First of all, if you were a man, you were outta luck. The overall survival rate for men was 20%. For women, it was 74%, and for children, 52%. Yes, it was indeed "women and children first."

But what about

*class*? Well, third class women were 41%

*more*likely to survive than first class men. And third class men were

*twice*as likely to survive as second class men.

Yes, class is a far weaker variable in determining survival rate than sex or age. Indeed, most of the variance in first class vs. third class survival rates can be attributed to sex alone. The reason for this is simple: 44% of the first class passengers were women, while only 23% of the third class passengers were women. Because the survival rate for women was far greater than the survival rate for men, we would thus expect a much higher survival rate for first class passengers as a whole than for third class passengers as a whole.

Although this analysis is incandescently obvious, it never seems to show up in mass media treatments of the Titanic disaster. Why is that?

And sex and age differences aside, why would anyone be surprised that passengers in steerage would have a lower survival rate than passengers topside close to the boat deck? (For the findings of Lord Mersey's Enquiry regarding the survival rate for third class passengers, see below under

*Lord Mersey's Report*.)

The table to the right,

*Actual survival rates by sex, age, and class compared to expected survival rates based on sex and age alone*, clarifies the variance in survival rates associated with (but not necessarily caused by) class. If sex and age were the only variables determining probability of survival, we would expect women in each class to have a 74.35% chance of survival, children to have a 52.29% chance, and men to have a 20% chance. Applying these percentages to the actual number of women, children, and men in each class, we compute the expected number of survivors. We then compute how that number varies from the actual number of survivors for that sex, age, and class category.

This method shows that the expected overall survival rate for first class passengers was 44.68%, for second class 40.46%, for third class 36.32%, and for crew 21.38%. It also shows that the actual survival rate was 39.80% higher than expectation for first class as a whole, and 30.58% below expectation for third class as a whole.

The more primitive approach -- taken by most writers on this subject -- is to divide the first-class overall survival rate (62.46%) by the overall average survival rate (31.97%), conclude that first-class passengers were twice as likely to survive as the average passenger, and attribute all this variance to class. The folly of this approach is obvious.