¹ The text in this section is based on excerpts from the Baccalaureate and Beyond Longitudinal Study: 1993/97 Methodology Report (NCES 1999–159) (Washington, DC: U.S. Department of Education, National Center for Education Statistics, 1999). (return to text)

² The NPSAS universe excludes institutions offering only correspondence courses, institutions enrolling only their own employees, and U.S. service academies. For this B&B cohort, institutions were further stratified by the number of degrees in education they had awarded in the past. (return to text)

³ The B&B sample is not a simple random sample, 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)

⁴ A 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)

⁵ U.S. Department of Education, National Center for Education Statistics, A Note from the Chief Statistician, no. 2, 1993. (return to text)

⁶ Ibid. (return to text)

⁷ The 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)

⁸ More 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)

⁹ For more information about weighted 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)

¹⁰ Although the DAS simplifies the process of making regression models, it also limits the range of models. Analysts who wish to use an approach other than pairwise treatment of missing values or to estimate probit/logit models (which are the most appropriate for models with categorical dependent variables) can apply for a restricted data license from NCES. See 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). (return to text)

¹¹ The 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)