Totally OT: Statistics

[Guest guest]

Autumn <ck72198> wrote:

> No, according to my Research Methods Professor, statistics only gives

you a 95% probability that something is true, that means that

> there is still the other 5% that they are sure of thus, the + or - 3%

for marginal error.

 

That's not necessarily true. Depending on how you work the formulas,

you can adjust your margin of error up or down, Things are commonly

calculated to a 95% confidence level, but you can just as readily do the

calculations to another level of confidence, such as 80% or 99% or

whatever.

 

I had to take WAY too many stats/research methods/etc classes back in

school. For some reason they felt it was fairly important in molecular

genetics. Go figure! Probabilities, statistical significance, margins

of error - ugh, I did enough of that to last me a lifetime.

 

--

Sherri

 

" I'm such a good lover because I practice a lot on my own. " - Woody

Allen

 

 

 

 

 

 

[Guest guest]

That's very true but then again the sample has to be bias free and what the

popultion is suppose to represent . Keep in mind that the use of probability

sampling methods does not gurantee that a sample is representative of the

population from which it was selected. You still have to factor that there is

always some chance that the sample differs substantially from the population.

That is why I usually use inferential statistics to estimate how great this

cahnce is. Yes, you can vary your margin of error up or down but depending on

how confident you are about the method used for gathering the data. that what I

love about being a Sociology Major minoring in Criminology, all those

statistics. Being a 46-yer old 3rd year college student has it advantages in

alot of other ways to I am more interested and motivated to get my degree and to

not rush things reasearch takes time and paitience. Thanks for helping me to

clarify what I was saying about the stats, I didn't want to go in depth just was

basically trying to make lite of Mr. Stewarts comments. But is is nice to have a

different topic to talk about.

Hope you have a great day.

Autumn

Sherri <sherria wrote:Autumn wrote:

> No, according to my Research Methods Professor, statistics only gives

you a 95% probability that something is true, that means that

> there is still the other 5% that they are sure of thus, the + or - 3%

for marginal error.

 

That's not necessarily true. Depending on how you work the formulas,

you can adjust your margin of error up or down, Things are commonly

calculated to a 95% confidence level, but you can just as readily do the

calculations to another level of confidence, such as 80% or 99% or

whatever.

 

I had to take WAY too many stats/research methods/etc classes back in

school. For some reason they felt it was fairly important in molecular

genetics. Go figure! Probabilities, statistical significance, margins

of error - ugh, I did enough of that to last me a lifetime.

 

--

Sherri

 

" I'm such a good lover because I practice a lot on my own. " - Woody

Allen