Web surveys have replaced Face-to-Face and computer assisted telephone interviewing (CATI) as the main mode of data collection in most countries. This trend was reinforced as a consequence of COVID-19 pandemic-related restrictions. However, this mode still faces significant limitations in obtaining probability-based samples of the general population. For this reason, most web surveys rely on nonprobability survey designs. Whereas probability-based designs continue to be the gold standard in survey sampling, nonprobability web surveys may still prove useful in some situations. For instance, when small subpopulations are the group under study and probability sampling is unlikely to meet sample size requirements, complementing a small probability sample with a larger nonprobability one may improve the efficiency of the estimates. Nonprobability samples may also be designed as a mean for compensating for known biases in probability-based web survey samples by purposely targeting respondent profiles that tend to be underrepresented in these surveys. This is the case in the Survey on the impact of the COVID-19 pandemic in Spain (ESPACOV) that motivates this paper. In this paper, we propose a methodology for combining probability and nonprobability web-based survey samples with the help of machine-learning techniques. We then assess the efficiency of the resulting estimates by comparing them with other strategies that have been used before. Our simulation study and the application of the proposed estimation method to the second wave of the ESPACOV Survey allow us to conclude that this is the best option for reducing the biases observed in our data.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9538074PMC
http://dx.doi.org/10.1002/bimj.202200035DOI Listing

Publication Analysis

Top Keywords

survey samples
12
web surveys
12
probability nonprobability
8
survey
8
nonprobability survey
8
machine-learning techniques
8
survey impact
8
impact covid-19
8
covid-19 pandemic
8
pandemic spain
8

Similar Publications

Self -reported bruxism in patients undergoing Orthodontic treatment - a cross-sectional study.

BMC Oral Health

December 2024

Department of Surgery, Program Director Orthodontics Residency Program, Section of Dentistry, The Aga Khan University Hospital, Stadium Road, P.O Box 3500, Karachi, 74800, Pakistan.

Background: Orthodontic treatment involves manipulation of tooth position that can temporarily lead to occlusal interferences capable of inducing bruxism. The objectives of this study were to determine the self-reported incidence of bruxism and its association with factors such as facial type, divergence, skeletal and dental malocclusions in orthodontic patients.

Materials And Methods: 80 patients who underwent a minimum of six months of orthodontic treatment were surveyed using a validated self-reported bruxism questionnaire for the presence or absence of bruxism.

View Article and Find Full Text PDF

Introduction: It is estimated that 1.3 billion people in the world have hypertension and a large proportion of them are unaware. Waist circumference has emerged as Potential predictor of Cardiovascular Diseases (CVD) risk; however, fewer studies in Tanzania have evaluated its role in screening for CVD risk.

View Article and Find Full Text PDF

Background: The COVID-19 pandemic resulted in the rapid implementation of telemedicine for HIV care at federally qualified health centers (FQHCs) in the United States. We sought to understand use of telemedicine (telephone and video) at two FQHCs in Los Angeles, and the client attitudes towards and experiences with telemedicine as part of future HIV care.

Methods: We conducted surveys with 271 people living with HIV (PLHIV), with questions covering sociodemographic factors, telemedicine attitudes and experiences, technological literacy, and access to technological resources and privacy.

View Article and Find Full Text PDF

Introduction: Labor and delivery (L&D) nurses are often responsible for initiating the critical first steps of the Neonatal Resuscitation Program (NRP). Identification of knowledge gaps and underlying reasons for NRP non-adherence is crucial for designing educational interventions.

Methods: A convenience sample of 37 L&D nurses from two delivery hospitals were enrolled.

View Article and Find Full Text PDF

Cardiovascular risk markers in apparently healthy young adults: Evaluation according to optimal or non-optimal office blood pressure.

Hipertens Riesgo Vasc

December 2024

Facultad de Ciencias Médicas, Universidad Nacional de La Plata, Argentina; Centro de Investigaciones Cardiovasculares "Dr. Horacio E. Cingolani" CONICET-UNLP, Argentina. Electronic address:

Background: Blood pressure (BP) is linearly related to the incidence of cardiovascular disease from values as low as 115/75mmHg, even at young ages. A particularly concerning issue is the decrease representation of optimal BP among children and youth. The mechanisms by which minimal elevations in BP increase cardiovascular risk are not defined.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!