statistics Help

[Concubine]

---

Members don't see this ad.

So, I'm trying to work through some statistics problems for our public health class, and I'm having a hard time grasping significance. Is anyone else savvy with statistics?

The data looks like this:

Question 1:
Control sample: mean = 0.03, SD = 0.09
Test sample: mean = 0.31, SD = 0.43
p value is 0.0041.
Are the results significant?


Question 2:
Control sample: mean = 583, SD = 114
Test sample: mean = 389, SD = 66
p = < 0.0001
Are the results significant?

What does this all mean?

 

[mordounhas]

The simple rule is that p<.05 means that you can reject the null hypothesis. When you're comparing means, as they are doing in these questions, you would consider the null hypothesis to be that the two means are the same value. p stands for probability, and a value of p=.05 means that there is a 5% chance that the null hypothesis could be true.

Therefore, the p=.0041 in question #1 means that you can reject the null hypothesis, as it is highly unlikely that the means are the same (it's considerably less than p<.05). Doubly so in #2 with a p<.0001.

Now, it can get considerably more complicated than that (which is why biostatistics people are often consulted on journal papers), but those principles will probably help you at least in the beginning.

 

[austinap]

The simple rule is that p<.05 means that you can reject the null hypothesis. When you're comparing means, as they are doing in these questions, you would consider the null hypothesis to be that the two means are the same value. p stands for probability, and a value of p=.05 means that there is a 5% chance that the null hypothesis could be true.

Therefore, the p=.0041 in question #1 means that you can reject the null hypothesis, as it is highly unlikely that the means are the same (it's considerably less than p<.05). Doubly so in #2 with a p<.0001.

Now, it can get considerably more complicated than that (which is why biostatistics people are often consulted on journal papers), but those principles will probably help you at least in the beginning.

I hate to nitpick, but the p-value is the probability that a test statistic as extreme or more than that obtained would be observed if the null hypothesis were true. This is not the same as "the probability that the null hypothesis is true". So, a value of p=0.05 means that, if the null hypothesis is true, there is a 5% chance that the test statistic would be as extreme or more than the result observed, but does not mean that there is a 5% chance of the null hypothesis being true.

The reason that the is important is because you are only comparing against one possible hypothesis, rather than deciding between an infinite number of other possibly true hypotheses. Frequentist statistics cannot attach probabilities to hypotheses. For more details, see frequent misunderstandings of p-values.

 

[mordounhas]

Knew I shouldn't have tried to put up an explanation of p without looking it up (haven't taken a real statistics class since high school). Thanks.

 

[loveoforganic]

Yeah, those questions both provide unnecessary info. p is the sole determinant of significance.

 

[engineeredout]

Thats all you have to do for med school biostats?

Man my stats degree is going to ****.

 

[mjl1717]

I hate to nitpick, but the p-value is the probability that a test statistic as extreme or more than that obtained would be observed if the null hypothesis were true. This is not the same as "the probability that the null hypothesis is true". So, a value of p=0.05 means that, if the null hypothesis is true, there is a 5% chance that the test statistic would be as extreme or more than the result observed, but does not mean that there is a 5% chance of the null hypothesis being true.

The reason that the is important is because you are only comparing against one possible hypothesis, rather than deciding between an infinite number of other possibly true hypotheses. Frequentist statistics cannot attach probabilities to hypotheses. For more details, see frequent misunderstandings of p-values.

Excellent!! (Especially if one is new to stat -it can be a tortuous area!)