The power of online panel paradata to predict unit nonresponse and voluntary attrition in a longitudinal design

The objective of this study is to identify factors affecting participation rates, i.e., nonresponse and voluntary attrition rates, and their predictive power in a probability-based online panel. Participation for this panel had already been investigated in the literature according to the socio-demographic and socio-psychological characteristics of respondents and different types of paradata, such as device type or questionnaire navigation, had also been explored. In this study, the predictive power of online panel participation paradata was instead evaluated, which was expected (at least in theory) to offer even more complex insight into respondents' behavior over time. This kind of paradata would also enable the derivation of longitudinal variables measuring respondents' panel activity, such as survey outcome rates and consecutive waves with a particular survey outcome prior to a wave (e.g., response, noncontact, refusal), and could also be used in models controlling for unobserved heterogeneity. Using the Life in Australia™ participation data for all recruited members for the first 30 waves, multiple linear, binary logistic and panel random-effect logit regression analyses were carried out to assess socio-demographic and online panel paradata predictors of nonresponse and attrition that were available and contributed to the accuracy of prediction and the best statistical modeling. The proposed approach with the derived paradata predictors and random-effect logistic regression proved to be reasonably accurate for predicting nonresponse-with just 15 waves of online panel paradata (even without sociodemographics) and logit random-effect modeling almost four out of five nonrespondents could be correctly identified in the subsequent wave.

Modelling attrition and nonparticipation in a longitudinal study of prostate cancer

Background

Attrition occurs when a participant fails to respond to one or more study waves. The accumulation of attrition over several waves can lower the sample size and power and create a final sample that could differ in characteristics than those who drop out. The main reason to conduct a longitudinal study is to analyze repeated measures; research subjects who drop out cannot be replaced easily. Our group recently investigated factors affecting nonparticipation (refusal) in the first wave of a population-based study of prostate cancer. In this study we assess factors affecting attrition in the second wave of the same study. We compare factors affecting nonparticipation in the second wave to the ones affecting nonparticipation in the first wave.

Methods

Information available on participants in the first wave was used to model attrition. Different sources of attrition were investigated separately. The overall and race-stratified factors affecting attrition were assessed. Kaplan-Meier survival curve estimates were calculated to assess the impact of follow-up time on participation.

Results

High cancer aggressiveness was the main predictor of attrition due to death or frailty. Higher Charlson Comorbidity Index increased the odds of attrition due to death or frailty only in African Americans (AAs). Young age at diagnosis for AAs and low income for European Americans (EAs) were predictors for attrition due to lost to follow-up. High cancer aggressiveness for AAs, low income for EAs, and lower patient provider communication scores for EAs were predictors for attrition due to refusal. These predictors of nonparticipation were not the same as those in wave 1. For short follow-up time, the participation probability of EAs was higher than that of AAs.

Conclusions

Predictors of attrition can vary depending on the attrition source. Examining overall attrition (combining all sources of attrition under one category) instead of distinguishing among its different sources should be avoided. The factors affecting attrition in one wave can be different in a later wave and should be studied separately.

The effect of sociodemographic (mis)match between interviewers and respondents on unit and item nonresponse in Belgium

Interviewer characteristics affect nonresponse and measurement errors in face-to-face surveys. Some studies have shown that mismatched sociodemographic characteristics - for example gender - affect people's behavior when interacting with an interviewer at the door and during the survey interview, resulting in more nonresponse. We investigate the effect of sociodemographic (mis)matching on nonresponse in two successive rounds of the European Social Survey in Belgium. As such, we replicate the analyses of the effect of (mis)matching gender and age on unit nonresponse on the one hand, and of gender, age and education level (mis)matching on item nonresponse on the other hand. Recurring effects of sociodemographic (mis)match are found for both unit and item nonresponse.

A two-phase sampling survey for nonresponse and its paradata to correct nonresponse bias in a health surveillance survey

BACKGROUND

The decline in participation rates in surveys, including epidemiological surveillance surveys, has become a real concern since it may increase nonresponse bias. The aim of this study is to estimate the contribution of a complementary survey among a subsample of nonrespondents, and the additional contribution of paradata in correcting for nonresponse bias in an occupational health surveillance survey.

METHODS

In 2010, 10,000 workers were randomly selected and sent a postal questionnaire. Sociodemographic data were available for the whole sample. After data collection of the questionnaires, a complementary survey among a random subsample of 500 nonrespondents was performed using a questionnaire administered by an interviewer. Paradata were collected for the complete subsample of the complementary survey. Nonresponse bias in the initial sample and in the combined samples were assessed using variables from administrative databases available for the whole sample, not subject to differential measurement errors. Corrected prevalences by reweighting technique were estimated by first using the initial survey alone and then the initial and complementary surveys combined, under several assumptions regarding the missing data process. Results were compared by computing relative errors.

RESULTS

The response rates of the initial and complementary surveys were 23.6% and 62.6%, respectively. For the initial and the combined surveys, the relative errors decreased after correction for nonresponse on sociodemographic variables. For the combined surveys without paradata, relative errors decreased compared with the initial survey. The contribution of the paradata was weak.

CONCLUSION

When a complex descriptive survey has a low response rate, a short complementary survey among nonrespondents with a protocol which aims to maximize the response rates, is useful. The contribution of sociodemographic variables in correcting for nonresponse bias is important whereas the additional contribution of paradata in correcting for nonresponse bias is questionable.

In an occupational health surveillance study, auxiliary data from administrative health and occupational databases effectively corrected for nonresponse

OBJECTIVES

To show how reweighting can correct for unit nonresponse bias in an occupational health surveillance survey by using data from administrative databases in addition to classic sociodemographic data.

STUDY DESIGN AND SETTING

In 2010, about 10,000 workers covered by a French health insurance fund were randomly selected and were sent a postal questionnaire. Simultaneously, auxiliary data from routine health insurance and occupational databases were collected for all these workers. To model the probability of response to the questionnaire, logistic regressions were performed with these auxiliary data to compute weights for correcting unit nonresponse. Corrected prevalences of questionnaire variables were estimated under several assumptions regarding the missing data process. The impact of reweighting was evaluated by a sensitivity analysis.

RESULTS

Respondents had more reimbursement claims for medical services than nonrespondents but fewer reimbursements for medical prescriptions or hospitalizations. Salaried workers, workers in service companies, or who had held their job longer than 6 months were more likely to respond. Corrected prevalences after reweighting were slightly different from crude prevalences for some variables but meaningfully different for others.

CONCLUSION

Linking health insurance and occupational data effectively corrects for nonresponse bias using reweighting techniques. Sociodemographic variables may be not sufficient to correct for nonresponse.