The NHANES survey methodology uses a missing value entry (denoted as a question mark "?" in Weka).
Descriptive statistics will be used to summarise the data and detect outliers, missing values and data entry mistakes.
Main advantage of web-based administration [ 40] is completeness and reliability of data (no missing values, no data entry errors).
Any value that could not be inferred or obtained directly was set as missing values (empty data entries) in the learning dataset.
Note that P contains in fact missing values for all entries (P) ij with max i, j)> n.
Prior to analyses all variables were examined for accuracy of data entry, missing values, presence of outliers, and normality of distribution.
Prior to conducting analyses, all variables were examined through SPSS for accuracy of data entry, missing values, fit between their distributions, and assumptions of multivariate analyses.
Preliminary analyses of the raw country-specific datasets including descriptive work to identify out-of-range values for variables, conflicting results, missing values and possible data entry errors will be performed.
Prior to analysis, the variables were examined for accuracy of data entry, missing values, and violations of multivariate assumptions and passed all required tests (some variables were not normally distributed but bootstrapping is unaffected by this).
Before the analyses, we examined age, current alcohol use, current smoking, ES, and the lead exposure metrics using univariate descriptive statistics to check for accuracy of data entry, missing values, and assumptions underlying multivariate analysis.
