17.3.3.1 The Pair-Sample T-Test Dialog Box

Supporting Information

To open the Pair-Sample t-Test dialog box:

With a worksheet active, click Statistics: Hypothesis Testing: Pair-Sample t-Test.

More information:

Recalculate	Controls recalculation of analysis results None Auto Manual For more information, see: Recalculating Analysis Results

Recalculate

Controls recalculation of analysis results

Input Data Form	Raw Data Enter column with range format. For help with range controls, see: Specifying Your Input Data Summarized Data Enter value of mean difference, pooled standard deviation and sample size data for sample(s), in the form as "mean SD N" (space separated numbers) You can calculate the mean difference by: $\bar{d}=\frac{1}{N}\sum_{i=1}^N d_i$ and calculate the SD by: $s_d=\sqrt{\frac{1}{N-1}\sum_{i=1}^N(d_i-\bar{d})}$ The N is sample size. $d_i$ is $x_{1i}-x_{2i},for(i=1,2,...,N)\,\!$ .
Input	Enter the input data.

Input Data Form

Raw Data

Enter column with range format. For help with range controls, see: Specifying Your Input Data

Summarized Data

Enter value of mean difference, pooled standard deviation and sample size data for sample(s), in the form as "mean SD N" (space separated numbers)

You can calculate the mean difference by:

$\bar{d}=\frac{1}{N}\sum_{i=1}^N d_i$

and calculate the SD by:

$s_d=\sqrt{\frac{1}{N-1}\sum_{i=1}^N(d_i-\bar{d})}$

The N is sample size. $d_i$ is $x_{1i}-x_{2i},for(i=1,2,...,N)\,\!$ .

Input

Enter the input data.

Test Mean	The hypothetical mean difference to be tested (usually 0).
Null Hypothesis	The null hypothesis to be tested (usually Mean1 - Mean2 = 0).
Alternate Hypothesis	Options: Mean1-Mean2 <> Test Mean Perform a two-tailed paired-sample t-test: Is the mean difference between the two datasets more or less than the hypothetical mean difference? Mean1-Mean2 > Test Mean Perform an upper-tailed paired-sample t-test: Is the mean difference greater than the hypothetical mean difference? Mean1-Mean2 < Test Mean Perform a lower-tailed paired-sample t-test: Is the mean difference less than the hypothetical mean difference?
Significance Level	The significance level of the test.
Effect Size	Calculate Cohen's d value to show the effect size which is the magnitude of the difference between actual mean1-mean2 and Test Mean.
Confidence Interval(s)	Compute a confidence interval for the mean. For computation details, see Algorithms: Paired Sample t-Test.
Confidence Level(s) in %	Specify space-separated confidence levels, (1- $\sigma$ )%. Available when Confidence Interval(s) is selected.

Minimal Difference of Interest	Specify the minimum difference from the Test Mean you are interested in. Once you enter a value in this box, the Power Anlysis graph will show the minimum sample size to detect this difference in the different powers.
Actual Power	Compute the power of the test (the probability of rejecting the null hypothesis when it is false).
Significance Level	Only available when Hypothetical Power and/or Actual Power are checked. This is the significance level (alpha value) for power computation.
Hypothetical Power	Compute the power of the test at various sample sizes.
Hypothetical Sample Size(s)	Only available when Hypothetical Power is checked. Specify hypothetical sample size(s) for power computation.

Plot for Differences	Check this box to plot the box chart and histogram for mean differences.
Box Chart	Plot a box chart.
Histogram	Plot a histogram.
Arrange Plots of Same Type in One Graph	Note that for histograms, the layers will be arranged vertically. For box charts, layers will be arranged horizontally. This box will be hidden if you chose Plot For Differences box.

For help with the range controls, see: Output Results

Results Log Output	Select to output results to the Results Log.
Graphical Summary Report	Create a summary graph report, including P value bar, effect size bar and difference distribution graph. Three result tables will also been attached to the graph to show the statistical values. The footnote after the summary graph will diagnose these statistical results.
Graphical Power Analysis Report	Create a graphical power analysis report with power curve(s). If no value has been set for Minimal Difference of Interest editbox, only the power curve of current sample size will be output; and the power at 60%, 70%, 80% and 90% will be tagged to show the difference at these power values. If a difference value has been set for Minimal Difference of Interest editbox, five power curves will be output: power curves at around 60%, 70%, 80% and 90% and current sample size; and the sample size to detect the minimal difference will be listed. A footnote will be added after the power curve graph to show you the diagnosis results of power analysis.
Output Plot Data	Specify a worksheet for output of plotted data (Origin generates tabular data when plots are created).
Output Results	Specify the worksheet for output of test results.