Alternative hypothesis: what it means and how to read it
In the report: alternative — The hypothesis tested: two-sided — the means differ in either direction; greater — the first is larger; less — the first is smaller.
How to read it
Leave it two-sided unless you decided on the direction before looking at the data; a one-sided test halves p in the expected direction.
In plain words
A teacher tried a new method, and the new group got 4 points more than the old one. With alternative set to two-sided, the test looks for a gap in either direction, and the gap could still be luck. With alternative set to greater, the test asks only whether the new group is higher. It gives the same gap twice the weight, and now it looked hard to put down to chance alone. That second answer is fair only if she chose the direction before she saw any marks. She chose it after, so she keeps two-sided. Less works the same way, for a first group expected to be lower.
Example
The cell with alternative is selected in each report; point at a label to see its note.
- — the sample data (50 rows) the examples use
- —
=STATS_INDEPENDENT(Data!A1:T51, "y,x1", "welch",0.95)
Where you see it
STATS_INDEPENDENT— Independent-samples t-test for two groups: the two selected columns are the groups.
Related values
- Exact significance — The exact p of U (Dineen and Blakesley), shown only for small groups without ties, as SPSS; for a two-sided test twice the one-tailed p.
- Margin of error — Half the width of a two-sided interval: the t quantile times the standard error.
- Width of the interval — Upper − Lower.
- Estimate within the interval — TRUE (1) when the slope of the data lies inside the interval.
- p-value — The probability of a result at least this extreme if there were no effect (if the null hypothesis held).
Get it in your sheet
One formula in a cell gives the whole report, Alternative hypothesis included. ALS Stats is a free add-on for Google Sheets™.
Install ALS Stats from Google Workspace Marketplace (opens in a new tab)