A national health survey conducted exclusively online will miss the populations most affected by health disparities: older adults without smartphones, rural residents without broadband, low-income families without reliable internet, and people experiencing homelessness with no fixed address for email invitations. These are not edge cases. They represent hundreds of millions of people worldwide.

The term "hard-to-reach" in survey methodology refers to populations that are difficult to sample, contact, or engage using standard survey approaches. For online surveys, any population without consistent internet access is hard to reach by definition. The consequences for data quality are severe: coverage bias that no statistical adjustment can fully correct.

Here is what published research and field practice say about reaching these populations effectively.

Who Gets Missed by Online Surveys

The Pew Research Center (2024) reports that approximately 6% of U.S. adults do not use the internet at all. This rises to 13% among adults aged 65 and older and 17% among adults with less than a high school education. Among adults in households earning less than $30,000 per year, 12% are offline entirely.

Globally, the International Telecommunication Union (2023) estimates that 2.6 billion people lack internet access. In sub-Saharan Africa, internet penetration is approximately 36%. In South Asia, it is approximately 48%.

These are not evenly distributed gaps. The populations excluded by online-only surveys differ systematically from those included. They are older, poorer, less educated, more rural, and more likely to belong to racial and ethnic minorities. When your survey misses these groups, your data does not just have a smaller sample. It has a biased sample.

Groves (2006), in a foundational article published in Public Opinion Quarterly, demonstrated that nonresponse bias is a function not just of who fails to respond but of how nonrespondents differ from respondents on the variables being measured. If health surveys miss the unhealthiest populations, the bias is directly on the outcome of interest.

Older Adults and Residential Care Populations

Survey research with older adults consistently shows higher response rates with paper-based methods. Jorm et al. (2015), in a study of adults aged 65 and older, found that mailed paper surveys achieved response rates 15 to 25 percentage points higher than equivalent online surveys in this age group.

For residents of care facilities, assisted living communities, and nursing homes, paper is often the only feasible option. These settings typically lack individual internet access for residents. Staff-assisted paper survey administration is the standard methodology used by organizations including the Centers for Medicare and Medicaid Services for the CAHPS surveys.

The CAHPS (Consumer Assessment of Healthcare Providers and Systems) program, run by the Agency for Healthcare Research and Quality, uses mailed paper surveys as the primary mode for surveying Medicare beneficiaries. Their methodological research found that paper-only and paper-first mixed-mode designs produced the highest response rates among older enrollees (Elliott et al., 2009).

Rural and Remote Communities

In rural areas of both developing and developed countries, internet access is inconsistent, mobile signal is unreliable, and the nearest survey administration site may be hours away.

The WHO's STEPS survey methodology (WHO, 2017) for chronic disease risk factor surveillance in low- and middle-income countries uses paper-based data collection as the primary method. Trained enumerators visit households, administer structured questionnaires on paper, and transport completed forms to central processing sites.

UNICEF's Multiple Indicator Cluster Surveys (MICS), which have been conducted in over 100 countries since 1995, use paper questionnaires administered by trained interviewers. The program has collected data from more than 300 surveys, reaching some of the most remote communities in the world.

In developed countries, the U.S. Census Bureau maintains a paper option for the decennial census specifically because online-only collection would miss rural, elderly, and low-connectivity households. The 2020 Census allowed internet response but sent paper questionnaires to all non-responding households as follow-up (U.S. Census Bureau, 2021).

Refugee and Displaced Populations

Surveying refugee populations presents challenges that are qualitatively different from standard survey research. Respondents may not have fixed addresses, stable phone numbers, or any internet access. They may not speak the language of the host country. They may be wary of providing personal information to organizations they do not trust.

Jacobsen and Landau (2003), writing in the Journal of Refugee Studies, outlined the methodological challenges of refugee research and recommended community-based sampling with in-person paper survey administration. Digital alternatives assume infrastructure and trust that may not exist.

Paper surveys administered by community health workers or trusted local organizations can navigate these barriers. Questionnaires printed in the respondent's language, administered face-to-face by a trained enumerator, and collected in sealed envelopes preserve both accessibility and confidentiality.

PaperSurvey.io supports over 30 languages for form content, making it possible to create survey instruments in Arabic, Dari, Somali, Ukrainian, or any other language needed for refugee populations. All language versions can feed into a single dataset for unified analysis.

Incarcerated Populations

Correctional facilities generally prohibit internet-connected devices for inmates. Survey research with incarcerated populations, which is essential for criminal justice reform, health services research, and reentry program evaluation, must use paper.

The Bureau of Justice Statistics' National Inmate Survey uses paper-based self-administered questionnaires for sensitive topics and interviewer-administered paper instruments for other sections (BJS, 2018). The choice of paper is not preference but necessity.

Confidentiality is particularly important in correctional settings. Paper surveys placed in sealed envelopes by respondents provide a level of perceived privacy that is difficult to replicate with digital tools in a supervised environment.

People Experiencing Homelessness

Point-in-time counts and needs assessments for people experiencing homelessness rely almost entirely on in-person, paper-based survey administration. The U.S. Department of Housing and Urban Development's Annual Homeless Assessment Report methodology uses street-level enumeration with paper forms (HUD, 2023).

These surveys are typically conducted by trained volunteers who approach individuals at shelters, soup kitchens, and known encampment areas. A paper form and a clipboard are the standard tools. There is no email invitation to send, no web link to share, and no assumption of device ownership.

Children and Adolescents in School Settings

For research involving children and adolescents, school-based paper survey administration remains the most effective method. The CDC's Youth Risk Behavior Surveillance System (YRBSS) uses paper-and-pencil self-administered questionnaires in classrooms, achieving response rates above 60% consistently (Brener et al., 2013).

The classroom setting provides a controlled environment where surveys can be administered to large groups simultaneously. Students complete the questionnaire during a class period, place it in an unmarked envelope, and return it to the administrator. This procedure maintains anonymity while achieving near-complete participation among attendees.

Online alternatives administered outside of class time typically achieve response rates of 20-40% for the same populations, creating both response rate and representativeness problems.

Workers Without Desk Jobs

Manufacturing workers, agricultural laborers, construction crews, warehouse staff, delivery drivers, and other non-desk workers are systematically underrepresented in online employee surveys. They may not have company email addresses, dedicated workstations, or time during the workday to access web surveys.

Paper surveys distributed during shift meetings, training sessions, or break periods reach these workers where they are. Completion rates for paper surveys administered in workplace group settings consistently exceed those for email-distributed online surveys in the same organizations (Baruch & Holtom, 2008).

The Paper Workflow for Hard-to-Reach Populations

The challenge with paper survey research in hard-to-reach populations has never been the data collection. It has been getting the data from paper into a usable digital format. Modern OCR eliminates this bottleneck.

With PaperSurvey.io:

1. Design your instrument in the online builder with support for multiple question types and 30+ languages
2. Print on any paper, with any printer, in any quantity
3. Distribute to enumerators, field teams, community workers, or mail directly to respondents
4. Collect completed forms at field sites, no electricity or internet needed during collection
5. Scan completed forms with any scanner or phone camera when connectivity is available
6. Upload via browser, email, or Dropbox for automatic processing
7. Export structured data to Excel, CSV, or SPSS for analysis

The technology handles the data entry. Your field team focuses on reaching the people who matter.

References

• Baruch, Y., & Holtom, B. C. (2008). Survey response rate levels and trends in organizational research. Human Relations, 61(8), 1139-1160.
• Brener, N. D., Kann, L., Shanklin, S., et al. (2013). Methodology of the Youth Risk Behavior Surveillance System. Morbidity and Mortality Weekly Report, 62(RR-1), 1-20.
• Bureau of Justice Statistics. (2018). National Inmate Survey: Survey Methodology. U.S. Department of Justice.
• Elliott, M. N., Zaslavsky, A. M., Goldstein, E., et al. (2009). Effects of survey mode, patient mix, and nonresponse on CAHPS hospital survey scores. Health Services Research, 44(2p1), 501-518.
• Groves, R. M. (2006). Nonresponse rates and nonresponse bias in household surveys. Public Opinion Quarterly, 70(5), 646-675.
• HUD. (2023). Annual Homeless Assessment Report to Congress. U.S. Department of Housing and Urban Development.
• International Telecommunication Union. (2023). Facts and Figures: Focus on Least Developed Countries.
• Jacobsen, K., & Landau, L. B. (2003). The dual imperative in refugee research. Journal of Refugee Studies, 16(2), 185-205.
• Jorm, L. R., et al. (2015). Participation in health surveys by older adults. BMC Medical Research Methodology, 15, 104.
• Pew Research Center. (2024). Internet/Broadband Fact Sheet.
• U.S. Census Bureau. (2021). 2020 Census Operational Quality Metrics.
• WHO. (2017). STEPwise Approach to NCD Risk Factor Surveillance (STEPS). World Health Organization.

Start your free trial and design a multilingual survey for your next field project.