My Math Forum Help with interpreting confidence intervals

 Probability and Statistics Basic Probability and Statistics Math Forum

 September 13th, 2017, 07:06 AM #1 Member   Joined: May 2015 From: Australia Posts: 77 Thanks: 7 Help with interpreting confidence intervals I need to do a two sample t test and I have shown my working out below: State: Does taking an elderberry supplement reduce the number of 'cold' days? Formulate: We want to compare mean reduction of cold days with elderberry (µ1) vs a placebo (µ2) We will test the hypotheses: H0: µ1 = µ2 H1: µ1 ≠ µ2 α = 0.05 Solve: The samples are random and independent. From the Normal probability plots, the P-values for both groups are > 0.05. Thus, assume Normality. We have the requirements for a two-sample t-test. I used Minitab to obtain the t-test statistic and P-value: From the Minitab output, the t-test statistic is -2.07 which follows a t distribution with 26 degrees of freedom. The corresponding P-value is 0.049 Decision: Since the P-value = 0.049 < 0.05, Ho is rejected Conclusion: At 5% significance level, there is enough evidence to conclude that there is a difference in the mean number of cold days between the two treatment groups. The result was significant at 5% level. Interpreting 95% confidence interval: 95% confidence interval (-4.54, -0.01) obtained from Minitab We are 95% confident that the difference in the population mean number of cold days of elderberry and placebo is between -4.54 and -0.01 days. Since both confidence limits are negative, the elderberry appears to be effective in reducing the population mean number of cold days. I'm not sure if my interpretation of the confidence interval was correct. How do i know if the 'negative' confidence limits favours elderberry or the placebo?

 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post calypso Probability and Statistics 3 May 28th, 2016 09:36 AM Artus Advanced Statistics 0 November 14th, 2012 02:39 PM jackinda Algebra 1 December 4th, 2011 02:39 AM mochisushi01 Advanced Statistics 1 April 13th, 2010 12:54 AM symmetry Advanced Statistics 1 March 29th, 2007 08:59 PM

 Contact - Home - Forums - Cryptocurrency Forum - Top