I get the impression that in general, there is a stronger correlation between age and salary than there is between merit and salary.

Suddenly, I am no longer unnerved by the emergence of grey hair.

Decius wrote:

Here is the definition of "mid-career:"

Full-time employees with 10 or more years of experience in their career or field who are Bachelors graduates.

For the graduates in this data set, the typical (median) mid-career employee is 42 years old and has 15.5 years of experience.

I get the impression that in general, there is a stronger correlation between age and salary than there is between merit and salary.

Could you expand on this impression? I gather you're saying that most of the 10th percentile mid-career people are the people who are only 10 years out of school, and the 90th percentile of "mid-career" people is mostly the ones who are a few years away from retirement.

When I look at the mid-career "salary spread", which I define as "Mid-Career 90th Percentile Salary" - "Mid-Career 10th Percentile Salary" for each major, I find that salary spread is always greater than "early gain", which I define as "Mid-Career Median Salary" - "Starting Median Salary". But it varies widely from one major to another.

This data suggests that age and merit may be weighted differently in each major/field. In music, for example, the median early gain is only $19,100, whereas the salary spread is $107,300. In computer engineering, the early gain is $43,600 and the salary spread is $95,900.

I found an interesting result when I sorted the list of majors according to the ratio between "mid-career" salary spread and early gain. At the top of the list, with the largest ratios, are music (5.6), drama (5.5), hospitality & tourism (4.5), and education (4.25). At the bottom of the list, with the lowest ratios, are computer engineering (2.2), aerospace engineering (2.3), civil engineering (2.3), electrical engineering (2.3), and computer science (2.4). The majors with the highest mid-career median salaries have the lowest ratios.

Although major-dependent variations in career duration could also be a factor, it seems that merit has a much bigger effect on the salaries of musicians and actors than it does on the salaries of engineers.

Seeing that, one might argue that merit naturally varies more widely in the performing arts, and thus the greater influence on salary is warranted. Of course, it's also possible that although merit varies widely in many fields, only a few have developed effective methods of performance evaluation.

it's also possible that although merit varies widely in many fields, only a few have developed effective methods of performance evaluation.

I'd lean more towards this being the truth. In my experience, the people setting the salaries (HR) have no clue how to judge the effectiveness of the technical employees, and thus rely more on outside factors such as searching for salary ranges.

I have personally been lectured by 3 HR managers at 3 different jobs that I am "within the expected pay range for my job" and then pointed to Yahoo! Jobs or whatever is the fashionable website at the time. (Of course, this is generally the point at which I start looking for a new job.)

At each job I find myself less and less expecting of merit-based incentives. No matter how above-and-beyond I go, they don't materialize. I could have demonstrably saved the company hundreds of thousands or millions of dollars, but thanks to readily-available pay ranges it's all too easy for the bean-counters to throw up their hands dramatically and exclaim how there is nothing they can do to help me out.