Wednesday, January 03, 2018

Smoking Causes Criminal Behavior.

You would probably say that is ridiculous and it is.

I remember taking a research class where we were discussing research methods and the professor was talking about selecting your research population. He said if you went to a prison and did a survey of the number of prisoners who smoked (this was before the ban on smoking in prisons) you will find the vast majority of inmates smoked and you could draw an assumption that smoking cause criminal behavior. But if you did a random sample of smokers in the general populations you would get different results.

This was to show how us that the sampling method is an important part of any research study. Some of the types of samples are,
Reliance On Available Subjects. Relying on available subjects, such as stopping people on a street corner as they pass by, is one method of sampling, although it is extremely risky and comes with many cautions. [Also known as a convenience sample]

Snowball Sample. A snowball sample is appropriate to use in research when the members of a population are difficult to locate, such as homeless individuals, migrant workers, or undocumented immigrants. A snowball sample is one in which the researcher collects data on the few members of the target population he or she can locate, then asks those individuals to provide information needed to locate other members of that population whom they know.

Simple Random Sample. The simple random sample is the basic sampling method assumed in statistical methods and computations.
Okay why do I bring this us?

There is a lot of research that is being used against us, many of the studies have flaws. Some research used samples who they labels as trans but did not conform to the diagnostic criteria or they considered those who dropped out of the study as detranstioning.

The survey 2015 Transgender Survey used a form of convenience sample and the research project that I took part in I was one of the Project’s Principal Investigators and we used a modified snowball sample.

Okay, why am I writing this post today, well on the web there has been a lot of discussion on a recent study (I wrote about here) and there are worried trans people about the study findings.

Relax take a deep breath and let it out slowly to a count of ten.

Let’s look at how the survey was conducted,

First the sample was taking from a medical database of Medicare recipients so first off that means everyone had medical problems to begin with. It is kind of like the prison example. The findings would be different if they looked at the general cis-gender population verse the trans gender population.

Second we do not know any complicating factors such as preexisting conditions nor do we know if they really had the disease listed in their medical file. Before the mandated insurance coverage for us many healthcare providers fudge their diagnoses in order to get insurance coverage for their patient.

Lastly we do not know how society’s pressure to conform caused any of the stress related findings.

So I don’t think the report is all that alarming especially because the data came from a metadata examination of the Medicare records.

1 comment:

  1. Interestingly, smoking IS criminal behavior in many places these days. Smoking guns, however, may or may not be. :-)

    I have responded to many trans surveys over the years. I am sometimes left suspicious by them, and there have been a few that were obviously made up so as to reach a conclusion to support a biased "hypothesis." Even studies that are pro-trans can be used against us if they are not done properly.