The statistics in this report are estimates derived from a sample. Two broad categories of error occur in such estimates: sampling and nonsampling errors. Sampling errors occur because observations are made only on samples of populations rather than on entire populations. Nonsampling errors occur not only in sample surveys but also in complete censuses of entire populations. Nonsampling errors can be attributed to a number of sources: inability to obtain complete information about all sample members (e.g., some students or institutions refused to participate, or students participated but answered only certain items); ambiguous definitions; differences in interpreting questions; inability or unwillingness to give correct information; mistakes in recording or coding data; and other errors of collecting, processing, sampling, and imputing missing data.

Weighted item response rates were calculated for all variables used in this report. The weighted item response rates were calculated by dividing the final weighted number of valid responses by the weighted population for which the item was applicable. Most of the items had very high response rates (at least 85 percent). For these variables, it is unlikely that reported differences between low- and middle-income students are biased because of missing data. Two variables had an item response rate below 85 percent: NDCRDBAL (the balance due on all credit cards according to their last statement for students who reported that they usually carried a balance) and NDSMRSAV (the amount students who worked during the summer saved for their education expenses). Since both of these variables are related to income, it is important to consider whether the response rates differ for low- and middle-income students. In the case of NCCRDBAL, both low- and middle-income students had response rates of 64 percent. For NDSMRSAV, the response rate for low-income students was slightly lower for low-income students (76 percent) than for middle-income students (82 percent).

For NCCRDBA, the low item response rate (65 percent) is due in part to the fact that the question was applicable to a relatively small proportion of the sample (33 percent). Given the methodology for calculating the item response rates, all students with incomplete interviews (9 percent) are assumed to have been eligible to answer the question and not responded, which is very unlikely. When students with incomplete interviews are excluded from the calculation, the item response rate for NCCRDBAL is 89 percent. NDSMRSAV applied to a relatively larger number of students (66 percent), which means that incomplete interviews have a smaller effect on the response rate. Excluding students with incomplete interviews from the calculation increases the item response rate to 93 percent overall and also for low- and middle-income students separately. Given the similarity in response rates for low- and middle-income students for these variables, it is unlikely that bias was introduced due to differential response rates for the two income groups.