Aspects Of Data Quality Research Articles

Research data management in agricultural sciences in Germany: We are not yet where we want to be.

To meet the future challenges and foster integrated and holistic research approaches in agricultural sciences, new and sustainable methods in research data management (RDM) are needed. The involvement of scientific users is a critical success factor for their development. We conducted an online survey in 2020 among different user groups in agricultural sciences about their RDM practices and needs. In total, the questionnaire contained 52 questions on information about produced and (re-)used data, data quality aspects, information about the use of standards, publication practices and legal aspects of agricultural research data, the current situation in RDM in regards to awareness, consulting and curricula as well as needs of the agricultural community in respect to future developments. We received 196 (partially) completed questionnaires from data providers, data users, infrastructure and information service providers. In addition to the diversity in the research data landscape of agricultural sciences in Germany, the study reveals challenges, deficits and uncertainties in handling research data in agricultural sciences standing in the way of access and efficient reuse of valuable research data. However, the study also suggests and discusses potential solutions to enhance data publications, facilitate and secure data re-use, ensure data quality and develop services (i.e. training, support and bundling services). Therefore, our research article provides the basis for the development of common RDM, future infrastructures and services needed to foster the cultural change in handling research data across agricultural sciences in Germany and beyond.

Users Perception and Factors Affecting Data Quality in Nyarugenge Public Health Facility, Rwanda

The study evaluated the users' opinions on data quality and related characteristics in ten public-health institutions of Nyarugenge district. A study used a cross-sectional design, data was collected through quantitative (n=150) methodology while qualitative data was obtained using interviews (n=20) and focus group talks (n=3). A checklist was utilized to examine the completeness, accuracy, and timeliness of data quality aspects. Collected quantitative data was analyzed through logistic regression by SPSS to examine the association of variables, while qualitative data was analyzed using the summative content analysis (SCA) to summarize the key themes. A 95 percent confidence level, Odds Ratio (AOR) were used to establish the strength of correlation among study variables, while a p-value of less than (p<0.05) was utilized to identify the variables which were statistically significant associated to HMIS data quality. The study finding showed that the majority (53.0%) was female while, 33 years was an average age, the majority of health practitioner (52.7%) had an A1 diploma. Approximately 90.6% of respondents have a positive perceptions on the system usage. Limited ability and a lack of relevant technology equipment such as computers and the internet have been cited as the challenges while using the system. Factors such as training AOR:2.62(95% CI:1.45, supervision AOR:1.81(95% CI:1.02, AOR:2.50(95% CI:0.85 for education background, works-experience AOR:1.60(95% CI:090 are factors associated with data quality. Maintaining, supportive supervision, regular training and refresher courses should be regularly offered to public health professionals to improve their knowledge in order to maximize the use of health information.

Performance of Multi-GNSS in the Asia-Pacific Region: Signal Quality, Broadcast Ephemeris and Precise Point Positioning (PPP)

Since BeiDou Navigation Satellite System (BDS) and Japan’s Quasi-Zenith Satellite System (QZSS) have more visible satellites in the Asia-Pacific region, and navigation satellites of Global Positioning System (GPS), Galileo satellite navigation system (Galileo), and GLONASS satellite navigation system (GLONASS) are uniformly distributed globally, the service level of multi-mode Global Navigation Satellite System (GNSS) in the Asia-Pacific region should represent the best service capability. Based on the observation data of 10 Multi-GNSS Experiment (MGEX) stations, broadcast ephemeris and precision ephemeris from 13 to 19 October 2021, this paper comprehensively evaluated the service capability of multi-GNSS in the Asia-Pacific region from three aspects of observation data quality, broadcast ephemeris performance, and precision positioning level. The results show that: (1) the carrier-to-noise-density ratio (C/N0) quality of the GPS and Galileo is the best, followed by BDS and GLONASS, and QZSS is the worst. GPS, BDS-2, GLONASS, and QZSS pseudorange multipath values range from 0 to 0.6 m, while Galileo system and BDS-3 pseudorange multipath values range from 0 to 0.8 m. (2) In terms of broadcast ephemeris accuracy, BDS-3 broadcast ephemeris has the best orbit, and the three-dimensional (3D) Root Mean Square (RMS) is 0.21 m; BDS-2 was the worst, with a 3D RMS of 1.99 m. The broadcast ephemeris orbits of GPS, Galileo, QZSS, and GLONASS have 3D RMS of 0.60 m, 0.62 m, 0.83 m, and 1.27 m, respectively. For broadcast ephemeris clock offset: Galileo has the best performance, 0.61 ns, GLONASS is the worst, standard deviation (STD) is 3.10 ns, GPS, QZSS, BDS-3 and BDS-2 are 0.65 ns, 0.75 ns, and 1.72 ns, respectively. For signal-in-space ranging errors (SISRE), the SISRE results of GPS and Galileo systems are the best, fluctuating in the range of 0 m–2 m, followed by QZSS, BDS-3, Galileo, and BDS-2. (3) GPS, BDS, GLONASS, Galileo, GPS/QZSS, and BDS/QZSS were used for positioning experiments. In static PPP, the convergence time and positioning accuracy of GPS show the best performance. The positioning accuracy of GPS/QZSS and BDS/QZSS is improved compared with that of GPS and BDS. In terms of kinematic PPP, the convergence time and positioning accuracy of GPS/QZSS and BDS/QZSS are improved compared with that of GPS and BDS. In addition to GLONASS and Galileo systems, the other combinations outperformed 3 cm, 3 cm, and 5 cm in the east, north, and up directions.

If They Don’t Understand the Question, They Don’t answer. Language Mismatch in Face-to-Face Interviews

Abstract The provision of translated field instruments is a crucial aspect to reduce response burden and subsequently increase data quality in surveys with a multi-linguistic target population such as surveys on recent immigrants. Failure to address this can result in a mismatch between the survey language and the respondent’s mother tongue. By using a survey on refugees in Germany, this article explores the correlation of the absence of the respondents’ mother tongue on item nonresponse – a crucial aspect of data quality. Further, this article investigates whether supplementary audio recordings in the same language as the written questions can reduce item nonresponse when the mother tongue is not available. To answer the research questions, all missing answers per individual are counted and analyzed by means of poisson regression analyses. In a second step, the likelihood of item-nonresponse for single items is estimated as well. Results show that a language mismatch as well as the usage of audio recordings increase item nonresponse.

A CONCEPTUAL MODEL FOR CONVERTING OPENSTREETMAP CONTRIBUTION TO GEOSPATIAL MACHINE LEARNING TRAINING DATA

Abstract. In the recent decade, Volunteered Geographical Information (VGI), in particular the OpenStreetMap (OSM), has helped to fill substantial data gaps in base maps, especially in Global South, thus has become a promising source of massive, free training data together with rich and detailed semantic information for geospatial artificial intelligence (GeoAI) applications. Although intensive works have explored the potential of generating training data from OSM, a systematic approach of harvesting OSM contribution as quality-aware training data for different GeoAI tasks is still missing. To fill this research gap, we proposed a conceptual model consisting of three major components: historical OSM and external datasets, quality indicators, and GeoAI models. As a proof of concept, we validated our conceptual model with an example task of detecting OSM missing buildings in Mozambique, where the impact of different error sources (e.g., completeness, alignment, rotation) in training data were compared and investigated in a quantitative manner. The lessons learned in this paper shed important lights on cooperating OSM data quality aspects with the development of more explainable GeoAI models.

Potentials of Low-Budget Microdrones: Processing 3D Point Clouds and Images for Representing Post-Industrial Landmarks in Immersive Virtual Environments.

Post-industrial areas in Europe, such as the Rhine-Ruhr Metropolitan region in Germany, include cultural heritage sites fostering local and regional identities with the industrial past. Today, these landmarks are popular places of interest for visitors. In addition to portable camera devices, low-budget ultra-lightweight unmanned aerial vehicles, such as micro quadcopter drones, are on their way to being established as mass photography equipment. This low-cost hardware is not only useful for recreational usage but also supports individualized remote sensing with optical images and facilitates the acquisition of 3D point clouds of the targeted object(s). Both data sets are valuable and accurate geospatial data resources for further processing of textured 3D models. To experience these 3D models in a timely way, these 3D visualizations can directly be imported into game engines. They can be extended with modern interaction techniques and additional (semantic) information. The visualization of the data can be explored in immersive virtual environments, which allows, for instance, urban planners to use low-cost microdrones to 3D map the human impact on the environment and preserve this status in a 3D model that can be analyzed and explored in following steps. A case example of the old wage hall of the Zeche “Bonifacius” (Essen, Germany) with its simple building structure showed that it is possible to generate a detailed and accurate 3D model based on the microdrone data. The point cloud which the 3D model of the old wage hall was based on represented partly better data accuracy than the point clouds derived from airborne laser scanning and offered by public agencies as open data. On average, the distance between the point clouds was 0.7 m, while the average distance between the airborne laser scanning point cloud and the 3D model was −0.02 m. Matching high-quality textures of the building facades brings in a new aspect of 3D data quality which can be adopted when creating immersive virtual environments using the Unity engine. The example of the wage hall makes it clear that the use of low-cost drones and the subsequent data processing can result in valuable sources of point clouds and textured 3D models.

EHealth Tools That Assess and Track Health and Well-being in Children and Young People: Systematic Review.

BackgroundeHealth tools that assess and track health outcomes in children or young people are an emerging type of technology that has the potential to reform health service delivery and facilitate integrated, interdisciplinary care.ObjectiveThe aim of this review is to summarize eHealth tools that have assessed and tracked health in children or young people to provide greater clarity around the populations and settings in which they have been used, characteristics of digital devices (eg, health domains, respondents, presence of tracking, and connection to care), primary outcomes, and risks and challenges of implementation.MethodsA search was conducted in PsycINFO, PubMed or MEDLINE, and Embase in April 2020. Studies were included if they evaluated a digital device whose primary purpose was to assess and track health, focused on children or young people (birth to the age of 24 years), reported original research, and were published in peer-reviewed journals in English.ResultsA total of 39 papers were included in this review. The sample sizes ranged from 7 to 149,329 participants (median 163, mean 5155). More studies were conducted in urban (18/39, 46%) regions than in rural (3/39, 8%) regions or a combination of urban and rural areas (8/39, 21%). Devices were implemented in three main settings: outpatient health clinics (12/39, 31%), hospitals (14/39, 36%), community outreach (10/39, 26%), or a combination of these settings (3/39, 8%). Mental and general health were the most common health domains assessed, with a single study assessing multiple health domains. Just under half of the devices tracked children’s health over time (16/39, 41%), and two-thirds (25/39, 64%) connected children or young people to clinical care. It was more common for information to be collected from a single informant (ie, the child or young person, trained health worker, clinician, and parent or caregiver) than from multiple informants. The health of children or young people was assessed as a primary or secondary outcome in 36% (14/39) of studies; however, only 3% (1/39) of studies assessed whether using the digital tool improved the health of users. Most papers reported early phase research (formative or process evaluations), with fewer outcome evaluations and only 3 randomized controlled trials. Identified challenges or risks were related to accessibility, clinical utility and safety, uptake, data quality, user interface or design aspects of the device, language proficiency or literacy, sociocultural barriers, and privacy or confidentiality concerns; ways to address these barriers were not thoroughly explored.ConclusionseHealth tools that assess and track health in children or young people have the potential to enhance health service delivery; however, a strong evidence base validating the clinical utility, efficacy, and safety of tools is lacking, and more thorough investigation is needed to address the risks and challenges of using these emerging technologies in clinical care. At present, there is greater potential for the tools to facilitate multi-informant, multidomain assessments and longitudinally track health over time and room for further implementation in rural or remote regions and community settings around the world.

ATA2022 Annual Conference & Expo May 1–3, 2022Boston, Massachusetts

ATA2022 Annual Conference & Expo May 1–3, 2022Boston, Massachusetts

Is Google Trends a quality data source?

ABSTRACT Google Trends (GT) has become a popular data source among researchers in a wide variety of fields. In economics, its main use has been to forecast other economic variables such as tourism demand, unemployment or sales. This paper questions the quality of these data by discussing the main data quality aspects according to the literature. Our analysis evidences some non-negligible issues related to the measurement accuracy of GT, which potentially affects the results obtained with GT data and therefore the decisions made with this information. These issues are illustrated with an example in which some queries to GT are repeated on six different days.

Data quality of platforms and panels for online behavioral research.

We examine key aspects of data quality for online behavioral research between selected platforms (Amazon Mechanical Turk, CloudResearch, and Prolific) and panels (Qualtrics and Dynata). To identify the key aspects of data quality, we first engaged with the behavioral research community to discover which aspects are most critical to researchers and found that these include attention, comprehension, honesty, and reliability. We then explored differences in these data quality aspects in two studies (N ~ 4000), with or without data quality filters (approval ratings). We found considerable differences between the sites, especially in comprehension, attention, and dishonesty. In Study 1 (without filters), we found that only Prolific provided high data quality on all measures. In Study 2 (with filters), we found high data quality among CloudResearch and Prolific. MTurk showed alarmingly low data quality even with data quality filters. We also found that while reputation (approval rating) did not predict data quality, frequency and purpose of usage did, especially on MTurk: the lowest data quality came from MTurk participants who report using the site as their main source of income but spend few hours on it per week. We provide a framework for future investigation into the ever-changing nature of data quality in online research, and how the evolving set of platforms and panels performs on these key aspects.

GBIF Integration of Open Data

The Global Biodiversity Information Facility (GBIF) runs a global data infrastructure that integrates data from more than 1700 institutions. Combining data at this scale has been achieved by deploying open Application Programming Interfaces (API) that adhere to the open data standards provided by Biodiversity Information Standards (TDWG). In this presentation, we will provide an overview of the GBIF infrastructure and APIs and provide insight into lessons learned while operating and evolving the systems, such as long-term API stability, ease of use, and efficiency. This will include the following topics: The registry component provides RESTful APIs for managing the organizations, repositories and datasets that comprise the network and control access permissions. Stability and ease of use have been critical to this being embedded in many systems. Changes within the registry trigger data crawling processes, which connect to external systems through their APIs and deposit datasets into GBIF's central data warehouse. One challenge here relates to the consistency of data across a distributed network. Once a dataset is crawled, the data processing infrastructure organizes and enriches data using reference catalogues accessed through open APIs, such as the vocabulary server and the taxonomic backbone. Being able to process data quickly as source data and reference catalogues change is a challenge for this component. The data access APIs provide search and download services. Asynchronous APIs are required for some of these aspects, and long-term stability is a requirement for widespread adoption. Here we will talk about policies for schema evolution to avoid incompatible changes, which would cause failures in client systems. The APIs that drive the user interface have specific needs such as efficient use of the network bandwidth. We will present how we approached this, and how we are currently adopting GraphQL as the next generation of these APIs. There are several APIs that we believe are of use for the data publishing community. These include APIs that will help in data quality aspects, and new data of interest thanks to the data clustering algorithms GBIF deploys. The registry component provides RESTful APIs for managing the organizations, repositories and datasets that comprise the network and control access permissions. Stability and ease of use have been critical to this being embedded in many systems. Changes within the registry trigger data crawling processes, which connect to external systems through their APIs and deposit datasets into GBIF's central data warehouse. One challenge here relates to the consistency of data across a distributed network. Once a dataset is crawled, the data processing infrastructure organizes and enriches data using reference catalogues accessed through open APIs, such as the vocabulary server and the taxonomic backbone. Being able to process data quickly as source data and reference catalogues change is a challenge for this component. The data access APIs provide search and download services. Asynchronous APIs are required for some of these aspects, and long-term stability is a requirement for widespread adoption. Here we will talk about policies for schema evolution to avoid incompatible changes, which would cause failures in client systems. The APIs that drive the user interface have specific needs such as efficient use of the network bandwidth. We will present how we approached this, and how we are currently adopting GraphQL as the next generation of these APIs. There are several APIs that we believe are of use for the data publishing community. These include APIs that will help in data quality aspects, and new data of interest thanks to the data clustering algorithms GBIF deploys.

Capturing Data in Rare Disease Registries to Support Regulatory Decision Making: A Survey Study Among Industry and Other Stakeholders.

IntroductionIn rare diseases, registry-based studies can be used to provide natural history data pre-approval and complement drug efficacy and/or safety knowledge post-approval.ObjectiveThe objective of this study was to investigate the opinion of stakeholders about key aspects of rare disease registries that are used to support regulatory decision making and to compare the responses of employees from industry to other stakeholders.MethodsA web-based survey was used to gauge the importance of (1) common data elements (including safety outcomes), (2) data quality and (3) governance aspects that are generic across different rare diseases. The survey included 47 questions. The data were collected in the period April-October 2019.ResultsSeventy-three respondents completed ≥ 80% of the survey. Most of the respondents were from the industry (n = 42, 57%). For safety data, 31 (42%) respondents were in favour of collecting all adverse events. For data quality, the respondents found a level of 30% reasonable for source data verification. For missing data, a level of 20% was considered acceptable. Compared to responders from industry, the other stakeholders found it less relevant to share data with industry and found it less acceptable if the registry is financed by industry.ConclusionsThis study showed that the opinion towards data and governance is well aligned across parties, and issues of data and governance on their own should not pose a barrier to collaboration. This finding is supportive of the European Medicines Agency’s efforts to encourage stakeholders to work with existing registries when collecting data to support regulatory decision making.Supplementary InformationThe online version contains supplementary material available at 10.1007/s40264-021-01081-z.

Evaluation of the EQ-5D-3L and 5L versions in low back pain patients

BackgroundThe EuroQol EQ-5D is one of the most widely researched and applied patient-reported outcome measures worldwide. The original EQ-5D-3L and more recent EQ-5D-5L include three and five response categories respectively. Evidence from healthy and sick populations shows that the additional two response categories improve measurement properties but there has not been a concurrent comparison of the two versions in patients with low back pain (LBP).MethodsLBP patients taking part in a multicenter randomized controlled trial of lumbar total disc replacement and conservative treatment completed the EQ-5D-3L and 5L in an eight-year follow-up questionnaire. The 3L and 5L were assessed for aspects of data quality including missing data, floor and ceiling effects, response consistency, and based on a priori hypotheses, associations with the Oswestry Disability Index (ODI), Pain-Visual Analogue Scales and Hopkins Symptom Checklist (HSCL-25).ResultsAt the eight-year follow-up, 151 (87%) patients were available and 146 completed both the 3L and 5L. Levels of missing data were the same for the two versions. Compared to the EQ-5D-5L, the 3L had significantly higher floor (pain discomfort) and ceiling effects (mobility, self-care, pain/discomfort, anxiety/depression). For these patients the EQ-5D-5L described 73 health states compared to 28 for the 3L. Shannon’s indices showed the 5L outperformed the 3L in tests of classification efficiency. Correlations with the ODI, Pain-VAS and HSCL-25 were largely as hypothesized, the 5L having slightly higher correlations than the 3L.ConclusionThe EQ-5D assesses important aspect of health in LBP patients and the 5L improves upon the 3L in this respect. The EQ-5D-5L is recommended in preference to the 3L version, however, further testing in other back pain populations together with additional measurement properties, including responsiveness to change, is recommended.Trial registration: retrospectively registered: https://clinicaltrials.gov/ct2/show/NCT01704677.

Data Freshness in Ecology and Conservation

Evolving capabilities in environmental data collection, sharing, and processing, are enabling unprecedented use of data from a wide range of sources. Yet data freshness, an important quality dimension associated with the age of data, is a poorly reported aspect of data quality that can lead to additional uncertainty in research findings.

Facilitating harmonized data quality assessments. A data quality framework for observational health research data collections with software implementations in R

BackgroundNo standards exist for the handling and reporting of data quality in health research. This work introduces a data quality framework for observational health research data collections with supporting software implementations to facilitate harmonized data quality assessments.MethodsDevelopments were guided by the evaluation of an existing data quality framework and literature reviews. Functions for the computation of data quality indicators were written in R. The concept and implementations are illustrated based on data from the population-based Study of Health in Pomerania (SHIP).ResultsThe data quality framework comprises 34 data quality indicators. These target four aspects of data quality: compliance with pre-specified structural and technical requirements (integrity); presence of data values (completeness); inadmissible or uncertain data values and contradictions (consistency); unexpected distributions and associations (accuracy). R functions calculate data quality metrics based on the provided study data and metadata and R Markdown reports are generated. Guidance on the concept and tools is available through a dedicated website.ConclusionsThe presented data quality framework is the first of its kind for observational health research data collections that links a formal concept to implementations in R. The framework and tools facilitate harmonized data quality assessments in pursue of transparent and reproducible research. Application scenarios comprise data quality monitoring while a study is carried out as well as performing an initial data analysis before starting substantive scientific analyses but the developments are also of relevance beyond research.

Factors Affecting the Quality of Person-Generated Wearable Device Data and Associated Challenges: Rapid Systematic Review.

BackgroundThere is increasing interest in reusing person-generated wearable device data for research purposes, which raises concerns about data quality. However, the amount of literature on data quality challenges, specifically those for person-generated wearable device data, is sparse.ObjectiveThis study aims to systematically review the literature on factors affecting the quality of person-generated wearable device data and their associated intrinsic data quality challenges for research.MethodsThe literature was searched in the PubMed, Association for Computing Machinery, Institute of Electrical and Electronics Engineers, and Google Scholar databases by using search terms related to wearable devices and data quality. By using PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, studies were reviewed to identify factors affecting the quality of wearable device data. Studies were eligible if they included content on the data quality of wearable devices, such as fitness trackers and sleep monitors. Both research-grade and consumer-grade wearable devices were included in the review. Relevant content was annotated and iteratively categorized into semantically similar factors until a consensus was reached. If any data quality challenges were mentioned in the study, those contents were extracted and categorized as well.ResultsA total of 19 papers were included in this review. We identified three high-level factors that affect data quality—device- and technical-related factors, user-related factors, and data governance-related factors. Device- and technical-related factors include problems with hardware, software, and the connectivity of the device; user-related factors include device nonwear and user error; and data governance-related factors include a lack of standardization. The identified factors can potentially lead to intrinsic data quality challenges, such as incomplete, incorrect, and heterogeneous data. Although missing and incorrect data are widely known data quality challenges for wearable devices, the heterogeneity of data is another aspect of data quality that should be considered for wearable devices. Heterogeneity in wearable device data exists at three levels: heterogeneity in data generated by a single person using a single device (within-person heterogeneity); heterogeneity in data generated by multiple people who use the same brand, model, and version of a device (between-person heterogeneity); and heterogeneity in data generated from multiple people using different devices (between-person heterogeneity), which would apply especially to data collected under a bring-your-own-device policy.ConclusionsOur study identifies potential intrinsic data quality challenges that could occur when analyzing wearable device data for research and three major contributing factors for these challenges. As poor data quality can compromise the reliability and accuracy of research results, further investigation is needed on how to address the data quality challenges of wearable devices.

MTurk, Prolific or Panels? Choosing the Right Audience for Online Research

We examine key aspects of data quality for online behavioral research between selected platforms (Amazon Mechanical Turk, CloudResearch and Prolific) and panels (Qualtrics and Dynata). To identify the key aspects of data quality, we first engaged with the behavioral research community to discover which aspects are most critical to researchers and found these include attention, comprehension, honesty, and reliability. We then explored differences in these data quality aspects in two studies (N~4,000), with or without data quality filters (approval rating). We found considerable differences between audiences, especially in comprehension, attention, and dishonesty. In Study 1 (without filters) we found that only Prolific provided high data quality on all measures. In Study 2 (with filters), we found high data quality among CloudResearch and Prolific. MTurk showed alarmingly low data quality even with data quality filters. We also found that while reputation (approval rating) did not predict data quality, frequency and purpose of usage did, especially on MTurk: the lowest data quality came from MTurk participants that report using the site as their main source of income but spend few hours on it per week. We provide a framework for future investigation into the ever-changing nature of data quality in online research, and how the evolving set of platforms and panels performs on these key aspects.

Data Quality Measures and Efficient Evaluation Algorithms for Large-Scale High-Dimensional Data

Machine learning has been proven to be effective in various application areas, such as object and speech recognition on mobile systems. Since a critical key to machine learning success is the availability of large training data, many datasets are being disclosed and published online. From a data consumer or manager point of view, measuring data quality is an important first step in the learning process. We need to determine which datasets to use, update, and maintain. However, not many practical ways to measure data quality are available today, especially when it comes to large-scale high-dimensional data, such as images and videos. This paper proposes two data quality measures that can compute class separability and in-class variability, the two important aspects of data quality, for a given dataset. Classical data quality measures tend to focus only on class separability; however, we suggest that in-class variability is another important data quality factor. We provide efficient algorithms to compute our quality measures based on random projections and bootstrapping with statistical benefits on large-scale high-dimensional data. In experiments, we show that our measures are compatible with classical measures on small-scale data and can be computed much more efficiently on large-scale high-dimensional datasets.

Data quality-aware genomic data integration

Genomic data are growing at unprecedented pace, along with new protocols, update polices, formats and guidelines, terminologies and ontologies, which are made available every day by data providers. In this continuously evolving universe, enforcing quality on data and metadata is increasingly critical. While many aspects of data quality are addressed at each individual source, we focus on the need for a systematic approach when data from several sources are integrated, as such integration is an essential aspect for modern genomic data analysis. Data quality must be assessed from many perspectives, including accessibility, currency, representational consistency, specificity, and reliability.In this article we review relevant literature and, based on the analysis of many datasets and platforms, we report on methods used for guaranteeing data quality while integrating heterogeneous data sources. We explore several real-world cases that are exemplary of more general underlying data quality problems and we illustrate how they can be resolved with a structured method, sensibly applicable also to other biomedical domains. The overviewed methods are implemented in a large framework for the integration of processed genomic data, which is made available to the research community for supporting tertiary data analysis over Next Generation Sequencing datasets, continuously loaded from many open data sources, bringing considerable added value to biological knowledge discovery.

Collecting Online Survey Data: A Comparison of Data Quality among a Commercial Panel & MTurk

Scholars seeking online data collection have sought the assistance of professional panels, which are commercial entities that recruit prospective research participants. Although not intended for recruiting participants, Amazon’s Mechanical Turk (MTurk) service is a crowdsourcing platform that is being utilized for a lower cost. This study compares data collected from MTurk (N = 760) and a professional panel (N = 1,110) to assess aspects of data quality and respondent features. Results show that MTurk may produce better data quality based on completion rate and success in passing manipulation checks. Participants also differ in terms of educational attainment, age, and preferences of certain devices used for participation. Overall, results provide insight that researchers collecting online data should take into consideration before choosing a data collection platform.