Click for next page ( 214


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 213
Appendix E National Center for Science and Engineering Statistics Research Abstracts 2012 Linda Cohen, University of California - Irvine “Doctoral Dissertation Research: The Price Elasticity of R&D: Evidence From State Tax Policies” In tax year 2008 the federal research and development (R&D) tax credit paid out over $8 billion to businesses, which was 7% of total federal expenditures on research. The intent of the tax credit is to provide an incentive for firms to raise their private level of R&D funding. This project will investigate how effective tax incentives are at increasing R&D. This evaluation is difficult because, while we can observe R&D spending before and after a tax policy change, we can only speculate on what R&D would have been without the tax policy change. Because policymakers implement tax incentives in response to current and/or expected economic conditions, a simple comparison of R&D before and after a tax incentive is implemented will lead to inaccurate inferences about the effects of the tax incentive. For example, if R&D in a given year is low, then policymakers may respond with a tax incentive. While a rebound in the following year could be due to the tax incentive, it might also reflect R&D simply returning to its mean value. Alternatively, policymakers might foresee a decline in R&D and implement a tax incentive to prevent the decline. Subsequently observing no change in R&D after the tax incentive takes place would be evidence supporting the efficacy of the tax incentive. To correct for the endogeneity of tax incentives, we will use state-level tax variation driven by changes in the U.S. federal R&D tax credit. While state governments are attentive to state-level economic conditions when forming their idiosyncratic state-level tax policies, the federal government sets a uniform national tax policy and is less attentive to individual state economic conditions. In addition, changes in the federal R&D tax credit have differential impacts on state-level tax incentives across states due to the interaction of federal and state taxes. These two features imply a regression mode that can generate an unbiased estimate of the effect of tax incentives on R&D. Broader Impacts: This study will make several important contributions, in addition to supporting the training of a doctoral candidate. First, the project will create a dataset on state corporate tax laws that will be more detailed than any existing dataset on state R&D tax incentives. These data will allow a descriptive analysis of how the overall tax burden for R&D has changed over time and across states/regions. Second, the project will generate an unbiased estimate of how tax incentives affect R&D. The final contribution will be an estimate of the endogeneity bias driven by self-selection of tax policies, which will help future economic research on tax incentives and uncover evidence on mechanisms behind the implementation of tax policies. PREPUBLICATION COPY: UNCORRECTED PROOFS E-1

OCR for page 213
E-2 CAPTURING CHANGE IN SCIENCE, TECHNOLOGY, AND INNOVATION: IMPROVING INDICATORS TO INFORM POLICY Frank Dobbin, Harvard University “The Retention and Promotion of Women and Minority Faculty Members: Effects of Institutional Hiring, Promotion, Diversity and Work-Life Initiatives, 1993-2008” U.S. colleges and universities have implemented a wide range of programs to promote diversity in the professoriate. Special recruitment programs, tenure extension policies for new parents, paid maternity leaves, mentoring programs targeting female and minority faculty members, dual- career hiring initiatives, and ethnic affinity networks for faculty are but a few of the initiatives. Which of these programs work? It is anyone’s guess, and critics argue that many of the programs may have no effects, or even adverse effects. This goal of this project is to understand the role of university recruitment, promotion, diversity, and work-family programs in attracting and retaining female and minority professors. The project will address questions such as: Do tenure extension programs for new parents help female faculty members win tenure, or do they do more to help male faculty members, who more often have spouses with low-demand careers? Do networking programs help African-American and Latino faculty members to succeed, or do they stigmatize and isolate those faculty members? Do formal promotion requirements help women and minorities to win promotion, or do they serve as window-dressing? NCSES' Survey of Doctorate Recipients (SDR) data from 1993 to 2008, as well as new survey data, will be used to address these questions. The project will develop and pilot a questionnaire designed to obtain historical data on university recruitment, promotion, diversity, and work- family programs, develop a sample of colleges and universities and matched SDR respondents, and build methods for analyzing the data. The goal is to show the effects of the presence, and adoption, of different programs on the career progression of male and female, majority and minority Ph.D.s. New methods will be developed for analyzing individual-level data from the SDR panels, using multinomial logit methods in hierarchical linear models, in which individuals are embedded in institutions. We will develop models that account for both left and right censoring in the data, but which make use of the multiple years of observation available for SDR respondents. Broader impacts: The result of this project will be to show which types of programs help schools to attract, retain, and promote women and minority faculty members, and will guide future administrators in making choices about program utilization and design. This project will also train 3 to 5 doctoral students to analyze data from the Survey of Doctorate Recipients, and to use advanced statistical techniques to examine factors shaping the careers of U.S. scientists and engineers. Jeffery Gibeling, University of California - Davis “Doctoral Dissertation Research: Analysis of Institutional Characteristics that Contribute to Extended Time to Doctoral Degree” The purpose of this research is to identify institutional factors that impact time to degree for the doctoral students who take the longest to complete their studies and graduate. Comparisons are made relative to their disciplinary peers, across nationally representative samples, without disclosing the identity of any institution or student. This study merges data from two nationally collected sources: (1) The Survey of Earned Doctorates (SED); and (2) The supplemental data (not rankings) from the National Research Council's (NRC) A Data-based Assessment of Research-Doctorate Programs in the United States. The SED and NRC data are merged to PREPUBLICATION COPY: UNCORRECTED PROOFS

OCR for page 213
NATIONAL CENTER FOR SCIENCE AND ENGINEERING STATISTICS E-3 RESEARCH ABSTRACTS 2012 determine the patterns of time to degree and the point of extended time to degree within each discipline using the NRC taxonomy. The analysis then looks for interactions between the different levels of data--student qualities, socio-demographic factors, and institutional factors--to identify which factors influence extended time to degree. The driving force behind the research is a void in the current literature. We know that the length of time to doctoral degree varies widely within and across disciplines. While other research has evaluated interactions between various individual and program factors on time to doctoral degree, the impact of institutional factors on extended time to degree has not been specifically investigated. Furthermore, a statistical analysis has not previously been conducted using merged SED and NRC data to evaluate extended time to doctoral degree. This research seeks to fill that void and to add new information to the body of knowledge. Broader impacts: One significant outcome from this study will be the research-based identification of institutional factors associated with extended time to degree. Institutions, doctoral students, and researchers will all be able to identify different fields and populations impacted by the phenomenon of extended time to degree and thereby make more informed decisions about effective strategies to promote timely doctoral degree completion. This project will also train a new researcher in the use of multiple large-scale national datasets. Alan Karr, National Institute of Statistical Sciences “Value-Added Postdoctoral Research on the Scientific Workforce” This postdoctoral research program at the National Institute of Statistical Sciences (NISS) comprises performing innovative research and creating usable products that not only support the mission of the National Center for Science and Engineering Statistics (NCSES) but also address the needs of the nation. From a technical perspective, the research is framed by two statistical themes and two key societal issues. The first statistical theme is characterization of uncertainties arising from novel methods of integrating and analyzing data, addressing a critical need in an era of declining data collection budgets and decreasing participation in government surveys. The second theme centers on conducting experiments with real data, simulating phenomena of interest in order to evaluate, and in some cases enable, methodological advances. Key issues regarding surveys, such as how many times and by what means to contact nonrespondents, are too complex to be treated analytically, and infeasible to address with real world experiments; therefore simulation is effectively the only laboratory available. Specific research topics include data integration, prediction, model to design feedback, data-quality-aware statistical disclosure limitation and cost data quality tradeoffs. All Federal statistical agencies stand to benefit from the research, which will produce innovative theory, novel, methodology and algorithmic implementations, together with datasets, analyses, software and insights that inform future data collections. Broader Impacts: The societal issues are labor economics as it relates to the science, engineering and health workforce (SEHW). Understanding phenomena such as salaries, fringe benefits, mobility and training/job relationships is crucial to maintaining the United States' competitiveness in a global economy, as well as to facing the challenges of difficult economic times. The second issue is aging, because other than the role of students born outside of the US, aging is the most important phenomenon taking place in the SEHW (and, arguably, in society as a whole). For both issues, understanding the dramatically increasing richness of observed behaviors within the SEHW is a profound opportunity. New kinds of family structures, shared PREPUBLICATION COPY: UNCORRECTED PROOFS

OCR for page 213
E-4 CAPTURING CHANGE IN SCIENCE, TECHNOLOGY, AND INNOVATION: IMPROVING INDICATORS TO INFORM POLICY positions, and an array of forms of post-first-retirement employment are among the central social trends of our times. This project will generate new insights that inform both future research and sound policy. Anne Marie Knott, Washington University “The Impact of R&D Practices on R&D effectiveness (RQ)” In January 2011, President Obama signed into law the America COMPETES Reauthorization Act of 2010. The goal of the act was to invest in innovation through R&D and to improve the competitiveness of the United States. However, increasing investment in and of itself is unlikely to produce desired results. We need to understand who should increase spending and how. NCSES is well-positioned to provide that understanding through its data on firm innovative activities in the Survey of Industrial Research and Development (SIRD) (1987-2007), and its successor, the Business R&D and Innovation Survey (BRDIS) (2008-2011). The proposed study empirically examines the impact of US firms' innovative activities on economic outcomes by matching a new measure of economic performance, firms' Research Quotient (RQ) to the SIRD and BRDIS data. This matching enables us to test major hypotheses within the economics of innovation literature that have been unresolved previously due to lack of reliable firm-level measures of innovative outcomes. These hypotheses pertain to the impact of firm size, market structure, firm heterogeneity, innovation type, innovation source, and appropriability on the incentives to conduct R&D as well as the effectiveness of that R&D. Broader Impact: At the policy-level, the study provides theoretically motivated and empirically rigorous insights for directing investment in innovation for the America COMPETES Reorganization Act of 2010: characteristics of firms likely to generate the highest returns to that investment (who). Second, for practitioners, the study offers firms prescriptions for increasing their R&D effectiveness (how). Thus the study has the potential to increase the aggregate R&D productivity in the US. Finally, for academics, the study will answer long-standing questions in the economics of innovation literature to support future theory development on the optimal conditions for innovation. Peter Miller, Northwestern University “Doctoral Dissertation Research: Testing Information and Communication Technology (ICT) Recall Aids for Surveys of Personal Networks” This study seeks to develop recall aids for the name generator procedure from the General Social Survey and examine empirically whether these aids can improve the recall accuracy of the information about who comprises their personal networks from survey participants. It hypothesizes that researchers can obtain more comprehensive personal network data by encouraging survey respondents to consult the actual records that they keep in the contact directories provided by various ICTs (such as the phone book stored in a mobile phone and the address book functionality of email applications). Thus far, although the past literature has suggested a few techniques to reduce respondents’ burden in the survey setting, there is little work addressing the issue of the recall accuracy for personal network data collection. This study employs a survey experiment; a Web survey will be administered to college students to gather information about their personal networks. College students consist of a homogeneous sample appropriate for this study, given the concern of internal validity in the expected findings. PREPUBLICATION COPY: UNCORRECTED PROOFS

OCR for page 213
NATIONAL CENTER FOR SCIENCE AND ENGINEERING STATISTICS E-5 RESEARCH ABSTRACTS 2012 Students who agree to participate in the survey will be randomly assigned to three conditions. The control group will take the questionnaire without any recall aid, while the two experimental groups will take the survey with two different forms of prompts and probes respectively. Broader impacts: The study will provide an effective technique to collect personal network information from individuals such as scientists and engineers. The proposed technique can then be used in surveys to collect information to develop new social capital indicators for the science and engineering workforce. As a result, researchers can use these indicators to investigate how various dimensions (e.g. advice, support, etc.) of personal networks may explain the productivity and career outcomes of scientists and engineers. More generally, this project will advance the understanding about individuals' personal networks as well as the data collection technique for personal network research. It will also offer new insights into the understanding of the psychology of survey response. Sharon Sassle, Cornell University “Race and Gender Variation in STEM Employment and Retention: A Cohort Analysis Using SESTAT Data” The continuing underrepresentation of women, Blacks and Hispanics in science and engineering occupations impedes efforts to increase the size of the science, technology, engineering and mathematics (STEM) workforce, a concern for policy makers who view science and scientists as critical to the future of the U.S. economy. Existing research shows that one factor contributing to this underrepresentation is that gains in women's, Blacks' and Hispanics' representation among STEM college majors do not necessarily translate into equal gains in STEM employment. Additionally, women, Blacks, and Hispanics remain far less likely than White or Asian men to be employed in most STEM occupations, particularly outside the life sciences. But little existing research studies trends over time in gender and race-ethnic differences in STEM employment or factors underlying these patterns. The proposed project will use seven waves of the National Science Foundation's (NSF's) Scientists and Engineers Statistical Data System (SESTAT) to study gender and race-ethnic differences in employment in STEM occupations among college graduates who hold a STEM degree. Broader Impacts: Government spending to educate and train STEM workers is considerable, reaching nearly 900 million by NSF in 2011. These investments and the need to increase the numbers of women and underrepresented minority scientists to maintain the future health of the STEM workforce make retention of STEM workers in related occupations a critical policy issue. This project will enhance the ability of public- and private sector policy makers and program directors to develop and implement practices that encourage the retention of women and underrepresented minorities in STEM occupations. PREPUBLICATION COPY: UNCORRECTED PROOFS