Quote:
|
Originally Posted by Astrocloud
Actually the study is flawed in that there is a sampling error. The study only picks readers of some public paper -and then only volunteers. To really make an unconditional and overwhelming study, however, -would be to pick bisexual, heterosexual and homosexual males completely by random. Since many people (especially homo and bisexual males) keep their identity a secret -it would be very hard if not impossible to make a proper study.
|
Sample bias is always difficult if not impossible to eliminate in a study like this. It is possible to argue that only the more ‘gay’ bisexual males would be looking in the magazines/paper where they posted they were looking for test subjects. Another issue is who would show up. Its possible that those with confusion to their own sexuality would be the ones more willing to ‘test’ it out of their own desire to understand themselves better. That being said the biggest issue is the sample size. They have around 30 of each 'type' and that is barely adequate to be a statistically significant sample. Its enough but it could be better.
Despite possible design flaws, one shouldn't dismiss the data as false just because one disagrees emotionally with the results. It gives an avenue for further exploration into the subject.