mypubs.bib

%% This BibTeX bibliography file was created using BibDesk.
%% https://bibdesk.sourceforge.io/

%% Created for SM at 2020-06-01 22:04:23 -0400 


%% Saved with string encoding Unicode (UTF-8) 


@article{aggarwal20232,
	  Abstract = {We present the results of a massive, \$8.9 million campaign-wide field experiment, conducted among 2 million moderate and low-information ``persuadable'' voters in five battleground states during the 2020 US Presidential election. Treatment group subjects were exposed to an eight-month-long advertising program delivered via social media, designed to persuade people to vote against Donald Trump and for Joe Biden. On average, the program neither increased or decreased turnout. We find evidence of differential turnout effects by modeled level of Trump support: the campaign increased voting among Biden leaners by 0.4 percentage points (SE: 0.2pp) and decreased voting among Trump leaners by 0.3 percentage points (SE: 0.3pp), for a difference-in-CATES of 0.7 points ($t(1035571) = -2.09$, $p = 0.036$, $\widehat{DIC} = 0.7$ points, 95\% CI = $[-0.014, -0.00]$). An important but exploratory finding is that the strongest differential effects appear in \textit{early voting} data, which may inform future work on early campaigning in a post-COVID electoral environment. Our results indicate that differential mobilization effects of even large digital advertising campaigns in presidential elections are likely to be modest.},
	  title={A 2 million-person, campaign-wide field experiment shows how digital advertising affects voter turnout},
	  author={Aggarwal, Minali and Allen, Jennifer and Coppock, Alexander and Frankowski, Dan and Messing, Solomon and Zhang, Kelly and Barnes, James and Beasley, Andrew and Hantman, Harry and Zheng, Sylvan},
	  journal={Nature Human Behaviour},
	  pages={1--10},
	  year={2023},
	  publisher={Nature Publishing Group}
}

@article{messing2016bias,
	Abstract = {There is strong evidence linking skin complexion to negative stereotypes and adverse real-world outcomes. We extend these findings to political ad campaigns, in which skin complexion can be easily manipulated in ways that are difficult to detect. Devising a method to measure how dark a candidate appears in an image, this paper examines how complexion varied with ad content during the 2008 presidential election campaign (study 1). Findings show that darker images were more frequent in negative ads---especially those linking Obama to crime---which aired more frequently as Election Day approached. We then conduct an experiment to document how these darker images can activate stereotypes, and show that a subtle darkness manipulation is sufficient to activate the most negative stereotypes about Blacks---even when the candidate is a famous counter-stereotypical exemplar---Barack Obama (study 2). Further evidence of an evaluative penalty for darker skin comes from an observational study measuring affective responses to depictions of Obama with varying skin complexion, presented via the Affect Misattribution Procedure in the 2008 American National Election Study (study 3). This study demonstrates that darker images are used in a way that complements ad content, and shows that doing so can negatively affect how individuals evaluate candidates and think about politics.},
	Author = {Messing, Solomon and Jabon, Maria and Plaut, Ethan},
	Date-Modified = {2020-06-01 22:04:18 -0400},
	Journal = {Public opinion quarterly},
	Number = {1},
	Pages = {44--65},
	Publisher = {Oxford University Press US},
	Title = {Bias in the flesh: Skin complexion and stereotype consistency in political campaigns},
	Volume = {80},
	Year = {2016}}

@article{iyengar2008explicit,
	Abstract = {Skin color is an explicit racial cue. Although there is strong evidence linking darker skin complexion to the activation of racial stereotypes and adverse societal outcomes, little is known about the extent to which this effect is in play during political campaigns. If white voters make use of this skin complexion cue, we would expect exposure to darker images of a minority candidate to result in a ``dark-skin penalty'' at the ballot box. We investigate the impact of skin complexion on support for Barack Obama at two different stages of the 2008 campaign: Study 1 occurred during the primary campaign and Study 2 during the closing stages of the general election. Our findings suggest that when citizens are still learning about a minority candidate's personal background, subtle changes in skin complexion can have an effect on evaluations of that candidate and that citizens with higher levels of implicit racial bias are less likely to prefer a darker-skinned minority candidate.

},
	Author = {Iyengar, Shanto and Hahn, Kyu S and Messing, Solomon and Bailenson, Jeremy N},
	Date-Modified = {2020-06-01 22:01:02 -0400},
	Journal = {The Case of Skin Complexion in the},
	Title = {Do explicit racial cues influence candidate preference? The case of skin complexion in the 2008 campaign},
	Year = {2008}}

@techreport{messing2011measuring,
	Author = {Messing, Solomon},
	Institution = {Stanford University},
	Title = {Measuring issue salience: Using supervised machine learning to generate data from free responses to the ``most important problem'' question},
	Year = {2011}}

@incollection{lew2011course,
	Author = {Lew, Letitia and Nguyen, Truc and Messing, Solomon and Westwood, Sean},
	Booktitle = {CHI'11 Extended Abstracts on Human Factors in Computing Systems},
	Pages = {419--428},
	Title = {Of course I wouldn't do that in real life: advancing the arguments for increasing realism in HCI experiments},
	Year = {2011}}

@article{aalberg2012deserving,
	Author = {Aalberg, Toril and Iyengar, Shanto and Messing, Solomon},
	Journal = {Scandinavian Political Studies},
	Number = {2},
	Pages = {97--116},
	Publisher = {Blackwell Publishing Ltd Oxford, UK},
	Title = {Who is a `deserving'immigrant? An experimental study of Norwegian attitudes},
	Volume = {35},
	Year = {2012}}

@article{gawande2012differentiation,
	Author = {Gawande, Rakhee S and Khurana, Aman and Messing, Solomon and Zhang, Dong and Casta{\~n}eda, Rosalinda T and Goldsby, Robert E and Hawkins, Randall A and Daldrup-Link, Heike E},
	Journal = {Radiology},
	Number = {2},
	Pages = {613--622},
	Publisher = {Radiological Society of North America},
	Title = {Differentiation of Normal Thymus from Anterior Mediastinal Lymphoma and Lymphoma Recurrence at Pediatric PET/CT},
	Volume = {262},
	Year = {2012}}

@article{khurana2012intravenous,
	Author = {Khurana, Aman and Nejadnik, Hossein and Gawande, Rakhee and Lin, Guiting and Lee, Sungmin and Messing, Solomon and Castaneda, Rosalinda and Derugin, Nikita and Pisani, Laura and Lue, Tom F and others},
	Journal = {Radiology},
	Number = {3},
	Pages = {803--811},
	Publisher = {Radiological Society of North America, Inc.},
	Title = {Intravenous ferumoxytol allows noninvasive MR imaging monitoring of macrophage migration into stem cell transplants},
	Volume = {264},
	Year = {2012}}

@article{messing2014selective,
	Abstract = {Much of the literature on polarization and selective exposure presumes that the internet exacerbates the fragmentation of the media and the citizenry. Yet this ignores how the widespread use of social media changes news consumption. Social media provide readers a choice of stories from different sources that come recommended from politically heterogeneous individuals, in a context that emphasizes social value over partisan affiliation. Building on existing models of news selectivity to emphasize information utility, we hypothesize that social media's distinctive feature, social endorsements, trigger several decision heuristics that suggest utility. In two experiments, we demonstrate that stronger social endorsements increase the probability that people select content and that their presence reduces partisan selective exposure to levels indistinguishable from chance.},
	Author = {Messing, Solomon and Westwood, Sean J},
	Date-Modified = {2020-06-01 21:53:26 -0400},
	Journal = {Communication research},
	Number = {8},
	Pages = {1042--1063},
	Publisher = {Sage Publications Sage CA: Los Angeles, CA},
	Title = {Selective exposure in the age of social media: Endorsements trump partisan source affiliation when selecting news online},
	Volume = {41},
	Year = {2014}}

@article{shi2013evaluation,
	Abstract = {Tumor-associated macrophages (TAM) maintain a chronic inflammation in cancers, which is associated with tumor aggressiveness and poor prognosis. The purpose of this study was to: (1) evaluate the pharmacokinetics and tolerability of the novel ultrasmall superparamagnetic iron oxide nanoparticle (USPIO) compound GEH121333; (2) assess whether GEH121333 can serve as a MR imaging biomarker for TAM; and (3) compare tumor MR enhancement profiles between GEH121333 and ferumoxytol. Blood half-lives of GEH121333 and ferumoxytol were measured by relaxometry (n = 4 each). Tolerance was assessed in healthy rats injected with high dose GEH121333, vehicle or saline (n = 4 each). Animals were monitored for 7 days regarding body weight, complete blood counts and serum chemistry, followed by histological evaluation of visceral organs. MR imaging was performed on mice harboring MMTV-PyMT-derived breast adenocarcinomas using a 7 T scanner before and up to 72 h post-injection (p.i.) of GEH121333 (n = 10) or ferumoxytol (n = 9). Tumor R1, R2* relaxation rates were compared between different experimental groups and time points, using a linear mixed effects model with a random effect for each animal. MR data were correlated with histopathology. GEH121333 showed a longer circulation half-life than ferumoxytol. Intravenous GEH121333 did not produce significant adverse effects in rats. All tumors demonstrated significant enhancement on T1, T2 and T2*-weighted images at 1, 24, 48 and 72 h p.i. GEH121333 generated stronger tumor T2* enhancement than ferumoxytol. Histological analysis verified intracellular compartmentalization of GEH121333 by TAM at 24, 48 and 72 h p.i. MR imaging with GEH121333 nanoparticles represents a novel biomarker for TAM assessment. This new USPIO MR contrast agent provides a longer blood half-life and better TAM enhancement compared with the iron supplement ferumoxytol. },
	Author = {Shi, Qiaoyun and Pisani, Laura J and Lee, Yauk K and Messing, Solomon and Ansari, Celina and Bhaumik, Srabani and Lowery, Lisa and Lee, Brian D and Meyer, Dan E and Daldrup-Link, Heike E},
	Date-Modified = {2020-06-01 21:54:36 -0400},
	Journal = {Contrast media \& molecular imaging},
	Number = {3},
	Pages = {281--288},
	Title = {Evaluation of the novel USPIO GEH121333 for MR imaging of cancer immune responses},
	Volume = {8},
	Year = {2013}}

@article{khurana2013ferumoxytol,
	Abstract = {Aim
To develop a clinically applicable MRI technique for tracking stem cells in matrix-associated stem-cell implants, using the US FDA-approved iron supplement ferumoxytol.

Materials & methods
Ferumoxytol-labeling of adipose-derived stem cells (ADSCs) was optimized in vitro. A total of 11 rats with osteochondral defects of both femurs were implanted with ferumoxytol- or ferumoxides-labeled or unlabeled ADSCs, and underwent MRI up to 4 weeks post matrix-associated stem-cell implant. The signal-to-noise ratio of different matrix-associated stem-cell implant was compared with t-tests and correlated with histopathology.

Results
An incubation concentration of 500 µg iron/ml ferumoxytol and 10 µg/ml protamine sulfate led to significant cellular iron uptake, T2 signal effects and unimpaired ADSC viability. In vivo, ferumoxytol-and ferumoxides-labeled ADSCs demonstrated significantly lower signal-to-noise ratio values compared with unlabeled controls (p < 0.01). Histopathology confirmed engraftment of labeled ADSCs, with slow dilution of the iron label over time.

Conclusion
Ferumoxytol can be used for in vivo tracking of stem cells with MRI.

},
	Author = {Khurana, Aman and Nejadnik, Hossein and Chapelin, Fanny and Lenkov, Olga and Gawande, Rakhee and Lee, Sungmin and Gupta, Sandeep N and Aflakian, Nooshin and Derugin, Nikita and Messing, Solomon and others},
	Date-Modified = {2020-06-01 21:55:21 -0400},
	Journal = {Nanomedicine},
	Number = {12},
	Pages = {1969--1983},
	Publisher = {Future Medicine Ltd London, UK},
	Title = {Ferumoxytol: a new, clinically applicable label for stem-cell tracking in arthritic joints with MRI},
	Volume = {8},
	Year = {2013}}

@article{gawande2013role,
	Abstract = {We retrospectively analyzed DWI scans of 68 consecutive children with 39 benign and 34 malignant abdominal masses. To calculate the apparent diffusion coefficient (ADC) maps and ADC values, we used 1.5-T sequences at TR/TE/b-value of 5,250--7,500/54--64/b = 0, 500 and 3-T sequences at 3,500--4,000/66--73/b = 0, 500, 800. ADC values were compared between benign and malignant and between data derived at 1.5 tesla (T) and at 3 tesla magnetic field strength, using the Mann-Whitney-Wilcoxon test, ANOVA and a receiver operating curve (ROC) analysis.

Results
There was no significant difference in ADC values obtained at 1.5 T and 3 T (P = 0.962). Mean ADC values (× 10−3 mm2/s) were 1.07 for solid malignant tumors, 1.6 for solid benign tumors, 2.9 for necrotic portions of malignant tumors and 3.1 for cystic benign lesions. The differences between malignant and benign solid tumors were statistically significant (P = 0.000025). ROC analysis revealed an optimal cut-off ADC value for differentiating malignant and benign solid tumors as 1.29 with excellent inter-observer reliability (alpha score 0.88).

Conclusion
DWI scans and ADC values can contribute to distinguishing between benign and malignant pediatric abdominal tumors.

},
	Author = {Gawande, Rakhee S and Gonzalez, Gabriel and Messing, Solomon and Khurana, Aman and Daldrup-Link, Heike E},
	Date-Modified = {2020-06-01 21:57:13 -0400},
	Journal = {Pediatric radiology},
	Number = {7},
	Pages = {836--845},
	Publisher = {Springer-Verlag},
	Title = {Role of diffusion-weighted imaging in differentiating benign and malignant pediatric abdominal tumors},
	Volume = {43},
	Year = {2013}}

@article{gawande2012fdg,
	Author = {Gawande, Rakhee S and Behr, Spencer and Messing, Solomon and Goldsby, Robert E and Hawkins, Randall A and Daldrup-Link, Heike E},
	Journal = {RADIOLOGY},
	Number = {3},
	Pages = {919--920},
	Publisher = {RADIOLOGICAL SOC NORTH AMERICA 820 JORIE BLVD, OAK BROOK, IL 60523 USA},
	Title = {FDG PET/CT for the Evaluation of Normal Thymus, Lymphoma Recurrence, and Mediastinal Lymphoma in Pediatric Patients Response},
	Volume = {264},
	Year = {2012}}

@article{grimmer2017estimating,
	Abstract = {Randomized experiments are increasingly used to study political phenomena because they can credibly estimate the average effect of a treatment on a population of interest. But political scientists are often interested in how effects vary across subpopulations---heterogeneous treatment effects---and how differences in the content of the treatment affects responses---the response to heterogeneous treatments. Several new methods have been introduced to estimate heterogeneous effects, but it is difficult to know if a method will perform well for a particular data set. Rather than using only one method, we show how an ensemble of methods---weighted averages of estimates from individual models increasingly used in machine learning---accurately measure heterogeneous effects. Building on a large literature on ensemble methods, we show how the weighting of methods can contribute to accurate estimation of heterogeneous treatment effects and demonstrate how pooling models lead to superior performance to individual methods across diverse problems. We apply the ensemble method to two experiments, illuminating how the ensemble method for heterogeneous treatment effects facilitates exploratory analysis of treatment effects.

},
	Author = {Grimmer, Justin and Messing, Solomon and Westwood, Sean J},
	Date-Modified = {2020-06-01 21:52:41 -0400},
	Journal = {Political Analysis},
	Number = {4},
	Pages = {413--434},
	Publisher = {Cambridge University Press},
	Title = {Estimating heterogeneous treatment effects and the effects of heterogeneous treatments with ensemble methods},
	Volume = {25},
	Year = {2017}}

@article{khurana2013iron,
	Abstract = {In vivo--labeled MSCs demonstrated significantly higher
ferumoxytol uptake compared with ex vivo--labeled cells.
With electron microscopy, iron oxide nanoparticles were
localized in secondary lysosomes. In vivo--labeled cells
demonstrated significant T2 shortening effects in vitro
and in vivo when they were compared with unlabeled control cells (T2 in vivo, 15.4 vs 24.4 msec; P < .05) and could be tracked in osteochondral defects for 4 weeks.
Histologic examination confirmed the presence of iron in
labeled transplants and defect remodeling.

Conclusion: Intravenous ferumoxytol can be used to effectively label
MSCs in vivo and can be used for tracking of stem cell
transplants with MR imaging. This method eliminates
risks of contamination and biologic alteration of MSCs
associated with ex vivo--labeling procedures.},
	Author = {Khurana, Aman and Chapelin, Fanny and Beck, Graham and Lenkov, Olga D and Donig, Jessica and Nejadnik, Hossein and Messing, Solomon and Derugin, Nikita and Chan, Ray Chun-Fai and Gaur, Amitabh and others},
	Date-Modified = {2020-06-01 21:56:12 -0400},
	Journal = {Radiology},
	Number = {1},
	Pages = {186--197},
	Publisher = {Radiological Society of North America, Inc.},
	Title = {Iron administration before stem cell harvest enables MR imaging tracking after transplantation},
	Volume = {269},
	Year = {2013}}

@article{iyengar2013attitudes,
	Abstract = {This paper demonstrates that citizens in seven advanced industrialized democracies generally oppose more open immigration policies, but stand ready to admit individual immigrants. Using an experimental design, we demonstrate the applicability of the ``person-positivity bias'' to immigration and investigate the effects of economic and cultural ``deservingness'' on evaluations of individual immigrants. Our results show that immigrants from professional backgrounds elicit higher levels of support than unskilled workers. The bias against unskilled workers is enlarged among immigrants accompanied by families. In comparison with occupational status and the number of family dependents, the target immigrant's cultural attributes---as measured by Middle Eastern nationality and Afrocentric appearance---prove relatively inconsequential as criteria for evaluating immigrants.},
	Author = {Iyengar, Shanto and Jackman, Simon and Messing, Solomon and Valentino, Nicholas and Aalberg, Toril and Duch, Raymond and Hahn, Kyu S and Soroka, Stuart and Harell, Allison and Kobayashi, Tetsuro},
	Journal = {Public opinion quarterly},
	Number = {3},
	Pages = {641--665},
	Publisher = {Oxford University Press UK},
	Title = {Do attitudes about immigration predict willingness to admit individual immigrants? A cross-national test of the person-positivity bias},
	Volume = {77},
	Year = {2013}}

@article{bond2015quantifying,
	Abstrac = {We demonstrate that social media data represent a useful resource for testing models of legislative and individual-level political behavior and attitudes. First, we develop a model to estimate the ideology of politicians and their supporters using social media data on individual citizens' endorsements of political figures. Our measure allows us to place politicians and more than 6 million citizens who are active in social media on the same metric. We validate the ideological estimates that result from the scaling process by showing they correlate highly with existing measures of ideology from Congress, and with individual-level self-reported political views. Finally, we use these measures to study the relationship between ideology and age, social relationships and ideology, and the relationship between friend ideology and turnout.},
	Author = {Bond, Robert and Messing, Solomon},
	Journal = {American Political Science Review},
	Number = {1},
	Pages = {62--78},
	Publisher = {Cambridge University Press},
	Title = {Quantifying social media's political space: Estimating ideology from publicly revealed preferences on Facebook},
	Volume = {109},
	Year = {2015}}

@article{bakshy2015exposure,
	Abstract = {Exposure to news, opinion, and civic information increasingly occurs through social media. How do these online networks influence exposure to perspectives that cut across ideological lines? Using deidentified data, we examined how 10.1 million U.S. Facebook users interact with socially shared news. We directly measured ideological homophily in friend networks and examined the extent to which heterogeneous friends could potentially expose individuals to cross-cutting content. We then quantified the extent to which individuals encounter comparatively more or less diverse content while interacting via Facebook's algorithmically ranked News Feed and further studied users' choices to click through to ideologically discordant content. Compared with algorithmic ranking, individuals' choices played a stronger role in limiting exposure to cross-cutting content.},
	Author = {Bakshy, Eytan and Messing, Solomon and Adamic, Lada A},
	Journal = {Science},
	Number = {6239},
	Pages = {1130--1132},
	Publisher = {American Association for the Advancement of Science},
	Title = {Exposure to ideologically diverse news and opinion on Facebook},
	Volume = {348},
	Year = {2015}}

@article{solomon2017partisan,
	Abstract = {Partisan criticism generates most engagement in social media; most
liberal and conservative legislators, party leadership more likely to
`go negative'},
	Author = {Solomon Messing, Patrick van Kessel, Adam G Hughes, Nick Judd, Rachel Blum, Brian Broderick},
	Journal = {Pew Research Center},
	Title = {Partisan Conflict and Congressional Outreach},
	Year = {2017}}

@phdthesis{messing2013friends,
	Author = {Messing, Solomon},
	School = {Stanford University},
	Title = {Friends that Matter: How Social Transmission of Elite Discourse Shapes Political Knowledge, Attitudes, and Behavior},
	Year = {2013}}

@article{westwood2019projecting,
	Abstract = {Recent years have seen a dramatic change in horserace coverage of elections in the
U.S.---shifting focus from late-breaking poll numbers to sophisticated meta-analytic
forecasts that emphasize candidates' chance of victory. Could this shift in the political information environment affect election outcomes? We use experiments to show
that forecasting increases certainty about an election's outcome, confuses many, and
decreases turnout. Furthermore, we show that election forecasting has become prominent in the media, particularly in outlets with liberal audiences, and show that such
coverage tends to more strongly affect the candidate who is ahead---raising questions
about whether they contributed to Trump's victory over Clinton in 2016. We bring
empirical evidence to this question, using ANES data to show that Democrats and
Independents expressed unusual confidence in a decisive 2016 election outcome---and
that the same measure of confidence is associated with lower reported turnout.},
	Author = {Westwood, Sean and Messing, Solomon and Lelkes, Yphtach},
	Journal = {Journal of Politics},
	Title = {Projecting confidence: How the probabilistic horse race confuses and demobilizes the public},
	Website = {https://www.journals.uchicago.edu/doi/abs/10.1086/708682},
	Year = {2019}}

@article{chapelin2019tumor,
	Abstract = {Purpose: While imaging matrix-associated stem cell transplants aimed for cartilage repair in a rodent arthritis model, we noticed that some transplants formed locally destructive tumors. The purpose of this study was to determine the cause for this tumor formation in order to avoid this complication for future transplants.

Procedures: Adipose-derived stem cells (ADSC) isolated from subcutaneous adipose tissue were implanted into 24 osteochondral defects of the distal femur in ten athymic rats and two immunocompetent control rats. All transplants underwent serial magnetic resonance imaging (MRI) up to 6 weeks post-transplantation to monitor joint defect repair. Nine transplants showed an increasing size over time that caused local bone destruction (group 1), while 11 transplants in athymic rats (group 2) and 4 transplants in immunocompetent rats did not. We compared the ADSC implant size and growth rate on MR images, macroscopic features, histopathologic features, surface markers, and karyotypes of these presumed neoplastic transplants with non-neoplastic ADSC transplants.

Results: Implants in group 1 showed a significantly increased two-dimensional area at week 2 (p = 0.0092), 4 (p = 0.003), and 6 (p = 0.0205) compared to week 0, as determined by MRI. Histopathological correlations confirmed neoplastic features in group 1 with significantly increased size, cellularity, mitoses, and cytological atypia compared to group 2. Six transplants in group 1 were identified as malignant chondrosarcomas and three transplants as fibromyxoid sarcomas. Transplants in group 2 and immunocompetent controls exhibited normal cartilage features. Both groups showed a normal ADSC phenotype; however, neoplastic ADSC demonstrated a mixed population of diploid and tetraploid cells without genetic imbalance.

Conclusions: ADSC transplants can form tumors in vivo. Preventive actions to avoid in vivo tumor formations may include karyotyping of culture-expanded ADSC before transplantation. In addition, serial imaging of ADSC transplants in vivo may enable early detection of abnormally proliferating cell transplants.

},
	Author = {Chapelin, Fanny and Khurana, Aman and Moneeb, Mohammad and Hazard, Florette K Gray and Chan, Chun Fai Ray and Nejadnik, Hossein and Gratzinger, Dita and Messing, Solomon and Erdmann, Jason and Gaur, Amitabh and others},
	Date-Modified = {2020-06-01 21:52:16 -0400},
	Journal = {Molecular Imaging and Biology},
	Number = {1},
	Pages = {95--104},
	Publisher = {Springer International Publishing},
	Title = {Tumor formation of adult stem cell transplants in rodent arthritic joints},
	Volume = {21},
	Year = {2019}}

@article{kifer2020guidelines,
	Abstract = {Differential privacy is an information theoretic constraint on algorithms and code. It provides quantification of privacy leakage and formal privacy guarantees that are currently considered the gold standard in privacy protections. In this paper we provide an initial set of "best practices" for developing differentially private platforms, techniques for unit testing that are specific to differential privacy, guidelines for checking if differential privacy is being applied correctly in an application, and recommendations for parameter settings. The genesis of this paper was an initiative by Facebook and Social Science One to provide social science researchers with programmatic access to a URL-shares dataset. In order to maximize the utility of the data for research while protecting privacy, researchers should access the data through an interactive platform that supports differential privacy.
The intention of this paper is to provide guidelines and recommendations that can generally be re-used in a wide variety of systems. For this reason, no specific platforms will be named, except for systems whose details and theory appear in academic papers.},
	Author = {Kifer, Daniel and Messing, Solomon and Roth, Aaron and Thakurta, Abhradeep and Zhang, Danfeng},
	Date-Modified = {2020-06-01 21:51:35 -0400},
	Journal = {arXiv preprint arXiv:2002.04049},
	Title = {Guidelines for Implementing and Auditing Differentially Private Systems},
	Year = {2020}}

@manual{messing2020,
	Abstract = {One of the largest social science data sets ever constructed, meant to facilitate research on misinformation from across the web, shared and spread on Facebook. It contains exposure data describing external links that billions of users saw and read while using the site. The data set goes beyond URL-level data, breaking down exposure and interactions by month, country, age, gender, and in the U.S., political page affinity (see Barbera et al 2015).



},
	Author = {Messing, Solomon and DeGregorio, Christina and Hillenbrand, Bennett and King, Gary and Mahanti, Saurav and Nayak, Chaya and Persily, Nate},
	Date-Modified = {2020-06-01 21:47:59 -0400},
	Organization = {Social Science One},
	Title = {Facebook Privacy-Protected Full URLs Data Set},
	Website = {https://solomonmg.github.io/assets/pdf/Facebook_DP_URLs_Dataset.pdf},
	Year = {2020}}

@article{grimmer_how_2012,
	Abstract = {Particularistic spending, a large literature argues, builds support for incumbents. This literature equates money spent in the district with the credit constituents allocate. Yet, constituents lack the necessary information and motivation to allocate credit in this way. We use extensive observational and experimental evidence to show how legislators' credit claiming messages---and not just money spent in the district---affect how constituents allocate credit. Legislators use credit claiming messages to influence the expenditures they receive credit for and to affect how closely they are associated with spending in the district. Constituents are responsive to credit claiming messages---they build more support than other nonpartisan messages. But contrary to expectations from other studies, constituents are more responsive to the total number of messages sent rather than the amount claimed. Our results have broad implications for political representation, the personal vote, and the study of U.S. Congressional elections.

},
	Author = {Grimmer, Justin and Messing, Solomon and Westwood, Sean J.},
	Date-Modified = {2020-06-01 21:59:15 -0400},
	Doi = {10.1017/S0003055412000457},
	Journal = {American Political Science Review},
	Number = {04},
	Pages = {703--719},
	Shorttitle = {How {Words} and {Money} {Cultivate} a {Personal} {Vote}},
	Title = {How {Words} and {Money} {Cultivate} a {Personal} {Vote}: {The} {Effect} of {Legislator} {Credit} {Claiming} on {Constituent} {Credit} {Allocation}},
	Url = {http://bit.ly/XOuZi1},
	Volume = {106},
	Year = {2012},
	Bdsk-Url-1 = {http://bit.ly/XOuZi1},
	Bdsk-Url-2 = {https://doi.org/10.1017/S0003055412000457}}

@book{GriMesWes14,
	Abstract = {Constituents often fail to hold their representatives accountable for federal spending decisions―even though those very choices have a pervasive influence on American life. Why does this happen? Breaking new ground in the study of representation, The Impression of Influence demonstrates how legislators skillfully inform constituents with strategic communication and how this facilitates or undermines accountability. Using a massive collection of Congressional texts and innovative experiments and methods, the book shows how legislators create an impression of influence through credit claiming messages.

Anticipating constituents' reactions, legislators claim credit for programs that elicit a positive response, making constituents believe their legislator is effectively representing their district. This spurs legislators to create and defend projects popular with their constituents. Yet legislators claim credit for much more―they announce projects long before they begin, deceptively imply they deserve credit for expenditures they had little role in securing, and boast about minuscule projects. Unfortunately, legislators get away with seeking credit broadly because constituents evaluate the actions that are reported, rather than the size of the expenditures.

The Impression of Influence raises critical questions about how citizens hold their political representatives accountable and when deception is allowable in a democracy.},
	Author = {Grimmer, Justin and Westwood, Sean J. and Messing, Solomon},
	Date-Modified = {2020-06-01 21:53:57 -0400},
	Publisher = {Princeton University Press},
	Title = {The Impression of Influence: Legislator Communication, Representation, and Democratic Accountability},
	Year = {2014}}

@article{Messing:2012ts,
	Author = {Messing, Solomon and Marlow, Cameron and Bakshy, Eytan},
	Date-Modified = {2020-06-01 21:59:33 -0400},
	Title = {{The 2012 Election Day Through the Facebook Lens}},
	Url = {http://on.fb.me/Vmkovb},
	Year = {2012},
	Bdsk-Url-1 = {http://on.fb.me/Vmkovb}}