Attrition analysed in five waves of a longitudinal yearly survey of smokers: findings from the ITC Netherlands survey
Bibliographic record
Abstract
BACKGROUND: Attrition bias can affect the external validity of findings. This article analyses attrition bias and assesses the effectiveness of replenishment samples on demographic and smoking-related characteristics for the International Tobacco Control Netherlands Survey, a longitudinal survey among smokers. METHODS: Attrition analyses were conducted for the first five survey waves (2008-12). We assessed, including and excluding replenishment samples, whether the demographic composition of the samples changed between the first and fifth waves. Replenishment samples were tailored to ensure the sample remained representative of the smoking population. We also constructed a multivariable survival model of attrition that included all five waves with replenishment samples. RESULTS: Of the original 1820 respondents recruited in 2008, 46% participated again in 2012. Demographic differences between waves due to attrition were generally small and replenishment samples tended to minimize them further. The multivariable survival analysis revealed that only two of the 10 variables analysed were significant predictors of attrition: a weak effect for gender (men dropped out more often) and weak to moderate effects for age (respondents aged 15-24 years dropped out more than aged 25-39 years, who dropped out more than those aged 40+ years). CONCLUSIONS: Weak to moderate attrition effects were found for men and younger age groups. This information could be used to minimize respondent attrition. Our findings suggest that sampling weights and tailored replenishment samples can effectively compensate for attrition effects. This is already being done for the International Tobacco Control Netherlands Survey, including the categories that significantly predicted attrition in this study.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.037 | 0.110 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.004 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".