The use of personal digital assistants for data entry at the point of collection in a large household survey in southern Tanzania

Shirima, Kizito; Mukasa, Oscar; Schellenberg, Joanna Armstrong; Manzi, Fatuma; John, Davis; Mushi, Adiel K; Mrisho, Mwifadhi; Tanner, Marcel; Mshinda, Hassan; Schellenberg, David

doi:10.1186/1742-7622-4-5

Cited by 109 publications

(118 citation statements)

References 2 publications

Supporting

Mentioning

115

Contrasting

Unclassified

Order By: Relevance

“…In other evaluations, respondents tended to have accepted the new methods [8,23], but some concerns were raised over ‘outsider’ technology [4,14], and one study found a higher refusal rate with EDC [15]; the respondents we spoke to had no reservations regarding the technology. This acceptance in our setting may be due to the long-running nature of the project and good relationships with the community.…”

Section: Discussionmentioning

confidence: 99%

“…Other commentators also reported that users tended to prefer EDC over paper [8–13] and that interviewers without prior computing experience readily adapted to tablet use [2,8,23]. The experience has been less positive for data processing staff who feared for employment security, especially those without the higher level of skills and experience required to work on EDC.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Implementing electronic data capture at a well-established health and demographic surveillance site in rural northern Malawi

McLean

Dube

Saul

et al. 2017

Global Health Action

View full text Add to dashboard Cite

This article aims to assess multiple issues of resources, staffing, local opinion, data quality, cost, and security while transitioning to electronic data collection (EDC) at a long-running community research site in northern Malawi. Levels of missing and error fields, delay from data collection to availability, and average number of interviews per day were compared between EDC and paper in a complex, repeated annual household survey. Three focus groups with field and data staff with experience using both methods, and in-depth interviews with participants were carried out. Cost for each method were estimated and compared. Missing data was more common on paper questionnaires than on EDC, and a similar number were carried out per day. Fieldworkers generally preferred EDC, but data staff feared for their employment. Most respondents had no strong preference for a method. The cost of the paper system was estimated to be higher than using EDC. The existing infrastructure and technical expertise could be adapted to using EDC, but changes have an impact on data processing jobs as fewer, and better qualified staff are required. EDC is cost-effective, and, for a long-running site, may offer further savings, as devices can be used in multiple studies and perform several other functions. EDC is accepted by fieldworkers and respondents, has good levels of quality and timeliness, and security can be maintained. EDC is well-suited for use in a well-established research site using and developing existing infrastructure and expertise.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Implementing electronic data capture at a well-established health and demographic surveillance site in rural northern Malawi

McLean

Dube

Saul

et al. 2017

Global Health Action

View full text Add to dashboard Cite

show abstract

“…Responses to questions about the child's health, vaccination and IPTi status and any admissions, including those associated with a rash, were directly entered into personal digital assistants (PDA), as described previously. 25 These data were used to estimate the coverage of IPTi during 2007, powered to detect a coverage of 50% to within five percentage points, as well as the incidence of hospital admission with a rash as part of the safety evaluation.…”

Section: Rash-related Hospital Admissions In Children <2 Years Of Agementioning

confidence: 99%

“…The training lasted 2 weeks and consisted of detailed questionnaire review, case studies and a 3-day pilot survey. Work lists were produced from the cleaned census data, which had been collected using PDAs, 25 and provided to the VA interviewers within 4 weeks of the original household visit. The flow of VA questionnaires and job lists between the data unit and field was managed by the VA team supervisor whose responsibility it was to conduct quality control checks.…”

Section: Rash-related Deaths In Children Aged 2-11 Monthsmentioning

confidence: 99%

Safety of sulfadoxine/pyrimethamine for intermittent preventive treatment of malaria in infants: evidence from large-scale operational research in southern Tanzania

Maokola

Chemba

Hamisi

et al. 2011

International Health

Self Cite

View full text Add to dashboard Cite

“…In 1991, Forster et al [15] in the Gambia were among the first to apply digital survey methods. They reported a 31% shorter enumeration time using electronic compared with paper questionnaires, and later in Tanzania, in 2005, a personal digital assistant (PDA) model of electronic devices was shown to achieve over 99% data completeness in a cross-sectional survey of over 21,000 rural households in southern Tanzania [16]. Initial analyses were possible within 24 hours after the last day of the survey, and there were no noteworthy experiences of device-related problems or data loss.…”

Section: Introductionmentioning

confidence: 99%

Do surveys with paper and electronic devices differ in quality and cost? Experience from the Rufiji Health and demographic surveillance system in Tanzania

Mukasa

Mushi

Maire

et al. 2017

Global Health Action

Self Cite

View full text Add to dashboard Cite

Background: Data entry at the point of collection using mobile electronic devices may make data-handling processes more efficient and cost-effective, but there is little literature to document and quantify gains, especially for longitudinal surveillance systems. Objective: To examine the potential of mobile electronic devices compared with paper-based tools in health data collection. Methods: Using data from 961 households from the Rufiji Household and Demographic Survey in Tanzania, the quality and costs of data collected on paper forms and electronic devices were compared. We also documented, using qualitative approaches, field workers, whom we called ‘enumerators’, and households’ members on the use of both methods. Existing administrative records were combined with logistics expenditure measured directly from comparison households to approximate annual costs per 1,000 households surveyed. Results: Errors were detected in 17% (166) of households for the paper records and 2% (15) for the electronic records (p < 0.001). There were differences in the types of errors (p = 0.03). Of the errors occurring, a higher proportion were due to accuracy in paper surveys (79%, 95% CI: 72%, 86%) compared with electronic surveys (58%, 95% CI: 29%, 87%). Errors in electronic surveys were more likely to be related to completeness (32%, 95% CI 12%, 56%) than in paper surveys (11%, 95% CI: 7%, 17%).The median duration of the interviews (‘enumeration’), per household was 9.4 minutes (90% central range 6.4, 12.2) for paper and 8.3 (6.1, 12.0) for electronic surveys (p = 0.001). Surveys using electronic tools, compared with paper-based tools, were less costly by 28% for recurrent and 19% for total costs. Although there were technical problems with electronic devices, there was good acceptance of both methods by enumerators and members of the community. Conclusions: Our findings support the use of mobile electronic devices for large-scale longitudinal surveys in resource-limited settings.

show abstract

The use of personal digital assistants for data entry at the point of collection in a large household survey in southern Tanzania

Cited by 109 publications

References 2 publications

Implementing electronic data capture at a well-established health and demographic surveillance site in rural northern Malawi

Implementing electronic data capture at a well-established health and demographic surveillance site in rural northern Malawi

Safety of sulfadoxine/pyrimethamine for intermittent preventive treatment of malaria in infants: evidence from large-scale operational research in southern Tanzania

Do surveys with paper and electronic devices differ in quality and cost? Experience from the Rufiji Health and demographic surveillance system in Tanzania

Contact Info

Product

Resources

About