In MS Excel, we can easily test statistical significance by using the function TTEST. Let's first understand the parameters of this function.
Ttest function syntax
=TTEST(array1, array2, tails, type)
Array1 (Required): Range for dataset 1
Array2 (Required): Range for dataset 2
Tails (Required): This determines if it is a one-tail T-test or two tail T-test.
- 1: One tail T-test
- 2: Two tail T test
Type (required): Type of t-test
- 1: Paired T-test, i.e. if the two data sets are related to each other
- 2: Unpaired T-test, the two samples have equal variance
- 3: Unpaired T-test , the two samples have unequal variance
Ttest formula example
1. Consider the following dataset for T-test.
2. Select the two –datasets in array 1 and array 2.
3. On an Empty Cell, write the formula to perform 1-tail Paired T test
One-tailed probability obtained
4. Similarly, perform a two–tailed paired T test
Two-tailed probability obtained
Based on the obtained value you can accept or reject the null hypothesis. If the T-test gives the probability less than 0.05, then reject the null hypothesis. This means there is less than 5% probability that the null hypothesis is true.
Here, in both cases, p-value> 0.05. So, we accept the null hypothesis.
- #NUM! – If the arguments for tail or type are not correct. Tail argument must be 1 or 2. Type argument must be 1,2 or 3.
- #N/A! – The datasets are of different length
- #VALUE – The tail-argument or the type argument are non-numeric
Further reading: How to run f test in Excel