Using Paradata to Evaluate Youth Participation in a Digital Diary Study

Tienda, Marta; Koffman, Dawn

doi:10.1177/0894439320929272

Cited by 2 publications

(1 citation statement)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, our analyses cannot ascertain whether "other" reports are indicative of survey fatigue with the instrument or a result of a slower learning curve during the beginning of time-use record completion (Chatzitheochari and Mylona, 2021). Better understanding of the specificities of app instruments could be achieved with paradata (Tienda and Koffman, 2020), including timestamps for data entry, which are not made available in the MCS survey.…”

Section: Discussionmentioning

confidence: 98%

Does Diary Mode Matter in Time-Use Research?

Chatzitheochari¹,

Mylona²

2022

JTUR

View full text Add to dashboard Cite

Recent years have witnessed an increasing interest in the use of new technologies for time-use data collection, driven by their potential to reduce survey administration costs and improve data quality. However, despite the steady growth of studies that employ web and app time diaries, there is little research on their comparability with traditional paper-administered diaries that have long been regarded as the “gold standard” for measurement in time-use research. This paper investigates diary mode effects on data quality and measurement, drawing on data from a mixed-mode large-scale time diary study of adolescents in the United Kingdom. After controlling for observable characteristics associated with diary mode selection and adolescent time-use, we find that web and app diaries yield higher quality data than paper diaries, which attests to the potential of new technologies in facilitating diary completion. At the same time, our analysis of broad time-use domains does not find substantial mode effects on measurement for most daily activity categories. We conclude by discussing avenues for future methodological research and implications for time-use data collection.

show abstract

Section: Discussionmentioning

confidence: 98%

Does Diary Mode Matter in Time-Use Research?

Chatzitheochari¹,

Mylona²

2022

JTUR

View full text Add to dashboard Cite

show abstract

Longitudinal Nonresponse Prediction with Time Series Machine Learning

Collins,

Kern

2024

Journal of Survey Statistics and Methodology

View full text Add to dashboard Cite

Panel surveys are an important tool for social science researchers, but nonresponse in any panel wave can significantly reduce data quality. Panel managers then attempt to identify participants who may be at risk of not participating using predictive models to target interventions before data collection through adaptive designs. Previous research has shown that these predictions can be improved by accounting for a sample member’s behavior in past waves. These past behaviors are often operationalized through rolling average variables that aggregate information over the past two, three, or all waves, such as each participant’s nonresponse rate. However, it is possible that this approach is too simple. In this paper, we evaluate models that account for more nuanced temporal dependency, namely recurrent neural networks (RNNs) and feature-, interval-, and kernel-based time series classification techniques. We compare these novel techniques’ performances to more traditional logistic regression and tree-based models in predicting future panel survey nonresponse. We apply these algorithms to predict nonresponse in the GESIS Panel, a large-scale, probability-based German longitudinal study, for surveys conducted between 2013 and 2021. Our findings show that RNNs perform similar to tree-based approaches, but the RNNs do not require the analyst to create rolling average variables. More complex feature-, interval-, and kernel-based techniques are not more effective at classifying future respondents and nonrespondents than RNNs or traditional logistic regression or tree-based methods. We find that predicting nonresponse of newly recruited participants is a more difficult task, and basic RNN models and penalized logistic regression performed best in this situation. We conclude that RNNs may be better at classifying future response propensity than traditional logistic regression and tree-based approaches when the association between time-varying characteristics and survey participation is complex but did not do so in the current analysis when a traditional rolling averages approach yielded comparable results.

show abstract

Using Paradata to Evaluate Youth Participation in a Digital Diary Study

Cited by 2 publications

References 42 publications

Does Diary Mode Matter in Time-Use Research?

Does Diary Mode Matter in Time-Use Research?

Longitudinal Nonresponse Prediction with Time Series Machine Learning

Contact Info

Product

Resources

About