Details of this Paper

nside Excel, the sample means are compared by evaluating the probability of equal

Description

solution


Question

Question;Information;Inside Excel, the sample means are;compared by evaluating the probability of equal or more extreme differences;assuming a null hypothesis (H0) of equality of population means. As in;hypothesis testing, when the probability of equal or more extreme test statistics;occurs with a probability less than our type 1 error alpha, rather than assume;the hypothesis true and assert the data to be rare, we take the position that;the hypothesis is not true with the possibility of a type 1 error = probability;alpha. Also, when the analysis fails to detect a difference, a type 2 error is;possible. So it is wise not to say that the data prove anything. Instead, when;the test statistic exceeds the critical value, or the p-level is < alpha;say the data provide statistically significant evidence against H0 and H0 is;rejected. Otherwise, say the data do not provide significant evidence against;H0 and H0 is accepted.;The;relevant Excel functions in Data Analysis are:First use F-Test Two-Sample for;Variances to decide which of the two t-tests to use. Regarding the F test;output;In the Output, if f < 1 "P(F 1, "P(F <= f);one-tail" gives the probability of observing a value of the;F-statistic greater than f when population variances are equal, and;F Critical one-tail" gives the critical value in the high end;of the F distribution for alpha.If the F test statistic is in the;reject region, or p-level < alpha, use t-Test: Two-Sample Assuming Unequal;Variances. Otherwise, use t-Test: Two-Sample Assuming Equal Variance. A test of;equality is a two-tailed test with an accept region in between lower and upper;reject regions. The means are considered equal if lower critical value;tstat < upper critical value, which is the accept region. Otherwise, the;means are considered unequal.;For Bonus points,complete the following;1) Open the link Data for 2 sample tests and select the;tab containing the first letter of your last name.;2) For Problem 1, use the Data;Analysis functions above, perform the F-test on the data, and then perform the;appropriate t test (equal or pooled variance, or unequal variance). StatCrunch;output will not be accepted for step 2. Is there evidence that the population;means are unequal?;3) For Problem 2, answer the;question and complete the table. In the yellow cells, let Excel compute results;by beginning the cell entry with =. Use relative addressing to reference the;data in columns M and N and intermediate results. Is there evidence of a;difference?;4) Attach your Excel work to your;posting.

 

Paper#62281 | Written in 18-Jul-2015

Price : $27
SiteLock