Variable

A variable is any characteristic that can take different values across the units in your study — age, income, satisfaction, treatment condition. Anything that does not vary is a constant and cannot explain variation in an outcome.

Variables are classified by role and by measurement level. By role: independent (predictor), dependent (outcome), mediator (mechanism), moderator (condition), and control or confounding variables held constant to rule out alternative explanations. By measurement level: nominal, ordinal, interval and ratio.

The measurement level determines which statistics are legitimate. Nominal data supports frequencies, mode and chi-square; ordinal supports median and non-parametric tests; interval and ratio support means, standard deviations and the full parametric toolkit. Build a variable table in your methodology chapter listing each variable, its role, level and source.