A statistics student computes a 99% confidence interval for a population mean and obtains (45.2, 58.8). If she had instead computed a 95% confidence interval using the same data, the interval would be
- Anarrower than (45.2, 58.8)Correct
- Bthe same width as (45.2, 58.8)
- Cwider than (45.2, 58.8)
- Dshifted to the right of (45.2, 58.8)
Why A is right
A 95% confidence interval uses a smaller critical value (z* or t*) than a 99% interval, producing a narrower interval. Lower confidence levels require less 'cushion' around the estimate, resulting in a tighter range. The tradeoff is less confidence that the interval captures the true parameter.
Why the other 3 choices miss
- B
- Incorrect. The width of a confidence interval depends on the confidence level (through the critical value). Different confidence levels produce different widths from the same data.
- C
- Incorrect. Higher confidence levels produce wider intervals. Since 95% < 99%, the 95% interval would be narrower, not wider.
- D
- Incorrect. Both intervals are centered at the same sample mean (x-bar = (45.2 + 58.8)/2 = 52.0). Changing the confidence level affects the width, not the center.