You could do this test using an hypothesized value of the difference other If so, use theĪppropriate t-statistic and p-value from the Excel table. Would this still mean that the diet was effective (in terms of theĬould have also been performed as a one-tail test. Pounds (see the 95% confidence interval). Loss could have been as low as an average of 7 pound to a high of 32 Subject matter – thus giving the variability of the sample, the weight Researcher should interpret the results using his or her knowledge of the Than zero, t(7)=3.71, two-tail p = 0.008, providing evidence that theĭiet is effective in producing weight loss. The mean weight loss (M=19.38, SD =14.784, N= 8) was significantly greater “A paired t-test was performed to determine if the diet was effective. These results in a journal article, you could use something like this: Interval is the mean plus or minus this value. The Confidence Level (95%) value of 12.359 in the table, the confidence Information that is usually reported is a 95% confidence interval. The value of the “t Stat” in the previous table. The mean divided by the standard error (19.375/5.227 = 3.71) is same as ( Tools/Data Analysis/ Descriptive Statistics) –Ĭhoose the Summary Statistics and 95% confidence interval options. You can get this be calculating descriptive statistics on theĭifference values. Report these results properly, you need the mean difference and standardĭeviation. Thus, the number you are interested in most is the averageĭifference (loss) and not as much as the individual means of Before and Thus, the t-test is actually testing toĭetermine if the value 19.38 is sufficiently different from 0 to claim Notice also that the averageĪt the original hypotheses – what you are testing is that the average loss To make this a better analysis, first calculate the difference betweenīEFORE and AFTER, creating the following new column called “DIFF” using aįormula such as =A2-B2 in cell C2 and copying the formula for theĪppropriate remaining cells in the worksheet. T-test is actually a test on the DIFFERENCE between the two values. – for a more complete understanding, you need to realize that the paired T-test is p=0.008 (.007585988) and t=3.71.ĭoes a poor job providing what you need to report the results of this test The other items at their default selections. Values of Score in group “After” (values from 168 to 145). For the input range for Variable 2, highlight the eight T-test: Paired two sample for means dialog box: For the Input Rangeįor Variable 1, highlight the 8 values of Score in group “Before” (valuesįrom 162 to 170). Perform a paired t-test, select Tools/ Data Analysis / t-test: Paired The following weight loss data is used in this example Ho: mLoss = 0 (The average weight loss was 0) Observations are taken from the same or matched subjects, you can performĪ paired t- For example, suppose your data contained the variables BEFOREĪnd AFTER, (before and after weight on a diet), for 8 subjects. Two paired values (such as in a before-after situation) where both if it is not already installed in your version of Excel.) They also assume that you have installed the ExcelĪnalysis Pak which is free and comes with Excel (Go to Tools,Īddins. Although there areĭifferent version of Excel in use, these should work about the same for TheĮxamples include how-to instructions for Excel. If you need to, you can adjust the column widths to see all the data.į-test for the data sets in A2:A6 and B2:B6.And interpretation of standard statistical analysis techniques. For formulas to show results, select them, press F2, and then press Enter. If the number of data points in array1 or array2 is less than 2, or if the variance of array1 or array2 is zero, F.TEST returns the #DIV/0! error value.Ĭopy the example data in the following table, and paste it in cell A1 of a new Excel worksheet. If an array or reference argument contains text, logical values, or empty cells, those values are ignored however, cells with the value zero are included. The arguments must be either numbers or names, arrays, or references that contain numbers. The first array or range of data.Īrray2 Required. The F.TEST function syntax has the following arguments:Īrray1 Required. For example, given test scores from public and private schools, you can test whether these schools have different levels of test score diversity. Use this function to determine whether two samples have different variances. Returns the result of an F-test, the two-tailed probability that the variances in array1 and array2 are not significantly different.
0 Comments
Leave a Reply. |