A study of large carnivores analyzes whether there is a linear association between body length (meters) and weight (kilograms). Which method should be used?

A study of large carnivores analyzes whether there is a linear association between body length (meters) and weight (kilograms). Which method should be used?

• A. Chi-square test of independence
• B. Spearman correlation
• C. ANOVA
• D. Linear regression t-test

Answer: (D)

When you want to examine a straight-line relationship between two continuous measurements, use simple linear regression. You model weight as a function of body length, fitting a line that predicts weight from length. The slope tells how much weight changes with each additional meter of length. The t-test on that slope asks: is the slope different from zero? If yes, there is a significant linear association between length and weight, meaning longer carnivores tend to weigh more in a predictable, linear way. This approach also gives you the intercept, the slope estimate, and measures of fit to understand how well length explains weight. Chi-square is for relationships between categorical variables, so it’s not appropriate here. Spearman correlation checks monotonic (rank-based) associations and doesn’t directly test a linear slope. ANOVA compares means across categories, which isn’t the right framework when both variables are continuous.

When you want to examine a straight-line relationship between two continuous measurements, use simple linear regression. You model weight as a function of body length, fitting a line that predicts weight from length. The slope tells how much weight changes with each additional meter of length. The t-test on that slope asks: is the slope different from zero? If yes, there is a significant linear association between length and weight, meaning longer carnivores tend to weigh more in a predictable, linear way. This approach also gives you the intercept, the slope estimate, and measures of fit to understand how well length explains weight.

Chi-square is for relationships between categorical variables, so it’s not appropriate here. Spearman correlation checks monotonic (rank-based) associations and doesn’t directly test a linear slope. ANOVA compares means across categories, which isn’t the right framework when both variables are continuous.