A review of Brower Psychological Services' evaluation practices from 2018-2022 found:
1) No violations of adverse impact standards or statistically significant correlations between race/ethnicity and pass rates when looking at all agencies served in aggregate.
2) No adverse impact violations for the Aurora Police Department in any year, though 2021 first-stage evaluations showed a possible significant relationship between race and pass rates.
3) Additional analyses suggest the 2021 Aurora PD results were likely due to an undisclosed special recruiting program, not inherent bias in evaluation practices. Uniform processes across time and agencies were observed.
Show Me One or Six? A Field Study Examining How Lineup Format Affects Real-Wo...Heather Flowe
1) This field study examined how intoxicated witnesses' identification performance is affected by identification format (lineup vs. showup).
2) In target-present conditions, alcohol consumption and identification format had no effect on witnesses' ability to identify the perpetrator.
3) In target-absent conditions, identification format affected choosing behavior, with showups increasing the likelihood of choosing, but this did not significantly enhance false identifications.
The document analyzes demographic data and donor registry statistics for counties in Texas. It finds that the percentage of the population on the donor registry varies between TOSA counties from 15.79% to 23.64%. Additionally, it calculates odds ratios comparing the likelihood of Hispanics and Caucasians being on the donor registry. The odds ratios show Hispanics are less likely than non-Hispanics to be registered across all TOSA counties, while Caucasians are more likely than non-Caucasians to be registered.
1) The Oakland Police Department collects stop data to analyze potentially biased policing behaviors and ensure stops are conducted fairly. During a recent 8-month period, African Americans comprised 62% of stops despite being a smaller portion of the population.
2) African Americans and Hispanics had higher search rates than other groups but recovery rates were consistent across races. Excluding searches incident to arrest, Hispanics had the highest recovery rate.
3) Compared to other races, African Americans received more felony arrests and field investigation reports from stops but the lowest percentage of citations. Ongoing review aims to identify any unexplained disparities.
This presentation was provided to the Philadelphia EMA HIV Integrated Planning Council by Briana Morgan of the Office of HIV Planning. It includes data related to population-level data, race/ethnicity, STIs, risk behaviors, HIV, and more.
Descriptive and Inferential Statistical Methods: Analysis of Voting and Elect...Toni Menninger
This document discusses descriptive and inferential statistical methods for analyzing voting and elections. It provides graphs and charts showing descriptive statistics on voter turnout rates by income, age, and gender using data from 2008-2012. It also discusses inferential concepts like how election polls can be viewed as binomial experiments and the factors that influence poll accuracy such as sample size. More sophisticated election forecasting models aggregate data from multiple polls to estimate probabilities of different election outcomes.
Assignment (2- to 3-page case study analysis)Scenario 6.docxjane3dyson92312
Assignment (2- to 3-page case study analysis)
Scenario:
67-year-old female presents with chief complaint of shortness of breath, fatigue, weakness, unintentional weight loss, and mild numbness in her feet. She states she feels unsteady when she walks. PMH includes hypothyroidism well controlled on Synthroid 100 mcg/day. No hx of HTN or CHF.
Vital signs: Temp 98.7 F, pulse 118, Respirations 22, BP 108/64, PaO2 95% on room air.
Physical exam revealed pale, anxious female appearing older than stated years.
HEENT- pale conjunctiva of eyes and pale palate. Tongue beefy red and slightly swollen with loss of normal rugae. Turbinates pale but no swelling. Thyroid palpable but no nodules felt. No lymph nodes palpated.
Cardiac-regular rate and rhythm with soft II/VI systolic murmur. Respiratory- lungs clear with no adventitious breath sounds. Abdomen-soft, non-tender with positive bowel sounds. Liver edge palpated two finger breadths below right costal margin. Lab data- hemoglobin (hgb), hematocrit (hct), reticulocyte count, serum B12 levels low, mean corpuscle volume (MCV), plasma iron, and ferritin levels high, folate, Total Iron Binding Capacity (TIBC) are normal.
In your Case Study Analysis related to the scenario provided, explain the following as it applies to the scenario you were provided
(
not all bullet points below may apply to each scenario- only include those that apply.
):
The factors that affect fertility (STDs).
Why inflammatory markers rise in STD/PID.
Why prostatitis and infection happens. Also explain the causes of systemic reaction.
Why a patient would need a splenectomy after a diagnosis of ITP.
Anemia and the different kinds of anemia (i.e., micro and macrocytic).
Rubric:
Develop a 1- to 2-page case study analysis, examining the patient symptoms presented in the case study. Be sure to address the following as it relates to the case you were assigned (omit section that does not pertain to your case, faculty will give full points for that section):
Explain the factors that affect fertility (STDs)
Points Range:23 (23.00%) - 25 (25.00%)
The response accurately and thoroughly describes the patient symptoms.
The response includes accurate, clear, and detailed explanations of the processes related to women's and men's health, infections, and hematologic disorders and is supported by evidence and/or research, as appropriate, to support the explanation.
Points Range:20 (20.00%) - 22 (22.00%)
The response describes the patient symptoms.
The response includes accurate, explanations of the processes related to women's and men's health, infections, and hematologic disorders and is supported by evidence and/or research, as appropriate, to support the explanation.
Points Range:18 (18.00%) - 19 (19.00%)
The response describes the patient symptoms in a manner that is vague or inaccurate.
The response includes explanations of the processes related to women's and men's health, infections, and hematologic disorders,.
This document provides an overview of different types of statistical tests used for data analysis and interpretation. It discusses scales of measurement, parametric vs nonparametric tests, formulating hypotheses, types of statistical errors, establishing decision rules, and choosing the appropriate statistical test based on the number and types of variables. Key statistical tests covered include t-tests, ANOVA, chi-square tests, and correlations. Examples are provided to illustrate how to interpret and report the results of these common statistical analyses.
Show Me One or Six? A Field Study Examining How Lineup Format Affects Real-Wo...Heather Flowe
1) This field study examined how intoxicated witnesses' identification performance is affected by identification format (lineup vs. showup).
2) In target-present conditions, alcohol consumption and identification format had no effect on witnesses' ability to identify the perpetrator.
3) In target-absent conditions, identification format affected choosing behavior, with showups increasing the likelihood of choosing, but this did not significantly enhance false identifications.
The document analyzes demographic data and donor registry statistics for counties in Texas. It finds that the percentage of the population on the donor registry varies between TOSA counties from 15.79% to 23.64%. Additionally, it calculates odds ratios comparing the likelihood of Hispanics and Caucasians being on the donor registry. The odds ratios show Hispanics are less likely than non-Hispanics to be registered across all TOSA counties, while Caucasians are more likely than non-Caucasians to be registered.
1) The Oakland Police Department collects stop data to analyze potentially biased policing behaviors and ensure stops are conducted fairly. During a recent 8-month period, African Americans comprised 62% of stops despite being a smaller portion of the population.
2) African Americans and Hispanics had higher search rates than other groups but recovery rates were consistent across races. Excluding searches incident to arrest, Hispanics had the highest recovery rate.
3) Compared to other races, African Americans received more felony arrests and field investigation reports from stops but the lowest percentage of citations. Ongoing review aims to identify any unexplained disparities.
This presentation was provided to the Philadelphia EMA HIV Integrated Planning Council by Briana Morgan of the Office of HIV Planning. It includes data related to population-level data, race/ethnicity, STIs, risk behaviors, HIV, and more.
Descriptive and Inferential Statistical Methods: Analysis of Voting and Elect...Toni Menninger
This document discusses descriptive and inferential statistical methods for analyzing voting and elections. It provides graphs and charts showing descriptive statistics on voter turnout rates by income, age, and gender using data from 2008-2012. It also discusses inferential concepts like how election polls can be viewed as binomial experiments and the factors that influence poll accuracy such as sample size. More sophisticated election forecasting models aggregate data from multiple polls to estimate probabilities of different election outcomes.
Assignment (2- to 3-page case study analysis)Scenario 6.docxjane3dyson92312
Assignment (2- to 3-page case study analysis)
Scenario:
67-year-old female presents with chief complaint of shortness of breath, fatigue, weakness, unintentional weight loss, and mild numbness in her feet. She states she feels unsteady when she walks. PMH includes hypothyroidism well controlled on Synthroid 100 mcg/day. No hx of HTN or CHF.
Vital signs: Temp 98.7 F, pulse 118, Respirations 22, BP 108/64, PaO2 95% on room air.
Physical exam revealed pale, anxious female appearing older than stated years.
HEENT- pale conjunctiva of eyes and pale palate. Tongue beefy red and slightly swollen with loss of normal rugae. Turbinates pale but no swelling. Thyroid palpable but no nodules felt. No lymph nodes palpated.
Cardiac-regular rate and rhythm with soft II/VI systolic murmur. Respiratory- lungs clear with no adventitious breath sounds. Abdomen-soft, non-tender with positive bowel sounds. Liver edge palpated two finger breadths below right costal margin. Lab data- hemoglobin (hgb), hematocrit (hct), reticulocyte count, serum B12 levels low, mean corpuscle volume (MCV), plasma iron, and ferritin levels high, folate, Total Iron Binding Capacity (TIBC) are normal.
In your Case Study Analysis related to the scenario provided, explain the following as it applies to the scenario you were provided
(
not all bullet points below may apply to each scenario- only include those that apply.
):
The factors that affect fertility (STDs).
Why inflammatory markers rise in STD/PID.
Why prostatitis and infection happens. Also explain the causes of systemic reaction.
Why a patient would need a splenectomy after a diagnosis of ITP.
Anemia and the different kinds of anemia (i.e., micro and macrocytic).
Rubric:
Develop a 1- to 2-page case study analysis, examining the patient symptoms presented in the case study. Be sure to address the following as it relates to the case you were assigned (omit section that does not pertain to your case, faculty will give full points for that section):
Explain the factors that affect fertility (STDs)
Points Range:23 (23.00%) - 25 (25.00%)
The response accurately and thoroughly describes the patient symptoms.
The response includes accurate, clear, and detailed explanations of the processes related to women's and men's health, infections, and hematologic disorders and is supported by evidence and/or research, as appropriate, to support the explanation.
Points Range:20 (20.00%) - 22 (22.00%)
The response describes the patient symptoms.
The response includes accurate, explanations of the processes related to women's and men's health, infections, and hematologic disorders and is supported by evidence and/or research, as appropriate, to support the explanation.
Points Range:18 (18.00%) - 19 (19.00%)
The response describes the patient symptoms in a manner that is vague or inaccurate.
The response includes explanations of the processes related to women's and men's health, infections, and hematologic disorders,.
This document provides an overview of different types of statistical tests used for data analysis and interpretation. It discusses scales of measurement, parametric vs nonparametric tests, formulating hypotheses, types of statistical errors, establishing decision rules, and choosing the appropriate statistical test based on the number and types of variables. Key statistical tests covered include t-tests, ANOVA, chi-square tests, and correlations. Examples are provided to illustrate how to interpret and report the results of these common statistical analyses.
Patterns of political elite formation in post soviet Armenia (ASCN funded res...Hamazasp Danielyan
This document summarizes research on patterns of political elite formation in post-Soviet Armenia. It describes the project's objectives to describe and conceptualize how political elites are formed, recruited, and their motivations and values. The methodology included surveys, interviews, and biography analysis. Key findings from the survey showed that most see their future as better and have higher trust in local government than political parties. The document concludes that possible political elite representatives include civil servants, members of political parties who are already involved in politics, and those in local government.
QNT 275 FINAL EXAM NEW 2016
Buy Solutions: http://hwsoloutions.com/downloads/qnt-275-final-exam-new-2016/
What is the name of the variable that’s used to predict another variable?
Explanatory
Standard error of the estimate
Response
Coefficient of determination
Professors at a local university earn an average salary of $80000 with a standard deviation of $6000. The salary distribution is approximately bell-shaped. What can be said about the percentage of salaries that are at least $74,000?
About 97.5 percent
About 84 percent
About 68 percent
About 95 percent
What type of relationship is indicated in the scatterplot?
A positive linear or curvilinear relationship
No relationship
A negative curvilinear relationship
A negative linear relationship
The study of statistics can be defined as:
the art and science of getting information from data.
All of the answers
the language of data.
the study of collecting, analyzing, presenting, and interpreting data.
In the accompanying stem-and-leaf diagram the values in the stem and leaf portions represent 10s and 1s digits, respectively.
The stem-and-leaf diagram shows that the distribution is ___________.
symmetric
positively skewed
negatively skewed
None of the ans
Reuters/Ipsos Core Political Survey: Coronavirus Tracker (04/01/2020)Ipsos Public Affairs
This document provides a summary of an Ipsos poll conducted for Thomson Reuters between March 30-31, 2020. It includes data from a survey of 1,114 Americans, including registered Democratic, Republican, and Independent voters. The document summarizes responses on views of the direction of the country, most important problems facing America, approval of President Trump, responses to the coronavirus pandemic, and political identity. It also includes information on how the data was weighted and credibility intervals.
Magellan Strategies Montana US Senate Survey Release April 10, 2014Magellan Strategies
Magellan Strategies survey of 2,490 likely Montana general election voters. The survey includes a voter mood question, US Senate ballot test of Steve Daines, John Walsh, and Roger Roots, and a Congressional generic ballot test.
1) The document discusses using frequency distributions and crosstabulations in SPSS to analyze relationships between categorical variables.
2) It provides an example of a frequency distribution analyzing the ratio of Democratic and Republican candidates.
3) It then demonstrates using a crosstabulation and chi-square test to determine if there is a statistically significant relationship between ethnicity and political party affiliation in a data set.
This collection of slides are meant as a starting point and tutorial for the ones who want to understand AI Ethics and in particular the challenges around bias and fairness. Furthermore, I have also included studies on how we as humans perceive AI influence in our private as well as working lives.
Tunisie sondage : Une économie en manque de performance et la corruption cont...Jamaity
Tunis, Tunisie— Un nouveau sondage réalisé par le centre de l’Institut Républicain International (IRI) pour les sondages révèle que les mauvaises conditions économiques et la corruption continuent de susciter une grande insatisfaction publique en Tunisie.
« Ce sondage renforce une nécessité vitale de résoudre les problèmes les plus préoccupants des tunisiens : la corruption et l’économie, » déclare Scott Mastic, Directeur Régional pour l’Afrique du Nord et le Moyen Orient. » Nos résultats renforcent le besoin de faire face à la corruption aux niveaux local et national, et espérons que le gouvernement intègre cette approche dans sa ‘guerre contre la corruption’ lancée tout récemment »
Un total de 87% des tunisiens décrivent la situation économique comme mauvaise (26%) ou très mauvaise (61%). (44%) disent que le chômage est le plus grand problème auquel fait face le pays, suivi de l’économie et la crise financière à 24%. Une majorité claire de répondants (72%) pense que la lutte contre la corruption est le meilleur moyen pour améliorer l’économie de leurs communautés, suivi de 59% qui disent que les collectivités locales vont « rendre plus facile de lancer des entreprises pour les entrepreneurs »
La corruption continue d’être une préoccupation majeure, avec 89% de tunisiens qui affirment que la corruption est plus élevée aujourd’hui qu’avant la révolution démocratique de 2011. 46% pensent que les relations personnelles (wasta) mèneraient le « plus probablement » vers la richesse en Tunisie. Particulièrement, pour la deuxième fois dans l’histoire des sondages en Tunisie, les répondants ont mis les hôpitaux en tête des institutions où ils auraient fait face à la corruption (17%).
Le sondage indique aussi in déclin de l’intérêt à la participation au processus électoral. 50% des répondants disent qu’ils sont « peu » (9%) ou « très peu » (41%) enclins à aller voter aux élections municipales-une augmentation de 7 point par rapport à avril 2017. En outre, 83% ne savent pas ou refusent de dire à quels partis ils voteraient si les élections allaient être organisées demain.
Méthodologie
Ce sondage a été mené pour le compte du Centre de l’Institut Républicain International (IRI) pour les sondages par la société tunisienne, ELKA Consulting sous la supervision de Chesapeake Beach Consulting. Les questions avaient focalisé en premier sur l’économie afin de fournir aux différentes parties prenantes des données et de permettre aussi de leur indiquer les priorités publiques.
Les données ont été collectées du 11 au 17 Août 2017 moyennant des interviews en face à face. Un échantillon national de 1226 Tunisiens âgés de 18 ans et plus. Un sur-échantillonnage a été effectué dans trois régions connaissant des difficultés économiques. La marge d’erreur étant de plus ou moins 2, 85% au milieu de gamme de niveau de confiance de 95%. Les chiffres indiqués dans les graphs et des tableaux peuvent ne pas faire la somme de 100% en raison de l’arrondissement.
A 2016 Election Post-Mortem: The ABC News/Washington Post Tracking PollLangerResearch
This document summarizes the findings of a post-mortem analysis of the 2016 ABC News/Washington Post tracking poll conducted after Donald Trump's unexpected election victory. The analysis found the final poll estimate of Hillary Clinton leading by 4 points was accurate based on the poll's historical average error of 2 points. While some state polls underestimated Trump support, the national poll found no evidence of "shy" Trump voters or other issues. Overall, the national popular vote estimate was sound despite missing Trump's electoral college victory.
Reuters/Ipsos Core Political: Presidential Approval Tracker (06/10/2020)Ipsos Public Affairs
This document summarizes the results of an Ipsos poll conducted for Thomson Reuters between June 8-9, 2020. It provides data on Americans' views on various political issues including:
- Most think the country is headed in the wrong direction rather than the right direction.
- The economy, healthcare, and unemployment are seen as the most important problems facing the country.
- Trump's approval ratings are around 40% with strong partisan divides. A majority disapprove of his handling of coronavirus.
- In a hypothetical 2020 election, Biden leads Trump 43% to 35% nationally.
Reuters/Ipsos Core Political: Coronavirus Tracker (05/20/2020)Ipsos Public Affairs
This document summarizes the results of an Ipsos poll conducted for Thomson Reuters between May 18-19, 2020. It provides data on Americans' views on various political issues including: approval of President Trump's job performance; most important problems facing the country; concerns about COVID-19; and views on the 2020 general election between Trump and Biden. The document also describes Ipsos' methodology, sample size, and credibility intervals for the survey results.
Reuters/Ipsos Core Political: Coronavirus Tracker (04/08/2020)Ipsos Public Affairs
This 3 sentence summary provides the key details from the document:
The document summarizes the results of an Ipsos poll conducted for Thomson Reuters between April 6-7, 2020. The poll surveyed 1,116 American adults, including 480 Democratic voters, 376 Republican voters, and 73 Independent voters. The poll findings covered topics like approval ratings for President Trump, views on important issues, and behavioral changes in response to the coronavirus pandemic.
Nearly 3 in 4 New Yorkers favor offering half-price MetroCards to low-income residents. Support cuts across party lines and a majority would be more likely to vote for a mayoral candidate supporting this. The survey of over 1,700 city residents found widespread backing for reducing public transit costs for low-income New Yorkers across incomes, ethnicities, and opinions of the current mayor.
Reuters/Ipsos Core Political: Coronavirus Tracker (06/03/2020)Ipsos Public Affairs
This document summarizes the results of an Ipsos poll conducted for Thomson Reuters between June 1-2, 2020. It provides data on key political topics from a sample of over 1,100 American adults, including views on the direction of the country, most important issues, approval of President Trump, and the 2020 election matchup between Trump and Biden. The document also outlines the methodology used in the poll and provides credibility intervals for the results.
A Correction for the Full-Supplement Imputation Bias in the Current Populatio...soder145
The document describes research analyzing bias in estimates of uninsured rates in the Current Population Survey (CPS) due to incomplete health insurance data from households. It finds the CPS method of imputing full health insurance coverage for households without complete data introduces around a 1% upward bias in uninsured rates. The researchers created alternative CPS weights that remove households without complete data to eliminate this bias, finding lower uninsured rates more consistent with other data sources.
This document provides an overview of a basic statistics workshop for language teachers. It includes definitions of key statistical concepts like descriptive statistics, the normal distribution, reliability and validity. It also summarizes how to compare means using t-tests and ANOVA, and how to analyze correlations between variables. The workshop aims to help teachers calculate and interpret common statistical measures.
- Austin, Texas has experienced strong job growth over the past several years, with 44 consecutive months of total nonfarm job growth of 4.0% or better annually. The unemployment rate in Austin is currently the lowest it's been since the late 1990s.
- Austin has gained many high-earning residents in recent years, with 39,197 new residents holding graduate degrees between 2006-2014. The city has also seen a 41% growth in households earning over $150,000 annually.
- While Austin sees significant startup activity, businesses owned by women, Hispanics, Blacks tend to be smaller and earn less on average compared to those owned by
2016 Democrat Primary: Prediction of results For New York CountiesSanjita Jain
Predicting the number of counties won by the Democrats in the primary US Presidential Election for the New York state based on demographic data of all other US counties using Logistic regression model, Random Forest, Principal Component Analysis (PCA) and Factor Analysis
Reuters/Ipsos Core Political Survey: 2020 Democratic Primary (07/03/2019)Ipsos Public Affairs
This document provides the results of an Ipsos poll conducted for Thomson Reuters between June 28 and July 2, 2019. It summarizes data on various political topics from a survey of over 2,000 American adults, including nearly 1,000 Democratic and 800 Republican registered voters. Key findings include that 31% of adults feel the country is heading in the right direction, while 56% say it is on the wrong track. Immigration, healthcare, and the economy were seen as the most important problems facing America. Approval ratings for President Trump were 41% among all adults and 43% among registered voters. The document also lists familiarity and favorability ratings for 2020 Democratic presidential candidates among Democratic voters.
The document reports on statistical tests performed on survey data related to brand characteristics of L'Oreal. Key findings include:
- Ethnicity was found to influence familiarity with the L'Oreal brand, with the white ethnic group being most familiar.
- Statistically significant relationships were found between ethnicity and attitudes of the brand being good, likeable, and desirable. Spend category was also related to attitudes of the brand being high quality, attractive, and desirable.
- Relationships were investigated between demographic variables (ethnicity, spend category) and variables measuring brand awareness, attitudes toward the brand and products, desirability, behavioral tendencies, and value proposition. Many significant relationships were found using
The document reports the results of a survey about organizational whistleblowing conducted by SHRM. It finds that the most common ways organizations inform employees about reporting unethical behavior are through employee handbooks and new employee orientation. Most organizations encourage employees to report issues to HR or their direct supervisor. Over 80% of organizations investigate all reports of unethical behavior, usually handled by HR or senior management. Larger organizations are more likely than smaller ones to have formal reporting methods like ethics hotlines.
Patterns of political elite formation in post soviet Armenia (ASCN funded res...Hamazasp Danielyan
This document summarizes research on patterns of political elite formation in post-Soviet Armenia. It describes the project's objectives to describe and conceptualize how political elites are formed, recruited, and their motivations and values. The methodology included surveys, interviews, and biography analysis. Key findings from the survey showed that most see their future as better and have higher trust in local government than political parties. The document concludes that possible political elite representatives include civil servants, members of political parties who are already involved in politics, and those in local government.
QNT 275 FINAL EXAM NEW 2016
Buy Solutions: http://hwsoloutions.com/downloads/qnt-275-final-exam-new-2016/
What is the name of the variable that’s used to predict another variable?
Explanatory
Standard error of the estimate
Response
Coefficient of determination
Professors at a local university earn an average salary of $80000 with a standard deviation of $6000. The salary distribution is approximately bell-shaped. What can be said about the percentage of salaries that are at least $74,000?
About 97.5 percent
About 84 percent
About 68 percent
About 95 percent
What type of relationship is indicated in the scatterplot?
A positive linear or curvilinear relationship
No relationship
A negative curvilinear relationship
A negative linear relationship
The study of statistics can be defined as:
the art and science of getting information from data.
All of the answers
the language of data.
the study of collecting, analyzing, presenting, and interpreting data.
In the accompanying stem-and-leaf diagram the values in the stem and leaf portions represent 10s and 1s digits, respectively.
The stem-and-leaf diagram shows that the distribution is ___________.
symmetric
positively skewed
negatively skewed
None of the ans
Reuters/Ipsos Core Political Survey: Coronavirus Tracker (04/01/2020)Ipsos Public Affairs
This document provides a summary of an Ipsos poll conducted for Thomson Reuters between March 30-31, 2020. It includes data from a survey of 1,114 Americans, including registered Democratic, Republican, and Independent voters. The document summarizes responses on views of the direction of the country, most important problems facing America, approval of President Trump, responses to the coronavirus pandemic, and political identity. It also includes information on how the data was weighted and credibility intervals.
Magellan Strategies Montana US Senate Survey Release April 10, 2014Magellan Strategies
Magellan Strategies survey of 2,490 likely Montana general election voters. The survey includes a voter mood question, US Senate ballot test of Steve Daines, John Walsh, and Roger Roots, and a Congressional generic ballot test.
1) The document discusses using frequency distributions and crosstabulations in SPSS to analyze relationships between categorical variables.
2) It provides an example of a frequency distribution analyzing the ratio of Democratic and Republican candidates.
3) It then demonstrates using a crosstabulation and chi-square test to determine if there is a statistically significant relationship between ethnicity and political party affiliation in a data set.
This collection of slides are meant as a starting point and tutorial for the ones who want to understand AI Ethics and in particular the challenges around bias and fairness. Furthermore, I have also included studies on how we as humans perceive AI influence in our private as well as working lives.
Tunisie sondage : Une économie en manque de performance et la corruption cont...Jamaity
Tunis, Tunisie— Un nouveau sondage réalisé par le centre de l’Institut Républicain International (IRI) pour les sondages révèle que les mauvaises conditions économiques et la corruption continuent de susciter une grande insatisfaction publique en Tunisie.
« Ce sondage renforce une nécessité vitale de résoudre les problèmes les plus préoccupants des tunisiens : la corruption et l’économie, » déclare Scott Mastic, Directeur Régional pour l’Afrique du Nord et le Moyen Orient. » Nos résultats renforcent le besoin de faire face à la corruption aux niveaux local et national, et espérons que le gouvernement intègre cette approche dans sa ‘guerre contre la corruption’ lancée tout récemment »
Un total de 87% des tunisiens décrivent la situation économique comme mauvaise (26%) ou très mauvaise (61%). (44%) disent que le chômage est le plus grand problème auquel fait face le pays, suivi de l’économie et la crise financière à 24%. Une majorité claire de répondants (72%) pense que la lutte contre la corruption est le meilleur moyen pour améliorer l’économie de leurs communautés, suivi de 59% qui disent que les collectivités locales vont « rendre plus facile de lancer des entreprises pour les entrepreneurs »
La corruption continue d’être une préoccupation majeure, avec 89% de tunisiens qui affirment que la corruption est plus élevée aujourd’hui qu’avant la révolution démocratique de 2011. 46% pensent que les relations personnelles (wasta) mèneraient le « plus probablement » vers la richesse en Tunisie. Particulièrement, pour la deuxième fois dans l’histoire des sondages en Tunisie, les répondants ont mis les hôpitaux en tête des institutions où ils auraient fait face à la corruption (17%).
Le sondage indique aussi in déclin de l’intérêt à la participation au processus électoral. 50% des répondants disent qu’ils sont « peu » (9%) ou « très peu » (41%) enclins à aller voter aux élections municipales-une augmentation de 7 point par rapport à avril 2017. En outre, 83% ne savent pas ou refusent de dire à quels partis ils voteraient si les élections allaient être organisées demain.
Méthodologie
Ce sondage a été mené pour le compte du Centre de l’Institut Républicain International (IRI) pour les sondages par la société tunisienne, ELKA Consulting sous la supervision de Chesapeake Beach Consulting. Les questions avaient focalisé en premier sur l’économie afin de fournir aux différentes parties prenantes des données et de permettre aussi de leur indiquer les priorités publiques.
Les données ont été collectées du 11 au 17 Août 2017 moyennant des interviews en face à face. Un échantillon national de 1226 Tunisiens âgés de 18 ans et plus. Un sur-échantillonnage a été effectué dans trois régions connaissant des difficultés économiques. La marge d’erreur étant de plus ou moins 2, 85% au milieu de gamme de niveau de confiance de 95%. Les chiffres indiqués dans les graphs et des tableaux peuvent ne pas faire la somme de 100% en raison de l’arrondissement.
A 2016 Election Post-Mortem: The ABC News/Washington Post Tracking PollLangerResearch
This document summarizes the findings of a post-mortem analysis of the 2016 ABC News/Washington Post tracking poll conducted after Donald Trump's unexpected election victory. The analysis found the final poll estimate of Hillary Clinton leading by 4 points was accurate based on the poll's historical average error of 2 points. While some state polls underestimated Trump support, the national poll found no evidence of "shy" Trump voters or other issues. Overall, the national popular vote estimate was sound despite missing Trump's electoral college victory.
Reuters/Ipsos Core Political: Presidential Approval Tracker (06/10/2020)Ipsos Public Affairs
This document summarizes the results of an Ipsos poll conducted for Thomson Reuters between June 8-9, 2020. It provides data on Americans' views on various political issues including:
- Most think the country is headed in the wrong direction rather than the right direction.
- The economy, healthcare, and unemployment are seen as the most important problems facing the country.
- Trump's approval ratings are around 40% with strong partisan divides. A majority disapprove of his handling of coronavirus.
- In a hypothetical 2020 election, Biden leads Trump 43% to 35% nationally.
Reuters/Ipsos Core Political: Coronavirus Tracker (05/20/2020)Ipsos Public Affairs
This document summarizes the results of an Ipsos poll conducted for Thomson Reuters between May 18-19, 2020. It provides data on Americans' views on various political issues including: approval of President Trump's job performance; most important problems facing the country; concerns about COVID-19; and views on the 2020 general election between Trump and Biden. The document also describes Ipsos' methodology, sample size, and credibility intervals for the survey results.
Reuters/Ipsos Core Political: Coronavirus Tracker (04/08/2020)Ipsos Public Affairs
This 3 sentence summary provides the key details from the document:
The document summarizes the results of an Ipsos poll conducted for Thomson Reuters between April 6-7, 2020. The poll surveyed 1,116 American adults, including 480 Democratic voters, 376 Republican voters, and 73 Independent voters. The poll findings covered topics like approval ratings for President Trump, views on important issues, and behavioral changes in response to the coronavirus pandemic.
Nearly 3 in 4 New Yorkers favor offering half-price MetroCards to low-income residents. Support cuts across party lines and a majority would be more likely to vote for a mayoral candidate supporting this. The survey of over 1,700 city residents found widespread backing for reducing public transit costs for low-income New Yorkers across incomes, ethnicities, and opinions of the current mayor.
Reuters/Ipsos Core Political: Coronavirus Tracker (06/03/2020)Ipsos Public Affairs
This document summarizes the results of an Ipsos poll conducted for Thomson Reuters between June 1-2, 2020. It provides data on key political topics from a sample of over 1,100 American adults, including views on the direction of the country, most important issues, approval of President Trump, and the 2020 election matchup between Trump and Biden. The document also outlines the methodology used in the poll and provides credibility intervals for the results.
A Correction for the Full-Supplement Imputation Bias in the Current Populatio...soder145
The document describes research analyzing bias in estimates of uninsured rates in the Current Population Survey (CPS) due to incomplete health insurance data from households. It finds the CPS method of imputing full health insurance coverage for households without complete data introduces around a 1% upward bias in uninsured rates. The researchers created alternative CPS weights that remove households without complete data to eliminate this bias, finding lower uninsured rates more consistent with other data sources.
This document provides an overview of a basic statistics workshop for language teachers. It includes definitions of key statistical concepts like descriptive statistics, the normal distribution, reliability and validity. It also summarizes how to compare means using t-tests and ANOVA, and how to analyze correlations between variables. The workshop aims to help teachers calculate and interpret common statistical measures.
- Austin, Texas has experienced strong job growth over the past several years, with 44 consecutive months of total nonfarm job growth of 4.0% or better annually. The unemployment rate in Austin is currently the lowest it's been since the late 1990s.
- Austin has gained many high-earning residents in recent years, with 39,197 new residents holding graduate degrees between 2006-2014. The city has also seen a 41% growth in households earning over $150,000 annually.
- While Austin sees significant startup activity, businesses owned by women, Hispanics, Blacks tend to be smaller and earn less on average compared to those owned by
2016 Democrat Primary: Prediction of results For New York CountiesSanjita Jain
Predicting the number of counties won by the Democrats in the primary US Presidential Election for the New York state based on demographic data of all other US counties using Logistic regression model, Random Forest, Principal Component Analysis (PCA) and Factor Analysis
Reuters/Ipsos Core Political Survey: 2020 Democratic Primary (07/03/2019)Ipsos Public Affairs
This document provides the results of an Ipsos poll conducted for Thomson Reuters between June 28 and July 2, 2019. It summarizes data on various political topics from a survey of over 2,000 American adults, including nearly 1,000 Democratic and 800 Republican registered voters. Key findings include that 31% of adults feel the country is heading in the right direction, while 56% say it is on the wrong track. Immigration, healthcare, and the economy were seen as the most important problems facing America. Approval ratings for President Trump were 41% among all adults and 43% among registered voters. The document also lists familiarity and favorability ratings for 2020 Democratic presidential candidates among Democratic voters.
The document reports on statistical tests performed on survey data related to brand characteristics of L'Oreal. Key findings include:
- Ethnicity was found to influence familiarity with the L'Oreal brand, with the white ethnic group being most familiar.
- Statistically significant relationships were found between ethnicity and attitudes of the brand being good, likeable, and desirable. Spend category was also related to attitudes of the brand being high quality, attractive, and desirable.
- Relationships were investigated between demographic variables (ethnicity, spend category) and variables measuring brand awareness, attitudes toward the brand and products, desirability, behavioral tendencies, and value proposition. Many significant relationships were found using
The document reports the results of a survey about organizational whistleblowing conducted by SHRM. It finds that the most common ways organizations inform employees about reporting unethical behavior are through employee handbooks and new employee orientation. Most organizations encourage employees to report issues to HR or their direct supervisor. Over 80% of organizations investigate all reports of unethical behavior, usually handled by HR or senior management. Larger organizations are more likely than smaller ones to have formal reporting methods like ethics hotlines.
Similar to Impact of Race and Ethnicity on Preemployment Psychological Assessment (20)
hematic appreciation test is a psychological assessment tool used to measure an individual's appreciation and understanding of specific themes or topics. This test helps to evaluate an individual's ability to connect different ideas and concepts within a given theme, as well as their overall comprehension and interpretation skills. The results of the test can provide valuable insights into an individual's cognitive abilities, creativity, and critical thinking skills
Mending Clothing to Support Sustainable Fashion_CIMaR 2024.pdfSelcen Ozturkcan
Ozturkcan, S., Berndt, A., & Angelakis, A. (2024). Mending clothing to support sustainable fashion. Presented at the 31st Annual Conference by the Consortium for International Marketing Research (CIMaR), 10-13 Jun 2024, University of Gävle, Sweden.
The technology uses reclaimed CO₂ as the dyeing medium in a closed loop process. When pressurized, CO₂ becomes supercritical (SC-CO₂). In this state CO₂ has a very high solvent power, allowing the dye to dissolve easily.
Authoring a personal GPT for your research and practice: How we created the Q...Leonel Morgado
Thematic analysis in qualitative research is a time-consuming and systematic task, typically done using teams. Team members must ground their activities on common understandings of the major concepts underlying the thematic analysis, and define criteria for its development. However, conceptual misunderstandings, equivocations, and lack of adherence to criteria are challenges to the quality and speed of this process. Given the distributed and uncertain nature of this process, we wondered if the tasks in thematic analysis could be supported by readily available artificial intelligence chatbots. Our early efforts point to potential benefits: not just saving time in the coding process but better adherence to criteria and grounding, by increasing triangulation between humans and artificial intelligence. This tutorial will provide a description and demonstration of the process we followed, as two academic researchers, to develop a custom ChatGPT to assist with qualitative coding in the thematic data analysis process of immersive learning accounts in a survey of the academic literature: QUAL-E Immersive Learning Thematic Analysis Helper. In the hands-on time, participants will try out QUAL-E and develop their ideas for their own qualitative coding ChatGPT. Participants that have the paid ChatGPT Plus subscription can create a draft of their assistants. The organizers will provide course materials and slide deck that participants will be able to utilize to continue development of their custom GPT. The paid subscription to ChatGPT Plus is not required to participate in this workshop, just for trying out personal GPTs during it.
ESR spectroscopy in liquid food and beverages.pptxPRIYANKA PATEL
With increasing population, people need to rely on packaged food stuffs. Packaging of food materials requires the preservation of food. There are various methods for the treatment of food to preserve them and irradiation treatment of food is one of them. It is the most common and the most harmless method for the food preservation as it does not alter the necessary micronutrients of food materials. Although irradiated food doesn’t cause any harm to the human health but still the quality assessment of food is required to provide consumers with necessary information about the food. ESR spectroscopy is the most sophisticated way to investigate the quality of the food and the free radicals induced during the processing of the food. ESR spin trapping technique is useful for the detection of highly unstable radicals in the food. The antioxidant capability of liquid food and beverages in mainly performed by spin trapping technique.
When I was asked to give a companion lecture in support of ‘The Philosophy of Science’ (https://shorturl.at/4pUXz) I decided not to walk through the detail of the many methodologies in order of use. Instead, I chose to employ a long standing, and ongoing, scientific development as an exemplar. And so, I chose the ever evolving story of Thermodynamics as a scientific investigation at its best.
Conducted over a period of >200 years, Thermodynamics R&D, and application, benefitted from the highest levels of professionalism, collaboration, and technical thoroughness. New layers of application, methodology, and practice were made possible by the progressive advance of technology. In turn, this has seen measurement and modelling accuracy continually improved at a micro and macro level.
Perhaps most importantly, Thermodynamics rapidly became a primary tool in the advance of applied science/engineering/technology, spanning micro-tech, to aerospace and cosmology. I can think of no better a story to illustrate the breadth of scientific methodologies and applications at their best.
Immersive Learning That Works: Research Grounding and Paths ForwardLeonel Morgado
We will metaverse into the essence of immersive learning, into its three dimensions and conceptual models. This approach encompasses elements from teaching methodologies to social involvement, through organizational concerns and technologies. Challenging the perception of learning as knowledge transfer, we introduce a 'Uses, Practices & Strategies' model operationalized by the 'Immersive Learning Brain' and ‘Immersion Cube’ frameworks. This approach offers a comprehensive guide through the intricacies of immersive educational experiences and spotlighting research frontiers, along the immersion dimensions of system, narrative, and agency. Our discourse extends to stakeholders beyond the academic sphere, addressing the interests of technologists, instructional designers, and policymakers. We span various contexts, from formal education to organizational transformation to the new horizon of an AI-pervasive society. This keynote aims to unite the iLRN community in a collaborative journey towards a future where immersive learning research and practice coalesce, paving the way for innovative educational research and practice landscapes.
The cost of acquiring information by natural selectionCarl Bergstrom
This is a short talk that I gave at the Banff International Research Station workshop on Modeling and Theory in Population Biology. The idea is to try to understand how the burden of natural selection relates to the amount of information that selection puts into the genome.
It's based on the first part of this research paper:
The cost of information acquisition by natural selection
Ryan Seamus McGee, Olivia Kosterlitz, Artem Kaznatcheev, Benjamin Kerr, Carl T. Bergstrom
bioRxiv 2022.07.02.498577; doi: https://doi.org/10.1101/2022.07.02.498577
Impact of Race and Ethnicity on Preemployment Psychological Assessment
1. Page 1 of 23
Introduction and Summary
A thorough review of Brower Psychological Services’ (BPS) evaluation practices from 2018 through 03/2022 was
completed to uncover the possible presence of racially and/or ethnically driven adverse impact against police applicants
for the Aurora Police Department considered in insolation and for all of the agencies serviced by BPS. Typically, adverse
impact is determined by using the four-fifths or eighty percent rule. The four-fifths or 80% rule is described by the
Uniform Guidelines for Employee Selection Procedures as “a selection rate for any race, sex, or ethnic group which is less
than four-fifths (or 80%) of the rate for the group with the highest rate will generally be regarded by the Federal
enforcement agencies as evidence of adverse impact, while a greater than four-fifths rate will generally not be regarded
by Federal enforcement agencies as evidence of adverse impact.” Since the 80% test does not involve probability
distributions to determine whether the disparity is a “beyond chance” occurrence, we have additionally included tests of
statistical significance where the 80% rule may have been violated, as well as observations to assess practical significance
of the results.
No violations of the 80% rule or any statistically significant correlation between race/ethnicity and pass rate were
found when all of the area agencies serviced by BPS were viewed in aggregate, regardless of stage of evaluation. No 80%
violations were found for the Aurora Police Department in any year or stage of evaluation; however, Fisher’s Exact Test of
first-stage evaluations (“JSAs”) conducted in 2021 suggested a possibly significant relationship between race/ethnicity and
pass rate. Though there was no violation of the 80% rule despite this possible relationship, additional practical analyses
were conducted to uncover the presence of confounds or unaccounted for factors that may explain the disparity found in
this one department during a singular year. These analyses suggest that Aurora PD was engaged in either an official or
informal special recruiting program resulting in a demographic distribution that was atypical of the normal pool of
applicants from that group. Such demographic shifts in an applicant pool have been established in the literature and are
referenced specifically in the Uniform Guidelines on Employee Selection Procedures (1978) as a known source of data
distortion likely to create a specious impression of adverse impact. We find a potentially undisclosed hiring initiative to be
2. Page 2 of 23
the most likely explanation for this deviation that is confined both temporally and organizationally, as the uniformity and
stability of our processes across time and agency are observable throughout our data.
3. Page 3 of 23
Analyses
All Agency Applicants 01/2021-03/2022
Filtered By: All Agencies and Type (First)
Descriptives
Introduction
Frequencies and percentages were calculated for Pass_Fail split by Race_Ethnicity.
Results
Frequencies and Percentages
The most frequently observed category of Pass_Fail within the White category of Race_Ethnicity was Pass (n =
419, 79.06%). The most frequently observed category of Pass_Fail within the Hispanic category of Race_Ethnicity was Pass
(n = 140, 74.07%). The most frequently observed category of Pass_Fail within the Black category of Race_Ethnicity was
Pass (n = 57, 67.86%). The most frequently observed category of Pass_Fail within the Other category of Race_Ethnicity
was Pass (n = 50, 79.37%). Frequencies and percentages are presented in Table 1.
Table 1
Frequency Table for Nominal Variables
Race_Ethnicity
Variable White Hispanic Black Other Missing
Pass_Fail
Fail 111 (20.94%) 49 (25.93%) 27 (32.14%) 13 (20.63%) 0 (0.00%)
Pass 419 (79.06%) 140 (74.07%) 57 (67.86%) 50 (79.37%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 530 (100.00%) 189 (100.00%) 84 (100.00%) 63 (100.00%) 0 (100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
Filtered By: All Agencies and Type (Second)
Descriptives
Introduction
Frequencies and percentages were calculated for Pass_Fail split by Race_Ethnicity.
Results
Frequencies and Percentages
The most frequently observed category of Pass_Fail within the White category of Race_Ethnicity was Pass (n =
158, 90.80%). The most frequently observed category of Pass_Fail within the Hispanic category of Race_Ethnicity was Pass
4. Page 4 of 23
(n = 48, 88.89%). The most frequently observed category of Pass_Fail within the Black category of Race_Ethnicity was Pass
(n = 20, 86.96%). The most frequently observed category of Pass_Fail within the Other category of Race_Ethnicity was
Pass (n = 18, 90.00%). Frequencies and percentages are presented in Table 2.
Table 2
Frequency Table for Nominal Variables
Race_Ethnicity
Variable White Hispanic Black Other Missing
Pass_Fail
Fail 16 (9.20%) 6 (11.11%) 3 (13.04%) 2 (10.00%) 0 (0.00%)
Pass 158 (90.80%) 48 (88.89%) 20 (86.96%) 18 (90.00%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 174 (100.00%) 54 (100.00%) 23 (100.00%) 20 (100.00%) 0 (100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
Filtered By: All Agencies and Type (Post)
Descriptives
Introduction
Frequencies and percentages were calculated for Pass_Fail split by Race_Ethnicity.
Results
Frequencies and Percentages
The most frequently observed category of Pass_Fail within the White category of Race_Ethnicity was Pass (n =
355, 74.27%). The most frequently observed category of Pass_Fail within the Hispanic category of Race_Ethnicity was Pass
(n = 47, 66.20%). The most frequently observed category of Pass_Fail within the Black category of Race_Ethnicity was Pass
(n = 20, 66.67%). The most frequently observed category of Pass_Fail within the Other category of Race_Ethnicity was
Pass (n = 13, 59.09%). Frequencies and percentages are presented in Table 3.
5. Page 5 of 23
Table 3
Frequency Table for Nominal Variables
Race_Ethnicity
Variable White Hispanic Black Other Missing
Pass_Fail
Fail 123 (25.73%) 24 (33.80%) 10 (33.33%) 9 (40.91%) 0 (0.00%)
Pass 355 (74.27%) 47 (66.20%) 20 (66.67%) 13 (59.09%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 478 (100.00%) 71 (100.00%) 30 (100.00%) 22 (100.00%) 0 (100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
Filtered By: All Agencies and Type (First)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Pass_Fail and Race_Ethnicity were independent. There
were 2 levels in Pass_Fail: Fail and Pass. There were 4 levels in Race_Ethnicity: White, Hispanic, Black, and Other.
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = .100, suggesting that
Pass_Fail and Race_Ethnicity could be independent of one another. This implies that the observed frequencies were not
significantly different than the expected frequencies. Table 4 presents the results of the Fisher's exact test.
Table 4
Observed and Expected Frequencies
Pass_Fail
Race_Ethnicity Fail Pass p
White 111[122.40] 419[407.60] .100
Hispanic 49[43.65] 140[145.35]
Black 27[19.40] 57[64.60]
Other 13[14.55] 50[48.45]
Note. Values formatted as Observed[Expected].
Filtered By: All Agencies and Type (Second)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Pass_Fail and Race_Ethnicity were independent. There
were 2 levels in Pass_Fail: Fail and Pass. There were 4 levels in Race_Ethnicity: White, Hispanic, Black, and Other.
6. Page 6 of 23
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = .870, suggesting that
Pass_Fail and Race_Ethnicity could be independent of one another. This implies that the observed frequencies were not
significantly different than the expected frequencies. Table 5 presents the results of the Fisher's exact test.
Table 5
Observed and Expected Frequencies
Pass_Fail
Race_Ethnicity Fail Pass p
White 16[17.34] 158[156.66] .870
Hispanic 6[5.38] 48[48.62]
Black 3[2.29] 20[20.71]
Other 2[1.99] 18[18.01]
Note. Values formatted as Observed[Expected].
7. Page 7 of 23
Aurora PD Police Applicants 2018 – 03/2022
Filtered By: Stage (First)
Descriptives
Introduction
Frequencies and percentages were calculated for Bivariate_Pass split by Year_Nominal.
Results
Frequencies and Percentages
The most frequently observed category of Bivariate_Pass within the 2018 category of Year_Nominal was Pass (n =
151, 70.56%). The most frequently observed category of Bivariate_Pass within the 2019 category of Year_Nominal was
Pass (n = 263, 70.70%). The most frequently observed category of Bivariate_Pass within the 2020 category of
Year_Nominal was Pass (n = 160, 68.67%). The most frequently observed category of Bivariate_Pass within the 2021
category of Year_Nominal was Pass (n = 245, 69.80%). The most frequently observed category of Bivariate_Pass within the
2022 category of Year_Nominal was Pass (n = 34, 66.67%). Frequencies and percentages are presented in Table 1.
Table 1
Frequency Table for Nominal Variables
Year_Nominal
Variable 2018 2019 2020 2021 2022 Missing
Bivariate_Pass
Fail 63 (29.44%) 109 (29.30%) 73 (31.33%) 106 (30.20%) 17 (33.33%) 0 (0.00%)
Pass 151 (70.56%) 263 (70.70%) 160 (68.67%) 245 (69.80%) 34 (66.67%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 214 (100.00%) 372 (100.00%) 233 (100.00%) 351 (100.00%) 51 (100.00%) 0 (100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
8. Page 8 of 23
Filtered By: Stage (Second)
Descriptives
Introduction
Frequencies and percentages were calculated for Bivariate_Pass split by Year_Nominal.
Results
Frequencies and Percentages
All observations were missing for the 2018 category of Year_Nominal within each category of Bivariate_Pass. The
most frequently observed category of Bivariate_Pass within the 2019 category of Year_Nominal was Pass (n = 79,
96.34%). The most frequently observed category of Bivariate_Pass within the 2020 category of Year_Nominal was Pass (n
= 77, 81.05%). The most frequently observed category of Bivariate_Pass within the 2021 category of Year_Nominal was
Pass (n = 83, 90.22%). The most frequently observed category of Bivariate_Pass within the 2022 category of Year_Nominal
was Pass (n = 12, 92.31%). Frequencies and percentages are presented in Table 2.
Table 2
Frequency Table for Nominal Variables
Year_Nominal
Variable 2018 2019 2020 2021 2022 Missing
Bivariate_Pass
Fail 0 (0.00%) 3 (3.66%) 18 (18.95%) 9 (9.78%) 1 (7.69%) 0 (0.00%)
Pass 0 (0.00%) 79 (96.34%) 77 (81.05%) 83 (90.22%) 12 (92.31%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 0 (100.00%) 82 (100.00%) 95 (100.00%) 92 (100.00%) 13 (100.00%) 0 (100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
Filtered By: Stage (First)
Descriptives
Introduction
Frequencies and percentages were calculated for Race_Ethnicity_Simplified split by Year_Nominal.
Results
Frequencies and Percentages
The most frequently observed category of Race_Ethnicity_Simplified within the 2018 category of Year_Nominal
was White (n = 128, 59.81%). The most frequently observed category of Race_Ethnicity_Simplified within the 2019
category of Year_Nominal was White (n = 238, 63.98%). The most frequently observed category of
Race_Ethnicity_Simplified within the 2020 category of Year_Nominal was White (n = 139, 59.66%). The most frequently
9. Page 9 of 23
observed category of Race_Ethnicity_Simplified within the 2021 category of Year_Nominal was White (n = 194, 55.27%).
The most frequently observed category of Race_Ethnicity_Simplified within the 2022 category of Year_Nominal was
White (n = 27, 52.94%). Frequencies and percentages are presented in Table 3.
Table 3
Frequency Table for Nominal Variables
Year_Nominal
Variable 2018 2019 2020 2021 2022 Missing
Race_Ethnicity_Simplified
White 128 (59.81%) 238 (63.98%) 139 (59.66%) 194 (55.27%) 27 (52.94%) 0 (0.00%)
Hispanic 51 (23.83%) 67 (18.01%) 47 (20.17%) 90 (25.64%) 11 (21.57%) 0 (0.00%)
Other 13 (6.07%) 39 (10.48%) 23 (9.87%) 31 (8.83%) 2 (3.92%) 0 (0.00%)
Black 22 (10.28%) 28 (7.53%) 24 (10.30%) 36 (10.26%) 11 (21.57%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total
214
(100.00%)
372
(100.00%)
233
(100.00%)
351
(100.00%)
51
(100.00%)
0
(100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
Filtered By: Stage (Second)
Descriptives
Introduction
Frequencies and percentages were calculated for Race_Ethnicity_Simplified split by Year_Nominal.
Results
Frequencies and Percentages
All observations were missing for the 2018 category of Year_Nominal within each category of
Race_Ethnicity_Simplified. The most frequently observed category of Race_Ethnicity_Simplified within the 2019 category
of Year_Nominal was White (n = 60, 73.17%). The most frequently observed category of Race_Ethnicity_Simplified within
the 2020 category of Year_Nominal was White (n = 60, 63.16%). The most frequently observed category of
Race_Ethnicity_Simplified within the 2021 category of Year_Nominal was White (n = 58, 63.04%). The most frequently
observed category of Race_Ethnicity_Simplified within the 2022 category of Year_Nominal was White (n = 9, 69.23%).
Frequencies and percentages are presented in Table 4.
10. Page 10 of 23
Table 4
Frequency Table for Nominal Variables
Year_Nominal
Variable 2018 2019 2020 2021 2022 Missing
Race_Ethnicity_Simplified
White 0 (0.00%) 60 (73.17%) 60 (63.16%) 58 (63.04%) 9 (69.23%) 0 (0.00%)
Hispanic 0 (0.00%) 13 (15.85%) 20 (21.05%) 22 (23.91%) 3 (23.08%) 0 (0.00%)
Other 0 (0.00%) 6 (7.32%) 11 (11.58%) 6 (6.52%) 1 (7.69%) 0 (0.00%)
Black 0 (0.00%) 3 (3.66%) 4 (4.21%) 6 (6.52%) 0 (0.00%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 0 (100.00%) 82 (100.00%) 95 (100.00%) 92 (100.00%) 13 (100.00%) 0 (100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
Filtered By: Stage (First)
Descriptives
Introduction
Frequencies and percentages were calculated for Year_Nominal and Bivariate_Pass split by
Race_Ethnicity_Simplified.
Results
Frequencies and Percentages
The most frequently observed category of Year_Nominal within the White category of Race_Ethnicity_Simplified
was 2019 (n = 238, 32.78%). The most frequently observed category of Year_Nominal within the Hispanic category of
Race_Ethnicity_Simplified was 2021 (n = 90, 33.83%). The most frequently observed category of Year_Nominal within the
Other category of Race_Ethnicity_Simplified was 2019 (n = 39, 36.11%). The most frequently observed category of
Year_Nominal within the Black category of Race_Ethnicity_Simplified was 2021 (n = 36, 29.75%). The most frequently
observed category of Bivariate_Pass within the White category of Race_Ethnicity_Simplified was Pass (n = 519, 71.49%).
The most frequently observed category of Bivariate_Pass within the Hispanic category of Race_Ethnicity_Simplified was
Pass (n = 183, 68.80%). The most frequently observed category of Bivariate_Pass within the Other category of
Race_Ethnicity_Simplified was Pass (n = 80, 74.07%). The most frequently observed category of Bivariate_Pass within the
Black category of Race_Ethnicity_Simplified was Pass (n = 71, 58.68%). Frequencies and percentages are presented in
Table 5.
11. Page 11 of 23
Table 5
Frequency Table for Nominal Variables
Race_Ethnicity_Simplified
Variable White Hispanic Other Black Missing
Year_Nominal
2018 128 (17.63%) 51 (19.17%) 13 (12.04%) 22 (18.18%) 0 (0.00%)
2019 238 (32.78%) 67 (25.19%) 39 (36.11%) 28 (23.14%) 0 (0.00%)
2020 139 (19.15%) 47 (17.67%) 23 (21.30%) 24 (19.83%) 0 (0.00%)
2021 194 (26.72%) 90 (33.83%) 31 (28.70%) 36 (29.75%) 0 (0.00%)
2022 27 (3.72%) 11 (4.14%) 2 (1.85%) 11 (9.09%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 726 (100.00%) 266 (100.00%) 108 (100.00%) 121 (100.00%) 0 (100.00%)
Bivariate_Pass
Fail 207 (28.51%) 83 (31.20%) 28 (25.93%) 50 (41.32%) 0 (0.00%)
Pass 519 (71.49%) 183 (68.80%) 80 (74.07%) 71 (58.68%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 726 (100.00%) 266 (100.00%) 108 (100.00%) 121 (100.00%) 0 (100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
Filtered By: Stage (Second)
Descriptives
Introduction
Frequencies and percentages were calculated for Year_Nominal and Bivariate_Pass split by
Race_Ethnicity_Simplified.
Results
Frequencies and Percentages
The most frequently observed categories of Year_Nominal within the White category of Race_Ethnicity_Simplified
were 2019 and 2020 (n = 60, 32.09%). The most frequently observed category of Year_Nominal within the Hispanic
category of Race_Ethnicity_Simplified was 2021 (n = 22, 37.93%). The most frequently observed category of
Year_Nominal within the Other category of Race_Ethnicity_Simplified was 2020 (n = 11, 45.83%). The most frequently
observed category of Year_Nominal within the Black category of Race_Ethnicity_Simplified was 2021 (n = 6, 46.15%). The
most frequently observed category of Bivariate_Pass within the White category of Race_Ethnicity_Simplified was Pass (n =
165, 88.24%). The most frequently observed category of Bivariate_Pass within the Hispanic category of
Race_Ethnicity_Simplified was Pass (n = 55, 94.83%). The most frequently observed category of Bivariate_Pass within the
Other category of Race_Ethnicity_Simplified was Pass (n = 20, 83.33%). The most frequently observed category of
Bivariate_Pass within the Black category of Race_Ethnicity_Simplified was Pass (n = 11, 84.62%). Frequencies and
percentages are presented in Table 6.
12. Page 12 of 23
Table 6
Frequency Table for Nominal Variables
Race_Ethnicity_Simplified
Variable White Hispanic Other Black Missing
Year_Nominal
2018 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
2019 60 (32.09%) 13 (22.41%) 6 (25.00%) 3 (23.08%) 0 (0.00%)
2020 60 (32.09%) 20 (34.48%) 11 (45.83%) 4 (30.77%) 0 (0.00%)
2021 58 (31.02%) 22 (37.93%) 6 (25.00%) 6 (46.15%) 0 (0.00%)
2022 9 (4.81%) 3 (5.17%) 1 (4.17%) 0 (0.00%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 187 (100.00%) 58 (100.00%) 24 (100.00%) 13 (100.00%) 0 (100.00%)
Bivariate_Pass
Fail 22 (11.76%) 3 (5.17%) 4 (16.67%) 2 (15.38%) 0 (0.00%)
Pass 165 (88.24%) 55 (94.83%) 20 (83.33%) 11 (84.62%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 187 (100.00%) 58 (100.00%) 24 (100.00%) 13 (100.00%) 0 (100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
Filtered By: Year_Nominal (2018) and Stage (First)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Bivariate_Pass and Race_Ethnicity_Simplified were
independent. There were 2 levels in Bivariate_Pass: Fail and Pass. There were 4 levels in Race_Ethnicity_Simplified: White,
Hispanic, Other, and Black.
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = .734, suggesting that
Bivariate_Pass and Race_Ethnicity_Simplified could be independent of one another. This implies that the observed
frequencies were not significantly different than the expected frequencies. Table 7 presents the results of the Fisher's
exact test.
Table 7
Observed and Expected Frequencies
Bivariate_Pass
Race_Ethnicity_Simplified Fail Pass p
White 41[37.68] 87[90.32] .734
Hispanic 12[15.01] 39[35.99]
Other 4[3.83] 9[9.17]
Black 6[6.48] 16[15.52]
Note. Values formatted as Observed[Expected].
13. Page 13 of 23
Filtered By: Year_Nominal (2019) and Stage (First)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Bivariate_Pass and Race_Ethnicity_Simplified were
independent. There were 2 levels in Bivariate_Pass: Fail and Pass. There were 4 levels in Race_Ethnicity_Simplified: White,
Hispanic, Other, and Black.
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = .453, suggesting that
Bivariate_Pass and Race_Ethnicity_Simplified could be independent of one another. This implies that the observed
frequencies were not significantly different than the expected frequencies. Table 8 presents the results of the Fisher's
exact test.
Table 8
Observed and Expected Frequencies
Bivariate_Pass
Race_Ethnicity_Simplified Fail Pass p
White 67[69.74] 171[168.26] .453
Hispanic 19[19.63] 48[47.37]
Other 11[11.43] 28[27.57]
Black 12[8.20] 16[19.80]
Note. Values formatted as Observed[Expected].
Filtered By: Year_Nominal (2020) and Stage (First)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Bivariate_Pass and Race_Ethnicity_Simplified were
independent. There were 2 levels in Bivariate_Pass: Fail and Pass. There were 4 levels in Race_Ethnicity_Simplified: White,
Hispanic, Other, and Black.
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = .680, suggesting that
Bivariate_Pass and Race_Ethnicity_Simplified could be independent of one another. This implies that the observed
frequencies were not significantly different than the expected frequencies. Table 9 presents the results of the Fisher's
exact test.
14. Page 14 of 23
Table 9
Observed and Expected Frequencies
Bivariate_Pass
Race_Ethnicity_Simplified Fail Pass p
White 43[43.55] 96[95.45] .680
Hispanic 13[14.73] 34[32.27]
Other 7[7.21] 16[15.79]
Black 10[7.52] 14[16.48]
Note. Values formatted as Observed[Expected].
Filtered By: Year_Nominal (2021) and Stage (First)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Bivariate_Pass and Race_Ethnicity_Simplified were
independent. There were 2 levels in Bivariate_Pass: Fail and Pass. There were 4 levels in Race_Ethnicity_Simplified: White,
Hispanic, Other, and Black.
Results
The results of the Fisher exact test were significant based on an alpha value of .05, p = .006, suggesting that
Bivariate_Pass and Race_Ethnicity_Simplified are related to one another. The following level combinations had observed
values that were greater than their expected values: Race_Ethnicity_Simplified (Hispanic):Bivariate_Pass (Fail),
Race_Ethnicity_Simplified (Black):Bivariate_Pass (Fail), Race_Ethnicity_Simplified (White):Bivariate_Pass (Pass), and
Race_Ethnicity_Simplified (Other):Bivariate_Pass (Pass). The following level combinations had observed values that were
less than their expected values: Race_Ethnicity_Simplified (White):Bivariate_Pass (Fail), Race_Ethnicity_Simplified
(Other):Bivariate_Pass (Fail), Race_Ethnicity_Simplified (Hispanic):Bivariate_Pass (Pass), and Race_Ethnicity_Simplified
(Black):Bivariate_Pass (Pass). Table 10 presents the results of the Fisher's exact test.
Table 10
Observed and Expected Frequencies
Bivariate_Pass
Race_Ethnicity_Simplified Fail Pass p
White 48[58.59] 146[135.41] .006
Hispanic 35[27.18] 55[62.82]
Other 6[9.36] 25[21.64]
Black 17[10.87] 19[25.13]
Note. Values formatted as Observed[Expected].
15. Page 15 of 23
Filtered By: Year_Nominal (2022) and Stage (First)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Bivariate_Pass and Race_Ethnicity_Simplified were
independent. There were 2 levels in Bivariate_Pass: Fail and Pass. There were 4 levels in Race_Ethnicity_Simplified: White,
Hispanic, Other, and Black.
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = .698, suggesting that
Bivariate_Pass and Race_Ethnicity_Simplified could be independent of one another. This implies that the observed
frequencies were not significantly different than the expected frequencies. Table 11 presents the results of the Fisher's
exact test.
Table 11
Observed and Expected Frequencies
Bivariate_Pass
Race_Ethnicity_Simplified Fail Pass p
White 8[9.00] 19[18.00] .698
Hispanic 4[3.67] 7[7.33]
Other 0[0.67] 2[1.33]
Black 5[3.67] 6[7.33]
Note. Values formatted as Observed[Expected].
Filtered By: Year_Nominal (2019) and Stage (Second)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Bivariate_Pass and Race_Ethnicity_Simplified were
independent. There were 2 levels in Bivariate_Pass: Fail and Pass. There were 4 levels in Race_Ethnicity_Simplified: White,
Hispanic, Other, and Black.
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = 1.000, suggesting that
Bivariate_Pass and Race_Ethnicity_Simplified could be independent of one another. This implies that the observed
frequencies were not significantly different than the expected frequencies. Table 12 presents the results of the Fisher's
exact test.
Table 12
16. Page 16 of 23
Observed and Expected Frequencies
Bivariate_Pass
Race_Ethnicity_Simplified Fail Pass p
White 3[2.20] 57[57.80] 1.000
Hispanic 0[0.48] 13[12.52]
Other 0[0.22] 6[5.78]
Black 0[0.11] 3[2.89]
Note. Values formatted as Observed[Expected].
Filtered By: Year_Nominal (2020) and Stage (Second)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Bivariate_Pass and Race_Ethnicity_Simplified were
independent. There were 2 levels in Bivariate_Pass: Fail and Pass. There were 4 levels in Race_Ethnicity_Simplified: White,
Hispanic, Other, and Black.
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = .227, suggesting that
Bivariate_Pass and Race_Ethnicity_Simplified could be independent of one another. This implies that the observed
frequencies were not significantly different than the expected frequencies. Table 13 presents the results of the Fisher's
exact test.
Table 13
Observed and Expected Frequencies
Bivariate_Pass
Race_Ethnicity_Simplified Fail Pass p
White 13[11.37] 47[48.63] .227
Hispanic 1[3.79] 19[16.21]
Other 3[2.08] 8[8.92]
Black 1[0.76] 3[3.24]
Note. Values formatted as Observed[Expected].
Filtered By: Year_Nominal (2021) and Stage (Second)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Bivariate_Pass and Race_Ethnicity_Simplified were
independent. There were 2 levels in Bivariate_Pass: Fail and Pass. There were 4 levels in Race_Ethnicity_Simplified: White,
Hispanic, Other, and Black.
17. Page 17 of 23
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = .399, suggesting that
Bivariate_Pass and Race_Ethnicity_Simplified could be independent of one another. This implies that the observed
frequencies were not significantly different than the expected frequencies. Table 14 presents the results of the Fisher's
exact test.
Table 14
Observed and Expected Frequencies
Bivariate_Pass
Race_Ethnicity_Simplified Fail Pass p
White 6[5.67] 52[52.33] .399
Hispanic 1[2.15] 21[19.85]
Other 1[0.59] 5[5.41]
Black 1[0.59] 5[5.41]
Note. Values formatted as Observed[Expected].
18. Page 18 of 23
Filtered By: Year_Nominal (2022) and Stage (Second)
Fisher's Exact Test
Introduction
A Fisher's exact test was conducted to examine whether Bivariate_Pass and Race_Ethnicity_Simplified were
independent. There were 2 levels in Bivariate_Pass: Fail and Pass. There were 4 levels in Race_Ethnicity_Simplified: White,
Hispanic, Other, and Black.
Results
The results of the Fisher exact test were not significant based on an alpha value of .05, p = .308, suggesting that
Bivariate_Pass and Race_Ethnicity_Simplified could be independent of one another. This implies that the observed
frequencies were not significantly different than the expected frequencies. Table 15 presents the results of the Fisher's
exact test.
Table 15
Observed and Expected Frequencies
Bivariate_Pass
Race_Ethnicity_Simplified Fail Pass p
White 0[0.69] 9[8.31] .308
Hispanic 1[0.23] 2[2.77]
Other 0[0.08] 1[0.92]
Black 0[0.00] 0[0.00]
Note. Values formatted as Observed[Expected].
19. Page 19 of 23
Aurora PD Observations
Adverse Impact: 4/5ths - 80/20 Rule
The following figures were published/distributed by Aurora PD describing the demographic composition and
selection rates for police academies held 2018-2020 titled “Passing Job Suitability Interview”:
Table 16
Aurora PD’s Figures for Passing Job Suitability Interview, from “APD Academies from 2018-2020”
Race/Ethnicity
White or Caucasian Black or African American Hispanic or Latino Asian 2+ or Other
Passing Job
Suitability
Interview
413 (42.4%) 42 (34.4%) 112 (37.7%) 18 (37.5%) 80 (44.7%)
According to these figures, the 4/5ths – 80/20 rule would have been violated for “Black or African American”
applicants when compared to the group with the highest passing rate, “2+ or Other.” These figures are substantially
different than the figures we have for police applicants referred by Aurora PD for JSA’s from 2018-2020, which appear in
the table below.
Table 17
Frequency Table for Nominal Variables
Race_Ethnicity_Simplified
Variable White Hispanic Other Black Missing
Bivariate_Pass
Fail 151 (29.90%) 44 (26.67%) 22 (29.33%) 28 (37.84%) 0 (0.00%)
Pass 354 (70.10%) 121 (73.33%) 53 (70.67%) 46 (62.16%) 0 (0.00%)
Missing 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Total 505 (100.00%) 165 (100.00%) 75 (100.00%) 74 (100.00%) 0 (100.00%)
Note. Due to rounding error, percentages may not sum to 100%.
Not only do our figures represent a far higher passing percentage for applicants, there is no violation of the 4/5ths
– 80/20 rule when comparing the demographic group with the highest passing rate (Hispanic; 73.33%) to the
demographic group with the lowest passing rate (Black; 62.16%).
When 2018-2022 JSAs are viewed in aggregate, Other is the demographic group with the highest passing
percentage (~74%) and Black is the demographic group with the lowest passing percentage (~59%). White and Hispanic
20. Page 20 of 23
groupings showed passing rates within a few percent (~71% and ~69%, respectively). When the 4/5ths rule is applied to
measure the difference between Black and Other applicants, an ~20% difference is observed, suggesting that adverse
impact is not present at a level surpassing prevailing/traditional Federal guidelines. When Black applicants are compared
to White applicants, the demographic group comprising the majority of all applicants, a 17% difference is seen, again
suggesting that adverse impact is not present at a level surpassing prevailing/traditional Federal guidelines.
When the same aggregated approach is applied to Aurora PD 2nd
Stage evaluations, there is no observable
difference indicating violation of the 4/5ths rule. The demographic group with the highest passing rate amongst 2nd
Stage
evaluations from 2018-2022 is Hispanic (~95%) while the lowest passing rate being found in the Other demographic group
(~83%); a 13% difference. Among all 2nd
Stages from 2018-2022, members of the White group passed at ~88%, while
members of the Black group passed at approximately ~85%.
0%
10%
20%
30%
40%
50%
60%
70%
80%
2018-2022 JSA Passing Rates
White Black Hispanic Other
21. Page 21 of 23
Adverse Impact: Statistical Significance
Fisher’s Exact Tests split by year and stage of evaluation for Aurora PD suggest that there is no difference between
observed and expected frequencies of passing among the four demographic groups for JSAs conducted in 2018, 2019,
2020, and 2022 and for 2nd
Stage evaluations conducted from 2018-2022. Among 2021 JSAs there is a possible
relationship suggested between Race/Ethnicity and Passing; specifically, that members of the White and Other
demographic groups are more likely to pass, while members of the Black and Hispanic demographic groups are less likely
to pass.
Adverse Impact: Practical Significance (Demographic Shifts)
When comparing demographic distributions of applicants referred for JSAs by Aurora PD from 2019 to 2022, the
percentage of all applicants self-identifying as White, decreased by ~10%. Between 2020 and 2021, the percentage of
applicants self-identifying as Hispanic nearly doubled (~18% vs ~34%), and between 2019 and 2022, the percentage of
applicant’s self-identifying as Black more than doubled, increasing by ~96% (from ~8% to ~22%).
62%
96%
-10%
Applicant Demographic Changes JSAs
Hispanic Black White
76%
78%
80%
82%
84%
86%
88%
90%
92%
94%
96%
2018-2022 2nd Stage Passing Rates
White Black Hispanic Other
22. Page 22 of 23
When looking at 2nd Stage evaluations for Aurora PD, a similar 10% decrease in White applicants is observable
when comparing 2019’s ~73% proportion to 2020 and 2021’s ~63%. When comparing the number of Black applicants
referred for 2nd
Stage evaluations from 2019 to 2021, a 56% increase is observable. The percentage of Hispanic applicants
also increased by more than 50% from 2019 (~22%) to 2021 (~38%).
Adverse Impact: Practical Significance (Inclusion/Exclusion Shifts)
In 2021, Aurora PD referred 351 applicants for JSAs, ~40% more applicants than were referred in the preceding
year (233). The same increase in applicant referrals was not witnessed for 2nd
Stage evaluations; in fact, there was a slight
decrease observed from 2020 to 2021 (95 vs. 92; ~3% decrease). This absence of a difference between the number of
applicants referred for 2nd
Stage referrals, despite the marked increase in the number of applicants referred for JSAs is
notable. In 2020, ~41% of the applicants seen for a JSA returned for a 2nd
Stage, but in 2021, only 26% of applicants seen
for a JSA were referred for a 2nd
Stage evaluation, a decrease of ~45%.
51%
56%
-10%
Applicant Demographic Changes 2nd Stage
Hispanic Black White
23. Page 23 of 23
Adverse Impact: Conclusion
The observable dramatic demographic shifts which accompany the only indication of possible adverse impacts
suggests the presence of an overt or covert modification in the early portion of the hiring process creating an atypical
demographic distribution that is being reflected in the seemingly disproportionate fail rates of Black and Hispanic
applicants that becomes undetectable when examining the demographic distribution of Aurora PD applicants who
returned for 2nd
Stage evaluations.