An Alternative Approach to Addressing Missing Indicators in Parallel Datasets
Bibliographic record
Abstract
When doing secondary data analysis, it is not uncommon to find that a key variable was not measured. Often the researcher has no option but to do without the missing indicator, but when nearly parallel datasets exist, the researcher may have other options. In an earlier article leading up to this special issue, this research team was confronted with the problem that research utilization had been measured in only one of two similar datasets, namely, in the 1996 but not the 1998 Alberta Registered Nurse survey. The 1998 dataset had a larger sample size (6,526 compared to 600 nurse respondents in 1996) and a stronger set of measured variables, but was missing the key variable of interest--research utilization. To overcome this, a regression-based strategy was used to create a research utilization score for each nurse in the 1998 survey by exploiting the availability of several anticipated causes of research utilization in both datasets. Presented here is an alternative and more complicated procedure that might be applied in future investigations. The article presents a methodological understanding of how to use a phantom variable to account for the unmeasured research utilization variable in a two-group structural equation model. This approach could be used to overcome several of the limitations connected to using a regression-based approach to creating a key missing variable when nearly parallel datasets are available.
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 | 0.217 | 0.521 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.005 |
| Bibliometrics | 0.009 | 0.023 |
| Science and technology studies | 0.004 | 0.005 |
| Scholarly communication | 0.008 | 0.012 |
| Open science | 0.009 | 0.012 |
| Research integrity | 0.009 | 0.010 |
| Insufficient payload (model declined to judge) | 0.008 | 0.002 |
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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".