...
So, unfortunately, I am not on the probability side of things, but I'll do my best. I always start by defining any terms I do not know. So, I looked at these. http://en.wikipedia.org/wiki/Student's_t-test http://en.wikipedia.org/wiki/Student%27s_t-distribution
http://en.wikipedia.org/wiki/P-value Then I find an example : http://www.gla.ac.uk/sums/users/jdbmcdonald/PrePost_TTest/pandt1.html
Any help?
k Sorry, I'm more on the theoretical side of the spectrum, but I'll read up and see what I can find.
Also, don't read until you want the answer https://answers.yahoo.com/question/index?qid=20120327101905AAq4Xqs
Hmm, let's check it? What is the formula for t?
They have a whole section on it
Sorry I'm not much help, I haven't had prob and stats for like 7 years
I would either watch the Mit course or khan to find that
Can also try youtube
sure, there is actually a practice problem on that one site withthat
http://www.gla.ac.uk/sums/users/jdbmcdonald/PrePost_TTest/pandt1.html
which ti are you using?
ok
i'd say 50
well I'd say mu not is the claimed change of score, and that is what we want to show is either true or not true
watch this viideo: https://www.khanacademy.org/math/probability/statistics-inferential/hypothesis-testing/v/hypothesis-testing-and-p-values
it's ok I'm watching it too, this is starting to come back a little
I need to find the p-value. How do I start the t-test? The information is below in the prompt. I really need help. A company advertises that by using its SAT math prep class, students can increase their score by at least 50 points, A teacher is skeptical of this claim. Suppose μ represents the actual mean increase in math SAT scores of the population who have used the method and assume the distribution of their scores is approximately normal. The teacher collects a data sample of 16 students who took the class to see if the claim is true. The mean increase of scores is 45 points with a standard deviation of 20 points. Is this data enough evidence that the mean improvement is less than 50 points? H0: μ = 50 H1:Aμ < 50
It is almost correct. You subtract the estimate minus your null hypothesis value... But you got t = -1 anyway which was correct..
I would have divided 20 by √16 first
t=z i guess, but listen to kirby
Is your significance level just assumed to be 0.05?
I got \[\frac{50-45}{5}=1 ~standard~deiviation away\]
so p is high, and can accept the null
You can use the approximation t = z when you have |"large" values of n (large sample size). Usually it is considered to be "30" but is subjective. Also, you are not given the standard deviation of the population, but just of the sample.
You have to do 45 - 50, not 50 - 45
ah ok
(sample statistic ) - (null hypothesis value)
kirb deserves it, he actually knows the material, no hard feelings
Ya Fibonacci you did a good job even though you are not too familiar with the topic. Kudos
thanks guys, :) Now, I feel the need to find my prob and stats book.... haha but really great job kirb and Blank, you worked it out on your own for the most part, so major kudos
:)
Join our real-time social learning platform and learn together with your friends!