Cutting the costs of attrition: Results from the Indonesia Family Life Survey

Thomas, Duncan; Witoelar, Firman; Frankenberg, Elizabeth; Sikoki, Bondan; Strauss, John; Sumantri, Cecep; Suriastini, Wayan

doi:10.1016/j.jdeveco.2010.08.015

Cited by 101 publications

(62 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Family Life Survey, Kagera Health and Development Survey, and Kenya Life Panel Survey all successfully reinterview more than 80% of the target respondents at least ten years after the baseline (Baird et al, Forthcoming;Beegle et al, 2011;Thomas et al, 2012). However, these studies use tracking protocols that are unlikely to be feasible in a high-frequency panel.…”

Section: Benchmarks For Missed Interviewsmentioning

confidence: 99%

“…However, these studies use tracking protocols that are unlikely to be feasible in a high-frequency panel. Each round of these surveys may take several months to complete, with several more month spent tracking difficult-to-locate respondents or respondents who have migrated (Thomas et al, 2012). High-frequency panels inherently have a short timeframe for completing each interview, increasing the risk of missed interviews.…”

Section: Benchmarks For Missed Interviewsmentioning

confidence: 99%

See 1 more Smart Citation

Call Me Maybe: Experimental Evidence on Using Mobile Phones to Survey Microenterprises

2016

View full text Add to dashboard Cite

High-frequency data is useful to measure volatility, reduce recall bias, and measure dynamic treatment effects. We conduct the first experimental evaluation of high-frequency phone surveys in a developing country or with microenterprises. We randomly assign microenterprise owners to monthly in-person, weekly inperson, or weekly phone interviews. We find high-frequency phone surveys are useful and accurate. Phone and in-person surveys yield similar measurements, with few large or significant differences in reported outcome means or distributions. Neither interview frequency nor medium affects reported outcomes in a common in-person endline. Phone surveys reduce costs without increasing permanent attrition from the panel. * We are grateful for helpful comments from

show abstract

Section: Benchmarks For Missed Interviewsmentioning

confidence: 99%

Section: Benchmarks For Missed Interviewsmentioning

confidence: 99%

Call Me Maybe: Experimental Evidence on Using Mobile Phones to Survey Microenterprises

2016

View full text Add to dashboard Cite

show abstract

“…A potential worry associated with attrition is that there might be systematic differences between those who are reinterviewed and those who attrit. As Thomas et al (2012) 5 This assumption can be relaxed. Time-varying measurement errors are discussed in the instrument validity section (in Section 6.1).…”

Section: Data and Variablesmentioning

confidence: 99%

“…Summary statistics for all control variables are reported in the appendix. IFLS is not fielded Thomas et al (2012) find that attrition is selected on not only age, location and gender but also socio-economic characteristics in IFLS. They show that respondents who were found, but tracked over long distances resemble more attriters who were never found, than they do respondents who were found but never moved.…”

Section: Data and Variablesmentioning

confidence: 99%

The dynamics of health and its determinants among the elderly in developing countries

Kim

2015

Economics & Human Biology

View full text Add to dashboard Cite

“…Loss of cases throughout the study period is a common problem in longitudinal studies (Aneshensel et al 1989;Thomas et al 2012). However, if they could not be found in the second interview, the chance of finding them for the third interview reduced significantly and, therefore, potentially very valuable data would be completely lost.…”

mentioning

confidence: 99%

Rethinking late and lost to follow-up participants: the New Zealand youth transitions study

Thimasarn-Anwar¹,

Sanders²,

Munford³

et al. 2013

Journal of Youth Studies

View full text Add to dashboard Cite

This article examines the demographic characteristics of youth who were interviewed on time, located late or lost for a second interview in a three-year study of vulnerable adolescents called the New Zealand youth transitions study. It reports on the results of an analysis to determine whether or not the considerable effort invested in trying to locate these youth three times made a difference to the quality of the data. Chi-Square test, Fisher's exact test and Multivariate Analysis of Variance (MANOVA) were used to analyse the relationship between completion times, demographic factors and key variables of interest to the study (risk, resilience, functional outcomes, service-use history and service quality). Results revealed that both the key variables of interest and demographic factors significantly varied across completion times. Double interaction by age and completion time emerged across specific sub-scales that represented risk and service-use history. This suggests that the effort invested in locating youth was valuable in terms of reducing bias in results and that suitable techniques of handling 'late and lost to follow-up' data are required in order to fully understand the key factors influencing the lives of vulnerable youth.

show abstract

Cutting the costs of attrition: Results from the Indonesia Family Life Survey

Cited by 101 publications

References 22 publications

Call Me Maybe: Experimental Evidence on Using Mobile Phones to Survey Microenterprises

Call Me Maybe: Experimental Evidence on Using Mobile Phones to Survey Microenterprises

The dynamics of health and its determinants among the elderly in developing countries

Rethinking late and lost to follow-up participants: the New Zealand youth transitions study

Contact Info

Product

Resources

About