This experiment explored sex differences in behaviour using factor analysis to describe the relationship between different behavioral variables. A principal component solution with an orthogonal rotation of the factor matrix was used, ensuring that the extracted factors are independent of one another, and thus reflect separate processes. In the elevated plus-maze test of anxiety, in male rats factor 1 accounted for 75% of the variance and reflected anxiety, factor 2 represented activity, and accounted for 24% of the variance. This contrasted with the finding in female rats in which factor 1 was activity, accounting for 57% of the variance, with the anxiety factor accounting for only 34% of the variance. When behaviour in both the plus-maze and holeboard were analysed, a similar sex difference was found with anxiety emerging as factor 1 in males and holeboard activity as factor 1 in females. Locomotor activity in the inner portion of the holeboard loaded on the anxiety factor for males, but on activity for females. When behaviours in the plus-maze and sexual orientation tests were analysed, anxiety emerged as factor 1 in males, sexual preferences factor 2, and activity factor 3. In females, activity was factor 1, sexual preference factor 2, anxiety factor 3, and social interest factor 4. These results suggest caution should be exercised in interpreting the results from female rats in tests validated on males because the primary controlling factor may be different.