When you describe a research project involving sampling, you need to explain how your subjects were chosen. If the population you are studying is all the students of your college, for example, it is not enough to know that you sampled one percent of them. How did you choose that one percent? The ideal is generally to get a spread of variables that would appear to represent the whole population. So you take the figures from the registrar's office about the entire population, and you try to get proportions.
What variables? Race, that tricky thing to define, is a tempting one in any cultural questions, but it may be better just to ask people to describe themselves with "all that apply," and simply present those statistics for what they are worth. Too hard to define.
Gender and age are easier, although these days you may want to add "decline to state" to BOTH of these! But realistically, your data selection is limited to what the registrar's office can give you about the entire population of the school.
Then when you gather the data, you need to be able to explain exactly how you did it. Did you mail out questionnaires? This has the disadvantage that someone else in the household may fill it out rather than the person chosen for your sample. And to take the questions around and interview people personally is time-consuming and expensive. And there, you do not know exactly how much distortion is introduced by, for example, the gender and attractiveness of the questioner. We ARE talking a college population in our example.
You see, no research that deals with humans is very tidy, in terms of numbers. People are, despite modern attempts to dumb them down and homogenize them into a herd, extremely individual critters.
Cussed independents created this country, and most other countries by this time. And cussed independents are what it takes to keep the country as free as possible. Takes adults with backbones to run a nation -- any nation.
2007-01-10 21:47:20
·
answer #1
·
answered by auntb93again 7
·
0⤊
0⤋