STATSTESTREVIEW - O ne sample t test - p- value method ( 9...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
One sample t test - p- value method ( 9 steps) and CI method ( 9 steps) Step 1 : state assumptions Assume we have an SRS of ( n individuals), we do not know the population standard deviation and we assume the population distribution of ( whatever we are measuring) … n> 40 – has not outliers 15<n40<- is symmetric with no outliers n<15 is normal with no outliers Steo 2: Define parameter of interest U= the true mean ( whatever were measuring) for all ( individuals) Step 3: give point estimate Pe= xbar= Step 4: state ho and ha Get ha first because it is what we are trying to prove. Ha = same number as h0 Pvalue methong steps 6-9 Step 6-Calculate the test statistic T= xbar- ho value _________________ s/ square root of n step 7- determine the pvalue look up t test statistic on t table with DF= n-1 ( will probably be between 2 values_ - if ha > or < look up t on table and get one sided p - if ha look up t on table and get the 2 sided p ex. N=30 so 30-1 = 29 and test statistic =2.74 pvalue is between .01 and .02 so .01< pvalue<.02 step 8 and 9- make a decision and state the conclusion - Because the pvalue < a = .05, we reject h0 - The same date indicate that the mean amount of money spent of shitty beer is not equal to 8$ - If pvalue < a , reject ho - If pvalue > a , do not reject Ho - If do not reject Ho- the data do no rule out the possibility that Ho CI method- steps 6-9 Step 6: compute the margin of error ME= t*x s/square root of n ---- t* comes from t table- match conifidence level and df= n-1 Round ME to one more decimal than x bar Step 7: calcuate the interval lb and ub ( lb, ub)= ( xbar-ME, xbar+ME) --- round ib DOWN and ub UP with the same decimals as xbar step 8 and 9 Because the entire interval given is above/ not above the hypothesized mean of _____, we reject/do not reject H0 9. The same data do not rule out the possibility that the mean amount - if Ha > , reject Ho if the entire interval is obove the H0 value ( otherwise do not reject h0) - if ha < , reject Ho if entire interval is below H0 value ( otherwise , do not reject H0 - if ha doesn’t = , reject Ho if Ho value is NOT in interval ( otherwise, do not reject Ho) important note- if asked to JUST make a one sameple t-interval ( ie we are NOT asked to run a test using the C0 method just do steps 1-3 and 6-7 as out outlined above and then… last step: we are ( 95%) confident that the true mean ( whatever we are measuring) for all ( individuals) fa between ( lb) and ( up) EXAMPLE: suppose u = the true mean weight of all sumo wrestlers in pounds suppose our null and alternative are hp =u = 200 and ha:u > 200 if reject H0- the data indicate that the true mean weight of all sumo wrestlers is greater than 200 lbs if do not reject h0- The date do not rule of the possibility that the true mean weight of all sumo wresterlers is 200
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
sameple size ( always round up) if C goes up than ME goes up- of ME goes down C goes down formula n = (z*x s/me) ^2 if n goes up than ME goes down- if ME goes down M goes up
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 4

STATSTESTREVIEW - O ne sample t test - p- value method ( 9...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online