"Richard Ulrich" <Rich.Ulrich@comcast.net> wrote in message news:firstname.lastname@example.org... > The "bio" in biostatistics has that implication, a certain amount > of biology.
I agree, but the amount of "bio" emphasized varies by year. When I went into the masters program, computer skills were in fashion, and biology was out. Today, the emphasis is on biology, plus many schools have cut back funding, so I was turned down by a lot of schools that should have accepted me in a normal year. I'm debating whether to get the PhD over with now at a decent school, or wait a while to see whether I get into a better school.
> The emphasis of the biostatistics was different from that of the > psychologists that I worked with. Epidemiologists use survival > analyses and log-linear methods, a lot, for data which is most > often observational. Models for gene distribution are solved > by maximum likelihood. Thirty years ago, no one had heard > of "factor analysis" at the Pitt GSPH.
Yes, I had a genetics-oriented fellowship during my masters. The more I've learned about genetics, the more I've found out how much we don't know, and how much we won't know for a long time. Everyone was expecting microarray technology to have advanced several generations by now, but it's still the same expensive, unreliable technology as 10 years ago. As of now, there is a Wild West mentality in the genetics world, where people publish papers that claim to have found a gene for something; follow-up work is rare, and negative results are seldom published. Genetics has fallen out of fashion in the biostat world for this reason.
> Psychologists use a lot of designed studies with ANOVAs, and > pay attention to development of rating scales. Ferguson's > textbook is "Statistical Analysis in Psychology and Education" -- > I never noticed that any parts of it were particularly for one > specialty or the other.
I'll look into that. I'm thinking my genetics training will eventually come in handy, even if I do my dissertation on a psychology topic.
> Psychology papers can suffer from poor techniques, but they > are not the major techniques of biostatisticians, in my experience.