want to eat ice cream)

    • Spurious correlations

Spurious correlations

Correlations affected by confounding variables