Help with definition

[nnguyenc]

Can someone explain to me what a cross-sectional study is?

omg...i hate biostats/epidemiology...so confusing


i'm trying to understand what the advantages/disadvantages
of this study is...

---

Members don't see this ad.

 

[RxWildcat]

A cross sectional study is an observational study that looks at the prevalence of something in only one group (it can describe a specific characteristic of a population). Thats about all I know without digging into my notes.

 

[cy19861126]

Cross-sectional is basically looking at a "snapshot" of what was already occured at a specific time. For example, if you want to see whether alcohol causes high blood pressure, you can for instance look at the number of alcoholisms who have high blood pressure and the number of alcoholism who do not have high blood pressure. The problem of this is that it does not show causation. For example, from looking at this study, you can never tell whether alcohol caused high blood pressure, or high blood pressure caused people to drink more.

 

[nnguyenc]

lol...you guys are nerds

but i still don't really see any advantages vs. disadvantages?

 

[goldennemesis]

lol...you guys are nerds

but i still don't really see any advantages vs. disadvantages?

As far as disadvantages, someone already mentioned the issue of causation.

An advantage is that these are the easiest studies to do. You decide to do the study, pick your subjects, ask them some questions, voila, you're done.

 

[KUMoose]

Use wiki to do your homework, not other 3rd years.

http://en.wikipedia.org/wiki/Cross-sectional_study
http://en.wikipedia.org/wiki/Study_design
http://en.wikipedia.org/wiki/Design_of_experiments

You have to know other study types then the advantages and disadvantages should be easy to pick out for all the study types.

 

[nnguyenc]

you gotta admit that biostats is hard

then you gotta know all the freaking tests they do...chi square this, fisher test that...

blah!

 

[KUMoose]

Keep the studies separate from the stats part, and recognize strength and weeknesses in the statistical models used in studies. Unless you plan on a research intensive future I think it's worthless to focus on the minutia of stats (beyond what's needed to pass a test).

 

[Farmercyst]

Keep the studies separate from the stats part, and recognize strength and weeknesses in the statistical models used in studies. Unless you plan on a research intensive future I think it's worthless to focus on the minutia of stats (beyond what's needed to pass a test).

Did great on the differentiation of studies section, not so good on repeating chi square equations. Probably because my opinion matches the above. Last quiz over this stuff on Friday.

 

[bacillus1]

A cross-sectional study looks at one "snapshot" at one single time, as someone else above said. A longitudinal study follows people throughout it.

A problem with longitudinal studies is that it eventually gets quite nonrandom (if you follow people and then their children and whatnot). So that causes bias. Bias can also be caused because longitudinal studies often study similar people in a group (example: Framingham heart study studied the inhabitants of the town of Framingham). There may also be a large loss of data (people move away, don't want to be harassed by nerdy statisticians, etc...). Also groups can overlap (parents and their children studied at the same time).

Of course the advantage of a longitudinal study is that there is more data, and you can see progression in symptoms and whatnot.