Attitudes towards AI development
Kulani Abendroth-Dias - Statistics for International Relations Research II
- Introduction & Data
- Literature Review
- Methods
- Analysis & Results
- Discussion, Limitations, and Conclusion
- References
Data: revised_AI_dataset.csv
Introduction & Data
“We fear what we do not know.” - Anonymous / alternatively, this quote is attributable to many.
Artificial intelligence, physical security, and political security are inextricably linked in the Digital Age. The advent of COVID-19 has often loosened restrictions on the use of AI-driven technologies around the world, and made salient its use in combating emerging threats. However, the dual-use nature of AI leads to its use and prolific misuse. The global damages for AI-driven cyber-attacks totaled over five billion euro in 2017. These included the hacking of the United Kingdom’s National Health Service, where 20,000 appointments were cancelled, and the NotPetya ransomware hack costing Maersk, Merck, and FedEx approximately USD 300 million each in immediate damages. These attacks often lead the public to question the robustness of their political and public structures, and work to undermine trust in governments and political participation. While AI has significant beneficial use across political, social, and economic domains in our daily lives, its misuse due to a lack of regulations to support accountability have led to an erosion in attitudes towards its use.
For this final project, I am using a simplified version of the data shared by Baobao Zhang and Allan Dafoe in their 2018 study entitled Artificial Intelligence: American Attitudes and Trends to explore attitudes towards AI development in one case study: the United States. This study may help us identify association between demographic groups and distrust in AI development, which could make salient for policymakers specific areas where regulation is needed the most, in support of their own constituents. This is keeping in mind that the US tends to exhibit more liberal attitudes towards AI development than in the EU for example. It may be interesting to contrast this dataset with that looking at German (more risk averse) attitudes towards AI development versus regulation - and this is something I am interested in exploring in my dissertation research.
Researchers have looked at the effect of fear of automation and robots on support for development of AI technologies. According to the World Economic Forum, artificial intelligence holds the promise of making organisations 40% more efficient by 2035 and its development and use and lead to estimated $14 trillion in new economic value, all while improving the lives of billions. However, it will only achieve its potential if it is used responsibly, and its impact on billions will only be realised if it is developed in democratic ways. The only issue with democratising code and AI-driven technologies is its potential malicious use, which could then be used to engender fear regarding its development among constituents.
The data comprise a series of socio-demographic and subjective attitude measures and I will run multiple linear regressions to predict affinity for AI governance based on age, gender, race, income, and political affiliation of US respondents in a survey conducted in 2018.
Literature Review
There have been several studies conducted investigating public attitudes towards the use of Artificial Intelligence-driven technologies in the medical sciences and banking. In the former, for example, Dos Santos et al (2019) found that medical studies appreciated the use of AI in improving diagnoses in radiology, but disagreed in its use to fully replace human diagnoses. Similarly, within the banking sector, several studies (see Aitken, Ng, Toreini, van Moorsel, Coopamootoo, and Elliott, 2020) have found support for the use of AI in increasing the efficiency of processes in the banking sector, while disagreeing that it could be used to replace human judgment and oversight. Across domains, participants agree on the use of AI in improving processes, but raise ethics concerns with regard to misuse. This is where the governance of these technologies, to ensure their use to improve lives, and limit misuse, is particularly pertinent. However, the regulation of AI is a complex problem due to the stakeholders and so-called “pacing problem” (Taehagh & Ramesh, 2021: see below) between the private sector that develops the technologies and the public sector that are meant to govern these technologies.
The Role of the Private Sector
The defense sector has traditionally been a core driver of innovation. In recent years however, disruptive innovation has been driven by the private sector with the military (often problematically) vying not to be left behind. For example, according to the OECD, Google, Microsoft, Amazon, and Intel spent more than USD 50 billion a year on digital innovation in 2018 alone (OECD, 2017). This sum dwarfs the 13 billion euros over seven years originally put up by the European Defense Fund (for defense spending in general, not solely on the development of AI-driven technologies), let alone any other numbers put up by European governments (Bachmann et al, 2017).
In their seminal work, “Prediction machines: the simple economics of artificial intelligence,” Agrawal, Gans, and Golfarb (2018) argue that from an economic perspective, the advent of AI brings a far better future than envisioned by those wary of its development. They posit that cheaper data will lead to more, and better predictions, and predictions in turn will improve human judgement. Predictions are important because they make life for humans easier: exemplified by recommendations based on purchases on Amazon and improvements in data and prediction escalating the development of self-driving cars. The authors argue that machines don’t make judgements – humans do – but these judgments can be better informed by more data and better predictions e.g. in the case of identifying credit card fraud. In this way, the book describes assistive AI, and is aimed at general audiences trying to understand the economic trade-offs of using AI in the future. Most interesting is that while the authors discuss the economics of automation on employment, and the tradeoffs therein, they fail to look at the implications of automation on defense and security. This piece is an illustration of how the role of the private sector in developing AI and their effects on economics markets, versus the role of policy and security actors in building resilience to AI-driven hybrid threats has thus far largely been discussed in isolation from each other – more has to be done to look at ‘where the money goes’ in the development of AI and its implications within the defense and security sectors. This thesis project is a step in the direction of bringing these economic, political, and security lines together.
AI and the Labour Market
While the implications of the assistive functions of AI-driven technologies on improving human judgment and innovation across sectors are clear, its effects on the labor market on more short-term, practical levels form a large part of the current literature on its development. Muro, Maxim, and Whiton (2019) use government and private data, including from the McKinsey Global Institute, to map trends in automation between 1980 to 2016, and extrapolate these impacts to 2030 across 800 occupations potentially at risk of automation.
AI and ML technologies trained to play games, classify images, and guide vehicles have ramifications across sectors, from self-driving cars and traffic coordination within the civilian space to conducting reconnaissance missions and automated offensive strikes within the military space. In principle, tasks that could be transformed via AI are vast. Muro, Maxim, and Whiton illustrate that almost every occupation will be affected by the adoption of currently available technologies. The authors present that routine physical and cognitive tasks will be most at risk for automation, with jobs located in rural geographical locations most vulnerable. Lowest wage jobs will be most vulnerable to automation, with Black and Hispanic workers disproportionately currently in occupations at risk for automation. The report makes broad recommendations to policy-makers considering these trends and design programs advocating for a learning mindset that facilitate smoother adjustment through subsidized hiring programs. They further call for the reduction of hardships for workers who are vulnerable to the risks of automation on the labor market by improving resources for skill development and expanding traditional education programs in rural areas. This is of particular relevance to the current project, as it looks at AI discourse vis-a-vis specific demographic groups.
In fact, most research on the impact of AI almost exclusively focuses on its development across the private sector and the effect of automation on jobs. However, it is crucial to look at attitudes towards AI from the perspective of constituents, in order to operationalise actions that policy makers can take to address distrust in AI-driven technologies due to a fear of automation/lack of regulation/asymmetries in information.
Global studies looking at attitudes towards AI
Neudert, Knuutila & Howard (2020) in their “Global Attitudes Towards AI, Machine Learning & Automation Decision Making” study found that there are clear global divides in support for AI, with worries about the risks of an increased use of AI highest among those based in North and South America and Europe, and least among those in South Asia and South-East Asia. Here, we wonder if there is a mediating factor of governance structure or democracy in driving this support. I am not however able to unpack this within the confines of this dataset.
Asymmetries in Information: We fear what we don’t know
Given recent competitive private sector investment, AI is developing at an exponential rate. Indeed, most of the literature on AI is outdated at this point, with much of what was discussed by the authors above already coming to pass. This is due to what Taehagh & Ramesh (2021) call the “pacing problem,” i.e. the lag between a government’s or supranational body’s slow responses (due to their consensus-based model) to emerging challenges posed by a competition-fuelled private sector market for the development of emerging technologies. As it often comes to pass, dual-use technologies are developed in the market at a pace set by the private sector, and governance mechanisms struggle to attract the skills and capacity needed to keep up and regulate these developments. Often, a lack of technical literacy within governance actors, whose agencies are not structured in ways to attract younger talent, grossly contributes to this problem. This lack of ability to regulate against emerging threats leads to the erosion of public trust in governance bodies, as well as the use of AI in general. This may be an acute issue
There is no doubt that we need “effective, values-based digital frameworks” to govern emerging technologies and protect constituents. However, asymmetries in information, the pacing lag between the private sector and public sector, and a lack of accountability mechanisms all contribute to undermining regulatory protocols, and erode public trust in government as well as the emerging technologies themselves.
Biases in datasets
Much academic and popular attention has been devoted to discussing biases in datasets that skew algorithmic outputs (Noble, 2018). The private sector solution to this has been a drive to partner with public sector agents with the aim to collect more data on diverse populations. The response to this bias in objectives and outputs should rather be accountability mechanisms, where programmers and their organisations are held accountable for code that may not be trained on inclusive datasets
- in tandem with robust regulations on which types of data can be supplied to train these algorithms by public policy bodies. However, if these officials struggle to understand the problem themselves and lack the technical capacity to address it (always outsourcing the discussion to private sector actors who join the dialogue to support their own agenda) we end up with the tautological problem of a lack of governance frameworks eroding public trust in their government as well as the technologies themselves.
Unpacking the ‘public’ in public trust in AI
While much has been written on developing or addressing public trust in AI-driven technologies, it is crucial to investigate attitudes towards AI across demographic groups (Spiegeleire, Maas, & Sweijs, 2017; Taehagh & Ramesh, 2021). This could help us identify gaps in support for AI, and help drive targeted efforts to include these groups in the move to the Digital Age. Importantly, it may help increase the salience of the need to regulate AI within policy circles. If policy makers are aware which constituents are most likely to support - or not support - AI development, which is linked to economic development, they may be incentivised to work to build more robust check and improve the literacy of AI among those more likely not to support its development. Policymakers often need to balance not being left behind with constituent and ethical concerns, and understanding evolving public attitudes towards, and factors mediating these attitudes, is crucial to the public and private sector alike. This is of paricular importance in the onset of COVID-19, which has seen an increase in the adoption of AI-driven technologies by national governments and the public alike.
This study aims to investigate the effect of age, religion, gender, race, income, employment status, political party, and experience in Computer Science an Engineering respectively on support for AI via multiple regression analysis.
Hypotheses
- H1: Age is negatively associated with support for AI.
- H2: Those who are secular are more likely to support AI.
- H3: Education level is positively associated with support for AI.
- H4: Marital status is not associated with support for AI.
- H5: Males are more likely to support AI.
- H6: Race is not associated with support for AI.
- H7: Employment status is not associated with support for AI.
- H8: Political party is not associated with support for AI.
- H9: Political ideology is not associated with support for AI.
- H10: Experience with computer science is positively associated with support for AI.
- H11: An engineering degree is positively associated with support for AI.
Methods
The dataset comprises categorical data as ordered response variables (age groups, education levels, and income brackets) as well as unordered response variables (political party, ethnicity, and religion). The former is typically modeled using ordinal logistic regression, and the latter multinomial logistic regression. However, for this analysis, since I aim to predict several variables which can typically be any real number (do not add up to 1) I will run multiple regressions.
I aim to look at support for AI development by 11 demographic subgroup variables, including age, gender, race, and education based on the data available. I aim to perform a multiple linear regression to predict support for developing AI using all these demographic variables. My null hypotheses are such that age, gender, race, experience with CS programming, and education will not have an influence on support for development of AI within this sample in the US.
Analysis & Results
##Multiple Regressions
summary(AI_Importance)
## SupportAIdev USmilitary UScivilian_go NationalSecurity
## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.000
## 1st Qu.:2.000 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:4.000
## Median :3.000 Median :9.000 Median :9.000 Median :9.000
## Mean :3.048 Mean :7.009 Mean :7.036 Mean :6.885
## 3rd Qu.:4.000 3rd Qu.:9.000 3rd Qu.:9.000 3rd Qu.:9.000
## Max. :6.000 Max. :9.000 Max. :9.000 Max. :9.000
## FBI CIA NATO Techcomp
## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.000
## 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:4.000
## Median :9.000 Median :9.000 Median :9.000 Median :9.000
## Mean :7.058 Mean :6.846 Mean :7.026 Mean :6.924
## 3rd Qu.:9.000 3rd Qu.:9.000 3rd Qu.:9.000 3rd Qu.:9.000
## Max. :9.000 Max. :9.000 Max. :9.000 Max. :9.000
## Google Facebook Apple Microsoft Amazon
## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.00 Min. :1.000
## 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:4.00 1st Qu.:4.000
## Median :9.000 Median :9.000 Median :9.000 Median :9.00 Median :9.000
## Mean :7.064 Mean :7.254 Mean :6.913 Mean :7.17 Mean :6.928
## 3rd Qu.:9.000 3rd Qu.:9.000 3rd Qu.:9.000 3rd Qu.:9.00 3rd Qu.:9.000
## Max. :9.000 Max. :9.000 Max. :9.000 Max. :9.00 Max. :9.000
## NGO Uni Rank_US Rank_China gender
## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.00 Min. :1.000
## 1st Qu.:4.000 1st Qu.:4.000 1st Qu.:3.000 1st Qu.:2.00 1st Qu.:1.000
## Median :9.000 Median :9.000 Median :9.000 Median :3.00 Median :2.000
## Mean :7.046 Mean :6.928 Mean :6.001 Mean :3.36 Mean :1.524
## 3rd Qu.:9.000 3rd Qu.:9.000 3rd Qu.:9.000 3rd Qu.:4.00 3rd Qu.:2.000
## Max. :9.000 Max. :9.000 Max. :9.000 Max. :8.00 Max. :2.000
## race educ maritalstatus employ pid
## Min. :1.000 Min. :1.000 Min. :1.00 Min. :1.000 Min. :1.000
## 1st Qu.:1.000 1st Qu.:2.000 1st Qu.:1.00 1st Qu.:1.000 1st Qu.:1.000
## Median :1.000 Median :3.000 Median :3.00 Median :4.000 Median :2.000
## Mean :1.777 Mean :3.401 Mean :2.91 Mean :3.544 Mean :2.269
## 3rd Qu.:2.000 3rd Qu.:5.000 3rd Qu.:5.00 3rd Qu.:5.000 3rd Qu.:3.000
## Max. :8.000 Max. :6.000 Max. :6.00 Max. :9.000 Max. :5.000
## ideology religpew Collegelevelprogramming
## Min. :1.000 Min. : 1.000 Min. :1.00
## 1st Qu.:2.000 1st Qu.: 1.000 1st Qu.:1.00
## Median :3.000 Median : 2.000 Median :2.00
## Mean :3.304 Mean : 5.484 Mean :1.75
## 3rd Qu.:4.000 3rd Qu.:11.000 3rd Qu.:2.00
## Max. :8.000 Max. :12.000 Max. :2.00
## engineeringcompsciencedegree gradcompeng programexperience
## Min. :1.000 Min. :1.000 Min. :1.000
## 1st Qu.:2.000 1st Qu.:2.000 1st Qu.:2.000
## Median :2.000 Median :2.000 Median :2.000
## Mean :1.927 Mean :1.962 Mean :1.889
## 3rd Qu.:2.000 3rd Qu.:2.000 3rd Qu.:2.000
## Max. :2.000 Max. :2.000 Max. :2.000
## noeducationalcompexp birthyr age
## Min. :1.000 Min. :1927 Min. :1.000
## 1st Qu.:1.000 1st Qu.:1956 1st Qu.:2.000
## Median :1.000 Median :1972 Median :2.000
## Mean :1.368 Mean :1971 Mean :1.814
## 3rd Qu.:2.000 3rd Qu.:1986 3rd Qu.:2.000
## Max. :2.000 Max. :2000 Max. :2.000
hist(AI_Importance$SupportAIdev)
model <- lm(SupportAIdev ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = SupportAIdev ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.1955 -0.9770 -0.2531 0.8941 3.9685
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.393206 0.823541 0.477 0.633089
## age 0.061453 0.082362 0.746 0.455675
## gender 0.272564 0.061764 4.413 1.07e-05 ***
## race 0.041328 0.022888 1.806 0.071125 .
## educ -0.105489 0.022270 -4.737 2.32e-06 ***
## maritalstatus 0.002827 0.016760 0.169 0.866054
## employ 0.016064 0.012615 1.273 0.203034
## pid 0.101119 0.027108 3.730 0.000197 ***
## ideology 0.167102 0.022768 7.339 3.12e-13 ***
## religpew -0.011352 0.006870 -1.652 0.098614 .
## Collegelevelprogramming 0.201962 0.126111 1.601 0.109434
## engineeringcompsciencedegree 0.313521 0.134312 2.334 0.019680 *
## gradcompeng 0.174676 0.169877 1.028 0.303958
## programexperience 0.376103 0.114677 3.280 0.001057 **
## noeducationalcompexp -0.279152 0.134435 -2.076 0.037978 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.333 on 1985 degrees of freedom
## Multiple R-squared: 0.1674, Adjusted R-squared: 0.1615
## F-statistic: 28.51 on 14 and 1985 DF, p-value: < 2.2e-16
Simple Bar charts
library(ggplot2)
bargender<-ggplot(data=AI_Importance, aes (x=gender, y=SupportAIdev)) + geom_bar(stat="identity", fill="steelblue")+theme_minimal()
bargender + ggtitle("Support for AI development by gender: Male vs Female") +
xlab("Gender") + ylab("Support for AI development")
barage<-ggplot(data=AI_Importance, aes (x=age, y=SupportAIdev)) + geom_bar(stat="identity", fill="deepskyblue3")+theme_minimal()
barage + ggtitle("Support for AI development by age: Born on or after 1990 vs before 1990") +
xlab("Age") + ylab("Support for AI development")
This shows that those born before 1990 are more likely to support AI
development. However, it must be noted that this cut off point was
arbitrary, based on birth year data - and it would be interesting to
explore the effect of age buy smaller cohorts e.g. Millenials, Boomers,
etc.
barrace<-ggplot(data=AI_Importance, aes (x=race, y=SupportAIdev)) + geom_bar(stat="identity", fill="lightskyblue")+theme_minimal()
barrace + ggtitle("Support for AI development by race: White-Black-Hispanic-Asian-Native American-Mixed-Other-Middle Eastern-Skipped-Not Asked") +
xlab("Race") + ylab("Support for AI development")
bareduc<-ggplot(data=AI_Importance, aes (x=educ, y=SupportAIdev)) + geom_bar(stat="identity", fill="slateblue")+theme_minimal()
bareduc + ggtitle("Support for AI development by Education") +
xlab("Education level") + ylab("Support for AI development")
barmaritalstatus<-ggplot(data=AI_Importance, aes (x=maritalstatus, y=SupportAIdev)) + geom_bar(stat="identity", fill="springgreen2")+theme_minimal()
barmaritalstatus + ggtitle("Support for AI development by Marital Status") +
xlab("Marital Status") + ylab("Support for AI development")
baremployment<-ggplot(data=AI_Importance, aes (x=employ, y=SupportAIdev)) + geom_bar(stat="identity", fill="burlywood")+theme_minimal()
baremployment + ggtitle("Support for AI development by Employment Status") +
xlab("Employment Status") + ylab("Support for AI development")
barpoliticalidentity<-ggplot(data=AI_Importance, aes (x=pid, y=SupportAIdev)) + geom_bar(stat="identity", fill="lightseagreen")+theme_minimal()
barpoliticalidentity + ggtitle("Support for AI development by Political Identity") +
xlab("Political Identity") + ylab("Support for AI development")
barideology<-ggplot(data=AI_Importance, aes (x=ideology, y=SupportAIdev)) + geom_bar(stat="identity", fill="springgreen4")+theme_minimal()
barideology + ggtitle("Support for AI development by Political Ideology") +
xlab("Political Ideology") + ylab("Support for AI development")
barreligion<-ggplot(data=AI_Importance, aes (x=religpew, y=SupportAIdev)) + geom_bar(stat="identity", fill="navyblue")+theme_minimal()
barreligion + ggtitle("Support for AI development by Religion") +
xlab("Religious Identity") + ylab("Support for AI development")
barcollegeprogramming<-ggplot(data=AI_Importance, aes (x=Collegelevelprogramming, y=SupportAIdev)) + geom_bar(stat="identity", fill="turquoise4")+theme_minimal()
barcollegeprogramming + ggtitle("Support for AI development by Experience in College Level Programming") +
xlab("College Level Programming") + ylab("Support for AI development")
barengineering<-ggplot(data=AI_Importance, aes (x=engineeringcompsciencedegree, y=SupportAIdev)) + geom_bar(stat="identity", fill="mediumpurple4")+theme_minimal()
barengineering + ggtitle("Support for AI development by Engineering or Computer Science Bachelor's Degree") +
xlab("Comp Science or Engineering Bach degree") + ylab("Support for AI development")
barcompgrad<-ggplot(data=AI_Importance, aes (x=gradcompeng, y=SupportAIdev)) + geom_bar(stat="identity", fill="red4")+theme_minimal()
barcompgrad + ggtitle("Support for AI development by Graduate Engineering or Comp Science degree") +
xlab("Graduate degree in Engineering or Comp Science") + ylab("Support for AI development")
barprogramex<-ggplot(data=AI_Importance, aes (x=programexperience, y=SupportAIdev)) + geom_bar(stat="identity", fill="aquamarine4")+theme_minimal()
barprogramex + ggtitle("Support for AI development by prior programming experience") +
xlab("Programming Experience") + ylab("Support for AI development")
There is significance variance across demographic subgroups for support for the development of AI, with gender, education, income, and experience being key predictors.
Hypotheses - Results
Therefore:
- H1: Age is positively associated with support for AI.
- H2: There is no association between religion and support for AI.
- H3: Education level is positively associated with support for AI.
- H4: Marital status is not associated with support for AI.
- H5: Males are more likely to support AI.
- H6: Race is not associated with support for AI.
- H7: Employment status is not associated with support for AI.
- H8: You are more likely to support AI if you are a Democrat.
- H9: You are morel ikely to support AI if you are of a liberal ideology.
- H10: Experience with computer science is positively associated with support for AI.
- H11: An engineering degree is positively associated with support for AI.
Further, the current dataset allowed us to explore confidence in various organisations to develop AI in the best interests of the public. Similar to the above, the author conducted multiple regression analyses by age, education level, religious affiliation, marital status, gender, race, employment status, political party affiliation, political ideology, and experience with computer science.
Further analyses: Confidence in Organisations
model <- lm(USmilitary ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = USmilitary ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.049 -3.027 1.608 2.103 4.875
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.707931 0.935329 7.172 1.05e-12
## ageBefore before 1990 0.524379 0.209756 2.500 0.0125
## genderfemale 0.060806 0.146117 0.416 0.6773
## raceBlack -0.078927 0.233297 -0.338 0.7352
## raceHispanic -0.334210 0.203453 -1.643 0.1006
## raceAsian -0.007967 0.381283 -0.021 0.9833
## raceNative American -0.702441 0.782903 -0.897 0.3697
## raceMixed -0.845569 0.468653 -1.804 0.0713
## raceOther -0.512478 0.605690 -0.846 0.3976
## raceMiddle Eastern -0.218514 1.278417 -0.171 0.8643
## educ.L -0.006117 0.244476 -0.025 0.9800
## educ.Q 0.162656 0.217778 0.747 0.4552
## educ.C 0.165713 0.179058 0.925 0.3548
## educ^4 -0.036671 0.160303 -0.229 0.8191
## educ^5 0.093843 0.172332 0.545 0.5861
## maritalstatusSeparated 0.988856 0.566986 1.744 0.0813
## maritalstatusDivorced 0.152603 0.231840 0.658 0.5105
## maritalstatusWidowed 0.101690 0.339125 0.300 0.7643
## maritalstatusNever married 0.167012 0.176498 0.946 0.3441
## maritalstatusDomestic/civil partnership -0.200795 0.341762 -0.588 0.5569
## employPart-time 0.102479 0.241505 0.424 0.6714
## employTemporarily laid off 0.352764 0.785219 0.449 0.6533
## employUnemployed 0.346525 0.269471 1.286 0.1986
## employRetired -0.060650 0.203499 -0.298 0.7657
## employPermanently disabled -0.226650 0.279170 -0.812 0.4170
## employHomemaker 0.425984 0.283444 1.503 0.1330
## employStudent 0.537599 0.326426 1.647 0.0997
## employOther 0.578986 0.581162 0.996 0.3192
## pidRepublican -0.116978 0.230077 -0.508 0.6112
## pidIndependent 0.021356 0.184452 0.116 0.9078
## pidOther 0.314995 0.362813 0.868 0.3854
## pidNot sure 0.409888 0.313806 1.306 0.1916
## ideology.L 0.462177 1.739901 0.266 0.7905
## ideology.Q 0.605773 1.671064 0.363 0.7170
## ideology.C 0.201935 1.254361 0.161 0.8721
## ideology^4 -0.125724 0.764953 -0.164 0.8695
## ideology^5 0.195235 0.384167 0.508 0.6114
## ideology^6 0.094789 0.194743 0.487 0.6265
## religpewRoman Catholic -0.347459 0.200692 -1.731 0.0836
## religpewMormon 0.094665 0.535686 0.177 0.8597
## religpewEastern or Greek Orthodox -2.440518 1.086557 -2.246 0.0248
## religpewJewish 0.287570 0.403623 0.712 0.4763
## religpewMuslim -0.615888 0.655281 -0.940 0.3474
## religpewBuddhist -0.916553 0.755240 -1.214 0.2251
## religpewHindu 1.338137 1.000610 1.337 0.1813
## religpewAtheist 0.324750 0.301563 1.077 0.2817
## religpewAgnostic 0.225166 0.292230 0.771 0.4411
## religpewNothing in particular 0.041785 0.195991 0.213 0.8312
## religpewSomething else 0.187628 0.323322 0.580 0.5618
## CollegelevelprogrammingNo -0.322965 0.286556 -1.127 0.2599
## engineeringcompsciencedegreeNo -0.293387 0.306481 -0.957 0.3385
## gradcompengNo 0.114327 0.385718 0.296 0.7670
## programexperienceNo 0.416172 0.261922 1.589 0.1122
## noeducationalcompexpNo -0.158679 0.307259 -0.516 0.6056
##
## (Intercept) ***
## ageBefore before 1990 *
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed .
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q
## educ.C
## educ^4
## educ^5
## maritalstatusSeparated .
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent .
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic .
## religpewMormon
## religpewEastern or Greek Orthodox *
## religpewJewish
## religpewMuslim
## religpewBuddhist
## religpewHindu
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.985 on 1946 degrees of freedom
## Multiple R-squared: 0.03361, Adjusted R-squared: 0.007289
## F-statistic: 1.277 on 53 and 1946 DF, p-value: 0.08821
model <- lm(UScivilian_go ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = UScivilian_go ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.700 -2.969 1.629 2.027 3.422
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.65998 0.89046 8.602 <2e-16
## ageBefore before 1990 -0.26205 0.19969 -1.312 0.1896
## genderfemale 0.05297 0.13911 0.381 0.7034
## raceBlack -0.04305 0.22211 -0.194 0.8463
## raceHispanic 0.03172 0.19369 0.164 0.8699
## raceAsian 0.29007 0.36299 0.799 0.4243
## raceNative American 0.26216 0.74534 0.352 0.7251
## raceMixed -0.56893 0.44617 -1.275 0.2024
## raceOther -0.49658 0.57663 -0.861 0.3892
## raceMiddle Eastern 0.54089 1.21708 0.444 0.6568
## educ.L 0.30218 0.23275 1.298 0.1943
## educ.Q -0.23612 0.20733 -1.139 0.2549
## educ.C 0.23079 0.17047 1.354 0.1759
## educ^4 -0.30815 0.15261 -2.019 0.0436
## educ^5 -0.08833 0.16406 -0.538 0.5904
## maritalstatusSeparated -0.15585 0.53979 -0.289 0.7728
## maritalstatusDivorced -0.03303 0.22072 -0.150 0.8811
## maritalstatusWidowed 0.26360 0.32286 0.816 0.4143
## maritalstatusNever married 0.20338 0.16803 1.210 0.2263
## maritalstatusDomestic/civil partnership 0.35337 0.32537 1.086 0.2776
## employPart-time 0.02218 0.22992 0.096 0.9232
## employTemporarily laid off -0.10430 0.74755 -0.140 0.8890
## employUnemployed -0.10813 0.25654 -0.421 0.6734
## employRetired 0.24247 0.19374 1.252 0.2109
## employPermanently disabled 0.22056 0.26578 0.830 0.4067
## employHomemaker -0.10379 0.26985 -0.385 0.7005
## employStudent -0.39328 0.31077 -1.266 0.2058
## employOther 0.62061 0.55328 1.122 0.2621
## pidRepublican 0.22440 0.21904 1.024 0.3057
## pidIndependent 0.11242 0.17560 0.640 0.5221
## pidOther 0.02121 0.34541 0.061 0.9510
## pidNot sure 0.19696 0.29875 0.659 0.5098
## ideology.L 1.78016 1.65643 1.075 0.2826
## ideology.Q 1.41554 1.59089 0.890 0.3737
## ideology.C 1.09048 1.19418 0.913 0.3613
## ideology^4 0.69616 0.72825 0.956 0.3392
## ideology^5 0.38127 0.36574 1.042 0.2973
## ideology^6 0.05991 0.18540 0.323 0.7466
## religpewRoman Catholic -0.28961 0.19106 -1.516 0.1297
## religpewMormon 0.59441 0.50999 1.166 0.2439
## religpewEastern or Greek Orthodox 1.09038 1.03443 1.054 0.2920
## religpewJewish 0.24672 0.38426 0.642 0.5209
## religpewMuslim -0.14136 0.62384 -0.227 0.8208
## religpewBuddhist 0.98854 0.71901 1.375 0.1693
## religpewHindu 0.36880 0.95261 0.387 0.6987
## religpewAtheist -0.19109 0.28710 -0.666 0.5057
## religpewAgnostic 0.14491 0.27821 0.521 0.6025
## religpewNothing in particular 0.17065 0.18659 0.915 0.3605
## religpewSomething else -0.27570 0.30781 -0.896 0.3705
## CollegelevelprogrammingNo -0.19747 0.27281 -0.724 0.4692
## engineeringcompsciencedegreeNo 0.06633 0.29178 0.227 0.8202
## gradcompengNo -0.21420 0.36721 -0.583 0.5597
## programexperienceNo 0.12322 0.24936 0.494 0.6212
## noeducationalcompexpNo -0.37447 0.29252 -1.280 0.2006
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q
## educ.C
## educ^4 *
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim
## religpewBuddhist
## religpewHindu
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.842 on 1946 degrees of freedom
## Multiple R-squared: 0.02342, Adjusted R-squared: -0.003173
## F-statistic: 0.8807 on 53 and 1946 DF, p-value: 0.7155
model <- lm(NationalSecurity ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = NationalSecurity ~ age + gender + race + educ +
## maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.758 -3.006 1.723 2.162 4.395
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.836122 0.916793 7.457 1.33e-13
## ageBefore before 1990 0.190624 0.205599 0.927 0.35396
## genderfemale -0.074117 0.143222 -0.517 0.60487
## raceBlack -0.035944 0.228674 -0.157 0.87512
## raceHispanic 0.240389 0.199421 1.205 0.22818
## raceAsian 0.455665 0.373726 1.219 0.22290
## raceNative American -1.435794 0.767388 -1.871 0.06149
## raceMixed 0.183606 0.459365 0.400 0.68942
## raceOther 0.982584 0.593687 1.655 0.09807
## raceMiddle Eastern 0.526033 1.253082 0.420 0.67468
## educ.L -0.051701 0.239631 -0.216 0.82920
## educ.Q -0.354937 0.213462 -1.663 0.09652
## educ.C 0.386513 0.175509 2.202 0.02777
## educ^4 0.036380 0.157126 0.232 0.81692
## educ^5 -0.157452 0.168917 -0.932 0.35139
## maritalstatusSeparated -0.065407 0.555750 -0.118 0.90632
## maritalstatusDivorced 0.040862 0.227246 0.180 0.85732
## maritalstatusWidowed 0.230590 0.332404 0.694 0.48795
## maritalstatusNever married 0.289609 0.173000 1.674 0.09428
## maritalstatusDomestic/civil partnership -0.346244 0.334989 -1.034 0.30145
## employPart-time -0.367002 0.236719 -1.550 0.12121
## employTemporarily laid off 0.080958 0.769658 0.105 0.91624
## employUnemployed -0.138417 0.264131 -0.524 0.60031
## employRetired -0.113742 0.199466 -0.570 0.56859
## employPermanently disabled -0.039571 0.273638 -0.145 0.88503
## employHomemaker 0.231413 0.277827 0.833 0.40498
## employStudent -0.002377 0.319957 -0.007 0.99407
## employOther -0.519141 0.569645 -0.911 0.36223
## pidRepublican -0.037895 0.225518 -0.168 0.86657
## pidIndependent 0.285583 0.180796 1.580 0.11437
## pidOther 0.144549 0.355623 0.406 0.68444
## pidNot sure -0.190612 0.307588 -0.620 0.53553
## ideology.L 1.286978 1.705420 0.755 0.45056
## ideology.Q 1.041290 1.637947 0.636 0.52503
## ideology.C 0.343213 1.229503 0.279 0.78016
## ideology^4 0.425562 0.749793 0.568 0.57039
## ideology^5 0.003510 0.376554 0.009 0.99256
## ideology^6 -0.035956 0.190884 -0.188 0.85061
## religpewRoman Catholic 0.002715 0.196714 0.014 0.98899
## religpewMormon -1.643955 0.525070 -3.131 0.00177
## religpewEastern or Greek Orthodox -0.649015 1.065024 -0.609 0.54234
## religpewJewish 0.143428 0.395624 0.363 0.71699
## religpewMuslim 0.413605 0.642295 0.644 0.51968
## religpewBuddhist -1.442714 0.740273 -1.949 0.05145
## religpewHindu 1.474217 0.980780 1.503 0.13297
## religpewAtheist 0.269035 0.295587 0.910 0.36285
## religpewAgnostic 0.348937 0.286439 1.218 0.22330
## religpewNothing in particular -0.238305 0.192107 -1.240 0.21495
## religpewSomething else -0.055021 0.316915 -0.174 0.86219
## CollegelevelprogrammingNo -0.229252 0.280877 -0.816 0.41449
## engineeringcompsciencedegreeNo -0.043785 0.300408 -0.146 0.88413
## gradcompengNo 0.208101 0.378074 0.550 0.58209
## programexperienceNo 0.150437 0.256731 0.586 0.55796
## noeducationalcompexpNo -0.205596 0.301170 -0.683 0.49490
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American .
## raceMixed
## raceOther .
## raceMiddle Eastern
## educ.L
## educ.Q .
## educ.C *
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married .
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon **
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim
## religpewBuddhist .
## religpewHindu
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.926 on 1946 degrees of freedom
## Multiple R-squared: 0.0335, Adjusted R-squared: 0.007175
## F-statistic: 1.273 on 53 and 1946 DF, p-value: 0.09124
model <- lm(FBI ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = FBI ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.688 -2.868 1.610 2.035 3.717
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 9.1291215 0.8961487 10.187 < 2e-16
## ageBefore before 1990 -0.0951433 0.2009694 -0.473 0.63597
## genderfemale 0.0841476 0.1399965 0.601 0.54786
## raceBlack 0.1423751 0.2235248 0.637 0.52423
## raceHispanic -0.1424300 0.1949308 -0.731 0.46507
## raceAsian -0.0002492 0.3653108 -0.001 0.99946
## raceNative American 0.6410718 0.7501079 0.855 0.39286
## raceMixed 0.0965096 0.4490214 0.215 0.82984
## raceOther 0.5524664 0.5803183 0.952 0.34121
## raceMiddle Eastern 0.4847458 1.2248648 0.396 0.69233
## educ.L -0.2757804 0.2342351 -1.177 0.23919
## educ.Q -0.0623032 0.2086550 -0.299 0.76528
## educ.C 0.0761957 0.1715573 0.444 0.65699
## educ^4 -0.2139002 0.1535881 -1.393 0.16387
## educ^5 -0.0381275 0.1651133 -0.231 0.81740
## maritalstatusSeparated -0.4234495 0.5432356 -0.779 0.43578
## maritalstatusDivorced 0.2268773 0.2221288 1.021 0.30720
## maritalstatusWidowed -0.0238415 0.3249191 -0.073 0.94151
## maritalstatusNever married 0.2253819 0.1691043 1.333 0.18275
## maritalstatusDomestic/civil partnership 0.4209490 0.3274461 1.286 0.19875
## employPart-time -0.1768968 0.2313881 -0.765 0.44466
## employTemporarily laid off -0.2660815 0.7523265 -0.354 0.72362
## employUnemployed -0.1295727 0.2581829 -0.502 0.61582
## employRetired -0.2588281 0.1949748 -1.327 0.18450
## employPermanently disabled -0.6374294 0.2674760 -2.383 0.01726
## employHomemaker -0.0210068 0.2715705 -0.077 0.93835
## employStudent -0.4933292 0.3127518 -1.577 0.11487
## employOther 0.2581390 0.5568176 0.464 0.64299
## pidRepublican -0.0906680 0.2204394 -0.411 0.68089
## pidIndependent 0.1602367 0.1767252 0.907 0.36468
## pidOther 0.4003086 0.3476145 1.152 0.24963
## pidNot sure 0.0100209 0.3006613 0.033 0.97342
## ideology.L -1.2669252 1.6670173 -0.760 0.44735
## ideology.Q -1.6697357 1.6010638 -1.043 0.29713
## ideology.C -1.0414153 1.2018167 -0.867 0.38630
## ideology^4 -0.7791202 0.7329095 -1.063 0.28789
## ideology^5 -0.4295515 0.3680750 -1.167 0.24334
## ideology^6 -0.3860999 0.1865852 -2.069 0.03865
## religpewRoman Catholic -0.0593627 0.1922847 -0.309 0.75757
## religpewMormon -0.6518067 0.5132468 -1.270 0.20425
## religpewEastern or Greek Orthodox 0.3596760 1.0410421 0.345 0.72976
## religpewJewish 0.2018237 0.3867153 0.522 0.60180
## religpewMuslim 0.2742860 0.6278322 0.437 0.66225
## religpewBuddhist -0.6631946 0.7236037 -0.917 0.35951
## religpewHindu 1.4496835 0.9586951 1.512 0.13066
## religpewAtheist -0.0648936 0.2889310 -0.225 0.82231
## religpewAgnostic -0.5084512 0.2799891 -1.816 0.06953
## religpewNothing in particular -0.1822453 0.1877806 -0.971 0.33191
## religpewSomething else -0.2496945 0.3097785 -0.806 0.42032
## CollegelevelprogrammingNo -0.7817005 0.2745521 -2.847 0.00446
## engineeringcompsciencedegreeNo -0.0321333 0.2936432 -0.109 0.91287
## gradcompengNo -1.1164000 0.3695608 -3.021 0.00255
## programexperienceNo -0.4454574 0.2509502 -1.775 0.07604
## noeducationalcompexpNo -0.7724574 0.2943881 -2.624 0.00876
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q
## educ.C
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled *
## employHomemaker
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6 *
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim
## religpewBuddhist
## religpewHindu
## religpewAtheist
## religpewAgnostic .
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo **
## engineeringcompsciencedegreeNo
## gradcompengNo **
## programexperienceNo .
## noeducationalcompexpNo **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.86 on 1946 degrees of freedom
## Multiple R-squared: 0.02875, Adjusted R-squared: 0.002298
## F-statistic: 1.087 on 53 and 1946 DF, p-value: 0.3119
model <- lm(CIA ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = CIA ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.475 -3.018 1.790 2.247 3.536
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.98219 0.91850 6.513 9.35e-11
## ageBefore before 1990 0.05265 0.20598 0.256 0.7983
## genderfemale -0.04865 0.14349 -0.339 0.7346
## raceBlack 0.05970 0.22910 0.261 0.7944
## raceHispanic 0.03983 0.19979 0.199 0.8420
## raceAsian -0.04762 0.37442 -0.127 0.8988
## raceNative American 1.30930 0.76881 1.703 0.0887
## raceMixed -0.09723 0.46022 -0.211 0.8327
## raceOther 0.26587 0.59479 0.447 0.6549
## raceMiddle Eastern 1.07392 1.25541 0.855 0.3924
## educ.L 0.06186 0.24008 0.258 0.7967
## educ.Q -0.03406 0.21386 -0.159 0.8735
## educ.C 0.40690 0.17584 2.314 0.0208
## educ^4 0.34465 0.15742 2.189 0.0287
## educ^5 0.38259 0.16923 2.261 0.0239
## maritalstatusSeparated 0.09777 0.55678 0.176 0.8606
## maritalstatusDivorced -0.17402 0.22767 -0.764 0.4447
## maritalstatusWidowed 0.16859 0.33302 0.506 0.6127
## maritalstatusNever married 0.11777 0.17332 0.679 0.4969
## maritalstatusDomestic/civil partnership -0.15497 0.33561 -0.462 0.6443
## employPart-time -0.21030 0.23716 -0.887 0.3753
## employTemporarily laid off 0.01787 0.77109 0.023 0.9815
## employUnemployed 0.04838 0.26462 0.183 0.8550
## employRetired -0.21466 0.19984 -1.074 0.2829
## employPermanently disabled -0.25226 0.27415 -0.920 0.3576
## employHomemaker -0.26480 0.27834 -0.951 0.3415
## employStudent 0.04469 0.32055 0.139 0.8891
## employOther -0.45094 0.57070 -0.790 0.4295
## pidRepublican 0.20607 0.22594 0.912 0.3618
## pidIndependent -0.17996 0.18113 -0.994 0.3206
## pidOther 0.02509 0.35628 0.070 0.9439
## pidNot sure -0.07221 0.30816 -0.234 0.8148
## ideology.L -1.79635 1.70859 -1.051 0.2932
## ideology.Q -1.95401 1.64099 -1.191 0.2339
## ideology.C -1.49019 1.23179 -1.210 0.2265
## ideology^4 -0.81250 0.75119 -1.082 0.2796
## ideology^5 -0.19849 0.37725 -0.526 0.5989
## ideology^6 -0.14903 0.19124 -0.779 0.4359
## religpewRoman Catholic 0.03923 0.19708 0.199 0.8423
## religpewMormon 0.24265 0.52605 0.461 0.6446
## religpewEastern or Greek Orthodox 0.52179 1.06700 0.489 0.6249
## religpewJewish 0.34986 0.39636 0.883 0.3775
## religpewMuslim 1.13571 0.64349 1.765 0.0777
## religpewBuddhist 0.81086 0.74165 1.093 0.2744
## religpewHindu 2.42864 0.98260 2.472 0.0135
## religpewAtheist 0.22427 0.29614 0.757 0.4489
## religpewAgnostic 0.34937 0.28697 1.217 0.2236
## religpewNothing in particular 0.06620 0.19246 0.344 0.7309
## religpewSomething else 0.62729 0.31750 1.976 0.0483
## CollegelevelprogrammingNo 0.11770 0.28140 0.418 0.6758
## engineeringcompsciencedegreeNo 0.05192 0.30097 0.173 0.8631
## gradcompengNo 0.19592 0.37878 0.517 0.6050
## programexperienceNo -0.06431 0.25721 -0.250 0.8026
## noeducationalcompexpNo 0.12510 0.30173 0.415 0.6785
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American .
## raceMixed
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q
## educ.C *
## educ^4 *
## educ^5 *
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim .
## religpewBuddhist
## religpewHindu *
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else *
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.932 on 1946 degrees of freedom
## Multiple R-squared: 0.02555, Adjusted R-squared: -0.0009884
## F-statistic: 0.9628 on 53 and 1946 DF, p-value: 0.5514
model <- lm(NATO ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = NATO ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.712 -2.814 1.607 2.029 4.773
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.568249 0.873221 7.522 8.2e-14
## ageBefore before 1990 0.187029 0.195828 0.955 0.3397
## genderfemale 0.259991 0.136415 1.906 0.0568
## raceBlack 0.304794 0.217806 1.399 0.1619
## raceHispanic 0.264084 0.189944 1.390 0.1646
## raceAsian -0.091161 0.355965 -0.256 0.7979
## raceNative American 0.843937 0.730917 1.155 0.2484
## raceMixed -0.243108 0.437533 -0.556 0.5785
## raceOther -0.025568 0.565471 -0.045 0.9639
## raceMiddle Eastern -1.257702 1.193527 -1.054 0.2921
## educ.L 0.154523 0.228242 0.677 0.4985
## educ.Q -0.401420 0.203317 -1.974 0.0485
## educ.C -0.177711 0.167168 -1.063 0.2879
## educ^4 -0.039079 0.149659 -0.261 0.7940
## educ^5 -0.037941 0.160889 -0.236 0.8136
## maritalstatusSeparated 0.507740 0.529337 0.959 0.3376
## maritalstatusDivorced -0.153005 0.216446 -0.707 0.4797
## maritalstatusWidowed -0.145864 0.316606 -0.461 0.6451
## maritalstatusNever married 0.124213 0.164778 0.754 0.4510
## maritalstatusDomestic/civil partnership -0.060287 0.319069 -0.189 0.8502
## employPart-time -0.087309 0.225468 -0.387 0.6986
## employTemporarily laid off -0.913875 0.733079 -1.247 0.2127
## employUnemployed 0.280090 0.251577 1.113 0.2657
## employRetired 0.176449 0.189986 0.929 0.3531
## employPermanently disabled 0.259482 0.260633 0.996 0.3196
## employHomemaker 0.065771 0.264623 0.249 0.8037
## employStudent -0.355774 0.304750 -1.167 0.2432
## employOther 1.016965 0.542572 1.874 0.0610
## pidRepublican -0.019322 0.214800 -0.090 0.9283
## pidIndependent 0.192343 0.172204 1.117 0.2642
## pidOther -0.010352 0.338721 -0.031 0.9756
## pidNot sure -0.052199 0.292969 -0.178 0.8586
## ideology.L 1.638496 1.624368 1.009 0.3132
## ideology.Q 1.451048 1.560102 0.930 0.3524
## ideology.C 0.801806 1.171069 0.685 0.4936
## ideology^4 0.552567 0.714158 0.774 0.4392
## ideology^5 0.426324 0.358658 1.189 0.2347
## ideology^6 0.097046 0.181812 0.534 0.5936
## religpewRoman Catholic 0.022113 0.187365 0.118 0.9061
## religpewMormon -0.443040 0.500116 -0.886 0.3758
## religpewEastern or Greek Orthodox 0.259061 1.014408 0.255 0.7985
## religpewJewish -0.512330 0.376821 -1.360 0.1741
## religpewMuslim -0.783218 0.611769 -1.280 0.2006
## religpewBuddhist 0.269238 0.705091 0.382 0.7026
## religpewHindu 0.645103 0.934167 0.691 0.4899
## religpewAtheist -0.107006 0.281539 -0.380 0.7039
## religpewAgnostic -0.051652 0.272826 -0.189 0.8499
## religpewNothing in particular 0.001431 0.182976 0.008 0.9938
## religpewSomething else 0.229906 0.301853 0.762 0.4464
## CollegelevelprogrammingNo 0.457952 0.267528 1.712 0.0871
## engineeringcompsciencedegreeNo 0.695533 0.286130 2.431 0.0152
## gradcompengNo -0.305348 0.360106 -0.848 0.3966
## programexperienceNo -0.361525 0.244530 -1.478 0.1394
## noeducationalcompexpNo 0.031266 0.286856 0.109 0.9132
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale .
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q *
## educ.C
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther .
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim
## religpewBuddhist
## religpewHindu
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo .
## engineeringcompsciencedegreeNo *
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.787 on 1946 degrees of freedom
## Multiple R-squared: 0.03053, Adjusted R-squared: 0.004123
## F-statistic: 1.156 on 53 and 1946 DF, p-value: 0.2074
model <- lm(Techcomp ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = Techcomp ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.987 -3.121 1.749 2.124 4.527
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.39836 0.94027 6.805 1.34e-11
## ageBefore before 1990 0.08861 0.21086 0.420 0.6744
## genderfemale 0.08297 0.14689 0.565 0.5722
## raceBlack 0.02351 0.23453 0.100 0.9202
## raceHispanic -0.12069 0.20453 -0.590 0.5552
## raceAsian -0.09133 0.38329 -0.238 0.8117
## raceNative American -0.59315 0.78704 -0.754 0.4511
## raceMixed 0.36990 0.47113 0.785 0.4325
## raceOther -0.31755 0.60889 -0.522 0.6021
## raceMiddle Eastern -0.24688 1.28516 -0.192 0.8477
## educ.L -0.11762 0.24577 -0.479 0.6323
## educ.Q 0.02227 0.21893 0.102 0.9190
## educ.C -0.06006 0.18000 -0.334 0.7387
## educ^4 -0.06512 0.16115 -0.404 0.6862
## educ^5 -0.23097 0.17324 -1.333 0.1826
## maritalstatusSeparated -0.26591 0.56998 -0.467 0.6409
## maritalstatusDivorced 0.07865 0.23306 0.337 0.7358
## maritalstatusWidowed -0.12126 0.34091 -0.356 0.7221
## maritalstatusNever married -0.09890 0.17743 -0.557 0.5773
## maritalstatusDomestic/civil partnership -0.45855 0.34357 -1.335 0.1821
## employPart-time 0.18508 0.24278 0.762 0.4459
## employTemporarily laid off -1.51787 0.78936 -1.923 0.0546
## employUnemployed -0.05995 0.27089 -0.221 0.8249
## employRetired -0.04883 0.20457 -0.239 0.8114
## employPermanently disabled 0.39173 0.28064 1.396 0.1629
## employHomemaker 0.20627 0.28494 0.724 0.4692
## employStudent 0.23149 0.32815 0.705 0.4806
## employOther -0.11744 0.58423 -0.201 0.8407
## pidRepublican 0.21142 0.23129 0.914 0.3608
## pidIndependent 0.15479 0.18543 0.835 0.4040
## pidOther 0.13633 0.36473 0.374 0.7086
## pidNot sure 0.31950 0.31546 1.013 0.3113
## ideology.L 0.96858 1.74908 0.554 0.5798
## ideology.Q 0.82568 1.67988 0.492 0.6231
## ideology.C 0.51809 1.26098 0.411 0.6812
## ideology^4 0.27919 0.76899 0.363 0.7166
## ideology^5 0.01143 0.38620 0.030 0.9764
## ideology^6 0.35656 0.19577 1.821 0.0687
## religpewRoman Catholic 0.24316 0.20175 1.205 0.2283
## religpewMormon 0.05315 0.53851 0.099 0.9214
## religpewEastern or Greek Orthodox 0.62125 1.09229 0.569 0.5696
## religpewJewish -0.01087 0.40575 -0.027 0.9786
## religpewMuslim 0.32928 0.65874 0.500 0.6172
## religpewBuddhist 0.77828 0.75923 1.025 0.3054
## religpewHindu -1.51865 1.00589 -1.510 0.1313
## religpewAtheist -0.10940 0.30315 -0.361 0.7182
## religpewAgnostic 0.17218 0.29377 0.586 0.5579
## religpewNothing in particular 0.11781 0.19702 0.598 0.5499
## religpewSomething else 0.26076 0.32503 0.802 0.4225
## CollegelevelprogrammingNo -0.08844 0.28807 -0.307 0.7589
## engineeringcompsciencedegreeNo -0.32076 0.30810 -1.041 0.2980
## gradcompengNo 0.45723 0.38775 1.179 0.2385
## programexperienceNo 0.44767 0.26330 1.700 0.0893
## noeducationalcompexpNo -0.17091 0.30888 -0.553 0.5801
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q
## educ.C
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off .
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6 .
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim
## religpewBuddhist
## religpewHindu
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo .
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.001 on 1946 degrees of freedom
## Multiple R-squared: 0.02189, Adjusted R-squared: -0.004751
## F-statistic: 0.8217 on 53 and 1946 DF, p-value: 0.8171
model <- lm(Google ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = Google ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.295 -2.860 1.558 2.005 4.776
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.5463121 0.9008162 7.267 5.29e-13
## ageBefore before 1990 -0.2214963 0.2020161 -1.096 0.27303
## genderfemale -0.3380874 0.1407256 -2.402 0.01638
## raceBlack 0.2111831 0.2246890 0.940 0.34739
## raceHispanic 0.2061817 0.1959461 1.052 0.29282
## raceAsian -0.1882285 0.3672135 -0.513 0.60830
## raceNative American 0.6429113 0.7540148 0.853 0.39396
## raceMixed 0.9400735 0.4513600 2.083 0.03740
## raceOther -0.7416432 0.5833408 -1.271 0.20375
## raceMiddle Eastern 0.3311618 1.2312444 0.269 0.78799
## educ.L 0.4586555 0.2354551 1.948 0.05156
## educ.Q -0.1513884 0.2097417 -0.722 0.47051
## educ.C 0.1508099 0.1724509 0.875 0.38195
## educ^4 -0.1344666 0.1543881 -0.871 0.38388
## educ^5 0.1129281 0.1659733 0.680 0.49633
## maritalstatusSeparated 0.3956257 0.5460650 0.725 0.46884
## maritalstatusDivorced 0.1571610 0.2232858 0.704 0.48161
## maritalstatusWidowed 0.3889302 0.3266114 1.191 0.23388
## maritalstatusNever married -0.1765194 0.1699850 -1.038 0.29919
## maritalstatusDomestic/civil partnership 0.4307835 0.3291516 1.309 0.19077
## employPart-time 0.2946519 0.2325933 1.267 0.20537
## employTemporarily laid off 0.3337871 0.7562450 0.441 0.65899
## employUnemployed -0.0003878 0.2595276 -0.001 0.99881
## employRetired 0.0408778 0.1959903 0.209 0.83481
## employPermanently disabled 0.3771827 0.2688691 1.403 0.16082
## employHomemaker 0.5933880 0.2729850 2.174 0.02985
## employStudent 0.2638119 0.3143807 0.839 0.40149
## employOther -0.0549436 0.5597178 -0.098 0.92181
## pidRepublican 0.4119402 0.2215875 1.859 0.06317
## pidIndependent 0.4366300 0.1776456 2.458 0.01406
## pidOther 0.2953015 0.3494250 0.845 0.39816
## pidNot sure 0.1471938 0.3022272 0.487 0.62629
## ideology.L 1.1468733 1.6756998 0.684 0.49379
## ideology.Q 1.0108941 1.6094028 0.628 0.53000
## ideology.C 0.7000102 1.2080762 0.579 0.56236
## ideology^4 -0.0479701 0.7367267 -0.065 0.94809
## ideology^5 0.0232723 0.3699920 0.063 0.94985
## ideology^6 -0.1095982 0.1875570 -0.584 0.55906
## religpewRoman Catholic -0.3765772 0.1932862 -1.948 0.05152
## religpewMormon -1.4824823 0.5159199 -2.873 0.00410
## religpewEastern or Greek Orthodox 1.0012332 1.0464642 0.957 0.33880
## religpewJewish -0.8594236 0.3887294 -2.211 0.02716
## religpewMuslim 0.0606188 0.6311022 0.096 0.92349
## religpewBuddhist -1.8015512 0.7273725 -2.477 0.01334
## religpewHindu -2.5177275 0.9636884 -2.613 0.00905
## religpewAtheist -0.0243374 0.2904358 -0.084 0.93323
## religpewAgnostic -0.7484454 0.2814474 -2.659 0.00790
## religpewNothing in particular -0.2758369 0.1887587 -1.461 0.14409
## religpewSomething else -0.8202172 0.3113920 -2.634 0.00850
## CollegelevelprogrammingNo 0.1285251 0.2759821 0.466 0.64148
## engineeringcompsciencedegreeNo 0.0446497 0.2951726 0.151 0.87978
## gradcompengNo 0.5576870 0.3714857 1.501 0.13346
## programexperienceNo 0.2805461 0.2522573 1.112 0.26621
## noeducationalcompexpNo 0.1225870 0.2959214 0.414 0.67873
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale *
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed *
## raceOther
## raceMiddle Eastern
## educ.L .
## educ.Q
## educ.C
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker *
## employStudent
## employOther
## pidRepublican .
## pidIndependent *
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic .
## religpewMormon **
## religpewEastern or Greek Orthodox
## religpewJewish *
## religpewMuslim
## religpewBuddhist *
## religpewHindu **
## religpewAtheist
## religpewAgnostic **
## religpewNothing in particular
## religpewSomething else **
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.875 on 1946 degrees of freedom
## Multiple R-squared: 0.03897, Adjusted R-squared: 0.0128
## F-statistic: 1.489 on 53 and 1946 DF, p-value: 0.01334
model <- lm(Facebook ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = Facebook ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.020 -2.884 1.485 1.849 3.610
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.410604 0.824962 8.983 < 2e-16
## ageBefore before 1990 0.210426 0.185005 1.137 0.25551
## genderfemale -0.001218 0.128876 -0.009 0.99246
## raceBlack 0.286905 0.205769 1.394 0.16338
## raceHispanic 0.245122 0.179446 1.366 0.17210
## raceAsian 0.483514 0.336292 1.438 0.15066
## raceNative American 1.454628 0.690522 2.107 0.03528
## raceMixed 0.422268 0.413353 1.022 0.30711
## raceOther 0.625629 0.534220 1.171 0.24170
## raceMiddle Eastern 1.911881 1.127566 1.696 0.09012
## educ.L 0.290040 0.215628 1.345 0.17875
## educ.Q -0.175288 0.192080 -0.913 0.36158
## educ.C -0.166644 0.157929 -1.055 0.29147
## educ^4 -0.105808 0.141388 -0.748 0.45434
## educ^5 -0.139770 0.151997 -0.920 0.35792
## maritalstatusSeparated 0.119682 0.500083 0.239 0.81088
## maritalstatusDivorced 0.299263 0.204484 1.464 0.14349
## maritalstatusWidowed 0.151548 0.299109 0.507 0.61245
## maritalstatusNever married 0.128183 0.155671 0.823 0.41037
## maritalstatusDomestic/civil partnership 0.195420 0.301435 0.648 0.51687
## employPart-time 0.083202 0.213007 0.391 0.69613
## employTemporarily laid off 0.370899 0.692564 0.536 0.59233
## employUnemployed 0.413473 0.237674 1.740 0.08208
## employRetired 0.201039 0.179487 1.120 0.26282
## employPermanently disabled -0.031864 0.246229 -0.129 0.89705
## employHomemaker 0.163883 0.249998 0.656 0.51220
## employStudent 0.267751 0.287908 0.930 0.35249
## employOther 0.287297 0.512586 0.560 0.57521
## pidRepublican 0.182978 0.202928 0.902 0.36733
## pidIndependent 0.021895 0.162687 0.135 0.89295
## pidOther 0.079855 0.320001 0.250 0.80297
## pidNot sure -0.116871 0.276778 -0.422 0.67289
## ideology.L 0.886561 1.534595 0.578 0.56352
## ideology.Q 0.887797 1.473881 0.602 0.54701
## ideology.C 1.002742 1.106348 0.906 0.36486
## ideology^4 0.520463 0.674690 0.771 0.44056
## ideology^5 -0.048482 0.338836 -0.143 0.88624
## ideology^6 -0.115390 0.171763 -0.672 0.50179
## religpewRoman Catholic -0.045467 0.177010 -0.257 0.79731
## religpewMormon 0.031444 0.472476 0.067 0.94695
## religpewEastern or Greek Orthodox 0.060839 0.958345 0.063 0.94939
## religpewJewish -0.467759 0.355996 -1.314 0.18902
## religpewMuslim -1.601794 0.577959 -2.771 0.00563
## religpewBuddhist 0.973014 0.666123 1.461 0.14426
## religpewHindu -1.352224 0.882539 -1.532 0.12564
## religpewAtheist -0.118119 0.265979 -0.444 0.65703
## religpewAgnostic 0.218103 0.257748 0.846 0.39755
## religpewNothing in particular -0.070756 0.172864 -0.409 0.68235
## religpewSomething else -0.082538 0.285171 -0.289 0.77228
## CollegelevelprogrammingNo 0.184930 0.252743 0.732 0.46444
## engineeringcompsciencedegreeNo -0.394342 0.270317 -1.459 0.14478
## gradcompengNo -0.267328 0.340204 -0.786 0.43209
## programexperienceNo 0.086325 0.231016 0.374 0.70869
## noeducationalcompexpNo -0.143368 0.271003 -0.529 0.59685
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American *
## raceMixed
## raceOther
## raceMiddle Eastern .
## educ.L
## educ.Q
## educ.C
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed .
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim **
## religpewBuddhist
## religpewHindu
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.633 on 1946 degrees of freedom
## Multiple R-squared: 0.02712, Adjusted R-squared: 0.0006224
## F-statistic: 1.023 on 53 and 1946 DF, p-value: 0.4284
model <- lm(Apple ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = Apple ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.530 -3.005 1.785 2.142 4.432
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.5029929 0.9252841 5.947 3.22e-09
## ageBefore before 1990 -0.0007201 0.2075032 -0.003 0.9972
## genderfemale 0.0576219 0.1445480 0.399 0.6902
## raceBlack -0.0445903 0.2307920 -0.193 0.8468
## raceHispanic 0.2322129 0.2012684 1.154 0.2487
## raceAsian -0.4559380 0.3771877 -1.209 0.2269
## raceNative American -0.3148938 0.7744953 -0.407 0.6844
## raceMixed -0.1074695 0.4636198 -0.232 0.8167
## raceOther 0.9136415 0.5991855 1.525 0.1275
## raceMiddle Eastern 1.4331715 1.2646873 1.133 0.2573
## educ.L 0.1352049 0.2418505 0.559 0.5762
## educ.Q 0.3752085 0.2154387 1.742 0.0817
## educ.C 0.1195627 0.1771350 0.675 0.4998
## educ^4 -0.1444602 0.1585816 -0.911 0.3624
## educ^5 0.1192257 0.1704814 0.699 0.4844
## maritalstatusSeparated -0.5017665 0.5608972 -0.895 0.3711
## maritalstatusDivorced -0.3768692 0.2293506 -1.643 0.1005
## maritalstatusWidowed -0.3017655 0.3354828 -0.899 0.3685
## maritalstatusNever married -0.0546502 0.1746022 -0.313 0.7543
## maritalstatusDomestic/civil partnership -0.0423043 0.3380920 -0.125 0.9004
## employPart-time 0.0285889 0.2389110 0.120 0.9048
## employTemporarily laid off -0.0112473 0.7767860 -0.014 0.9884
## employUnemployed 0.2522421 0.2665769 0.946 0.3442
## employRetired -0.1293587 0.2013138 -0.643 0.5206
## employPermanently disabled 0.1565883 0.2761721 0.567 0.5708
## employHomemaker 0.0296804 0.2803998 0.106 0.9157
## employStudent 0.3614840 0.3229199 1.119 0.2631
## employOther 0.6402740 0.5749208 1.114 0.2656
## pidRepublican 0.1211166 0.2276063 0.532 0.5947
## pidIndependent -0.0039186 0.1824708 -0.021 0.9829
## pidOther 0.1395037 0.3589161 0.389 0.6976
## pidNot sure 0.0418272 0.3104363 0.135 0.8928
## ideology.L -1.8103249 1.7212149 -1.052 0.2930
## ideology.Q -1.7353367 1.6531172 -1.050 0.2940
## ideology.C -1.2010309 1.2408899 -0.968 0.3332
## ideology^4 -0.6875323 0.7567376 -0.909 0.3637
## ideology^5 -0.5957827 0.3800417 -1.568 0.1171
## ideology^6 -0.1236086 0.1926514 -0.642 0.5212
## religpewRoman Catholic 0.2045844 0.1985362 1.030 0.3029
## religpewMormon -0.0171357 0.5299333 -0.032 0.9742
## religpewEastern or Greek Orthodox -0.3856758 1.0748882 -0.359 0.7198
## religpewJewish 0.1129535 0.3992881 0.283 0.7773
## religpewMuslim -0.7050971 0.6482441 -1.088 0.2769
## religpewBuddhist 0.1389536 0.7471293 0.186 0.8525
## religpewHindu -1.8714942 0.9898640 -1.891 0.0588
## religpewAtheist 0.1346430 0.2983246 0.451 0.6518
## religpewAgnostic -0.0919569 0.2890920 -0.318 0.7505
## religpewNothing in particular 0.0941078 0.1938857 0.485 0.6275
## religpewSomething else 0.1994959 0.3198500 0.624 0.5329
## CollegelevelprogrammingNo 0.2352599 0.2834782 0.830 0.4067
## engineeringcompsciencedegreeNo 0.2681810 0.3031900 0.885 0.3765
## gradcompengNo 0.4356348 0.3815759 1.142 0.2537
## programexperienceNo -0.0628079 0.2591090 -0.242 0.8085
## noeducationalcompexpNo 0.1272045 0.3039592 0.418 0.6756
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q .
## educ.C
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim
## religpewBuddhist
## religpewHindu .
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.953 on 1946 degrees of freedom
## Multiple R-squared: 0.02009, Adjusted R-squared: -0.006601
## F-statistic: 0.7526 on 53 and 1946 DF, p-value: 0.9061
model <- lm(Microsoft ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = Microsoft ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.024 -2.876 1.537 1.941 3.829
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.711577 0.902519 8.545 <2e-16
## ageBefore before 1990 -0.163051 0.202398 -0.806 0.4206
## genderfemale 0.171907 0.140992 1.219 0.2229
## raceBlack 0.115274 0.225114 0.512 0.6087
## raceHispanic -0.335556 0.196317 -1.709 0.0876
## raceAsian 0.145757 0.367908 0.396 0.6920
## raceNative American 0.883397 0.755440 1.169 0.2424
## raceMixed -0.562698 0.452213 -1.244 0.2135
## raceOther 0.852360 0.584444 1.458 0.1449
## raceMiddle Eastern 0.713284 1.233572 0.578 0.5632
## educ.L -0.055402 0.235900 -0.235 0.8143
## educ.Q 0.358860 0.210138 1.708 0.0878
## educ.C -0.219396 0.172777 -1.270 0.2043
## educ^4 0.245421 0.154680 1.587 0.1128
## educ^5 0.018307 0.166287 0.110 0.9123
## maritalstatusSeparated -0.424826 0.547097 -0.777 0.4375
## maritalstatusDivorced 0.148262 0.223708 0.663 0.5076
## maritalstatusWidowed -0.054372 0.327229 -0.166 0.8680
## maritalstatusNever married 0.006389 0.170306 0.038 0.9701
## maritalstatusDomestic/civil partnership 0.301820 0.329774 0.915 0.3602
## employPart-time 0.357271 0.233033 1.533 0.1254
## employTemporarily laid off -0.311113 0.757675 -0.411 0.6814
## employUnemployed 0.291790 0.260018 1.122 0.2619
## employRetired -0.055914 0.196361 -0.285 0.7759
## employPermanently disabled 0.184316 0.269377 0.684 0.4939
## employHomemaker -0.575679 0.273501 -2.105 0.0354
## employStudent -0.010761 0.314975 -0.034 0.9727
## employOther -0.246047 0.560776 -0.439 0.6609
## pidRepublican -0.041544 0.222006 -0.187 0.8516
## pidIndependent -0.151555 0.177981 -0.852 0.3946
## pidOther -0.162218 0.350086 -0.463 0.6432
## pidNot sure -0.411407 0.302799 -1.359 0.1744
## ideology.L 1.862905 1.678868 1.110 0.2673
## ideology.Q 1.463643 1.612446 0.908 0.3641
## ideology.C 1.055168 1.210360 0.872 0.3834
## ideology^4 0.432273 0.738120 0.586 0.5582
## ideology^5 0.277214 0.370692 0.748 0.4547
## ideology^6 -0.094851 0.187912 -0.505 0.6138
## religpewRoman Catholic -0.009898 0.193652 -0.051 0.9592
## religpewMormon 0.504176 0.516895 0.975 0.3295
## religpewEastern or Greek Orthodox 0.240930 1.048443 0.230 0.8183
## religpewJewish -0.914672 0.389464 -2.349 0.0189
## religpewMuslim -1.175654 0.632295 -1.859 0.0631
## religpewBuddhist 0.365033 0.728748 0.501 0.6165
## religpewHindu -0.419634 0.965510 -0.435 0.6639
## religpewAtheist 0.272592 0.290985 0.937 0.3490
## religpewAgnostic 0.096831 0.281979 0.343 0.7313
## religpewNothing in particular -0.051166 0.189116 -0.271 0.7868
## religpewSomething else 0.073368 0.311981 0.235 0.8141
## CollegelevelprogrammingNo -0.023476 0.276504 -0.085 0.9323
## engineeringcompsciencedegreeNo 0.034579 0.295731 0.117 0.9069
## gradcompengNo 0.383372 0.372188 1.030 0.3031
## programexperienceNo -0.217690 0.252734 -0.861 0.3892
## noeducationalcompexpNo -0.176117 0.296481 -0.594 0.5526
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic .
## raceAsian
## raceNative American
## raceMixed
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q .
## educ.C
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker *
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish *
## religpewMuslim .
## religpewBuddhist
## religpewHindu
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.881 on 1946 degrees of freedom
## Multiple R-squared: 0.02647, Adjusted R-squared: -4.609e-05
## F-statistic: 0.9983 on 53 and 1946 DF, p-value: 0.4787
model <- lm(Amazon ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = Amazon ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -7.556 -3.024 1.701 2.196 4.203
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.994255 0.924287 7.567 5.85e-14
## ageBefore before 1990 -0.269753 0.207280 -1.301 0.1933
## genderfemale 0.037552 0.144392 0.260 0.7948
## raceBlack -0.131573 0.230543 -0.571 0.5683
## raceHispanic -0.126456 0.201051 -0.629 0.5294
## raceAsian 0.190685 0.376781 0.506 0.6129
## raceNative American -0.339482 0.773661 -0.439 0.6609
## raceMixed 0.617662 0.463120 1.334 0.1825
## raceOther -0.482765 0.598540 -0.807 0.4200
## raceMiddle Eastern -2.090979 1.263325 -1.655 0.0981
## educ.L -0.045413 0.241590 -0.188 0.8509
## educ.Q 0.322786 0.215207 1.500 0.1338
## educ.C -0.400972 0.176944 -2.266 0.0236
## educ^4 0.232161 0.158411 1.466 0.1429
## educ^5 0.097838 0.170298 0.575 0.5657
## maritalstatusSeparated 0.290946 0.560293 0.519 0.6036
## maritalstatusDivorced 0.208697 0.229103 0.911 0.3624
## maritalstatusWidowed 0.320435 0.335121 0.956 0.3391
## maritalstatusNever married 0.069359 0.174414 0.398 0.6909
## maritalstatusDomestic/civil partnership -0.052549 0.337728 -0.156 0.8764
## employPart-time -0.213625 0.238654 -0.895 0.3708
## employTemporarily laid off 1.017577 0.775949 1.311 0.1899
## employUnemployed 0.542968 0.266290 2.039 0.0416
## employRetired 0.008036 0.201097 0.040 0.9681
## employPermanently disabled 0.323777 0.275875 1.174 0.2407
## employHomemaker -0.130313 0.280098 -0.465 0.6418
## employStudent 0.121323 0.322572 0.376 0.7069
## employOther -0.054084 0.574301 -0.094 0.9250
## pidRepublican 0.156373 0.227361 0.688 0.4917
## pidIndependent 0.055814 0.182274 0.306 0.7595
## pidOther 0.136343 0.358529 0.380 0.7038
## pidNot sure 0.455232 0.310102 1.468 0.1423
## ideology.L -1.506824 1.719360 -0.876 0.3809
## ideology.Q -1.288307 1.651336 -0.780 0.4354
## ideology.C -1.084393 1.239553 -0.875 0.3818
## ideology^4 -0.744295 0.755922 -0.985 0.3249
## ideology^5 -0.220266 0.379632 -0.580 0.5618
## ideology^6 0.027760 0.192444 0.144 0.8853
## religpewRoman Catholic -0.097761 0.198322 -0.493 0.6221
## religpewMormon 0.679433 0.529362 1.283 0.1995
## religpewEastern or Greek Orthodox 1.255167 1.073730 1.169 0.2426
## religpewJewish 0.339327 0.398858 0.851 0.3950
## religpewMuslim 0.110796 0.647546 0.171 0.8642
## religpewBuddhist -0.489827 0.746324 -0.656 0.5117
## religpewHindu -0.399571 0.988797 -0.404 0.6862
## religpewAtheist -0.447363 0.298003 -1.501 0.1335
## religpewAgnostic -0.150135 0.288781 -0.520 0.6032
## religpewNothing in particular 0.223283 0.193677 1.153 0.2491
## religpewSomething else 0.320550 0.319505 1.003 0.3159
## CollegelevelprogrammingNo 0.250964 0.283173 0.886 0.3756
## engineeringcompsciencedegreeNo 0.101892 0.302863 0.336 0.7366
## gradcompengNo -0.125651 0.381165 -0.330 0.7417
## programexperienceNo -0.473793 0.258830 -1.831 0.0673
## noeducationalcompexpNo -0.170783 0.303632 -0.562 0.5739
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed
## raceOther
## raceMiddle Eastern .
## educ.L
## educ.Q
## educ.C *
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed *
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim
## religpewBuddhist
## religpewHindu
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo .
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.95 on 1946 degrees of freedom
## Multiple R-squared: 0.02986, Adjusted R-squared: 0.003433
## F-statistic: 1.13 on 53 and 1946 DF, p-value: 0.2438
model <- lm(NGO ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = NGO ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.805 -2.671 1.538 2.049 4.628
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.562255 0.900027 8.402 < 2e-16
## ageBefore before 1990 0.018821 0.201839 0.093 0.925717
## genderfemale 0.125668 0.140602 0.894 0.371548
## raceBlack 0.082180 0.224492 0.366 0.714351
## raceHispanic -0.071924 0.195774 -0.367 0.713376
## raceAsian -0.428494 0.366892 -1.168 0.242989
## raceNative American -0.336634 0.753354 -0.447 0.655035
## raceMixed -0.309290 0.450965 -0.686 0.492894
## raceOther -0.841572 0.582830 -1.444 0.148916
## raceMiddle Eastern -1.745316 1.230166 -1.419 0.156128
## educ.L -0.305382 0.235249 -1.298 0.194398
## educ.Q -0.209567 0.209558 -1.000 0.317415
## educ.C -0.413641 0.172300 -2.401 0.016457
## educ^4 0.039339 0.154253 0.255 0.798729
## educ^5 0.037954 0.165828 0.229 0.818991
## maritalstatusSeparated 0.075808 0.545587 0.139 0.889506
## maritalstatusDivorced 0.008449 0.223090 0.038 0.969793
## maritalstatusWidowed -0.581186 0.326325 -1.781 0.075068
## maritalstatusNever married -0.499174 0.169836 -2.939 0.003330
## maritalstatusDomestic/civil partnership -1.097663 0.328863 -3.338 0.000861
## employPart-time -0.065312 0.232390 -0.281 0.778705
## employTemporarily laid off 0.500350 0.755582 0.662 0.507919
## employUnemployed -0.470848 0.259300 -1.816 0.069549
## employRetired -0.083318 0.195819 -0.425 0.670531
## employPermanently disabled -0.300221 0.268634 -1.118 0.263881
## employHomemaker 0.068591 0.272746 0.251 0.801466
## employStudent -0.152988 0.314105 -0.487 0.626270
## employOther -0.561978 0.559227 -1.005 0.315061
## pidRepublican -0.081025 0.221393 -0.366 0.714421
## pidIndependent 0.246183 0.177490 1.387 0.165593
## pidOther 0.648307 0.349119 1.857 0.063465
## pidNot sure 0.388502 0.301962 1.287 0.198391
## ideology.L 0.371693 1.674232 0.222 0.824331
## ideology.Q 0.676627 1.607993 0.421 0.673955
## ideology.C 0.442651 1.207018 0.367 0.713859
## ideology^4 0.433746 0.736081 0.589 0.555753
## ideology^5 0.156643 0.369668 0.424 0.671802
## ideology^6 -0.039928 0.187393 -0.213 0.831294
## religpewRoman Catholic -0.202943 0.193117 -1.051 0.293443
## religpewMormon 0.279804 0.515468 0.543 0.587319
## religpewEastern or Greek Orthodox -0.005763 1.045547 -0.006 0.995602
## religpewJewish -0.067380 0.388389 -0.173 0.862288
## religpewMuslim -0.287803 0.630549 -0.456 0.648130
## religpewBuddhist -0.260902 0.726735 -0.359 0.719630
## religpewHindu -2.104224 0.962844 -2.185 0.028977
## religpewAtheist -0.461695 0.290181 -1.591 0.111760
## religpewAgnostic -0.406141 0.281201 -1.444 0.148813
## religpewNothing in particular 0.282273 0.188593 1.497 0.134626
## religpewSomething else 0.414982 0.311119 1.334 0.182414
## CollegelevelprogrammingNo 0.067073 0.275740 0.243 0.807840
## engineeringcompsciencedegreeNo 0.398390 0.294914 1.351 0.176894
## gradcompengNo -0.436543 0.371160 -1.176 0.239676
## programexperienceNo -0.180115 0.252036 -0.715 0.474919
## noeducationalcompexpNo 0.049221 0.295662 0.166 0.867798
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q
## educ.C *
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed .
## maritalstatusNever married **
## maritalstatusDomestic/civil partnership ***
## employPart-time
## employTemporarily laid off
## employUnemployed .
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther
## pidRepublican
## pidIndependent
## pidOther .
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim
## religpewBuddhist
## religpewHindu *
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.873 on 1946 degrees of freedom
## Multiple R-squared: 0.03732, Adjusted R-squared: 0.0111
## F-statistic: 1.423 on 53 and 1946 DF, p-value: 0.0251
model <- lm(Uni ~ age + gender + race + educ + maritalstatus + employ + pid + ideology + religpew + Collegelevelprogramming + engineeringcompsciencedegree + gradcompeng + programexperience + noeducationalcompexp, data = AI_Importance)
summary(model)
##
## Call:
## lm(formula = Uni ~ age + gender + race + educ + maritalstatus +
## employ + pid + ideology + religpew + Collegelevelprogramming +
## engineeringcompsciencedegree + gradcompeng + programexperience +
## noeducationalcompexp, data = AI_Importance)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.636 -3.197 1.800 2.165 4.685
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 7.060252 0.951424 7.421 1.73e-13
## ageBefore before 1990 -0.098148 0.213365 -0.460 0.6456
## genderfemale -0.048731 0.148632 -0.328 0.7431
## raceBlack -0.202805 0.237312 -0.855 0.3929
## raceHispanic -0.096465 0.206954 -0.466 0.6412
## raceAsian 0.312496 0.387843 0.806 0.4205
## raceNative American -0.888122 0.796375 -1.115 0.2649
## raceMixed 0.213701 0.476717 0.448 0.6540
## raceOther -0.373559 0.616113 -0.606 0.5444
## raceMiddle Eastern 1.298750 1.300415 0.999 0.3181
## educ.L -0.227905 0.248683 -0.916 0.3595
## educ.Q 0.011626 0.221525 0.052 0.9581
## educ.C -0.046025 0.182139 -0.253 0.8005
## educ^4 0.071768 0.163062 0.440 0.6599
## educ^5 0.098936 0.175298 0.564 0.5726
## maritalstatusSeparated -0.010469 0.576743 -0.018 0.9855
## maritalstatusDivorced -0.006776 0.235830 -0.029 0.9771
## maritalstatusWidowed 0.237149 0.344960 0.687 0.4919
## maritalstatusNever married -0.039179 0.179535 -0.218 0.8273
## maritalstatusDomestic/civil partnership 0.374352 0.347643 1.077 0.2817
## employPart-time 0.027335 0.245660 0.111 0.9114
## employTemporarily laid off 1.021724 0.798731 1.279 0.2010
## employUnemployed -0.127916 0.274108 -0.467 0.6408
## employRetired 0.193504 0.207001 0.935 0.3500
## employPermanently disabled 0.250867 0.283974 0.883 0.3771
## employHomemaker -0.266355 0.288321 -0.924 0.3557
## employStudent -0.286030 0.332043 -0.861 0.3891
## employOther -0.328061 0.591163 -0.555 0.5790
## pidRepublican -0.483877 0.234036 -2.068 0.0388
## pidIndependent -0.111544 0.187626 -0.595 0.5522
## pidOther -0.221509 0.369056 -0.600 0.5484
## pidNot sure -0.127965 0.319206 -0.401 0.6885
## ideology.L 1.723728 1.769840 0.974 0.3302
## ideology.Q 1.183956 1.699819 0.697 0.4862
## ideology.C 0.846248 1.275946 0.663 0.5073
## ideology^4 0.730360 0.778116 0.939 0.3480
## ideology^5 0.474161 0.390778 1.213 0.2251
## ideology^6 0.184790 0.198094 0.933 0.3510
## religpewRoman Catholic 0.052490 0.204145 0.257 0.7971
## religpewMormon 0.755374 0.544904 1.386 0.1658
## religpewEastern or Greek Orthodox -1.102195 1.105254 -0.997 0.3188
## religpewJewish 0.613024 0.410568 1.493 0.1356
## religpewMuslim 0.329756 0.666557 0.495 0.6209
## religpewBuddhist 0.176801 0.768236 0.230 0.8180
## religpewHindu 0.729146 1.017828 0.716 0.4738
## religpewAtheist -0.189312 0.306752 -0.617 0.5372
## religpewAgnostic 0.262494 0.297259 0.883 0.3773
## religpewNothing in particular -0.012961 0.199363 -0.065 0.9482
## religpewSomething else 0.341807 0.328886 1.039 0.2988
## CollegelevelprogrammingNo -0.081790 0.291487 -0.281 0.7791
## engineeringcompsciencedegreeNo 0.570548 0.311755 1.830 0.0674
## gradcompengNo 0.059285 0.392356 0.151 0.8799
## programexperienceNo -0.061454 0.266429 -0.231 0.8176
## noeducationalcompexpNo 0.017983 0.312546 0.058 0.9541
##
## (Intercept) ***
## ageBefore before 1990
## genderfemale
## raceBlack
## raceHispanic
## raceAsian
## raceNative American
## raceMixed
## raceOther
## raceMiddle Eastern
## educ.L
## educ.Q
## educ.C
## educ^4
## educ^5
## maritalstatusSeparated
## maritalstatusDivorced
## maritalstatusWidowed
## maritalstatusNever married
## maritalstatusDomestic/civil partnership
## employPart-time
## employTemporarily laid off
## employUnemployed
## employRetired
## employPermanently disabled
## employHomemaker
## employStudent
## employOther
## pidRepublican *
## pidIndependent
## pidOther
## pidNot sure
## ideology.L
## ideology.Q
## ideology.C
## ideology^4
## ideology^5
## ideology^6
## religpewRoman Catholic
## religpewMormon
## religpewEastern or Greek Orthodox
## religpewJewish
## religpewMuslim
## religpewBuddhist
## religpewHindu
## religpewAtheist
## religpewAgnostic
## religpewNothing in particular
## religpewSomething else
## CollegelevelprogrammingNo
## engineeringcompsciencedegreeNo .
## gradcompengNo
## programexperienceNo
## noeducationalcompexpNo
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.037 on 1946 degrees of freedom
## Multiple R-squared: 0.01763, Adjusted R-squared: -0.009125
## F-statistic: 0.659 on 53 and 1946 DF, p-value: 0.9727
As above, graphs can be generated to visualise the relationships between these variables. I have selected two of some signifiance below. We see that there seems to be a higher probability that women have higher confidence in Google to develop AI in the best interests of the public - as are Protestants and Roman Catholics.
barGoogle<-ggplot(data=AI_Importance, aes (x=gender, y=Google)) + geom_bar(stat="identity", fill="aquamarine4")+theme_minimal()
barGoogle + ggtitle("Confidence in Google to develop AI in best interests of public by gender") +
xlab("Gender") + ylab("Confidence")
barGooglereligion<-ggplot(data=AI_Importance, aes (x=religpew, y=Google)) + geom_bar(stat="identity", fill="lightseagreen")+theme_minimal()
barGooglereligion + ggtitle("Confidence in Google to develop AI in best interests of public by religion") +
xlab("Gender") + ylab("Confidence")
Discussion, Limitations, and Conclusion
These results have significant implications on for policymakers. We can formally reject the null hypotheses and conclude that those who are male, liberal, older, non-religious, and have a programming background are more likely to support the development of AI. In identifying which demographics are less likely to support AI development, public officials can design programs that target buy-in across these individuals. If we know for example that women are less likely than men to support the development of AI, or that those without a college degree are less likely to support AI development, it may be useful to consider why, and create programs that allow these individuals to increase their literacy of AI technologies and capabilities, in order to engender support among constituencies.
This simple multiple regression does not allow us to to make causal arguments about support for or antagonism against AI development in the US. Overall, we see that support for AI development is greater among those who are wealthy, educated, male, or have background in CS experience.
In conclusion, many scholars discuss how open AI and increased data literacy may lead to an increased trust in AI and its development. This simple study shows support for this assertion as those with prior CS experience for example is more likely to support AI development. However, it also has implications in looking at populations to target to increase literacy in AI technologies - taking on a targeted approach to support uptake among female populations for example, may be of particular use to policymakers. However, this study should be replicated across contexts - e.g. France, Germany, and India, for example, or even across geographic locations across the US, to understand if these findings hold across populations.
Finally, this dataset comprises a vast array of attitudes towards AI beyond demographic variables e.g. understanding of key terms and public opinion of AI governance with regard to data privacy among others. A fuller project would run analyses across all these data. It would also be very interesting to contrast this data pre-COVID on support for AI with data collected after the onset of COVID, to investigate if there is an effect of the increased use and discussion of AI-driven technologies in contract tracing in mediating support for AI among different demographic groups across the public. In sum, I am limited by the data available and look forward to collecting my own and running the appropriate analyses therein, instead of working to fit analyses and a research question into existing datasets.
References
Coppock, Alexander, Emily Ekins, David Kirby, and others. 2018. “The Long-Lasting Effects of Newspaper Op-Eds on Public Opinion.” Quarterly Journal of Political Science 13 (1): 59–87.
Dafoe, Allan. 2018. “AI Governance: A Research Agenda.” Governance of AI Program, Future of Humanity Institute, University of Oxford. https://perma.cc/L35A-7W2X.
Eurobarometer. 2017. “Special Eurobarometer 460: Attitudes Towards the Impact of Digitisation and Automation on Daily Life.” Eurobarometer. https://perma.cc/9FRT-ADST.
Funk, Cary, and Lee Rainie. 2015. “Public and Scientists’ Views on Science and Society.” Survey report. Pew Research Center. https://perma.cc/9XSJ-8AJA.
Grace, Katja, John Salvatier, Allan Dafoe, Baobao Zhang, and Owain Evans. 2018. “Viewpoint: When Will AI Exceed Human Performance? Evidence from AI Experts.” Journal of Artificial Intelligence Research 62: 729–54. https://doi.org/ 10.1613/jair.1.11222.
Graham, Edward. 2018. “Views on Automation’s U.S. Workforce Impact Highlight Demographic Divide.” Morning Consult. https://perma.cc/544D-WRUM.
Ipsos and Reuters. 2018. “Facebook Poll 3.23.18.” Survey report. Ipsos; Reuters. https://perma.cc/QGH5-S3KE.
Ipsos MORI. 2018. “Public Views of Machine Learning: Findings from Public Research and Engagement Conducted on Behalf of the Royal Society.” Survey report. The Royal Society. https://perma.cc/79FE-TEHH.
McCorduck, Pamela. 2004. Machines Who Think: A Personal Inquiry into the History and Prospects of Artificial Intelligence. New York: A K Peters/CRC Press.
Morning Consult. 2017. “National Tracking Poll 170401.” Survey report. Morning Consult. https ://perma.cc/ TBJ9-CB5K.
Müller, Vincent C, and Nick Bostrom. 2014. “Future Progress in Artificial Intelligence: A Poll Among Experts.” AI Matters 1 (1). ACM: 9–11. https://perma.cc/ZY22-E5FF.
Nagelhout, Gera E, Marc C Willemsen, Mary E Thompson, Geoffrey T Fong, Bas Van den Putte, and Hein de Vries. 2010. “Is Web Interviewing a Good Alternative to Telephone Interviewing? Findings from the International Tobacco Control (Itc) Netherlands Survey.” BMC Public Health 10 (1): 351.
Nedelkoska, Ljubica, and Glenda Quintini. 2018. “Automation, Skills Use and Training.” Working Papers No. 202. Organisation for Economic Co-operation; Development. https://doi.org/10.1787/2e2f4eea-en.
Noble, S.U., 2018. Algorithms of oppression: How search engines reinforce racism. NYU Press.
Walsh, Toby. 2018. “Expert and Non-Expert Opinion About Technological Unemployment.” International Journal of Automation and Computing 15 (5): 637–42. https://doi.org/10.1007/s11633-018-1127-x.
Zhang, B. and Dafoe, A., 2019. Artificial intelligence: American attitudes and trends. Available at SSRN 3312874.
