Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (2010)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

96 Aggregation, aggregated: Refers to the practice of combining smaller groups into larger groups. In the present context this term is typically used in reference to the presentation of utilization, availability, or related statistics according to industry. For example, statistics pre- sented for the “Construction” sector as a whole are more aggregated than separate statistics for “Building Construction,” “Heavy Con- struction,” and “Special Trades Construction” industries. See also “Disaggregation, disaggregated.” ADOT: Arizona Department of Transportation. Alaska DOT&PF: Alaska Department of Transportation & Public Facilities. ALDOT: Alabama Department of Transportation. Anecdotal evidence: Qualitative data regarding business owners’ accounts of experiences with disparate treatment and other barriers to business success. Baseline Business Population: The underlying universe of business establishments used in an availability analysis. The denominator in a DBE availability measure. But-for: A term that refers to a hypothetical market that is unaffected by the presence of business discrimination. Often used to describe what level of DBE availability would be expected to be observed in a perfectly race-neutral marketplace. Caltrans: California Department of Transportation. Capacity: This term has no single definition. See Appendix B for discussion. CBP: The U.S. Census Bureau’s County Business Patterns statistical data series. Additional information about the CBP is available from the Census Bureau Web site: http://www.census.gov/epcd/cbp/view/ cbpview.html. CCD: (Colorado) Center for Community Development. CDOT: Colorado Department of Transportation. CWC: Concrete Works of Colorado, Inc. Constitutional Significance or Substantive Significance: Refers to a case where a disparity is “large.” Under the EEOC’s four-fifths rule, a disparity is large if it is 0.8 or less. For example, if DBE utilization in a given industry category is 8% and corresponding DBE availabil- ity is 10%, the resulting disparity ratio would be considered consti- tutionally or substantively significant. Crosswalk table: A table that links Standard Industrial Classification (SIC) codes to their corresponding North American Industrial Clas- sification System (NAICS) codes and vice-versa. DBE: Disadvantaged Business Enterprise. De novo: “Anew.” A de novo review is a completely new review of evi- dence held in a higher or appellate court as if the original trial court’s review had never taken place. Decennial: Refers to the census conducted every decade by the U.S. Census Bureau. The last decennial census was conducted in 2000. The next will be conducted in 2010. Demand-side: Refers to activity on the demand-side of an economic market. For example, when state DOT’s conduct bid-lettings and hire contractors, they are creating market demand. See also “Supply-side.” Dependent variable: In a regression analysis, a variable whose value is postulated to be influenced by one or more other “independent” or “exogenous” or “explanatory” variables. For example, in business owner earnings regressions, business owner earnings is the dependent variable, and other variables, such as industry, geographic location, or age are the explanatory variables. See also “Independent variable,” “Exogenous variable.” Disaggregation, disaggregated: Refers to the practice of splitting larger groups into smaller groups. In the present context this term is typi- cally used in reference to the presentation of utilization, availability, or related statistics according to industry. For example, statistics pre- sented for “Building Construction,” “Heavy Construction,” and “Spe- cial Trades Construction” industries are more disaggregated than statistics for the “Construction” sector as a whole. A P P E N D I X D Glossary

97 Disparate impact: A synonym for “disparity,” often used in the em- ployment discrimination litigation context. A disparate impact oc- curs when a “good” outcome for a given group occurs significantly less often than expected given that group’s relative size, or when a “bad” outcome occurs significantly more often than expected. DOD: Department of Defense. Econometrics, econometrically: Econometrics is the field of econom- ics that concerns itself with the application of statistical inference to the empirical measurement of relationships postulated by economic theory. See also “Regression.” Endogenous variable: A variable that is correlated with the residual in a regression analysis or equation. Endogenous variables should not be used in statistical tests for the presence of disparities. See also “Exogenous variable.” Exogenous variable: A variable that is uncorrelated with the residual in a regression analysis or equation. Exogenous variables are appro- priate for use in statistical tests for the presence of disparities. See also “Endogenous variable,” “Independent variable,” “Dependent variable.” FDOT: Florida Department of Transportation. FFY: Federal Fiscal Year. The Federal Fiscal Year runs from October 1 through September 30. First-tier subcontractors: Subcontractors or suppliers hired directly by the prime contractor. GDOT: Georgia Department of Transportation. HDOT: Hawaii Department of Transportation. IDOT: Illinois Department of Transportation. Independent variable: In a regression analysis, one or more variables that are postulated to influence or explain the value of another, “depen- dent” variable. For example, in business owner earnings regressions, business owner earnings is the dependent variable, and other variables, such as industry, geographic location, or age are the independent or explanatory variables. See also “Dependent variable,” “Exogenous variable.” ITD: Idaho Transportation Department. MBE: Minority-Owned Business Enterprise. A business establishment that is 51% or more owned and controlled by racial or ethnic minori- ties (i.e., African Americans, Hispanics, Asians or Pacific Islanders, American Indians, or Alaska Natives). Mean: A term of art in statistics, synonymous in this context with the arithmetic average. For example, the mean value of the series 1, 1, 2, 2, 2, 4, 5 is 2.43. This is derived by calculating the sum of all the val- ues in the series (i.e., 17) and dividing that sum by the number of elements in the series (i.e., 7). MDOT: Maryland Department of Transportation. MDT: Montana Department of Transportation. Median: A term of art in statistics, meaning the middle value of a series of numbers. For example, the median value of the series 1, 1, 2, 2, 2, 4, 5 is 2. Microdata or micro-level data: Quantitative data rendered at the level of the individual person or business, as opposed to data rendered for groups or aggregates of individuals or businesses. For example, Dun & Bradstreet provides micro-level data on business establishments. The Census Bureau’s Survey of Business Owners (SBO) provides grouped or aggregated data on businesses. Misclassification: In the present context, this term refers to a situation when a listing or directory of minority-owned or women-owned firms has incorrectly classified a firm’s race or gender status. For example, when a firm listed as Hispanic owned is actually African American owned, or when a firm listed as white female-owned is actually white male owned. See also “Nonclassification.” Mn/DOT: Minnesota Department of Transportation. MoDOT: Missouri Department of Transportation. MSA: Metropolitan Statistical Area. As defined by the federal Office of Management and Budget, an urban area that meets specified size cri- teria: either it has a core city of at least 50,000 inhabitants within its corporate limits, or it contains an urbanized area of at least 50,000 inhabitants and has a total population of at least 100,000. M/WDBE: Minority- and Women-Owned Disadvantaged Business Enterprise. NAICS: North American Industry Classification System. The standard system for classifying industry-based data in the U.S. NAICS super- seded the Standard Industrial Classification (SIC) System in 1997. See also “SIC.” NCDOT: North Carolina Department of Transportation. NDDOT: North Dakota Department of Transportation NDOR: Nebraska Department of Roads. NDOT: Nevada Department of Transportation. NIGP: National Institute of Government Purchasing. NMDOT: New Mexico Department of Transportation. Nonclassification: In the present context, this term refers to a type of misclassification when a listing or directory has not identified firms as minority owned or women owned when, in fact, they are. See “Misclassification.” Nonresponse bias: See “Response bias.”

NSSBF: National Survey of Small Business Finances. ODOT: Oregon Department of Transportation. One-sided statistical test, two-sided statistical test: A “two-sided” test means that one is testing the hypothesis that two values, say u (utiliza- tion) and a (availability), are equal against the alternate hypothesis that u is not equal to a. In contrast, a one-sided test means that you are testing the hypothesis that u and a are equal against the alternate hypothesis u is not equal to a in only one direction. That is, that it is either larger than a or smaller than a. OSDBU: Office of Small and Disadvantaged Business Utilization. PUMS: Public Use Microdata Samples. p-value: A standard measure used to represent the level of statistical significance. It states the numerical probability that the stated rela- tionship is due to chance alone. For example, a p-value of 0.05 or 5% indicates that the chance a given statistical difference is due purely to chance is 1-in-20. See also “Statistical Significance.” Regression, multiple regression, multivariate regression: A type of statistical analysis that examines the correlation between two vari- ables (“regression”) or three or more variables (“multiple regres- sion” or “multivariate regression”) in a mathematical model by determining the line of best fit through a series of data points. Econometric research typically employs regression analysis. See also “Econometrics.” Residual: The difference between an observed value and a value predicted by a regression. A positive residual occurs where the observed value exceeds the value computed by the regression; a negative residual is the opposite. Response bias: Response bias can occur in survey research when the survey answers provided by respondents differ in important ways from the answers that would have been provided by those who did not respond to the survey. In survey research for disparity or avail- ability studies, response bias can be tested by eliciting additional information from nonrespondents through supplemental survey methods. RFP: Request for Proposal. SBO: The U.S. Census Bureau’s Survey of Business Owners statistical data series. Part of the 5-year Economic Census series. Additional informa- tion about the SBO is available from the Census Bureau Web site: http://www.census.gov/csd/sbo. SCDOT: South Carolina Department of Transportation. SDB: Small Disadvantaged Business. Set-aside, set-asides: A contracting practice where certain contracts or classes of contracts are reserved for competitive bidding exclusively among a given subset of contractors, for example minority-owned and women-owned contractors. SIC: Standard Industrial Classification System. Prior to 1997, the stan- dard system for classifying industry-based data in the U.S. SIC was superseded by the North American Industry Classification System (NAICS). See also “NAICS.” Statistical significance: A statistical outcome or result that is unlikely to have occurred as the result of random chance alone. The greater the statistical significance, the smaller the probability that it resulted from random chance alone. See also “p-value.” Stratified: In the present context, this refers to a statistical practice where random samples are drawn within different categories or “strata,” such as time period, industry sector, or DBE status. Substantive significance or constitutional significance: An indication of the how large or small a given disparity is. Under the EEOC’s “four-fifths” rule, a disparity ratio is substantively significant if it is 0.8 or less on a scale of 0 to 1. Supply-side: Refers to activity on the supply-side of an economic mar- ket. For example, when new businesses are formed, other things equal, the supply of contractors to the market is increased. See also “Demand-side.” TEA-21: Transportation Equity Act for the 21st Century (1998). TDOT: Tennessee Department of Transportation. t-test, t-statistic, t-distribution: Often employed in disparity studies to determine the statistical significance of a particular disparity statistic. A t-test is a statistical hypothesis test based on a test statistic whose sampling distribution is a t-distribution. Various t-tests, strictly speak- ing, are aimed at testing hypotheses about populations with normal probability distributions. However, statistical research has shown that t-tests often provide quite adequate results for nonnormally dis- tributed populations as well. Two-sided statistical test, One-sided statistical test: A “two-sided” test means that one is testing the hypothesis that two values, say u (utiliza- tion) and a (availability), are equal against the alternate hypothesis that u is not equal to a. In contrast, a one-sided test means that you are testing the hypothesis that u and a are equal against the alternate hypothesis u is not equal to a in only one direction. That is, that it is either larger than a or smaller than a. TxDOT: Texas Department of Transportation. VDOT: Virginia Department of Transportation. WBE: Women-Owned Business Enterprise: A business establishment that is 51% or more owned and controlled by women. WSDOT: Washington State Department of Transportation. 98