# QNT/351 Confidence levels

Continuing your discussion from Question #1 this week. What will be your decision rule for deciding to accept or reject the null? The decision rule is based on your critical value. Your critical value can be easily calculated in Excel. However you must first determine the confidence level you want to test at. This is at the discretion of the researcher based on the risk of accepting or rejecting the null in error. There are only three confidence levels used in research. These are 99%, 95%, and 90%. If you chose to test at a 90% confidence level this means that there is a 10% change that you could reject or accept the null in error. Therefore the probability (a or alpha) is 0.10.For my example I am choosing 95% level of confidence which gives me a = .05. This is the most common confidence level in business research. My test is right tailed because my alternative hypothesis states ‘greater than’. This means my test statistic (z) must be greater than the critical value of z to reject the null. I calculate the critical value using Excel formula =ABS(NORMSINV(0.05))  Just type this in to any cell in Excell and use the probability you have selected for your test (.01, .05 or .10)This gives me 1.645. So z must be greater than 1.645 in order to reject the null. I write the decision rule as:Reject H0 if z > 1.645Here is the sales for each day of July this year:1-Jul2662-Jul2453-Jul2184-Jul2195-Jul3626-Jul2547-Jul2878-Jul2559-Jul29010-Jul18711-Jul30012-Jul21113-Jul34114-Jul21615-Jul24716-Jul29917-Jul28718-Jul26519-Jul31220-Jul34721-Jul21522-Jul36123-Jul37824-Jul25825-Jul29426-Jul21227-Jul21728-Jul27929-Jul23330-Jul21531-Jul288