Difference between two means: pooled t-test
Studying this course? Keep your progress in one place.
Key facts
- The pooled -test compares two means from independent Normal samples with unknown but equal variances.
- Pooled variance: .
- .
- The degrees of freedom are , for both the variance and the critical value.
- Confidence interval: . If it contains , no evidence of a difference.
When it applies
For two independent Normal samples with unknown but equal variances.
Use the pooled -test to compare means and from two independent random samples. The variances are unknown but assumed equal, and the populations must be Normal.
The common variance is estimated from both samples together, so the standardised difference follows a distribution, not a Normal.
Pooled test
- Two independent samples
- Variances equal, unknown
Paired test
- Each value linked to one in the other sample
- Take differences, one-sample test
Ten students sit a test before and after a revision course. Is the pooled test appropriate?
The pooled variance
Weight each sample variance by its degrees of freedom.
Combine the two unbiased sample variances, weighting each by its degrees of freedom. If you are given sums, use , so .
The pooled value always lies between the two sample variances, nearer to the one from the larger sample.
- 1
Sample variances
s_X² and s_Y²
- 2
Weight by degrees of freedom
(n_X − 1)s_X² and (n_Y − 1)s_Y²
- 3
Add
Combine both weighted values
- 4
Divide by n_X + n_Y − 2
This gives the pooled s²
- Pooled variance
with , and with . Which pooled value is plausible?
Statistic and degrees of freedom
Divide the difference of means by its estimated standard error; use degrees of freedom.
Under , . More generally, .
One degree of freedom is lost for each sample mean estimated. The denominator is the estimated standard error of .
- Statistic
- Degrees of freedom
Samples have sizes and . How many degrees of freedom does the pooled statistic have?
Carrying out the test
State the hypotheses, find then , compare with , and conclude in context.
State and (one-tailed or two-tailed). Find , then . Find the critical value of , halving the level for two tails. Compare, and conclude about the real situation, not just "reject ".
- 1
Hypotheses
- 2
Pooled
weighted by degrees of freedom
- 3
Calculate
with the estimated standard error
- 4
Compare and conclude
critical value of
Worked example
, , ; , , . Test at .
- 1
.
- 2
.
- 3
The critical value of is , and .
The test statistic is and the critical value is . What is the conclusion?
Interval for the difference
Use with the two-tailed critical value.
The value is the two-tailed critical value (for , the upper point).
If the interval contains , there is no evidence of a difference in means at the matching significance level. If it does not, there is.
Worked example
Using the data from the worked example, find a confidence interval for , given .
- 1
The standard error is .
- 2
.
A interval for is . What does it suggest?
Try an exam question
Two independent random samples from Normal populations with equal variances give: , , ; , , . Test, at the level, whether . The upper point of is .
[4 marks]
- [1], .
- [1]Pooled variance .
- [1].
- [1]: reject ; evidence that the mean of is greater than the mean of .
That's the notes covered.
Carry on to the next subtopic.
Try a question
How many degrees of freedom does the statistic for testing the difference between the two population means have?
1 mark, marked against the mark scheme
How many degrees of freedom does the statistic for testing the difference between the two population means have?
1 mark, marked against the mark scheme