[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y ]

# STATS

This is a blue board which means that it's for everybody (Safe For Work content only). If you see any adult content, please report it.

File: 001.jpg (268KB, 650x975px) Image search: [iqdb] [SauceNao] [Google]
268KB, 650x975px
>Null hypothesis: Population mean = 46.5
>Alternative hypothesis: Population mean > 46.5
>Sample mean: 46.3
>Sample standard deviation: 0.7
>Number of tries: 20

How do I find an approximation for the p-value? It should be greater than 10, yes?
>>
>>7591538
t-test
what confidence?
>>
>>7591547
Sorry, forgot to include: 95% confidence
>>
>>7591538
>>7591547
So t = 46.3-46.5/( (.7)/(sqrt(20))

You get a t-value

Df = 19
and if confidence is 95%

a = .975

If |t|>p value in table, reject the null
>>
>>7591558
a = .975 because the test is two tail

95% -> .05 ->.05/2 -> 1- .025-> .975
>>
>>7591558
So t-value is t = -1.27775, df=19.
How do I determine the p-value from that?

>>7591576
Isn't it one tail? The alternative hypothesis is that the population mean is greater than 46.5.
>>
>>7591585
Yeah my bad. You wrote the null hypothesis incorrect. its Pop mean less than or equal too.

Look at the t table. X is p vaule
Y is df

so t = -1.27775 and if t> p table value, reject null
>>
File: t-table.jpg (590KB, 879x1187px) Image search: [iqdb] [SauceNao] [Google]
590KB, 879x1187px
2.093 is table value. so you fail to reject null.
>>
>>7591592
So I'm looking in the table, at df=19, for a value similar to 1.27775?
I can find that for df=19 and p-value=0.1 the the t-value is 1.328.
So from that I can draw the conclusion that the p-value must be greater than 0.1 (10%)? Is that correct?
(My table doesn't include p-values above 0.1)

>>7591602
Both Anon in this thread and me landed at 1.27775 though? How did you get 2.093?
>>
>>7591614
.05 at top
19 on the left
Find the value that fits.
Compare that value to calculated t value
I fucked up and used .025 but look at .05
>>
>>7591625
If I use this (>>7591602) table rather than my own, I can see that our t-value (1.27775) must be between p-value 0.15 and 0.10, right?
So 10% < p-value < 15% ???
>>
>>7591629

what are you asking? if your |t|>p-value at .05, which is 2.093, then you reject your null.
>>
File: screen.png (3KB, 919x47px) Image search: [iqdb] [SauceNao] [Google]
3KB, 919x47px
>>7591731
The answer I need isn't if I reject the null hypothesis or not, but in what intervall I find the p-value.
These pic for the different options.

It's p-value > 10%, isn't it?
>>
>>7591770

Is the OP all of the data you're given?
>>
>>7591538

>>
File: CIk-WUfUAAA2YF3.jpg (21KB, 540x516px) Image search: [iqdb] [SauceNao] [Google]
21KB, 540x516px
>>7591547
>tfw no confidence
>>
>>7591806
>>7591770

Your t is < 1.729, thus yes, your p value will be in the >10%.
>>
>>7591806
Yeah I think that's all, though I summarized it.

Whole text:
>A shoe company that manufactures running shoes claims the wearer performs better with the use of their shoes. Markus is a promising 400-meter runner who have run a certain route so often that he knows that his average is 46.5 seconds. He now want to test the new shoes and does so by using them 20 times for the 400-meter route. If it can be shown that the new shoes are really better, he will buy them, otherwise he keeps the old ones. Markus test rounds gave the mean 46.3 seconds and standard deviation s = 0.7. Estimate the p-value.

The 400m stuff I reckoned was irrelevant information?

[Boards: 3 / a / aco / adv / an / asp / b / bant / biz / c / can / cgl / ck / cm / co / cock / d / diy / e / fa / fap / fit / fitlit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mlpol / mo / mtv / mu / n / news / o / out / outsoc / p / po / pol / qa / qst / r / r9k / s / s4s / sci / soc / sp / spa / t / tg / toy / trash / trv / tv / u / v / vg / vint / vip / vp / vr / w / wg / wsg / wsr / x / y] [Search | Top | Home]
Please support this website by donating Bitcoins to 16mKtbZiwW52BLkibtCr8jUg2KVUMTxVQ5
If a post contains copyrighted or illegal content, please click on that post's [Report] button and fill out a post removal request