Pros
Mathematica hires smart people with many skills. At all levels (whether you have Bachelor’s, Master’s or PhD degree), incoming staff are bright, motivated, hard workers, and interested in policy. They usually have a lot of expertise in their specific policy area and are eager when they join the company. Compensation is good. Good enough, in fact, to keep a lot of staff here despite much dissatisfaction with the job. Management appreciates the need to maintain good work-life balance. Except on rare occasions, it is generally accepted that you should not be working nights and weekends. Work hours are predictable. Most managers are flexible about time off. Colleagues are generally nice. The workplace is predictable.
Cons
The quality of the work is often lacking. The company’s deliverable reports (publicly available on the website if you want to see examples) seem produced by rote. Their structure is the same regardless of whether the methodology was quantitative or qualitative, and whether you’re on a project studying outcomes for low income kids enrolled in a health program, a nutrition study, a SNAP implementation study, an evaluation of an early childhood program, or a teacher salary study. Often, even words are the same across incredibly varied projects. Methodologically, the company does things a certain way because they’ve already done it that way once (or a hundred times) before and it was successful, so why take a risk doing it differently. Research methods and analyses are not tailored to the problem at hand. Related to the above, the research lacks theory. You can’t produce real research that isn’t *explicitly* guided by theory – it compromises the whole observation process. Also, lack of familiarity with qualitative research compromises quality of those projects. And senior managers in the survey division are not familiar with survey research, which makes it difficult to work for them on a project. Success at the company is highly dependent on how well you conform. See above about reports being identical. This is ironic given that research is about brainstorming, model building, and problem solving, all of which require imagination and creativity. There is no diversity. Despite office locations in cities with diverse ethnic and racial populations, the professional staff (and certainly the company leadership, especially in research) are overwhelmingly white. There is little discussion of cultural competence. Projects that require cultural competency are given short shrift. The company is very hierarchical. This can be a “pro” for some people, but I found it very constraining. The matrix-style organization means people in leadership positions easily evade accountability. I have worked in two organizations prior that were managed in a matrix style, and I had never come across this problem before. At Mathematica, responsibility for management decisions always seems to be someone else’s problem or fault. There is also a culture of blaming staff at the level below yours for project work not done properly (though, granted, this problem is not overwhelming). Mathematica does a terrible job of integrating new staff into projects. There is little understanding that while you are waiting for months to be integrated into the company, you have already begun to look for a new job and no longer feel loyal to the company. You also lose a sense of ownership over your work in this process. And it is disheatening when that lack of work during your first year is held against you at performance review time. At Mathematica, "survey research" means managing survey operations, not actual research. This totally changes the nature of the job! Applicants should be aware and ask many *specific* questions about the nature of the tasks during the interview process. If you're a survey researcher with a publication record, you will quickly find yourself going backwards here in terms of skill development, career development, and publication opportunities. You no longer have opportunity to conduct research, play with data, or use your stats background. The development of your CV suddenly halts. The survey research job is basically a glorified secretarial job. In terms of methods, Mathematica is not at all on the cutting edge. It often plays laggard to new developments in the field of survey research.