Some argue that the confidence interval achieves more than a hypothesis test. What do you think?

I maintain confidence interval achieves more than a hypothesis test.

The hypothesis test does not give statistical significance. Confidence intervals offer additional insight or information. For instance, the upper and lower limits can indicate the extent of an effect. A confidence interval gives a range of potential values and an estimate of the precision of the parameter value. If data has high variability, the margin of error may spread out of the point estimate.

Nonetheless, confidence intervals can help determine how well a sample estimates a value for an entire population. It can help you compare the exactness of different estimates, which is inconceivable with hypothesis tests.

In summary, confidence interval and hypothesis tests are inferential methods that use an approximated sampling distribution. Confidence intervals use data from a sample to estimate a specific population parameter, whereas hypothesis tests use sample data to test a defined hypothesis.

Hypothesis tests are preferable for a strict comparison with a pre-specified hypothesis and a certain significance level. Confidence intervals are better in describing a single sample and explaining the significance of an effect. They remind us that any estimate is subject to error and that precision is possible.

They also provide information not provided by hypothesis tests. For this reason, confidence intervals are superior to hypothesis tests in terms of what they can achieve.