Bardwell Consulting provided the statistical expert witness in the age discrimination class action in re Marvin Wilkerson, et. al. v. Martin Marietta Corporation, and in re Vaszlavik et al. v. Storage Technology Corporation. In both cases we authored multiple reports and provided expert presentations for mediation hearings resulting in successful settlement of the case.
The Minerals Management Service of the US Department of Interior needed a method for sampling all federal gas and oil royalty payments to estimate underpayments. We developed a Windows application which allowed auditors to design, draw and analyze sophisticated random and judgmental samples. The statistical algorithms used were developed by our staff and incorporate proprietary optimization routines for maximizing sampling efficiency. Auditors have been able to use the program with minimal training and have given it excellent reviews for ease of use.
Bardwell Consulting conducted statistical analyses of minority voting patterns in over 30 elections, demonstrating violation of the Voting Rights Act, using ecological regression, and homogeneous precinct analysis. We used the State’s GIS to design an alternative district, and provided comparative analyses of the racial composition and geographic compactness of the current and alternative State House District 60. Our reports and expert testimony were key to reversal by the Tenth Circuit Court of Appeals in Jennie Sanchez, et. al. v. Colorado, resulting in reapportionment of the Colorado General Assembly.
Development of Audit Sampling Program for Federal Election Commission
Bardwell Consulting designed and developed sampling program used by the Federal Election Commission to monitor contributions and expenditures for all Federal elections. Program was developed as a web-based application that can also run on auditors’ notebook computers. Program designs, draws and evaluates samples of transactions for audit. All results were tested against the American Institute of Certified Public Accountant statistical auditing programs.
Our staff developed an innovative probabilistic analysis which significantly strengthened the chemical analysis of shot, and shotgun shell evidence provided by the FBI.
The Colorado Lawyers Committee conducted a three-year study challenging the treatment received by children in the Colorado Child Welfare system. The client required statistical and information system expertise to analyze the State's computer databases, and to quantify the inadequate treatment of children in the system. We provided primarily pro-bono analysis of the Department's data systems which tracked child welfare cases, and prepared issues analyses which were instrumental in reaching a settlement. The settlement resulted in over 300 additional full-time equivalent employees and over $30 million in additional funding for the Child Welfare system, as well as a consent decree monitoring the Colorado Department of Human Services Child Welfare system, in re L.P.M., et. al. by their next friend David Littman v. Roy Romer and Karen Beye.
We were retained as consultant on survey execution, weighting, and estimation for a large and detailed national probability sample for the National Violence Against Women survey; conducted sensitivity analyses and theoretical explication of the impact of sample weighting, and revised methodology report, during review by the Center for Disease Control. We constructed an optimal weighting methodology, and documented the deviation of weighted from unweighted estimates. We were also retained to lead the project to compute statistical estimates and variances for incidence, prevalence, and total costs in The Economic Toll of Intimate Partner Violence against Women in the United States, Violence and Victims, 19 (3) (June 2004).
We conducted a detailed analysis of the jury selection process in the Denver Division of the United States District Court. We prepared an expert report for the defendant which revealed statistically significant race, ethnic and age and socio-economic disparities in the composition of the Qualified Jury Panel, indicating bias in the jury selection process. The analysis used an innovative technique by which unrecorded characteristics of prospective jurors were approximated from census demographics matching geocoded addresses.
Econometric Analysis of the Impact of Taxation on Cayman Islands Real Estate Sales
We conducted an econometric analysis of the impact of stamp duty rates on real estate transaction volume and value in the Cayman Islands from 1990 through 2004.
We were retained to design a stratified random sample of a nationwide class to estimate damages and provide evidence of commonality.
We testified on computer systems and statistical modeling for plaintiffs. CBMS was the result of over 4 years and 2 million dollars worth of programming services contracted from Electronic Data Services (EDS). However, the state and EDS failed to exercise best practices in complex computer system design resulting in thousands of benefit recipients not receiving benefits and millions of dollars of federal sanctions levied against the state of Colorado. We provided the statistical models of deficiencies in the system and projections of improperly denied benefits and testified regarding the deficiencies in the development process.
We analyzed a decade of educational testing, grade, discipline, and graduation data for a large school system, to evaluate the impact of race on the quality of education and the relative impacts of poverty and race in re William Crowley v. The Pinellas County School Board, et al.
Evaluation of Adverse Impact of Pre-employment Tests
We were retained as expert witness by plaintiffs to evaluate the adverse impact of alternative pre-employment tests on Minneapolis Fire Cadet Selection Process; we demonstrated adverse impact and proposed the remedy which was implemented, of augmenting the pool of Stage II candidates with 55% additional protected class applicants.
Colorado Center on Law and Policy
Testified as expert on computer systems and statistical modeling for plaintiffs, assessing adequacy of project management, testing, and preparation for release of the Colorado Benefits Management System (CBMS), which was designed to integrate administration of six Colorado and Federal benefit programs for all Colorado counties. Developed and presented model of caseload backlog resulting from CBMS implementation in re Valerie Imani Hawthorne-Bey, et. al., v. Karen Reinerstson, Executive Director of the Colorado Department of Health Care Policies and Financing, et. al., Case No. 04-CV-7059, District Court, City and County of Denver, Colorado.