I have Rosenberg self-esteem data from 45 morbid jealousy patients (male-29 and female-16) from a study I conducted at 4 centres. ALL the patients (76) who reported during a three month period was considered. 45 patients consented. I need to do a mean comparison between males and females on self-esteem scores. Self-esteem scores were normally distributed. Homogenity of variances could be assumed. Given the small sample size and self-esteem not being an absolute measure, can I use Independant sample t-test for the mean comparison of males and females on self-esteem or should I go for a non-parametric alternative?