On the other hand, incomplete outcome data (attrition bias) was found to be important, as missing data could potentially have an impact on the observed effect size.
And, acknowledged by the authors, they have an important amount of missing data and incorrect zero values (for countries with relatively small flows), covering, as a result, a limited fraction of total inflows (Mayda 2010, pp. 1258 1259).
These procedures illuminated two important causes of missing data.
The most important limitations involved missing data and inconsistent definitions of pivotal risk factors.
However this can only assess baseline covariate effects assuming particular values for the sensitivity parameter, and cannot quantify the evidence that the outcome itself is important in the missing data model.
To summarise the conclusions from the sensitivity analysis, the only baseline covariate that appears to be safely regarded as important in the missing data model is the age of participants.
The first α1 parameter, which we denote as α1,1 is the (adjusted) treatment effect, but here the focus of interest is on the covariates that are important in the missing data model, ie β1 and β2 are paramount.
The suspicion that participants with more educational qualifications may be more likely to respond to an email reminds us that the variables that are important in the missing data model are likely to be context specific, and can be anticipated to depend on the nature of the trial and how data are collected.
The most important aspects of missing data per section are: Section 1: visiting health care providers About 2/3 of the respondents from the breast cancer sample who ticked a CAM modality also filled in the item "No. of times" (please see valid N in 3rd column of Table 2).
Thresholds for important missing participant data will be determined on an outcome-by-outcome basis.
