²U.S. Department of Education, National Center for Education Statistics, Methodology Report for the National Postsecondary Student Aid Study, 1992–93. (return to text)

_³U.S. Department of Education, National Center for Education Statistics, National Postsecondary Student Aid Study, 1995–96 (NPSAS:96), Methodology Report. (return to text)

_⁴U.S. Department of Education, National Center for Education Statistics, National Postsecondary Student Aid Study, 1999–2000 (NPSAS:2000), Methodology Report. (return to text)

⁵For nonresponse bias analysis, see U.S. Department of Education, National Center for Education Statistics, National Postsecondary Student Aid Study, 1999–2000 (NPSAS:2000), CATI Nonresponse Bias Analysis Report (NCES 2002–03) (Washington, DC: 2002). (return to text)

⁶For more information on the BPS:96/98 survey, consult U.S. Department of Education, National Center for Education Statistics, Beginning Postsecondary Students Longitudinal Study First Follow-up 1996–98, Methodology Report (NCES 2000–157) (Washington, DC: 2000). (return to text)

⁷For more information on the BPS:1996/2001 survey, consult U.S. Department of Education, National Center for Education Statistics, Beginning Postsecondary Students Longitudinal Study: 1996–2001 Methodology Report (NCES 2002–171) (Washington, DC: 2002). (return to text)

⁸Ibid. (return to text)

⁹The NPSAS:2000 samples are not simple random samples, and therefore, simple random sample techniques for estimating sampling error cannot be applied to these data. The DAS takes into account the complexity of the sampling procedures and calculates standard errors appropriate for such samples. The method for computing sampling errors used by the DAS involves approximating the estimator by the linear terms of a Taylor series expansion. The procedure is typically referred to as the “Taylor series method.” (return to text)

¹⁰A Type I error occurs when one concludes that a difference observed in a sample reflects a true difference in the population from which the sample was drawn, when no such difference is present.(return to text)

¹¹U.S. Department of Education, National Center for Education Statistics, A Note from the Chief Statistician, no. 2, 1993. (return to text)

¹²Ibid. (return to text)

¹³The standard that p £ .05/k for each comparison is more stringent than the criterion that the significance level of the comparisons should sum to p £ .05. For tables showing the t statistic required to ensure that p £ .05/k for a particular family size and degrees of freedom, see Olive Jean Dunn, “Multiple Comparisons Among Means,” Journal of the American Statistical Association 56 (1961): 52–64. (return to text)

¹⁴More information about ANOVA and significance testing using the F statistic can be found in any standard textbook on statistical methods in the social and behavioral sciences. (return to text)

¹⁵Jacob Cohen, Statistical Power Analysis for the Behavioral Sciences, 2nd Edition (Hillsdale, NJ: Lawrence Erlbaum Associates, 1988). (return to text)

¹⁶See table 8.5 on page 338 for comparisons of r2s in G.S. Maddala, Introduction to Econometrics (New York: Macmillan Publishing Company, 1992). (return to text)

¹⁷For more information about least squares regression, see Michael S. Lewis-Beck, Applied Regression: An Introduction, Vol. 22 (Beverly Hills, CA: Sage Publications, Inc., 1980); William D. Berry and Stanley Feldman, Multiple Regression in Practice, Vol. 50 (Beverly Hills, CA: Sage Publications, Inc., 1987). (return to text)

¹⁸See John H. Aldrich and Forrest D. Nelson, “Linear Probability, Logit and Probit Models” (Quantitative Applications in Social Sciences, Vol. 45) (Beverly Hills, CA: Sage, 1984). Analysts who wish to estimate other types of models can apply for a restricted data license from NCES. (return to text)

¹⁹The adjustment procedure and its limitations are described in C.J. Skinner, D. Holt, and T.M.F. Smith, eds., Analysis of Complex Surveys (New York: John Wiley & Sons, 1989). (return to text)