Bias In Surveys Research Articles

Perceived Age Discrimination has Different Effects on Japanese People's Wellbeing Based on Age Groups.

This study elucidates the direct and indirect effects of perceived age discrimination on depressive symptoms and life satisfaction among young, middle-aged, and older adults-through the mediator variables of psychosocial resources and aging stereotypes. Data were collected through a web-based survey conducted in Japan. The total sample size was 2400: 400 men and women in each of the age groups 20 to 39, 40 to 59, and 60 to 79. Our findings indicate that perceived age discrimination is linked depressive symptoms, even among young adults, who report more age discrimination than older adults. Its effect on life satisfaction, however, is stronger among older adults. Across age groups, most of its effect on life satisfaction was explained by both psychosocial resources and aging stereotypes. By contrast, its effect on depressive symptoms did not differ significantly by age group, and was partly explained by the mediators. These differing results may be attributable to selection bias in the web survey.

Bridging Knowledge and Data Gaps in Odonata Rarity: A South Korean Case Study Using Multispecies Occupancy Models and the Rabinowitz Framework

Accurate assessment of species rarity and conservation status requires an approach that integrates data-driven models with established ecological knowledge. In this study, we applied multispecies occupancy (MSO) and latent factor multispecies occupancy (LFMSO) models to estimate the occurrence of 133 Odonata species in South Korea. Using the model outputs, we implemented the Rabinowitz rarity framework to conduct data-based rarity assessments, which were then compared with known ecological information, including geographic ranges, habitat preferences, regional Red List statuses, and citizen science observations. Our findings reveal both alignments and discrepancies between these data-driven rarity assessments and traditional ecological knowledge. For example, species classified as near threatened (NT) or vulnerable (VU) on the regional Red List generally corresponded with high-rarity classifications based on the Rabinowitz framework. However, significant inconsistencies were identified, particularly for certain lentic Odonata species traditionally considered common. These results suggest that spatial biases in field surveys, combined with limited access to data on legally protected species, can impede accurate rarity assessments. These findings underscore the need for standardized survey protocols and improved data-sharing policies for sensitive species to reduce biases and enhance the reliability of rarity assessments. This is essential for effective conservation planning and biodiversity management in freshwater ecosystems.

Public attitudes toward athlete activism in sport: utilizing a multivariate list experiment to address social desirability bias in survey responses

ABSTRACT Research Question: This study investigates the degree of hidden racial hatred towards athlete protests within the Black Lives Matter (BLM) movement by correcting for the Social Desirability Bias (SDB) in survey responses. Research Methods: Employing a multivariate list experiment on a nationally representative sample (n = 1,075), it assesses the extent of SDB and explores variations in support for on-field kneeling protests based on the race of the athlete, respondent demographics, and sport fan identification. Results and Findings: Findings reveal underreported disapproval for BLM-supportive athlete activism, particularly when the protesting athlete is identified as Black. Further analyses highlight hidden biases influenced by respondents’ race, region, political affiliation, and fan identification. Implications: Our findings offer valuable insight on applying list experiments to reduce social desirability biases in surveys, ensuring more accurate reflections of public opinion on contentious topics.

Pervasive biases in proxy genome-wide association studies based on parental history of Alzheimer's disease.

Almost every recent Alzheimer's disease (AD) genome-wide association study (GWAS) has performed meta-analysis to combine studies with clinical diagnosis of AD with studies that use proxy phenotypes based on parental disease history. Here, we report major limitations in current GWAS-by-proxy (GWAX) practices due to uncorrected survival bias and nonrandom participation in parental illness surveys, which cause substantial discrepancies between AD GWAS and GWAX results. We demonstrate that the current AD GWAX provide highly misleading genetic correlations between AD risk and higher education, which subsequently affects a variety of genetic epidemiological applications involving AD and cognition. Our study sheds light on potential issues in the design and analysis of middle-aged biobank cohorts and underscores the need for caution when interpreting genetic association results based on proxy-reported parental disease history.

Analysis of the 24-h biting patterns and human exposures to malaria vectors in south-eastern Tanzania

BackgroundAfrotropical malaria vectors are generally believed to bite nocturnally, leading to the predominant use of insecticide-treated nets (ITNs), which target indoor, nighttime-biting mosquitoes. This focus is reinforced by biases in entomological surveys, which largely overlook daytime mosquito activity. However, recent evidence challenges this paradigm, showing that Anopheles biting can extend way into the daytime, coinciding with human activities at dawn, daytime and evenings, suggesting a broader risk spectrum and potential protection gaps. We have therefore investigated the diurnal and nocturnal biting patterns of the malaria vectors Anopheles arabiensis and Anopheles funestus in south-eastern Tanzania, to better understand the scope of residual transmission and inform strategies for improved control.MethodsHost-seeking mosquitoes were collected hourly using miniaturized double net traps, both indoors and outdoors over 24-h periods between June 2023 and February 2024. Concurrently, human activities indoors and outdoors were monitored half-hourly to correlate with mosquito collections. A structured questionnaire was used to assess household members’ knowledge, perceptions and experiences regarding exposure to mosquito bites during both nighttime and daytime.ResultsNocturnal biting by An. arabiensis peaked between 7 p.m. and 11 p.m. while that of An. funestus peaked later, between 1 a.m. and 3 a.m. Daytime biting accounted for 15.03% of An. arabiensis catches, with peaks around 7–11 a.m. and after 4 p.m., and for 14.15% of An. funestus catches, peaking around mid-mornings, from 10 a.m. to 12 p.m. Nighttime exposure to An. arabiensis was greater outdoors (54.5%), while daytime exposure was greater indoors (80.4%). For An. funestus, higher exposure was observed indoors, both at nighttime (57.1%) and daytime (69%). Plasmodium falciparum sporozoites were detected in both day-biting and night-biting An. arabiensis. Common daytime activities potentially exposing residents during peak biting hours included household chores, eating, sleeping (including due to sickness), resting in the shade or under verandas and playing (children). From evenings onwards, exposures coincided with resting, socializing before bedtime and playtime (children). Nearly all survey respondents (95.6%) reported experiencing daytime mosquito bites, but only 28% believed malaria was transmissible diurnally.ConclusionsThis study updates our understanding of malaria vector biting patterns in south-eastern Tanzania, revealing considerable additional risk in the mornings, daytime and evenings. Consequently, there may be more gaps in the protection provided by ITNs, which primarily target nocturnal mosquitoes, than previously thought. Complementary strategies are needed to holistically suppress vectors regardless of biting patterns (e.g. using larval source management) and to extend personal protection limits (e.g. using repellents). Additionally, community engagement and education on mosquito activity and protective measures could significantly reduce malaria transmission risk.Graphical

Analyzing Potential Non-Ignorable Selection Bias in an Off-Wave Mail Survey Implemented in a Long-Standing Panel Study

Abstract Typical design-based methods for weighting probability samples rely on several assumptions, including the random selection of sampled units according to known probabilities of selection and ignorable unit nonresponse. If any of these assumptions are not met, weighting methods that account for the probabilities of selection, nonresponse, and calibration may not fully account for the potential selection bias in a given sample, which could produce misleading population estimates. This analysis investigates possible selection bias in the 2019 Health Survey Mailer (HSM), a sub-study of the longitudinal Health and Retirement Study (HRS). The primary HRS data collection has occurred in “even” years since 1992, but additional survey data collections take place in the “off-wave” odd years via mailed invitations sent to selected participants. While the HSM achieved a high response rate (83 percent), the assumption of ignorable probability-based selection of HRS panel members may not hold due to the eligibility criteria that were imposed. To investigate this possible non-ignorable selection bias, our analysis utilizes a novel analysis method for estimating measures of unadjusted bias for proportions (MUBP), introduced by Andridge et al. in 2019. This method incorporates aggregate information from the larger HRS target population, including means, variances, and covariances for key covariates related to the HSM variables, to inform estimates of proportions. We explore potential non-ignorable selection bias by comparing proportions calculated from the HSM under three conditions: ignoring HRS weights, weighting based on the usual design-based approach for HRS “off-wave” mail surveys, and using the MUBP adjustment. We find examples of differences between the weighted and MUBP-adjusted estimates in four out of ten outcomes we analyzed. However, these differences are modest, and while this result gives some evidence of non-ignorable selection bias, typical design-based weighting methods are sufficient for correcting for it and their use is appropriate in this case.

The Effectiveness of Tiktok in Increasing Brand Awareness Among Users

Technological advancements have revolutionized marketing strategies, particularly in the realm of social media. As businesses recognize the importance of brand awareness in consumer decision-making, platforms like TikTok have emerged as powerful tools for enhancing brand visibility. Despite its rapid rise in popularity, TikTok's effectiveness in increasing brand awareness remains underexplored. This study investigates "The Effectiveness of TikTok in Increasing Brand Awareness Among Users" with three primary objectives: (RO1) to identify the effect of Mobile Application Values, Information Quality, System Quality, and Influencer Trust on Brand Awareness; (RO2) to identify the effect between Brand Awareness and Word of Mouth; and (RO3) to analyze the effect between Brand Awareness on Purchase Intention. The research adopts a quantitative approach, utilizing a questionnaire distributed to B2C TikTok users. A sample size of 343 respondents was determined using stratified sampling. Descriptive and regression analyses were conducted to interpret the data. The results indicate that Mobile Application Value, System Quality, and Influencer Trust significantly influence Brand Awareness on TikTok, while Information Quality does not exhibit a significant effect. Moreover, Brand Awareness positively impacts both Word of Mouth and Purchase Intention, underscoring its pivotal role in consumer behavior. This study contributes to a deeper understanding of TikTok's potential as a branding tool and offers practical insights for marketers seeking to leverage the platform effectively. Despite limitations in sample size and survey biases, the findings provide a foundation for future research and the refinement of marketing strategies in the dynamic landscape of social media.

A stableness of resistance model for nonresponse adjustment with callback data

Abstract Nonresponse arises frequently in surveys, and follow-ups are routinely made to increase the response rate. In order to monitor the follow-up process, callback data have been used in social sciences and survey studies for decades. In modern surveys, the availability of callback data is increasing because the response rate is decreasing, and follow-ups are essential to collect maximum information. Although callback data are helpful to reduce the bias in surveys, such data have not been widely used in statistical analysis until recently. We propose a stableness of resistance assumption for nonresponse adjustment with callback data. We establish the identification and the semiparametric efficiency theory under this assumption, and propose a suite of semiparametric estimation methods including doubly robust estimators, which generalize existing parametric approaches for callback data analysis. We apply the approach to a Consumer Expenditure Survey dataset. The results suggest an association between nonresponse and high housing expenditures.

Are mobile phones predominant in dual-frame computer assisted telephone interviewing surveys?

The effectiveness of Computer Assisted Telephone Interviewing (CATI) surveys relying on fixed telephone numbers has been eroded by the rise in mobile phone usage and the growing population reachable exclusively through mobile phones. As a result, telephone survey methodologists have had to come up with research designs capable of dealing with the coverage bias in CATI surveys caused by incomplete fixed phone directories. The methodological and operational issues of dual-frame surveys, namely those combining fixed phones and mobile phones, have been extensively investigated, but a comprehensive overview of the magnitude of mobile phone use in CATI surveys is lacking. The present study explores this research gap by analysing data from 675 CATI surveys conducted in European countries. Although a generalised scenario of single frame mobile phone designs is not yet visible, the predominance of mobile phone units over fixed phone units in dual-frame CATI surveys is evident. The sampling method and the target population are the survey-related features most strongly associated to the percentage of mobile phone units in the sample. Survey organisations in countries which continue to have a high number of fixed phone subscriptions tend to give less weight to mobile phone frames in their dual-frame designs.

A topic model analysis of students' gendered expectations using surveyed critiques of lecturers

Student evaluations of teaching (SET) have been examined over the years to better understand the student experience, with an increasing portion of literature exploring the presence of implicit bias in SET surveys against minority gender and ethnic lecturers. This study explores free-text comments made by students from a large public university in Australia over the period 2010–2016, using a semi-supervised statistical approach. Data were collected via surveys administered online at the end of each course to every student officially enrolled in that course via the learning management system, and completion of the surveys was voluntary. We build a probabilistic topic model which incorporates student and lecturer characteristics into the topic formation process. We make statistical inference on the effects of gender and cultural or language backgrounds based on the topic and word prevalence probabilities. The results showed clear separation of topics discussed between male and female lecturers. From a gendered perspective, our topic analyses have found that students are significantly more likely to critique female lecturers to improve on structural aspects of the course, as well as aspects of time management and control of the lecturing environment. In comparison, male lecturers were significantly more likely to be critiqued on specific aspects related to lecture delivery. Lecturers from non-English speaking backgrounds were more likely to be both critiqued and praised for the clarity of their delivery.

Casting light out of shadows: Assessing the coverage of data on endemic firefly distribution in the Atlantic Forest

The Atlantic Forest holds the greatest diversity of fireflies on Earth. Although the knowledge about lampyrids in the Atlantic Forest has increased in the last 10 years thanks to the efforts of new taxonomists and ecologists, both the total number of species and their geographic ranges remain largely unknown. Furthermore, the Atlantic Forest has been facing intense deforestation since the XV century, and now only 23 % of the natural vegetation cover remains, resulting in high vulnerability of its endemic species. Here, we assess the quality and coverage of the information available on the distribution of Lampyridae fireflies; i.e. the extent of the so-called Wallacean shortfall. To do this, we used a database that comprises 3010 records with geospatial and taxonomic information to evaluate the overall coverage of the geographical and environmental distribution of fireflies in the Atlantic Forest. Our assessment shows that the knowledge of Atlantic Forest's firefly diversity and distribution presents significant spatial and climatic biases. Firefly species remain largely under-collected in the Atlantic Forest; only around 16 % of its grid cells at a 50 km scale are well-sampled, and most of them are located in the Serra do Mar bioregion. The location of well-sampled cells is conditioned by recorders’ survey bias and is particularly biased towards protected areas, large forest remnants, the proximity to institutions holding entomological collections, and light-polluted sites surrounding urban spaces. Our findings also highlight that a considerable number of firefly species may remain undiscovered, which adds to the lack of reliable information on the geographic location of many valid species. Based on this assessment, we further point to areas where more sampling effort would provide maximum benefits in terms of geographical and environmental coverage, thus maximizing its value for biodiversity knowledge and conservation.

Paired comparisons with quiet surface drones show evidence of fish behavioral response to motorized vessels during acoustic surveys in Lake Superior

Acoustic surveys are important for fish stock assessments, but fish responses to survey vessels can bias acoustic estimates. We leveraged quiet uncrewed surface vessels (USVs) to characterize potential bias in acoustic surveys. Five conventional motorized ships overtook USVs from astern over 2 km transects at night in Lake Superior in 2022. We examined the difference in acoustic backscatter, average target depth, and average target strength (TS) between USV and motorized vessels. Although sound level measurements from the motorized vessels sometimes exceeded recommendations for scientific vessels, we did not detect differences in acoustic measures among survey vessels. However, the USVs recorded 2 dB higher acoustic backscatter and TS than motorized vessels, leading to ∼15% higher fish densities with drones when using in situ TS and echo integration. Differences in fish density would increase to 30%–60% if a standard TS value was applied. Target depth did not differ between USVs and motorized ships. These results are consistent with a change in orientation but not depth of insonified fish and limited horizontal avoidance of motorized survey vessels.

Density matters: How population dynamics of house mice (Mus musculus) inform the epidemiology of Leptospira

Abstract Rodents are maintenance hosts of numerous pathogens, and both their density and the pathogen prevalence determine the risk they pose to other animals or humans. However, density is often overlooked. We investigated a capture‐mark‐recapture‐sampling strategy to study introduced mice (Mus musculus) and Leptospira as a model and demonstrate the advantages of a combined approach. We estimated population density and Leptospira prevalence in mice in a replicated longitudinal survey conducted between 2016 and 2018. Capture‐mark‐recapture sessions were undertaken at two sites in Spring and Autumn and blood and kidney samples were collected at the end of each session. Mouse density and areas of activity were estimated using spatially explicit capture–recapture (SECR) models and both were compared between Leptospira positive and negative mice. Leptospira exposure and shedding status were estimated using Microscopic Agglutination Test, and a combination of culture and lipL32 PCR on kidneys. Leptospira prevalence was higher in spring (83%–86%) than in autumn (31%–37%) and mouse densities simultaneously varied from 3.6 to 55.9/ha. However, despite these variations in prevalence and density, the density of infected animals remained relatively constant over time (3–8/ha). Shedding or being seropositive was also associated with the activity of mice. Shedding or seropositive mice had a larger activity area, and seropositive mice were trapped on average 1 day earlier than seronegative mice. Synthesis and applications: Our results show how understanding the population dynamics of pathogen‐carrying rodents is critical in epidemiology. The wider movement patterns and easier encounters of positive mice highlight the possibility of biases in classical prevalence surveys and have implications for disease transmission within and between species. Importantly, and quite counter‐intuitively, Leptospira prevalence was negatively associated with mouse density, resulting in a constant density of shedders that contradicts the conventional view of higher exposure risk at high rodent density. More broadly, such sampling designs can improve animal and disease control policies and better inform modelling studies by providing more parameter estimates than classical prevalence surveys.

Simultaneous alleviation of verification and reference standard biases in a community-based tuberculosis screening study using Bayesian latent class analysis.

Estimation of prevalence and diagnostic test accuracy in tuberculosis (TB) prevalence surveys suffer from reference standard and verification biases. The former is attributed to the imperfect reference test used to bacteriologically confirm TB disease. The latter occurs when only the participants screening positive for any TB-compatible symptom or chest X-ray abnormality are selected for bacteriological testing (verification). Bayesian latent class analysis (LCA) alleviates the reference standard bias but suffers verification bias in TB prevalence surveys. This work aims to identify best-practice approaches to simultaneously alleviate the reference standard and verification biases in the estimates of pulmonary TB prevalence and diagnostic test performance in TB prevalence surveys. We performed a secondary analysis of 9869 participants aged ≥15 years from a community-based multimorbidity screening study in a rural district of KwaZulu-Natal, South Africa (Vukuzazi study). Participants were eligible for bacteriological testing using Xpert Ultra and culture if they reported any cardinal TB symptom or had an abnormal chest X-ray finding. We conducted Bayesian LCA in five ways to handle the unverified individuals: (i) complete-case analysis, (ii) analysis assuming the unverified individuals would be negative if bacteriologically tested, (iii) analysis of multiply-imputed datasets with imputation of the missing bacteriological test results for the unverified individuals using multivariate imputation via chained equations (MICE), and simultaneous imputation of the missing bacteriological test results in the analysis model assuming the missing bacteriological test results were (iv) missing at random (MAR), and (v) missing not at random (MNAR). We compared the results of (i)-(iii) to the analysis based on a composite reference standard (CRS) of Xpert Ultra and culture. Through simulation with an overall true prevalence of 2.0%, we evaluated the ability of the models to alleviate both biases simultaneously. Based on simulation, Bayesian LCA with simultaneous imputation of the missing bacteriological test results under the assumption that the missing data are MAR and MNAR alleviate the reference standard and verification biases. CRS-based analysis and Bayesian LCA assuming the unverified are negative for TB alleviate the biases only when the true overall prevalence is <3.0%. Complete-case analysis produced biased estimates. In the Vukuzazi study, Bayesian LCA with simultaneous imputation of the missing bacteriological test results under the MAR and MNAR assumptions produced overall PTB prevalence of 0.9% (95% Credible Interval (CrI): 0.6-1.9) and 0.7% (95% CrI: 0.5-1.1) respectively alongside realistic estimates of overall diagnostic test sensitivity and specificity with substantially overlapping 95% CrI. The CRS-based analysis and Bayesian LCA assuming the unverified were negative for TB produced 0.7% (95% CrI: 0.5-0.9) and 0.7% (95% CrI: 0.5-1.2) overall PTB prevalence respectively with realistic estimates of overall diagnostic test sensitivity and specificity. Unlike CRS-based analysis, Bayesian LCA of multiply-imputed data using MICE mitigates both biases. The findings demonstrate the efficacy of these advanced techniques in alleviating the reference standard and verification biases, enhancing the robustness of community-based screening programs. Imputing missing values as negative for bacteriological tests is plausible under realistic assumptions.

Improving prevalence estimates of mental health and well-being indicators among sexual minority men: a propensity-weighting approach

Abstract The prevalence and relative disparities of mental health outcomes and well-being indicators are often inconsistent across studies of sexual minority men (SMM) due to selection biases in community-based surveys (nonprobability sample), as well as misclassification biases in population-based surveys where some SMM often conceal their sexual orientation identities. The present study estimated the prevalence of mental health related outcomes (depressive symptoms, mental health service use, anxiety) and well-being indicators (loneliness and self-rated mental health) among SMM, broken down by sexual orientation using the adjusted logistic propensity score (ALP) weighting. We applied the ALP to correct for selection biases in the 2019 Sex Now data (a community-based survey of SMMs in Canada) by reweighting it to the 2015-2018 Canadian Community Health Survey (a population survey from Statistics Canada). For all SMMs, the ALP-weighted prevalence of depressive symptoms was 15.96% (95% CI, 11.36%-23.83%), while for mental health service use, it was 32.13% (95% CI, 26.09%-41.20%). The ALP estimates lie in between the crude estimates from the two surveys. This method was successful in providing a more accurate estimate than relying on results from one survey alone. We recommend to the use of ALP on other minority populations under certain assumptions. This article is part of a Special Collection on Mental Health.

ALLEVIATING SURVEYOR BIAS IN REAL ESTATE: AN APPLICATION TO VACANCY AND PROPERTY PRICES

Although vacancies in the real estate market have always been of key interest to both private and public stakeholders, measuring the impact of vacancy rates on property prices is challenging. This study attempts to quantify the extent of the influence of vacancy rates on the price decline of commercial properties. The dataset used in this study was collected and collated by valuation experts. We attempt to alleviate the bias inherent in the price estimates provided by the surveyors. A Bayesian multilevel estimation model was employed, and the results revealed that while some experts do not show deviations from the standard tendency of a peer group, others do diverge from this norm. Based on these findings, a bias-controlling price decline rate was derived. As many real estate statistics are produced based on survey data compiled by experts, the approach adopted in this study is expected to be applied to various real estate practices to mitigate the inherent and inevitable surveyor bias.

Understanding, Detecting, and Adjusting for Response Bias in Surveys

Understanding, Detecting, and Adjusting for Response Bias in Surveys

Nonresponse Bias of Japanese Wage Statistics

This article analyzes unit nonresponse bias of establishment surveys, drawing on two major government wage surveys in Japan. We find that unit nonresponse is prevalent among establishments with fewer employees, in urban prefectures, and operating in the service industry. Further, while low-wage establishments are less likely to respond, the resulting bias is quantitatively negligible. Regarding establishment sampling of employees, we find that the sampled establishments randomly choose workers in an appropriate manner. Overall, with proper weighting, nonresponse bias is negligible in Japanese wage statistics to the extent that the mean estimate is concerned. In addition to nonresponse bias, we also assess potential bias due to limited coverage of the sampling population and find substantial bias at the tails of the wage distribution. In particular, excluding corporate executives and establishments with few employees results in an underestimation of wage inequality.

Quiet uncrewed surface vessels assess fish avoidance to motorized survey ships with varying noise levels in the Great Lakes

Acoustic surveys are a foundational component of many fisheries monitoring programs because they allow assessment of spatially extensive stocks. They are widely used to evaluate prey fish throughout the Great Lakes by numerous coordinating vessels. Traditionally, these surveys have been conducted by crewed and motorized vessels, but fish avoidance of motorized platforms has been reported in multiple studies and may bias survey estimates. Quiet uncrewed platforms are becoming increasingly available and offer the opportunity to explore bias in traditional surveys. Several identical quiet uncrewed surface vehicles (USVs) operated by Saildrone were equipped with 120 kHz Simrad EK80 transducers and deployed in Lakes Erie, Huron, Michigan, and Superior in the summers of 2021, 2022, and 2023. The USVs were overtaken by numerous motorized vessels over 2 km transects using transducers with the same frequency. Fishes showed a limited response to approaching vessels, and acoustic measures were similar between the USVs and motorized vessels. Therefore, acoustic surveys in the Great Lakes appear unbiased and are widely comparable. Findings from this work will inform interpretation of acoustic data in the Great Lakes and provide the largest scale testing of fish avoidance during acoustic surveys to date.

The Internet Usage Rate in Turkey: A Machine Learning Approach

Most studies on measuring coverage bias in internet surveys use internet access as a critical measurement variable. However, access to the internet does not mean that individuals are using it. Therefore, using the internet usage rate as a key variable is crucial to get an accurate overview of the internet coverage of a population. This study closes these gaps by using a better indicator for measuring the internet usage rate. It is the first study measuring the internet usage rate in Turkey by using the real internet usage rate of the population and applying a machine learning algorithm. The results exposed significant differences in socio-demographic characteristics when internet users were compared with non-users. Furthermore, the coverage bias associated with internet users remained different for several demographic categories. The results of web-based surveys based on the actual internet usage rate are crucial for the scientific community and marketers.