Degrees of freedom are different fro the t test for two matched samples because 11 equals the number of (7) and not the number of observations. One advantage of matching is that it eliminates one source of (8) . Confidence intervals are interpreted much the same for matched samples as for independent samples. Matching is desirable only when some uncontrolled variable appears to have a considerable impact on the variable being measured. Appropriate matching reduces (9) and the estimated (10) . Therefore, matching (11) the sensitivity of the hypothesis test. To identify a variable that should be matched, the researcher should be familiar with all previous research in the area and conduct (12) . A test could involve the use of the same subjects in both samples. This is referred to as (13) . A big advantage of repeated measures is that uncontrolled variability due to individual differences is eliminated. This advantage may be offset bv other concerns. such as fatigue. practice. or motivation when subiects must perform twice