Data Analytics and Policy

Permanent URI for this collection

https://jscholarship.library.jhu.edu/handle/1774.2/63122

Browse

Now showing 1 - 20 of 24

A QUANTITATIVE ANALYSIS OF ADAPTATIONS BY PERINATAL PSYCHIATRY ACCESS PROGRAMS TO PROMOTE MENTAL HEALTHCARE EQUITY
(2023-05) Alvey, Cody
Perinatal Psychiatry Access Programs (“access programs”) build the capacity of healthcare professionals in the perinatal field. As a systems-level intervention, the access programs aim to improve perinatal mental healthcare. This study uses multivariate regression analysis to examine (1) state and program characteristics associated with mental healthcare equity adaptations and (2) state factors associated with the creation of access programs. The analysis showed that both state and program characteristics correlated with mental healthcare equity adaptations. The state qualities that showed a positive relationship with creating an access program were federal funding, the legislative majority party, the percent of births to marginalized people, and median household income. These findings reveal that decisions regarding equity adaptations can lead to more targeted interventions within the national network of access programs as well as provide insight. Understanding the critical factors associated with states could inform the existing network to ensure success in creating new programs.
DRY TAPS: ANALYZING THE EFFECTIVENESS OF DROUGHT CONSERVATION ACTIONS ENACTED BY CALIFORNIA WATER SUPPLIERS
(2023-05) Cheung, Galen
Water conservation policy during a drought in California is often left to the water supplier to develop and implement with the goal of meeting a defined reduction. In December 2022, there were over 370 water suppliers in California. Most research has focused on the response of conservation actions by studying select individual suppliers. However, given California’s interconnected water systems, a statewide study is warranted. This paper analyzes the cumulative actions of water suppliers using multivariable regression models and identifies what conservation actions reduced residential water use and provides reasons why each supplier should implement their own strategies to reduce the water consumption of their customer base. The analysis shows that certain actions, such as water usage restrictions and drought surcharge pricing, were effective at a statewide level in obtaining a statistically significant reduction in residential water consumption. Additional findings show responsiveness to drought policy is complex and can vary by region, supporting local research and subsequent implementation.
THE EFFECTS OF HEALTH STATUS ON POLITICAL ENEGAGEMENT: A QUALITATIVE ANLAYSIS OF HOW SOCIAL CHARACTERISTICS IMPACT HELATH AND VOTING
(2022-08) Prisco, Rebecca
The connection between health status and its relationship to political engagement has not been frequently studied, though many past studies have indicated that social characteristics influence political engagement. Since social characteristics are also known to be a strong predictor of health, in this analysis, it was investigated whether or not there was a significant impact of health ailments on voting participation. The Voters variable (defined as the count of registered voters who showed up at the polls per State) was compared with Chronic Disease Prevalence (defined as counts of Chronic Disease Indicators (CDI) per state, such as: cancer, cardiovascular disease, etc.) utilizing liner models. There overall was a mild negative correlation found between voter turnout and CDIs, indicating that health has some undesirable effects on one’s political engagement. While future analyses are likely needed to test for outside factors, these results represent not only the need for improving health care in rural areas, but also the need for policies that encourage voting among populations with high chronic disease prevalence.
FULL HOSPITALS, EMPTY CLASSROOMS: THE IMPACT OF THE COVID-19 PANDEMIC ON CHRONIC ABSENTEEISM IN SCHOOLS
(2022-12) Young, Evan A.
Educators across the United States are holding their breath to see if their empty classrooms during the height of the COVID-19 pandemic remain so even after an end to widespread school closures. Existing research on chronic absenteeism has yielded insight into which students are particularly vulnerable to missing school and the debilitating long-term impacts for chronically absent students. This paper employs panel methods to analyze to explore how the pandemic impacted the rates and realities of chronic absenteeism in New York City public schools, relying on cross-sectional and panel observation data across several years from publicly available chronic absenteeism and demographic data of over 1,800 Department of Education schools, COVID-19 death and case rates from the NYC Department of Health and Mental Hygiene, and economic indicators from the US Census. This paper finds that the COVID-19 death rate of a community was associated with increased rates of chronic absenteeism within its schools, as were a range of economic and demographic characteristics that point to a sustained and amplified absenteeism crisis without targeted, immediate action from policymakers.
THE ENVIRONMENTAL IMPACT OF SUSTAINABLE INVESTMENT REGULATION: AN EVALUATION OF THREE ESG POLICIES IN THE UNITED STATES AND EUROPEAN UNION
(2022-12) Resheidat, Amina
The practice of weighting Environmental, Social, and Governance (ESG) risks in investing is thought to be critical in combating climate change by creating incentives for corporations to reduce their greenhouse gas (GHG) emissions. But existing literature lacks evidence of environmental-specific impact of regulation on corporate ESG performance and disclosures. This article reviews data on carbon dioxide (CO2) emissions between the years 1750 and 2020 in the United States and the 28 countries of the European Union to assess the impact of ESG regulatory policies. Findings show that the policies are associated with a decrease in CO2 emissions when controlling for changes in population, gross domestic product, and other country-specific attributes.
WHY CALIFORNIA ENACTED THE CCPA: ARE CONCERNS ABOUT DATA PRIVACY ENOUGH TO DRIVE STATE OR FEDERAL LEGISLATION?
(2022-12) Masamitsu, Casey
Data creation and usage is common practice for many corporations, governments, and consumers. Ample research suggests that Americans on both sides of the political spectrum seek data privacy protections, yet no robust federal data privacy law exists. However, California recently passed a thorough data privacy regulation called the California Consumer Protection Act (CCPA), strengthened by the California Privacy Rights Act (CPRA). This study first creates individual “data privacy scores” for a sample of Americans and uses the scores to conduct regression analysis. Next, further analysis determines that Californians do not hold differing views about who should be responsible for implementing data privacy protections. The analysis results demonstrate that Californians’ data privacy concerns are no more heightened than residents in other states. The finding that Americans share similar views on data privacy is substantial, as understanding what drives politicians to create policy is necessary to steer activists in a more efficacious direction.
Schools of Encanto: Puerto Rico's Schools Closure Policy Study Case
(2022-05) Rivera Rivera, Edward R.
Current research on school closures in the United States uses observational methods aiming to measure students’ academic performance; however, studies are limited to big cities and either elementary or high school students. Large-scale school closures in Puerto Rico during year 2018-19 allowed for a direct comparison of academic performance in Spanish, Mathematics, English, and Science on schools that received displaced students, in both big and small cities and across all educational levels. Using t-tests to compare difference in means of standardized testing scores on schools receiving displaced students and schools not receiving displaced students, this study’s findings revealed that, on average, schools receiving displaced students performed worse than their counterparts on all educational levels and, except for Mathematics, across all disciplines. Since local authorities cited only budget restructuration as justification for school closures unlike other cases studied, findings further advance the literature on school closure policies to attempt to diminish negative, academic effects on displaced students.
Patterns of Poverty: A Regional Exploration of Poverty in America
(2022-05) Clark, Jackie M.
Existing research on American poverty largely focuses on the national average experience of poverty, offering sweeping conclusions. This article describes regional differences in poverty outcomes, challenging the notion that there is a nearly singular experience of poverty. For example, while some research argues poverty occurs most commonly in minority communities, this research provides evidence for, and descriptions of, poverty in mostly white areas on average. 725 counties were grouped into four distinct regions based on proximity, and Census definitions of US Regions. K-means clustering was leveraged across each of the regions, producing two clusters per region. One cluster presented a notably higher mean poverty rate than the other in all cases. Employment, health, and sociodemographic cluster means varied, revealing unique patterns of poverty-relevant outcomes. The relationship of cluster and poverty was validated against an external measure reflective of poverty using Analysis of Variance (ANOVA). The relationship was found to be significant in two of four regions. Tukey’s post-hoc testing further detailed significant pairwise cluster differences. These results offer support for the utility of smaller-scale state or local anti-poverty policy, and support the hypothesis that poverty manifests differently across relatively small geographic regions.
Trump or Bust: How Political Identification Impacts Opinions on Health Reform
(2022-05) Duckwitz, Stefanie
Considerable research has been conducted on the development of one’s political ideology and its impact on opinions regarding policy. This study contributes to this body of research by studying factors which influence the formation of opinions on health care policy, through a framework of social and behavioral theories. The theories of political socialization, behavior and heuristics, motivated reasoning, and social identity were used as a foundation to examine this relationship. A logistic regression was conducted on the 2016 and 2020 American National Election Study (ANES) datasets to measure the effects of political identification on opinions of health care policy. Statistically significant results from the analysis support the original hypothesis—an expected direct, positive relationship between same-party political identification and health care approval—suggesting that approval for health policy is heavily politically aligned. Despite negligible changes in health policy between the two presidencies, cumulative approval ratings of same-party respondents were overwhelming in both cases.
The Price of Transgender Military Service: Do Public Opinions on Government Defense Spending Predict Support for Allowing Transgender People to Serve?
(2022-05) Sterling, Scott
While a significant body of research exists regarding public attitudes toward the transgender community, little has been conducted specifically concerning views on transgender military service. Existing scholarship suggests that, much as with the gay and lesbian community, public attitudes toward transgender people are tied to one’s age, sex, political identity, and guiding social values. This paper examines whether public support for transgender military service is tempered by concerns regarding the potential military budget impacts alleged by some in politics, or simply driven by the same factors that predict attitudes toward the transgender community more generally. To that end, this study compares Ordinary Least Squares (OLS) regression models of both public feelings toward the transgender community as a whole and public support for transgender military service. The results provide insufficient evidence to support the hypothesis that those who feel the government should increase military spending will be less likely to support transgender military service. Rather, predictors of public support for transgender military service parallel those for feelings toward transgender people more broadly, without regard for purported military budget impacts.
Shifting from Individual to Community Measures: The Impact of the Healthy Corner Stores Initiative on Obesity Rates in Washington, DC
(2021-12) Rayson, Lauren
In recent years local health departments have worked to address the impacts of food insecurity through environment-based food intervention programs. Program evaluations on intervention effectiveness typically focus on the individual-level impacts of these programs through measurements like store inventory and consumer purchasing patterns. This research expands the scope of these evaluations by examining the impacts of Washington, DC's Healthy Corner Stores Initiative on ward-level obesity rates. A difference-in-differences estimator was used to capture changes in average adult obesity rates by ward before and after the program's implementation. Although customer survey data suggests that consumers are purchasing fresh produce through participating stores, this research shows that, on the community level, this program has not had a measurable impact on average adult obesity rates. This finding is at odds with past findings of environment-based food intervention programs, suggesting that evaluating these programs on the community level instead of the individual level can lead to contradictory conclusions.
Is Social Media a Virus: How Social Media Influences Attitudes Toward COVID-19
(2021-12) Johnsen, Kiera
Recent research on the utilization of social media for receiving COVID-19 information finds most individuals utilize social media for knowledge on vaccines and evolving social distancing policies. Existing research on the influence of social media indicates these platforms can increase the likelihood of believing misinformation related to COVID-19. This paper utilizes regression modeling to analyze whether social media can increase negative sentiments toward COVID-19 policies. Results indicate social media does not display a strong relationship with more negative COVID-19 policies in isolation. Rather, higher social media usage can exacerbate existing negative COVID-19 sentiments when certain factors are present. These include lower levels of political knowledge, weaker social ties, and when users utilize specific social media platforms which better enable them to ignore conflicting information. However, this analysis found ideology also impacts the relationship between using social media and the development of COVID-19 policy sentiments, suggesting echo chambers and party tribalism likewise impact the internalization of COVID-19 information found on social media. These results offer partial support for the hypothesis that increasing social media usage can lead to negative sentiments toward COVID-19, especially when using social media platforms whose structure better enables the development of insulated environments.
Impacts of State Pretrial Detention Policies on Jail Populations and Crime Rates: Implications for Bail Reform
(2021-08) Sorensen, Jessica L.
Pretrial detention policies in the United States contribute to hundreds of thousands of unconvicted people being held in jail on any given day, either because they cannot afford bail or have been denied bail based on a judgement of their dangerousness. Low income and minority people are disproportionately incarcerated while unconvicted, and reform efforts aimed at countering injustices in pretrial systems are on the rise. However, these reforms face strong opposition from critics arguing that more lenient pretrial release policies will lead to increased crime. Research at the municipality level has generally found that pretrial detention may prevent someone from being rearrested in the short term, but that those detained pretrial are more likely to recidivate in the long term. The present study adds new evidence to the bail reform debate through state-level analysis of the impacts of pretrial detention policies on jail incarceration rates and crime rates. The pretrial detention policy found to have the strongest negative correlation with jail incarceration rates did not have any significant effect on violent or property crime rates, providing evidence that contradicts claims from bail reform opponents that more lenient pretrial detention policies will lead to increased crime.
Medicaid Expansion and Effects on Racial Disparities in Infant Mortality
(2021-08) Garcia, Julian
The Medicaid expansion has impacted several areas within the United States healthcare system including infant mortality rate. This research revisits racial disparities in infant mortality rate that were previously analyzed for the limited years past after 2014. By addressing newer years, this study assesses the impact of the Affordable Care Act's 2014 expansion of Medicaid on racial disparities in infant mortality. To analyze changes in racial disparities among infant mortality rate, difference-in-difference and multivariate regression models were used to study the impact of the Medicaid expansion on infant mortality rate over time and to evaluate the impact of the primary predictors in studies of infant mortality. The results of this research demonstrates that the 2014 Medicaid expansion decreased infant mortality rate, however, the expansion also continued existing racial disparities.
Other Social Effects: A Time-Series Analysis Comparing Social Vulnerability Changes Between Lower & Higher Income Communities from Flood Control Projects
(2021-08) Walter, Gerard
In January 2021, the United States Army Corps of Engineers (USACE) released a memorandum requiring future projects to be justified beyond National Economic Development benefits. The incorporation of Other Social Effects (OSE) benefits could be advantageous for better protecting low socioeconomic status (SES) neighborhoods in future flood and coastal storm projects. Past studies have analyzed different social vulnerability indices (SVI), but not in a historical context that analyzes the change in impacts during flood events. Three different methods of evaluating exposure indices (EI) were used within census tracts of Harris County, Texas using the 2019 American Community Survey and the CDC’s SVI. These EIs were assessed with historical flood data to calculate OSE benefits. A time series model indicated that high SES neighborhoods have had more protection projects implemented than low SES neighborhoods and that the EI used in previous studies by USACE is an acceptable method of evaluating OSE benefits.
Lowering Crime in Manhattan: The Effects of Implementing Predictive Policing Methods on Crime Rates
(2021-08) Noerenberg, Ellis
Current research shows an unclear picture of the impact predictive policing has on decreasing crime rates. Surging police officers to high-crime “hot spots” does lead to some deterrence; however, the practice is unsustainable over a long period and has led to new methods such as the “repeat victimization” method. The regression discontinuity design used to examine crime rates before and after the implementation of the predictive policing model focused on seven felony crimes: murder, rape, robbery, assault, burglary, larceny, and vehicular larceny. The results were not uniform across the board with murder, rape, assault, and vehicular larceny showing statistically significant differences post-implementation while robbery, burglary, larceny, and total arrests did not show significant differences. The crime rates for murder, rape, assault, and vehicular larceny were all higher post-implementation than pre-implementation. Overall, the results in this study are muddled and additional research is required before determining whether the specific policies effectively result in lower crime.
Mental Health and Employment in TANF Families
(2021-08) Aprile, Catherine Marie
One of the main goals of Temporary Assistance for Needy Families (TANF) is to return parents to work, yet parents receiving TANF experience various barriers to employment that are not easily remedied. One of those barriers is an increased incidence of mental health issues. Previous literature has established a strong connection between mental health and employment, but widespread rigorous studies have not yet determined that mental health care can improve employment outcomes. Using data from California’s CalWorks program and multiple regression analysis, this paper demonstrates that for every six individuals from two-parent families referred for mental health services, an additional TANF recipient received a job three, six, nine, and twelve months later. The same results were not demonstrated for non-two-parent families and the TANF population as whole. With this information, states can use their block grant to spend more on mental health to better reach their goal of returning parents to employment.
Do Current Use of Force Policies Reduce the Level of Injury for Civilians and Officers? A Study of the Effectiveness of Existing Police Policies and How They Contribute to Nation-Wide Upset
(2021-05) Stirn, Caroline
Current policing research evaluates effectiveness of added procedures boosting accountability, whether less lethal weapons are effective in replacing lethal weapons when force is required, or how policies differ across the nation. This article turns focus toward evaluating the existing use of force policing policy in one state. Six ordered logistic regression models were used to determine whether the policy as well as event characteristics reduce the level of severity of injury obtained in an encounter where force was applied for both civilians and officers and if the policy is effectively applied to all populations. The officer’s perception of the civilian and the officer’s intention in applying force were also considered. Given the tense climate between police and public, it is suspected and confirmed that the policy will not be effective nor universally applied. Following the policy does not slow the severity of injury sustained, nor impact all demographics equally.
Money Talks: The Role of Income & Financial Security in Determining Immigration Attitudes
(2020-12) Stastny, Kaitlin
Existing research on public attitudes toward immigration policy in the United States is divided. Lack of contact with immigrants and threat perception are two popular determinants that allegedly explain restrictionist attitudes; in contrast, individual financial standing is often overlooked. This paper challenges the claim that financial considerations play no role in the formation of public opinion toward immigration. Using 2020 Amazon Mechanical Turk survey data, an ordered logistical regression model reveals a robust link between financial security and immigration preferences, with those better off being more likely to favor greater restrictions. Although findings on income are less certain, they suggest that middle earners may be more preferential toward establishing greater restrictions than low or high earners. Surprisingly, proponents of immigration restrictions are found less likely to feel threatened by COVID-19 or election fraud. This contradicts threat theory and bolsters the argument that, in the context of competing influences on immigration attitudes, money talks.
Media Matters: Developing a Media Driven Model to Understand Bill Passage Rate in the Arizona State Legislature
(2020-12) Roth, Benjamin
The influence of media effects on policy is well documented. Existing research often points to clear relationships between media attention on national issues and legislative activity on those issues; increased attention on a policy area leads to an increase in bills passed on that topic. While state legislatures pass bills at a higher rate than Congress, less research has been focused on state-level activity. As state-based newspapers often focus on their state-relevant issues more so than national newspapers, this research seeks to develop a generalized media effects model by focusing on article activity of four Arizona newspapers: The Arizona Daily Star, the Arizona Republic, the Arizona Capitol Times, and the Phoenix New Times. While accounting for bill meta variables (number of sponsors, number of committees introduced to, etc.) and bill text data, a logit model was trained on per-bill-topic article counts from each of those four newspapers to determine their effect on the probability of bill passage in the Arizona Legislature. None of the article variables proved to be significant, suggesting that local media coverage on the topic of general legislative topics does not have the same impact on bill passage as national coverage on singular issues. Bill passage rate, however, positively statistically significantly increased with increases to bill meta variables; this finding confirms that these demographic variables are important in understanding why a bill passes. Finally, while initial analysis showed that text data is not inferential to the passage of bill, more work remains to confirm this finding.

Browse

Recent Submissions