references.bib

@article{kellingFindingSignalNoise2018,
  langid = {english},
  title = {Finding the Signal in the Noise of {{Citizen Science Observations}}},
  url = {https://www.biorxiv.org/content/10.1101/326314v1},
  doi = {10.1101/326314},
  abstract = {{$<$}p{$>$}While many observations of species are being collected by citizen science projects worldwide, it can be challenging to identify projects collecting data that effectively monitor biodiversity. Over the past several years the allure of taking a big data approach has provided the opportunity to gather massive quantities of observations via the Internet, too often with insufficient information to describe how the observations were made. Information about species populations, where and when they occur and how many of them are there (i.e., the signal) can be lost because insufficient information is gathered to account for the inherent biases in data collection (i.e., the noise). Here we suggest that citizen science projects that have succeeded in motivating large numbers of participants, must consider factors that influence the ecological process that affect species populations as well as the observation process that determines how observations are made. Those citizen science projects that collect sufficient contextual information describing the observation process can be used to generate increasingly accurate information about the distribution and abundance of organisms. We illustrate this using eBird as a case study, describing how this citizen science platform is able to collect vital contextual information on the observation process while maintaining a broad global constituency of participants. We highlight how eBird provides information with which to generate biodiversity indicators, specifically distribution, abundance, and habitat associations, across the entire annual cycle, even for populations of long distance migratory birds, a highly challenging taxon.{$<$}/p{$>$}},
  journaltitle = {bioRxiv},
  urldate = {2019-02-05},
  date = {2018-05-18},
  pages = {326314},
  author = {Kelling, Steve and Johnston, Alison and Fink, Daniel and Ruiz-Gutierrez, Viviana and Bonney, Rick and Bonn, Aletta and Fernandez, Miguel and Hochachka, Wesley and Julliard, Romain and Kraemer, Roland and Guralnick, Robert},
  file = {/Users/mes335/Zotero/storage/UUUFDMZ5/Kelling et al. - 2018 - Finding the signal in the noise of Citizen Science.pdf;/Users/mes335/Zotero/storage/RE8D4XIS/99654.html}
}

@article{sullivanEBirdEnterpriseIntegrated2014,
  title = {The {{eBird}} Enterprise: {{An}} Integrated Approach to Development and Application of Citizen Science},
  volume = {169},
  issn = {0006-3207},
  url = {http://www.sciencedirect.com/science/article/pii/S0006320713003820},
  doi = {10.1016/j.biocon.2013.11.003},
  shorttitle = {The {{eBird}} Enterprise},
  abstract = {Citizen-science projects engage volunteers to gather or process data to address scientific questions. But citizen-science projects vary in their ability to contribute usefully for science, conservation, or public policy. eBird has evolved from a basic citizen-science project into a collective enterprise, taking a novel approach to citizen science by developing cooperative partnerships among experts in a wide range of fields: population and distributions, conservation biologists, quantitative ecologists, statisticians, computer scientists, GIS and informatics specialists, application developers, and data administrators. The goal is to increase data quantity through participant recruitment and engagement, but also to quantify and control for data quality issues such as observer variability, imperfect detection of species, and both spatial and temporal bias in data collection. Advances at the interface among ecology, statistics, and computer science allow us to create new species distribution models that provide accurate estimates across broad spatial and temporal scales with extremely detailed resolution. eBird data are openly available and used by a broad spectrum of students, teachers, scientists, NGOs, government agencies, land managers, and policy makers. Feedback from this broad data use community helps identify development priorities. As a result, eBird has become a major source of biodiversity data, increasing our knowledge of the dynamics of species distributions, and having a direct impact on the conservation of birds and their habitats.},
  journaltitle = {Biological Conservation},
  shortjournal = {Biological Conservation},
  urldate = {2019-02-05},
  date = {2014-01-01},
  pages = {31-40},
  keywords = {Citizen-science,eBird},
  author = {Sullivan, Brian L. and Aycrigg, Jocelyn L. and Barry, Jessie H. and Bonney, Rick E. and Bruns, Nicholas and Cooper, Caren B. and Damoulas, Theo and Dhondt, André A. and Dietterich, Tom and Farnsworth, Andrew and Fink, Daniel and Fitzpatrick, John W. and Fredericks, Thomas and Gerbracht, Jeff and Gomes, Carla and Hochachka, Wesley M. and Iliff, Marshall J. and Lagoze, Carl and La Sorte, Frank A. and Merrifield, Matthew and Morris, Will and Phillips, Tina B. and Reynolds, Mark and Rodewald, Amanda D. and Rosenberg, Kenneth V. and Trautmann, Nancy M. and Wiggins, Andrea and Winkler, David W. and Wong, Weng-Keen and Wood, Christopher L. and Yu, Jun and Kelling, Steve},
  file = {/Users/mes335/Zotero/storage/MB5NPJZ4/S0006320713003820.html}
}

@article{lasorteOpportunitiesChallengesBig2018,
  title = {Opportunities and Challenges for Big Data Ornithology},
  volume = {120},
  number = {2},
  journaltitle = {The Condor},
  date = {2018},
  pages = {414--426},
  author = {La Sorte, Frank A. and Lepczyk, Christopher A. and Burnett, Jessica L. and Hurlbert, Allen H. and Tingley, Morgan W. and Zuckerberg, Benjamin},
  file = {/Users/mes335/Zotero/storage/92JNXMVI/aosjournals.html}
}

@article{greenwoodCitizensScienceBird2007,
  langid = {english},
  title = {Citizens, Science and Bird Conservation},
  volume = {148},
  issn = {1439-0361},
  url = {https://doi.org/10.1007/s10336-007-0239-9},
  doi = {10.1007/s10336-007-0239-9},
  abstract = {Collaborative research by networks of amateurs has had a major role in ornithology and conservation science and will continue to do so. It has been important in establishing the facts of migration, systematically recording distribution, providing insights into habitat requirements and recording variation in numbers, productivity and survival, thus allowing detailed demographic analyses. The availability of these data has allowed conservation work to be focussed on priority species, habitats and sites and enabled refined monitoring and research programmes aimed at providing the understanding necessary for sound conservation management and for evidence-based government policy. The success of such work depends on the independence of the science from those advocating particular policies in order to ensure that the science is unbiased. Wetland birds are surveyed in much of the world. Most countries also have a ringing scheme. Other forms of collaborative ornithology are strong in North America, Australia and Australasia, more patchily distributed in Asia (but with strong growth in some countries) and even patchier in Africa and South America. Such work is most successful where there is a strong partnership between the amateurs and the professional, based on their complementary roles. The participation of large numbers of volunteers not only enables work to be done that would otherwise be impossible but also facilitates democratic participation in the decisions made by society and builds social capital. The recruitment to and subsequent retention of people in the research networks are important skills. Surveys must be organized in ways that take into account the motives of the participants. It is useful to assess the skills of potential participants and, rather than rejecting those thought not to have adequate skills, to provide training. Special attention needs to be paid to ensuring that instructions are clear, that methods are standardized and that data are gathered in a form that is easily processed. Providing for the continuity of long-term projects is essential. There are advantages to having just one organization running most of the work in each country. Various sorts of organizations are possible: societies governed by their (amateur) members but employing professional staff to organize the work seem to be a particularly successful model. Independence from government and from conservation organizations is desirable.},
  number = {1},
  journaltitle = {Journal of Ornithology},
  shortjournal = {J Ornithol},
  urldate = {2019-02-06},
  date = {2007-12-01},
  pages = {77-124},
  keywords = {Amateur,Census,Collaboration,Conservation,Distribution,Habitat,Monitoring,Survival},
  author = {Greenwood, Jeremy J. D.}
}

@article{tullochBehaviouralEcologyApproach2012,
  title = {A Behavioural Ecology Approach to Understand Volunteer Surveying for Citizen Science Datasets},
  volume = {112},
  number = {4},
  journaltitle = {Emu-Austral Ornithology},
  date = {2012},
  pages = {313--325},
  author = {Tulloch, Ayesha IT and Szabo, Judit K.},
  file = {/Users/mes335/Zotero/storage/D4XGJ6X2/Tulloch and Szabo - 2012 - A behavioural ecology approach to understand volun.pdf;/Users/mes335/Zotero/storage/KTGL37AS/MU12009.html}
}

@article{luckAlleviatingSpatialConflict2004,
  langid = {english},
  title = {Alleviating Spatial Conflict between People and Biodiversity},
  volume = {101},
  issn = {0027-8424, 1091-6490},
  url = {https://www.pnas.org/content/101/1/182},
  doi = {10.1073/pnas.2237148100},
  abstract = {Human settlements are expanding in species-rich regions and pose a serious threat to biodiversity conservation. We quantify the degree to which this threat manifests itself in two contrasting continents, Australia and North America, and suggest how it can be substantially alleviated. Human population density has a strong positive correlation with species richness in Australia for birds, mammals, amphibians, and butterflies (but not reptiles) and in North America for all five taxa. Nevertheless, conservation investments could secure locations that harbor almost all species while greatly reducing overlap with densely populated regions. We compared two conservation-planning scenarios that each aimed to represent all species at least once in a minimum set of sampling sites. The first scenario assigned equal cost to each site (ignoring differences in human population density); the second assigned a cost proportional to the site's human population density. Under the equal-cost scenario, 13–40\% of selected sites occurred where population density values were highest (in the top decile). However, this overlap was reduced to as low as 0\%, and in almost all cases to {$<$}10\%, under the population-cost scenario, when sites of high population density were avoided where possible. Moreover, this reduction of overlap was achieved with only small increases in the total amount of area requiring protection. As densely populated regions continue to expand rapidly and drive up land values, the strategic conservation investments of the kind highlighted in our analysis are best made now.},
  number = {1},
  journaltitle = {Proceedings of the National Academy of Sciences},
  shortjournal = {PNAS},
  urldate = {2019-02-06},
  date = {2004-01-06},
  pages = {182-186},
  author = {Luck, Gary W. and Ricketts, Taylor H. and Daily, Gretchen C. and Imhoff, Marc},
  file = {/Users/mes335/Zotero/storage/29L6LUSM/Luck et al. - 2004 - Alleviating spatial conflict between people and bi.pdf;/Users/mes335/Zotero/storage/EWV8XJIJ/182.html},
  eprinttype = {pmid},
  eprint = {14681554}
}

@article{kadmonEffectRoadsideBias2004,
  title = {Effect of Roadside Bias on the Accuracy of Predictive Maps Produced by Bioclimatic Models},
  volume = {14},
  number = {2},
  journaltitle = {Ecological Applications},
  date = {2004},
  pages = {401--413},
  author = {Kadmon, Ronen and Farber, Oren and Danin, Avinoam},
  file = {/Users/mes335/Zotero/storage/42PEWUWY/02-5364.html}
}

@article{prendergastCorrectingVariationRecording1993,
  title = {Correcting for Variation in Recording Effort in Analyses of Diversity Hotspots},
  journaltitle = {Biodiversity Letters},
  date = {1993},
  pages = {39--53},
  author = {Prendergast, J. R. and Wood, S. N. and Lawton, J. H. and Eversham, B. C.},
  file = {/Users/mes335/Zotero/storage/5T93WN6Y/Prendergast et al. - 1993 - Correcting for variation in recording effort in an.pdf;/Users/mes335/Zotero/storage/DHUS3D6T/2999649.html}
}

@article{courterWeekendBiasCitizen2013,
  langid = {english},
  title = {Weekend Bias in {{Citizen Science}} Data Reporting: Implications for Phenology Studies},
  volume = {57},
  issn = {1432-1254},
  url = {https://doi.org/10.1007/s00484-012-0598-7},
  doi = {10.1007/s00484-012-0598-7},
  shorttitle = {Weekend Bias in {{Citizen Science}} Data Reporting},
  abstract = {Studies of bird phenology can help elucidate the effects of climate change on wildlife species but observations over broad spatial scales are difficult without a network of observers. Recently, networks of citizen volunteers have begun to report first arrival dates for many migratory species. Potential benefits are substantial (e.g., understanding ecological processes at broad spatial and temporal scales) if known biases of citizen data reporting are identified and addressed. One potential source of bias in bird phenology studies is the tendency for more “first” migratory arrivals to be reported on weekends than on weekdays. We investigated weekend bias in data reporting for five common bird species in North America (Baltimore Oriole, Icterus galbula; Barn Swallow, Hirundo rustica; Chimney Swift, Chaetura pelagica; Purple Martin, Progne subis; and Ruby-throated Hummingbird, Archilochus colubris), and assessed whether this bias affected mean arrival dates reported using data from historical (1880–1969; N = 25,555) and recent (1997–2010; N = 63,149) Citizen Science databases. We found a greater percentage of first arrivals reported on weekends and small but significant differences in mean arrival dates (approximately 0.5 days) for four of five species. Comparing time periods, this weekend bias decreased from 33.7 \% and five species in the historical time period to 32 \% and three species in the recent, perhaps related to changes in human activity patterns. Our results indicate that weekend bias in citizen data reporting is decreasing over time in North America and including a ‘day of week’ term in models examining changes in phenology could help make conclusions more robust.},
  number = {5},
  journaltitle = {International Journal of Biometeorology},
  shortjournal = {Int J Biometeorol},
  urldate = {2019-02-06},
  date = {2013-09-01},
  pages = {715-720},
  keywords = {Bird phenology,Climate change,Detectability,First arrival dates,Spring migration,Volunteers},
  author = {Courter, Jason R. and Johnson, Ron J. and Stuyck, Claire M. and Lang, Brian A. and Kaiser, Evan W.}
}

@article{kellingCanObservationSkills2015,
  title = {Can Observation Skills of Citizen Scientists Be Estimated Using Species Accumulation Curves?},
  volume = {10},
  number = {10},
  journaltitle = {PloS one},
  date = {2015},
  pages = {e0139600},
  author = {Kelling, Steve and Johnston, Alison and Hochachka, Wesley M. and Iliff, Marshall and Fink, Daniel and Gerbracht, Jeff and Lagoze, Carl and La Sorte, Frank A. and Moore, Travis and Wiggins, Andrea},
  file = {/Users/mes335/Zotero/storage/4EZGDF2V/article.html}
}

@article{johnstonEstimatesObserverExpertise2018,
  title = {Estimates of Observer Expertise Improve Species Distributions from Citizen Science Data},
  volume = {9},
  number = {1},
  journaltitle = {Methods in Ecology and Evolution},
  date = {2018},
  pages = {88--97},
  author = {Johnston, Alison and Fink, Daniel and Hochachka, Wesley M. and Kelling, Steve},
  file = {/Users/mes335/Zotero/storage/4NX5YAFM/Johnston et al. - 2018 - Estimates of observer expertise improve species di.pdf;/Users/mes335/Zotero/storage/DUNG8M8W/2041-210X.html}
}

@article{robinsonUsingCitizenScience2018,
  title = {Using Citizen Science Data in Integrated Population Models to Inform Conservation Decision-Making},
  journaltitle = {bioRxiv},
  date = {2018},
  pages = {293464},
  author = {Robinson, Orin J. and Ruiz-Gutierrez, Viviana and Fink, Daniel and Meese, Robert J. and Holyoak, Marcel and Cooch, Evan G.},
  file = {/Users/mes335/Zotero/storage/D7M55CYV/Robinson et al. - 2018 - Using citizen science data in integrated populatio.pdf;/Users/mes335/Zotero/storage/UMFC7IU4/90379.html}
}

@article{johnstonSpeciesTraitsExplain2014,
  title = {Species Traits Explain Variation in Detectability of {{UK}} Birds},
  volume = {61},
  number = {3},
  journaltitle = {Bird Study},
  date = {2014},
  pages = {340--350},
  author = {Johnston, Alison and Newson, Stuart E. and Risely, Kate and Musgrove, Andy J. and Massimino, Dario and Baillie, Stephen R. and Pearce-Higgins, James W.},
  file = {/Users/mes335/Zotero/storage/7HJGTJSG/00063657.2014.html;/Users/mes335/Zotero/storage/7NWIDLY8/00063657.2014.html}
}

@article{ellisEffectsWeatherTime2018,
  title = {Effects of Weather, Time of Day, and Survey Effort on Estimates of Species Richness in Temperate Woodlands},
  volume = {118},
  number = {2},
  journaltitle = {Emu-Austral Ornithology},
  date = {2018},
  pages = {183--192},
  author = {Ellis, Murray V. and Taylor, Jennifer E.},
  file = {/Users/mes335/Zotero/storage/LWUFQQRY/01584197.2017.html;/Users/mes335/Zotero/storage/MY6QZ8B7/01584197.2017.html}
}

@article{oliveiraObservationDiurnalSoaring2018,
  title = {Observation of {{Diurnal Soaring Raptors In Northeastern Brazil Depends On Weather Conditions}} and {{Time}} of {{Day}}},
  volume = {52},
  number = {1},
  journaltitle = {Journal of Raptor Research},
  date = {2018},
  pages = {56--65},
  author = {Oliveira, Camilo Viana and Olmos, Fabio and dos Santos-Filho, Manoel and Bernardo, Christine Steiner São},
  options = {useprefix=true},
  file = {/Users/mes335/Zotero/storage/Q7K2YX87/Oliveira et al. - 2018 - Observation of Diurnal Soaring Raptors In Northeas.pdf;/Users/mes335/Zotero/storage/ZJBJIB63/JRR-16-102.1.html}
}

@data{friedlMCD12Q1MODISTerra2015,
  title = {{{MCD12Q1 MODIS}}/{{Terra}}+{{Aqua Land Cover Type Yearly L3 Global}} 500m {{SIN Grid V006}}},
  url = {https://lpdaac.usgs.gov/node/1260},
  doi = {10.5067/MODIS/MCD12Q1.006},
  publisher = {{NASA EOSDIS Land Processes DAAC}},
  urldate = {2019-02-07},
  date = {2015},
  author = {Friedl, Mark and Sulla-Menashe, Damien}
}

@article{guillera-arroitaMySpeciesDistribution2015,
  title = {Is My Species Distribution Model Fit for Purpose? {{Matching}} Data and Models to Applications},
  volume = {24},
  shorttitle = {Is My Species Distribution Model Fit for Purpose?},
  number = {3},
  journaltitle = {Global Ecology and Biogeography},
  date = {2015},
  pages = {276--292},
  author = {Guillera-Arroita, Gurutzeta and Lahoz-Monfort, José J. and Elith, Jane and Gordon, Ascelin and Kujala, Heini and Lentini, Pia E. and McCarthy, Michael A. and Tingley, Reid and Wintle, Brendan A.},
  file = {/Users/mes335/Zotero/storage/7XL928WK/geb.html;/Users/mes335/Zotero/storage/EH6J9DTM/geb.html}
}

@article{sahrHexagonalDiscreteGlobal2011,
  title = {Hexagonal Discrete Global Grid Systems for Geospatial Computing},
  volume = {22},
  journaltitle = {Archiwum Fotogrametrii, Kartografii i Teledetekcji},
  date = {2011},
  pages = {363--376},
  author = {Sahr, Kevin},
  file = {/Users/mes335/Zotero/storage/MJMGFPCM/Sahr - 2011 - Hexagonal discrete global grid systems for geospat.pdf;/Users/mes335/Zotero/storage/XB9JWUDI/bwmeta1.element.html}
}

@article{phillipsPOCPlotsCalibrating2010,
  title = {{{POC}} Plots: Calibrating Species Distribution Models with Presence-Only Data},
  volume = {91},
  shorttitle = {{{POC}} Plots},
  number = {8},
  journaltitle = {Ecology},
  date = {2010},
  pages = {2476--2484},
  author = {Phillips, Steven J. and Elith, Jane},
  file = {/Users/mes335/Zotero/storage/TIUJFIWY/Phillips and Elith - 2010 - POC plots calibrating species distribution models.pdf;/Users/mes335/Zotero/storage/49R4RC7W/09-0760.html}
}

@article{bzdokPointsSignificanceStatistics2018,
  langid = {english},
  title = {Points of {{Significance}}: {{Statistics}} versus Machine Learning},
  volume = {15},
  issn = {1548-7105},
  url = {https://www.nature.com/articles/nmeth.4642},
  doi = {10.1038/nmeth.4642},
  shorttitle = {Points of {{Significance}}},
  abstract = {Statistics draws population inferences from a sample, and machine learning finds generalizable predictive patterns.},
  journaltitle = {Nature Methods},
  urldate = {2019-02-18},
  date = {2018-04-03},
  pages = {233-234},
  author = {Bzdok, Danilo and Altman, Naomi and Krzywinski, Martin},
  file = {/Users/mes335/Zotero/storage/MWUE7IJ8/Bzdok et al. - 2018 - Points of Significance Statistics versus machine .pdf;/Users/mes335/Zotero/storage/55MN6F96/nmeth.html}
}

@book{mackenzieOccupancyEstimationModeling2017,
  title = {Occupancy Estimation and Modeling: Inferring Patterns and Dynamics of Species Occurrence},
  shorttitle = {Occupancy Estimation and Modeling},
  publisher = {{Elsevier}},
  date = {2017},
  author = {MacKenzie, Darryl I. and Nichols, James D. and Royle, J. Andrew and Pollock, Kenneth H. and Bailey, Larissa and Hines, James E.},
  file = {/Users/mes335/Zotero/storage/AJ2TA93J/books.html}
}

@article{mackenzieAssessingFitSiteoccupancy2004,
  title = {Assessing the Fit of Site-Occupancy Models},
  volume = {9},
  number = {3},
  journaltitle = {Journal of Agricultural, Biological, and Environmental Statistics},
  date = {2004},
  pages = {300--318},
  author = {MacKenzie, Darryl I. and Bailey, Larissa L.},
  file = {/Users/mes335/Zotero/storage/PFM9KXT5/MacKenzie and Bailey - 2004 - Assessing the fit of site-occupancy models.pdf;/Users/mes335/Zotero/storage/YZXVEIB7/108571104X3361.html}
}

@article{collaborationEstimatingReproducibilityPsychological2015,
  langid = {english},
  title = {Estimating the Reproducibility of Psychological Science},
  volume = {349},
  issn = {0036-8075, 1095-9203},
  url = {http://science.sciencemag.org/content/349/6251/aac4716},
  doi = {10.1126/science.aac4716},
  abstract = {Empirically analyzing empirical evidence
One of the central goals in any scientific endeavor is to understand causality. Experiments that seek to demonstrate a cause/effect relation most often manipulate the postulated causal factor. Aarts et al. describe the replication of 100 experiments reported in papers published in 2008 in three high-ranking psychology journals. Assessing whether the replication and the original experiment yielded the same result according to several criteria, they find that about one-third to one-half of the original findings were also observed in the replication study.
Science, this issue 10.1126/science.aac4716
Structured Abstract
INTRODUCTIONReproducibility is a defining feature of science, but the extent to which it characterizes current research is unknown. Scientific claims should not gain credence because of the status or authority of their originator but by the replicability of their supporting evidence. Even research of exemplary quality may have irreproducible empirical findings because of random or systematic error.
RATIONALEThere is concern about the rate and predictors of reproducibility, but limited evidence. Potentially problematic practices include selective reporting, selective analysis, and insufficient specification of the conditions necessary or sufficient to obtain the results. Direct replication is the attempt to recreate the conditions believed sufficient for obtaining a previously observed finding and is the means of establishing reproducibility of a finding with new data. We conducted a large-scale, collaborative effort to obtain an initial estimate of the reproducibility of psychological science.
RESULTSWe conducted replications of 100 experimental and correlational studies published in three psychology journals using high-powered designs and original materials when available. There is no single standard for evaluating replication success. Here, we evaluated reproducibility using significance and P values, effect sizes, subjective assessments of replication teams, and meta-analysis of effect sizes. The mean effect size (r) of the replication effects (Mr = 0.197, SD = 0.257) was half the magnitude of the mean effect size of the original effects (Mr = 0.403, SD = 0.188), representing a substantial decline. Ninety-seven percent of original studies had significant results (P {$<$} .05). Thirty-six percent of replications had significant results; 47\% of original effect sizes were in the 95\% confidence interval of the replication effect size; 39\% of effects were subjectively rated to have replicated the original result; and if no bias in original results is assumed, combining original and replication results left 68\% with statistically significant effects. Correlational tests suggest that replication success was better predicted by the strength of original evidence than by characteristics of the original and replication teams.
CONCLUSIONNo single indicator sufficiently describes replication success, and the five indicators examined here are not the only ways to evaluate reproducibility. Nonetheless, collectively these results offer a clear conclusion: A large portion of replications produced weaker evidence for the original findings despite using materials provided by the original authors, review in advance for methodological fidelity, and high statistical power to detect the original effect sizes. Moreover, correlational evidence is consistent with the conclusion that variation in the strength of initial evidence (such as original P value) was more predictive of replication success than variation in the characteristics of the teams conducting the research (such as experience and expertise). The latter factors certainly can influence replication success, but they did not appear to do so here.Reproducibility is not well understood because the incentives for individual scientists prioritize novelty over replication. Innovation is the engine of discovery and is vital for a productive, effective scientific enterprise. However, innovative ideas become old news fast. Journal reviewers and editors may dismiss a new test of a published idea as unoriginal. The claim that “we already know this” belies the uncertainty of scientific evidence. Innovation points out paths that are possible; replication points out paths that are likely; progress relies on both. Replication can increase certainty when findings are reproduced and promote innovation when they are not. This project provides accumulating evidence for many findings in psychological research and suggests that there is still more work to do to verify whether we know what we think we know. {$<$}img class="fragment-image" aria-describedby="F1-caption" src="http://science.sciencemag.org/content/sci/349/6251/aac4716/F1.medium.gif"/{$>$} Download high-res image Open in new tab Download Powerpoint Original study effect size versus replication effect size (correlation coefficients).Diagonal line represents replication effect size equal to original effect size. Dotted line represents replication effect size of 0. Points below the dotted line were effects in the opposite direction of the original. Density plots are separated by significant (blue) and nonsignificant (red) effects.
Reproducibility is a defining feature of science, but the extent to which it characterizes current research is unknown. We conducted replications of 100 experimental and correlational studies published in three psychology journals using high-powered designs and original materials when available. Replication effects were half the magnitude of original effects, representing a substantial decline. Ninety-seven percent of original studies had statistically significant results. Thirty-six percent of replications had statistically significant results; 47\% of original effect sizes were in the 95\% confidence interval of the replication effect size; 39\% of effects were subjectively rated to have replicated the original result; and if no bias in original results is assumed, combining original and replication results left 68\% with statistically significant effects. Correlational tests suggest that replication success was better predicted by the strength of original evidence than by characteristics of the original and replication teams.
A large-scale assessment suggests that experimental reproducibility in psychology leaves a lot to be desired.
A large-scale assessment suggests that experimental reproducibility in psychology leaves a lot to be desired.},
  number = {6251},
  journaltitle = {Science},
  urldate = {2019-03-04},
  date = {2015-08-28},
  pages = {aac4716},
  author = {Collaboration, Open Science},
  file = {/Users/mes335/Zotero/storage/4NFZA5KX/Collaboration - 2015 - Estimating the reproducibility of psychological sc.pdf},
  eprinttype = {pmid},
  eprint = {26315443}
}

@article{simmonsFalsepositivePsychologyUndisclosed2011,
  title = {False-Positive Psychology: {{Undisclosed}} Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant},
  volume = {22},
  shorttitle = {False-Positive Psychology},
  number = {11},
  journaltitle = {Psychological science},
  date = {2011},
  pages = {1359--1366},
  author = {Simmons, Joseph P. and Nelson, Leif D. and Simonsohn, Uri},
  file = {/Users/mes335/Zotero/storage/7U45PIN5/Simmons et al. - 2011 - False-positive psychology Undisclosed flexibility.pdf;/Users/mes335/Zotero/storage/4K46B3MY/0956797611417632.html}
}

@article{ioannidisWhyMostPublished2005,
  title = {Why Most Published Research Findings Are False},
  volume = {2},
  number = {8},
  journaltitle = {PLoS medicine},
  date = {2005},
  pages = {e124},
  author = {Ioannidis, John PA},
  file = {/Users/mes335/Zotero/storage/459Q8K9C/Ioannidis - 2005 - Why most published research findings are false.pdf;/Users/mes335/Zotero/storage/DT8FT92R/journal.pmed.html}
}

@article{polandAgeoldStruggleAntivaccinationists2011,
  title = {The Age-Old Struggle against the Antivaccinationists},
  volume = {364},
  number = {2},
  journaltitle = {New England Journal of Medicine},
  date = {2011},
  pages = {97--99},
  author = {Poland, Gregory A. and Jacobson, Robert M.},
  file = {/Users/mes335/Zotero/storage/9G3QT8F5/NEJMp1010594.html}
}

@book{wakefieldRETRACTEDIleallymphoidnodularHyperplasia1998,
  title = {{{RETRACTED}}: {{Ileal}}-Lymphoid-Nodular Hyperplasia, Non-Specific Colitis, and Pervasive Developmental Disorder in Children},
  shorttitle = {{{RETRACTED}}},
  publisher = {{Elsevier}},
  date = {1998},
  author = {Wakefield, Andrew J. and Murch, Simon H. and Anthony, Andrew and Linnell, John and Casson, David M. and Malik, Mohsin and Berelowitz, Mark and Dhillon, Amar P. and Thomson, Michael A. and Harvey, Peter},
  file = {/Users/mes335/Zotero/storage/LWGKUFR5/fulltext.html;/Users/mes335/Zotero/storage/NEAF6IA4/S0140673697110960.html}
}

@article{nosekPromotingOpenResearch2015,
  langid = {english},
  title = {Promoting an Open Research Culture},
  volume = {348},
  issn = {0036-8075, 1095-9203},
  url = {http://science.sciencemag.org/content/348/6242/1422},
  doi = {10.1126/science.aab2374},
  abstract = {Author guidelines for journals could help to promote transparency, openness, and reproducibility
Author guidelines for journals could help to promote transparency, openness, and reproducibility},
  number = {6242},
  journaltitle = {Science},
  urldate = {2019-03-05},
  date = {2015-06-26},
  pages = {1422-1425},
  author = {Nosek, B. A. and Alter, G. and Banks, G. C. and Borsboom, D. and Bowman, S. D. and Breckler, S. J. and Buck, S. and Chambers, C. D. and Chin, G. and Christensen, G. and Contestabile, M. and Dafoe, A. and Eich, E. and Freese, J. and Glennerster, R. and Goroff, D. and Green, D. P. and Hesse, B. and Humphreys, M. and Ishiyama, J. and Karlan, D. and Kraut, A. and Lupia, A. and Mabry, P. and Madon, T. and Malhotra, N. and Mayo-Wilson, E. and McNutt, M. and Miguel, E. and Paluck, E. Levy and Simonsohn, U. and Soderberg, C. and Spellman, B. A. and Turitto, J. and VandenBos, G. and Vazire, S. and Wagenmakers, E. J. and Wilson, R. and Yarkoni, T.},
  file = {/Users/mes335/Zotero/storage/PNI6ZAET/Nosek et al. - 2015 - Promoting an open research culture.pdf;/Users/mes335/Zotero/storage/R4QYXD5T/1422.html},
  eprinttype = {pmid},
  eprint = {26113702}
}

@incollection{camazineSynchronizedFlashingFireflies2003,
  langid = {english},
  title = {Synchronized Flashing among Fireflies},
  isbn = {978-0-691-11624-2},
  abstract = {The synchronized flashing of fireflies at night. The spiraling patterns of an aggregating slime mold. The anastomosing network of army-ant trails. The coordinated movements of a school of fish. Researchers are finding in such patterns--phenomena that have fascinated naturalists for centuries--a fertile new approach to understanding biological systems: the study of self-organization. This book, a primer on self-organization in biological systems for students and other enthusiasts, introduces readers to the basic concepts and tools for studying self-organization and then examines numerous examples of self-organization in the natural world. Self-organization refers to diverse pattern formation processes in the physical and biological world, from sand grains assembling into rippled dunes to cells combining to create highly structured tissues to individual insects working to create sophisticated societies. What these diverse systems hold in common is the proximate means by which they acquire order and structure. In self-organizing systems, pattern at the global level emerges solely from interactions among lower-level components. Remarkably, even very complex structures result from the iteration of surprisingly simple behaviors performed by individuals relying on only local information. This striking conclusion suggests important lines of inquiry: To what degree is environmental rather than individual complexity responsible for group complexity? To what extent have widely differing organisms adopted similar, convergent strategies of pattern formation? How, specifically, has natural selection determined the rules governing interactions within biological systems? Broad in scope, thorough yet accessible, this book is a self-contained introduction to self-organization and complexity in biology--a field of study at the forefront of life sciences research.},
  booktitle = {Self-Organization in {{Biological Systems}}},
  publisher = {{Princeton University Press}},
  date = {2003},
  pages = {142-165},
  keywords = {Science / Life Sciences / Biology,Science / System Theory},
  author = {Camazine, Scott and Deneubourg, Jean-Louis and Franks, Nigel R. and Sneyd, James and Bonabeau, Eric and Theraula, Guy},
  file = {/Users/mes335/Zotero/storage/R9BMUANG/week4_camazineetal-ch10.pdf},
  eprinttype = {googlebooks},
  eprint = {zMgyNN6Ufj0C}
}

@article{visserStrictMastFruiting2011,
  langid = {english},
  title = {Strict Mast Fruiting for a Tropical Dipterocarp Tree: A Demographic Cost–Benefit Analysis of Delayed Reproduction and Seed Predation},
  volume = {99},
  issn = {1365-2745},
  url = {https://besjournals.onlinelibrary.wiley.com/doi/abs/10.1111/j.1365-2745.2011.01825.x},
  doi = {10.1111/j.1365-2745.2011.01825.x},
  shorttitle = {Strict Mast Fruiting for a Tropical Dipterocarp Tree},
  abstract = {1. Masting, the production of large seed crops at intervals of several years, is a reproductive adaptation displayed by many tree species. The predator satiation hypothesis predicts that starvation of seed predators between mast years and satiation during mast years decreases seed predation and thus enhances tree regeneration. 2. Mast fruiting comes at demographic costs such as missed reproduction opportunities and increased density-dependence of recruits, but it remains unknown if predator satiation constitutes a sufficiently large benefit for masting to evolve as a viable life-history strategy. So far, no studies have quantified the net fitness consequences of masting. 3. Using a long-term demographic data set of the dipterocarp Shorea leprosula in a Malaysian forest, we constructed stochastic matrix population models and performed a demographic cost–benefit analysis. 4. For observed values of mast frequency and seed predation rates, we show that strict masting strongly increases fitness compared with fruiting annually. Model results also show that the demographic costs of mast fruiting are very low compared to the demographic losses due to seed predation in a scenario of annual fruiting. Finally, we find that mast fruiting would still be selected for even at low levels of seed predation and when including additional costs such as decreased adult growth rates, limiting crop size and density-dependent seedling survival. 5. Synthesis. Our results are consistent with the predictions of the predator satiation hypothesis: mast fruiting increases fitness for a range of seed predation levels. Under seed predation pressure annually fruiting species are at a strong disadvantage and as a result a mast fruiting strategy may swiftly confer a fitness advantage. Our study shows that demographic modelling allows the weighing of fitness benefits and costs of life-history phenomena such as strict masting.},
  number = {4},
  journaltitle = {Journal of Ecology},
  urldate = {2019-03-13},
  date = {2011},
  pages = {1033-1044},
  keywords = {delayed reproduction,demography,Dipterocarpaceae,elasticity,life-history evolution,plant population and community dynamics,predator satiation hypothesis,stochastic matrix model},
  author = {Visser, Marco D. and Jongejans, Eelke and van Breugel, Michiel and Zuidema, Pieter A. and Chen, Yu-Yun and Kassim, Abdul Rahman and de Kroon, Hans},
  file = {/Users/mes335/Zotero/storage/KTM3AJCH/Visser et al. - 2011 - Strict mast fruiting for a tropical dipterocarp tr.pdf;/Users/mes335/Zotero/storage/WYZL93RI/j.1365-2745.2011.01825.html}
}

@article{petersInteractionMastingFire2005,
  langid = {english},
  title = {The {{Interaction Between Masting}} and {{Fire Is Key}} to {{White Spruce Regeneration}}},
  volume = {86},
  issn = {1939-9170},
  url = {https://esajournals.onlinelibrary.wiley.com/doi/abs/10.1890/03-0656},
  doi = {10.1890/03-0656},
  abstract = {We used the mast-seeding tree Picea glauca (white spruce) to examine whether the timing of mast years relative to fire had a lasting effect on the density and timing of regeneration. We studied 17 fires that occurred in mast years and in years with low cone production between 1941 and 1994. Trees were carefully aged by crossdating procedures. Over the 59-yr period studied, there was significantly more regeneration after fires that occurred in mast years than after fires that occurred in years of low cone production. Spruce density was significantly lower after fires that occurred 1–3 years before a mast year than after fires during mast years. The cohort of trees that regenerated in the first mast year after a fire was critical to white spruce regeneration for fires that occurred 0–1 year before a mast year, but mast years occurring three or more years after a fire contributed few recruits. Our results suggest that masting is a key process that interacts with fire to shape stand composition in boreal mixedwoods. For species like white spruce, for which establishment is linked to disturbance, masting may have a contingent, historical effect on succession and landscape structure.},
  number = {7},
  journaltitle = {Ecology},
  urldate = {2019-03-13},
  date = {2005},
  pages = {1744-1750},
  keywords = {cohorts,crossdating,fire,masting,mixedwood forest,Picea glauca,regeneration,seedbed,white spruce},
  author = {Peters, Vernon S. and Macdonald, S. Ellen and Dale, Mark R. T.},
  file = {/Users/mes335/Zotero/storage/AI9Y93NY/Peters et al. - 2005 - The Interaction Between Masting and Fire Is Key to.pdf;/Users/mes335/Zotero/storage/AT3NCH5I/03-0656.html}
}

@article{janzenHerbivoresNumberTree1970,
  title = {Herbivores and the {{Number}} of {{Tree Species}} in {{Tropical Forests}}},
  volume = {104},
  issn = {0003-0147},
  url = {https://www.journals.uchicago.edu/doi/abs/10.1086/282687},
  doi = {10.1086/282687},
  abstract = {A high number of tree species, low density of adults of each species, and long distances between conspecific adults are characteristic of many low-land tropical forest habitats. I propose that these three traits, in large part, are the result of the action of predators on seeds and seedlings. A model is presented that allows detailed examination of the effect of different predators, dispersal agents, seed-crop sizes, etc. on these three traits. In short, any event that increases the efficiency of the predators at eating seeds and seedlings of a given tree species may lead to a reduction in population density of the adults of that species and/or to increased distance between new adults and their parents. Either event will lead to more space in the habitat for other species of trees, and therefore higher total number of tree species, provided seed sources are available over evolutionary time. As one moves from the wet lowland tropics to the dry tropics or temperate zones, the seed and seedling predators in a habitat are hypothesized to be progressively less efficient at keeping one or a few tree species from monopolizing the habitat through competitive superiority. This lowered efficiency of the predators is brought about by the increased severity and unpredictability of the physical environment, which in turn leads to regular or erratic escape of large seed or seedling cohorts from the predators.},
  number = {940},
  journaltitle = {The American Naturalist},
  shortjournal = {The American Naturalist},
  urldate = {2019-03-13},
  date = {1970-11-01},
  pages = {501-528},
  author = {Janzen, Daniel H.},
  file = {/Users/mes335/Zotero/storage/KZCX73IK/282687.html}
}

@article{lecunDeepLearning2015,
  title = {Deep Learning},
  volume = {521},
  number = {7553},
  journaltitle = {nature},
  date = {2015},
  pages = {436},
  author = {LeCun, Yann and Bengio, Yoshua and Hinton, Geoffrey},
  file = {/Users/mes335/Zotero/storage/9D8I9U5V/LeCun et al. - 2015 - Deep learning.pdf;/Users/mes335/Zotero/storage/F2DWCRDM/nature14539.html}
}

@article{chingtraversOpportunitiesObstaclesDeep2018,
  title = {Opportunities and Obstacles for Deep Learning in Biology and Medicine},
  volume = {15},
  url = {https://royalsocietypublishing.org/doi/full/10.1098/rsif.2017.0387},
  doi = {10.1098/rsif.2017.0387},
  abstract = {Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine are data-rich disciplines, but the data are complex and often ill-understood. Hence, deep learning techniques may be particularly well suited to solve problems of these fields. We examine applications of deep learning to a variety of biomedical problems—patient classification, fundamental biological processes and treatment of patients—and discuss whether deep learning will be able to transform these tasks or if the biomedical sphere poses unique challenges. Following from an extensive literature review, we find that deep learning has yet to revolutionize biomedicine or definitively resolve any of the most pressing challenges in the field, but promising advances have been made on the prior state of the art. Even though improvements over previous baselines have been modest in general, the recent progress indicates that deep learning methods will provide valuable means for speeding up or aiding human investigation. Though progress has been made linking a specific neural network's prediction to input features, understanding how users should interpret these models to make testable hypotheses about the system under study remains an open challenge. Furthermore, the limited amount of labelled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning enabling changes at both bench and bedside with the potential to transform several areas of biology and medicine.},
  number = {141},
  journaltitle = {Journal of The Royal Society Interface},
  shortjournal = {Journal of The Royal Society Interface},
  urldate = {2019-03-16},
  date = {2018-04-30},
  pages = {20170387},
  author = {{Ching Travers} and {Himmelstein Daniel S.} and {Beaulieu-Jones Brett K.} and {Kalinin Alexandr A.} and {Do Brian T.} and {Way Gregory P.} and {Ferrero Enrico} and {Agapow Paul-Michael} and {Zietz Michael} and {Hoffman Michael M.} and {Xie Wei} and {Rosen Gail L.} and {Lengerich Benjamin J.} and {Israeli Johnny} and {Lanchantin Jack} and {Woloszynek Stephen} and {Carpenter Anne E.} and {Shrikumar Avanti} and {Xu Jinbo} and {Cofer Evan M.} and {Lavender Christopher A.} and {Turaga Srinivas C.} and {Alexandari Amr M.} and {Lu Zhiyong} and {Harris David J.} and {DeCaprio Dave} and {Qi Yanjun} and {Kundaje Anshul} and {Peng Yifan} and {Wiley Laura K.} and {Segler Marwin H. S.} and {Boca Simina M.} and {Swamidass S. Joshua} and {Huang Austin} and {Gitter Anthony} and {Greene Casey S.}},
  file = {/Users/mes335/Zotero/storage/5BV9C46X/Ching Travers et al. - 2018 - Opportunities and obstacles for deep learning in b.pdf;/Users/mes335/Zotero/storage/BR7WENQJ/rsif.2017.html}
}

@inproceedings{corbett-daviesAlgorithmicDecisionMaking2017,
  location = {{New York, NY, USA}},
  title = {Algorithmic {{Decision Making}} and the {{Cost}} of {{Fairness}}},
  isbn = {978-1-4503-4887-4},
  url = {http://doi.acm.org/10.1145/3097983.3098095},
  doi = {10.1145/3097983.3098095},
  abstract = {Algorithms are now regularly used to decide whether defendants awaiting trial are too dangerous to be released back into the community. In some cases, black defendants are substantially more likely than white defendants to be incorrectly classified as high risk. To mitigate such disparities, several techniques have recently been proposed to achieve algorithmic fairness. Here we reformulate algorithmic fairness as constrained optimization: the objective is to maximize public safety while satisfying formal fairness constraints designed to reduce racial disparities. We show that for several past definitions of fairness, the optimal algorithms that result require detaining defendants above race-specific risk thresholds. We further show that the optimal unconstrained algorithm requires applying a single, uniform threshold to all defendants. The unconstrained algorithm thus maximizes public safety while also satisfying one important understanding of equality: that all individuals are held to the same standard, irrespective of race. Because the optimal constrained and unconstrained algorithms generally differ, there is tension between improving public safety and satisfying prevailing notions of algorithmic fairness. By examining data from Broward County, Florida, we show that this trade-off can be large in practice. We focus on algorithms for pretrial release decisions, but the principles we discuss apply to other domains, and also to human decision makers carrying out structured decision rules.},
  booktitle = {Proceedings of the 23rd {{ACM SIGKDD International Conference}} on {{Knowledge Discovery}} and {{Data Mining}}},
  series = {{{KDD}} '17},
  publisher = {{ACM}},
  urldate = {2019-03-16},
  date = {2017},
  pages = {797--806},
  keywords = {algorithmic fairness,discrimination,disparate impact,pretrial detention,risk assessment},
  author = {Corbett-Davies, Sam and Pierson, Emma and Feller, Avi and Goel, Sharad and Huq, Aziz},
  file = {/Users/mes335/Zotero/storage/L7SP7676/Corbett-Davies et al. - 2017 - Algorithmic Decision Making and the Cost of Fairne.pdf},
  venue = {Halifax, NS, Canada}
}

@article{berkFairnessCriminalJustice2018,
  langid = {english},
  title = {Fairness in {{Criminal Justice Risk Assessments}}: {{The State}} of the {{Art}}},
  issn = {0049-1241},
  url = {https://doi.org/10.1177/0049124118782533},
  doi = {10.1177/0049124118782533},
  shorttitle = {Fairness in {{Criminal Justice Risk Assessments}}},
  abstract = {Objectives:Discussions of fairness in criminal justice risk assessments typically lack conceptual precision. Rhetoric too often substitutes for careful analysis. In this article, we seek to clarify the trade-offs between different kinds of fairness and between fairness and accuracy.Methods:We draw on the existing literatures in criminology, computer science, and statistics to provide an integrated examination of fairness and accuracy in criminal justice risk assessments. We also provide an empirical illustration using data from arraignments.Results:We show that there are at least six kinds of fairness, some of which are incompatible with one another and with accuracy.Conclusions:Except in trivial cases, it is impossible to maximize accuracy and fairness at the same time and impossible simultaneously to satisfy all kinds of fairness. In practice, a major complication is different base rates across different legally protected groups. There is a need to consider challenging trade-offs. These lessons apply to applications well beyond criminology where assessments of risk can be used by decision makers. Examples include mortgage lending, employment, college admissions, child welfare, and medical diagnoses.},
  journaltitle = {Sociological Methods \& Research},
  shortjournal = {Sociological Methods \& Research},
  urldate = {2019-03-16},
  date = {2018-07-02},
  pages = {0049124118782533},
  author = {Berk, Richard and Heidari, Hoda and Jabbari, Shahin and Kearns, Michael and Roth, Aaron},
  file = {/Users/mes335/Zotero/storage/C4PTKMED/Berk et al. - 2018 - Fairness in Criminal Justice Risk Assessments The.pdf}
}

@online{harwellWantedPerfectBabysitter,
  langid = {english},
  title = {Wanted: {{The}} ‘Perfect Babysitter.’ {{Must}} Pass {{AI}} Scan for Respect and Attitude.},
  url = {https://www.washingtonpost.com/technology/2018/11/16/wanted-perfect-babysitter-must-pass-ai-scan-respect-attitude/},
  shorttitle = {Wanted},
  abstract = {A start-up that requires prospective babysitters to hand over their social media accounts says it uses “advanced artificial intelligence” to assess a sitter's risk of drug abuse, bullying and more.},
  journaltitle = {Washington Post},
  urldate = {2019-03-16},
  author = {Harwell, Drew}
}

@article{wilsonPredictiveInequityObject2019,
  title = {Predictive {{Inequity}} in {{Object Detection}}},
  journaltitle = {arXiv preprint arXiv:1902.11097},
  date = {2019},
  author = {Wilson, Benjamin and Hoffman, Judy and Morgenstern, Jamie},
  file = {/Users/mes335/Zotero/storage/252P64LS/Wilson et al. - 2019 - Predictive Inequity in Object Detection.pdf;/Users/mes335/Zotero/storage/HCRL9XP8/1902.html}
}

@article{chingOpportunitiesObstaclesDeep2018,
  langid = {english},
  title = {Opportunities and Obstacles for Deep Learning in Biology and Medicine},
  volume = {15},
  issn = {1742-5662},
  doi = {10.1098/rsif.2017.0387},
  abstract = {Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine are data-rich disciplines, but the data are complex and often ill-understood. Hence, deep learning techniques may be particularly well suited to solve problems of these fields. We examine applications of deep learning to a variety of biomedical problems-patient classification, fundamental biological processes and treatment of patients-and discuss whether deep learning will be able to transform these tasks or if the biomedical sphere poses unique challenges. Following from an extensive literature review, we find that deep learning has yet to revolutionize biomedicine or definitively resolve any of the most pressing challenges in the field, but promising advances have been made on the prior state of the art. Even though improvements over previous baselines have been modest in general, the recent progress indicates that deep learning methods will provide valuable means for speeding up or aiding human investigation. Though progress has been made linking a specific neural network's prediction to input features, understanding how users should interpret these models to make testable hypotheses about the system under study remains an open challenge. Furthermore, the limited amount of labelled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning enabling changes at both bench and bedside with the potential to transform several areas of biology and medicine.},
  number = {141},
  journaltitle = {Journal of the Royal Society, Interface},
  shortjournal = {J R Soc Interface},
  date = {2018-04},
  keywords = {deep learning,genomics,machine learning,precision medicine},
  author = {Ching, Travers and Himmelstein, Daniel S. and Beaulieu-Jones, Brett K. and Kalinin, Alexandr A. and Do, Brian T. and Way, Gregory P. and Ferrero, Enrico and Agapow, Paul-Michael and Zietz, Michael and Hoffman, Michael M. and Xie, Wei and Rosen, Gail L. and Lengerich, Benjamin J. and Israeli, Johnny and Lanchantin, Jack and Woloszynek, Stephen and Carpenter, Anne E. and Shrikumar, Avanti and Xu, Jinbo and Cofer, Evan M. and Lavender, Christopher A. and Turaga, Srinivas C. and Alexandari, Amr M. and Lu, Zhiyong and Harris, David J. and DeCaprio, Dave and Qi, Yanjun and Kundaje, Anshul and Peng, Yifan and Wiley, Laura K. and Segler, Marwin H. S. and Boca, Simina M. and Swamidass, S. Joshua and Huang, Austin and Gitter, Anthony and Greene, Casey S.},
  file = {/Users/mes335/Zotero/storage/TIKEWP52/Ching et al. - 2018 - Opportunities and obstacles for deep learning in b.pdf},
  eprinttype = {pmid},
  eprint = {29618526},
  pmcid = {PMC5938574}
}

@book{bannisterSocialDarwinismScience2010,
  langid = {english},
  title = {Social {{Darwinism}}: {{Science}} and {{Myth}} in {{Anglo}}-{{American Social Thought}}},
  isbn = {978-1-4399-0605-7},
  shorttitle = {Social {{Darwinism}}},
  abstract = {"The most systematic and comprehensive effort yet made to assess the role played by Darwinian ideas in the writings of English-speaking social theorists of the late-nineteenth and early-twentieth centuries." --Isis "In seeking to set the record straight, Bannister cuts through the amalgam with an intellectual shredder, exposing the illogic and incompatibility involved in fusing Charles Darwin's On the Origin of Species with Herbert Spencer's Social Statics.... Bannister's familiarity with relevant texts and their reception by contemporary social theorists, scholars, and critics on both sides of the Atlantic is impressive." --Journal of Interdisciplinary History "A fine contribution to Anglo-American intellectual history." --Journal of American History},
  pagetotal = {336},
  publisher = {{Temple University Press}},
  date = {2010-06-09},
  keywords = {Social Science / General,Social Science / Sociology / General},
  author = {Bannister, Robert},
  eprinttype = {googlebooks},
  eprint = {bzANHSAo60cC}
}

@article{dennisSocialDarwinismScientific1995,
  title = {Social {{Darwinism}}, {{Scientific Racism}}, and the {{Metaphysics}} of {{Race}}},
  volume = {64},
  issn = {0022-2984},
  url = {https://www.jstor.org/stable/2967206},
  doi = {10.2307/2967206},
  abstract = {Tracing the philosophical underpinnings of scientific racism from the early work of hereditarians Darwin, Spencer, and Sumner, to the intelligence testing movement led by Galton and Binet, and lastly to the contemporary race and IQ studies of Jensen, Herrnstein, and Murray, this article maintains that science is often used as a justification to propose, project, and enact racist social policies. It begins with a review of the philosophy of Social Darwinism and of its assumptions about race and human abilities, and ends by analyzing a largely unbroached theme in this debate: the consequences of scientific racism for dominant groups.},
  number = {3},
  journaltitle = {The Journal of Negro Education},
  urldate = {2019-03-24},
  date = {1995},
  pages = {243-252},
  author = {Dennis, Rutledge M.}
}

@article{heyesceciliaNewThinkingEvolution2012,
  title = {New Thinking: The Evolution of Human Cognition},
  volume = {367},
  url = {https://royalsocietypublishing.org/doi/full/10.1098/rstb.2012.0111},
  doi = {10.1098/rstb.2012.0111},
  shorttitle = {New Thinking},
  abstract = {Humans are animals that specialize in thinking and knowing, and our extraordinary cognitive abilities have transformed every aspect of our lives. In contrast to our chimpanzee cousins and Stone Age ancestors, we are complex political, economic, scientific and artistic creatures, living in a vast range of habitats, many of which are our own creation. Research on the evolution of human cognition asks what types of thinking make us such peculiar animals, and how they have been generated by evolutionary processes. New research in this field looks deeper into the evolutionary history of human cognition, and adopts a more multi-disciplinary approach than earlier ‘Evolutionary Psychology’. It is informed by comparisons between humans and a range of primate and non-primate species, and integrates findings from anthropology, archaeology, economics, evolutionary biology, neuroscience, philosophy and psychology. Using these methods, recent research reveals profound commonalities, as well striking differences, between human and non-human minds, and suggests that the evolution of human cognition has been much more gradual and incremental than previously assumed. It accords crucial roles to cultural evolution, techno-social co-evolution and gene–culture co-evolution. These have produced domain-general developmental processes with extraordinary power—power that makes human cognition, and human lives, unique.},
  number = {1599},
  journaltitle = {Philosophical Transactions of the Royal Society B: Biological Sciences},
  shortjournal = {Philosophical Transactions of the Royal Society B: Biological Sciences},
  urldate = {2019-03-24},
  date = {2012-08-05},
  pages = {2091-2096},
  author = {{Heyes Cecilia}},
  file = {/Users/mes335/Zotero/storage/WACM8DSZ/Heyes Cecilia - 2012 - New thinking the evolution of human cognition.pdf;/Users/mes335/Zotero/storage/ZSP62HM5/rstb.2012.html}
}

@article{amatulliSuiteGlobalCrossscale2018,
  langid = {english},
  title = {A Suite of Global, Cross-Scale Topographic Variables for Environmental and Biodiversity Modeling},
  volume = {5},
  issn = {2052-4463},
  url = {https://www.nature.com/articles/sdata201840},
  doi = {10.1038/sdata.2018.40},
  abstract = {Topographic variation underpins a myriad of patterns and processes in hydrology, climatology, geography and ecology and is key to understanding the variation of life on the planet. A fully standardized and global multivariate product of different terrain features has the potential to support many large-scale research applications, however to date, such datasets are unavailable. Here we used the digital elevation model products of global 250 m GMTED2010 and near-global 90 m SRTM4.1dev to derive a suite of topographic variables: elevation, slope, aspect, eastness, northness, roughness, terrain roughness index, topographic position index, vector ruggedness measure, profile/tangential curvature, first/second order partial derivative, and 10 geomorphological landform classes. We aggregated each variable to 1, 5, 10, 50 and 100 km spatial grains using several aggregation approaches. While a cross-correlation underlines the high similarity of many variables, a more detailed view in four mountain regions reveals local differences, as well as scale variations in the aggregated variables at different spatial grains. All newly-developed variables are available for download at Data Citation 1 and for download and visualization at http://www.earthenv.org/topography.},
  journaltitle = {Scientific Data},
  urldate = {2019-08-28},
  date = {2018-03-20},
  pages = {180040},
  author = {Amatulli, Giuseppe and Domisch, Sami and Tuanmu, Mao-Ning and Parmentier, Benoit and Ranipeta, Ajay and Malczyk, Jeremy and Jetz, Walter},
  file = {/Users/mes335/Zotero/storage/SILIGGP8/Amatulli et al. - 2018 - A suite of global, cross-scale topographic variabl.pdf;/Users/mes335/Zotero/storage/TMA3GF7Z/sdata201840.html}
}

@article{chen2004using,
  title = {Using Random Forest to Learn Imbalanced Data},
  volume = {110},
  number = {1-12},
  journaltitle = {University of California, Berkeley},
  date = {2004},
  pages = {24},
  author = {Chen, Chao and Liaw, Andy and Breiman, Leo},
  file = {/Users/mes335/Zotero/storage/ECDFDVUX/Chen - Using Random Forest to Learn Imbalanced Data.pdf}
}

@article{robinsonCorrectingBiasDistribution2018,
  langid = {english},
  title = {Correcting for Bias in Distribution Modelling for Rare Species Using Citizen Science Data},
  volume = {24},
  issn = {1472-4642},
  url = {https://onlinelibrary.wiley.com/doi/abs/10.1111/ddi.12698},
  doi = {10.1111/ddi.12698},
  abstract = {Aim To improve the accuracy of inferences on habitat associations and distribution patterns of rare species by combining machine-learning, spatial filtering and resampling to address class imbalance and spatial bias of large volumes of citizen science data. Innovation Modelling rare species’ distributions is a pressing challenge for conservation and applied research. Often, a large number of surveys are required before enough detections occur to model distributions of rare species accurately, resulting in a data set with a high proportion of non-detections (i.e. class imbalance). Citizen science data can provide a cost-effective source of surveys but likely suffer from class imbalance. Citizen science data also suffer from spatial bias, likely from preferential sampling. To correct for class imbalance and spatial bias, we used spatial filtering to under-sample the majority class (non-detection) while maintaining all of the limited information from the minority class (detection). We investigated the use of spatial under-sampling with randomForest models and compared it to common approaches used for imbalanced data, the synthetic minority oversampling technique (SMOTE), weighted random forest and balanced random forest models. Model accuracy was assessed using kappa, Brier score and AUC. We demonstrate the method by evaluating habitat associations and seasonal distribution patterns using citizen science data for a rare species, the tricoloured blackbird (Agelaius tricolor). Main Conclusions Spatial under-sampling increased the accuracy of each model and outperformed the approach typically used to direct under-sampling in the SMOTE algorithm. Our approach is the first to characterize winter distribution and movement of tricoloured blackbirds. Our results show that tricoloured blackbirds are positively associated with grassland, pasture and wetland habitats, and negatively associated with high elevations or evergreen forests during both winter and breeding seasons. The seasonal differences in distribution indicate that individuals move to the coast during the winter, as suggested by historical accounts.},
  number = {4},
  journaltitle = {Diversity and Distributions},
  urldate = {2019-11-02},
  date = {2018},
  pages = {460-472},
  keywords = {citizen science,class imbalance,random forest,spatial bias,species distribution model,tricoloured blackbird},
  author = {Robinson, Orin J. and Ruiz‐Gutierrez, Viviana and Fink, Daniel},
  file = {/Users/mes335/Zotero/storage/DN5UMAN8/Robinson et al. - 2018 - Correcting for bias in distribution modelling for .pdf;/Users/mes335/Zotero/storage/2HQV8XTA/ddi.html}
}

@book{woodGeneralizedAdditiveModels2017,
  title = {Generalized Additive Models: An Introduction with {{R}}},
  edition = {2},
  shorttitle = {Generalized Additive Models},
  publisher = {{Chapman and Hall/CRC}},
  date = {2017},
  author = {Wood, Simon N.}
}

@article{murphyNewVectorPartition1973,
  title = {A {{New Vector Partition}} of the {{Probability Score}}},
  volume = {12},
  issn = {0021-8952},
  url = {https://journals.ametsoc.org/doi/abs/10.1175/1520-0450%281973%29012%3C0595%3AANVPOT%3E2.0.CO%3B2},
  doi = {10.1175/1520-0450(1973)012<0595:ANVPOT>2.0.CO;2},
  abstract = {A new vector partition of the probability, or Brier, score (PS) is formulated and the nature and properties of this partition are described. The relationships between the terms in this partition and the terms in the original vector partition of the PS are indicated. The new partition consists of three terms: 1) a measure of the uncertainty inherent in the events, or states, on the occasions of concern (namely, the PS for the sample relative frequencies); 2) a measure of the reliability of the forecasts; and 3) a new measure of the resolution of the forecasts. These measures of reliability and resolution are and are not, respectively, equivalent (i.e., linearly related) to the measures of reliability and resolution provided by the original partition. Two sample collections of probability forecasts are used to illustrate the differences and relationships between these partitions. Finally, the two partitions are compared, with particular reference to the attributes of the forecasts with which the partitions are concerned, the interpretation of the partitions in geometric terms, and the use of the partitions as the bases for the formulation of measures to evaluate probability forecasts. The results of these comparisons indicate that the new partition offers certain advantages vis-à-vis the original partition.},
  number = {4},
  journaltitle = {Journal of Applied Meteorology},
  shortjournal = {J. Appl. Meteor.},
  urldate = {2019-12-10},
  date = {1973-06-01},
  pages = {595-600},
  author = {Murphy, Allan H.},
  file = {/Users/mes335/Zotero/storage/5KSGAZMU/Murphy - 1973 - A New Vector Partition of the Probability Score.pdf;/Users/mes335/Zotero/storage/JVB2BWD5/1520-0450(1973)0120595ANVPOT2.0.html}
}

@article{vaughanContinuingChallengesTesting2005,
  langid = {english},
  title = {The Continuing Challenges of Testing Species Distribution Models},
  volume = {42},
  issn = {1365-2664},
  url = {https://besjournals.onlinelibrary.wiley.com/doi/abs/10.1111/j.1365-2664.2005.01052.x},
  doi = {10.1111/j.1365-2664.2005.01052.x},
  abstract = {1 Species distribution models could bring manifold benefits across ecology, but require careful testing to prove their reliability and guide users. Shortcomings in testing are often evident, failing to reflect recent methodological developments and changes in the way models are applied. We considered some of the fundamental issues. 2 Generalizability is a basic requirement for predictive models, describing their capacity to produce accurate predictions with new data, i.e. in real applications beyond model training. Tests of generalizability should be as rigorous as possible: ideally using a large number of independent test sites (≥ 200–300) that represent anticipated applications. Bootstrapping identifies the role of overfitting of the training data in limiting a model's generalizability. 3 Predictions from most distribution models are continuous variables. Their accuracy may be described by discrimination and calibration components. Discriminatory ability describes how well a model separates occupied from unoccupied sites. It is independent of species prevalence and is readily comparable between models. Rank correlation coefficients, such as the concordance index, are effective measures. 4 Calibration describes the numerical accuracy of predictions (e.g. whether 40\% of sites with predicted probabilities of 0·40 are occupied) but is frequently overlooked in model testing. Poor calibration could mislead any conservation efforts utilizing models to estimate the ‘value’ of different sites for a given species. Effective assessments can be made using smoothed calibration plots. 5 The effects of species prevalence on nominal presence–absence predictions are well known. The currently preferred accuracy measure, Cohen's κ, has weaknesses. We argue that mutual information measures, based in information theory, may be more appropriate. 6 Synthesis and applications. Model evaluation must be informative and should ideally: (i) define generalizability in detail; (ii) separate the discrimination and calibration components of accuracy and test both; (iii) adopt assessment techniques that permit more valid intermodel comparisons; (iv) avoid nominal presence–absence evaluation where possible and consider information-theoretic measures; and (v) utilize the full range of techniques to help diagnose the causes of prediction problems. Few modellers in applied ecology and conservation biology satisfy these needs, making it difficult for others to evaluate models and identify potential misuses. The problems are real, and if uncorrected will damage conservation efforts through the inaccurate assessment of distribution and habitat preferences of important organisms.},
  number = {4},
  journaltitle = {Journal of Applied Ecology},
  urldate = {2019-12-10},
  date = {2005},
  pages = {720-730},
  keywords = {discrimination,bootstrapping,calibration,generalizability,overfitting,presence–absence data,transportability},
  author = {Vaughan, I. P. and Ormerod, S. J.},
  file = {/Users/mes335/Zotero/storage/8ST959FG/Vaughan and Ormerod - 2005 - The continuing challenges of testing species distr.pdf;/Users/mes335/Zotero/storage/HTXTCC5G/j.1365-2664.2005.01052.html}
}

@inproceedings{niculescu-mizilPredictingGoodProbabilities2005,
  title = {Predicting Good Probabilities with Supervised Learning},
  booktitle = {Proceedings of the 22nd International Conference on {{Machine}} Learning},
  publisher = {{ACM}},
  date = {2005},
  pages = {625--632},
  author = {Niculescu-Mizil, Alexandru and Caruana, Rich},
  file = {/Users/mes335/Zotero/storage/JCIVTBPX/Niculescu-Mizil and Caruana - 2005 - Predicting good probabilities with supervised lear.pdf;/Users/mes335/Zotero/storage/HAIW3NVL/citation.html}
}

@article{plattProbabilisticOutputsSupport1999,
  title = {Probabilistic Outputs for Support Vector Machines and Comparisons to Regularized Likelihood Methods},
  volume = {10},
  number = {3},
  journaltitle = {Advances in large margin classifiers},
  date = {1999},
  pages = {61--74},
  author = {Platt, John},
  file = {/Users/mes335/Zotero/storage/HTL32WSW/Platt - 1999 - Probabilistic outputs for support vector machines .pdf}
}