Data Quality and Indicators

This study highlights the importance of collecting good quality data from multidisciplinary studies. Bias in data may be the result of instrument inaccuracies, imprecise data recording techniques, inaccurate data entry to computers or inappropriate statistical analysis and presentation. Recommendations for good data quality control are given. Different types of data are discussed: raw data, simple indicators and complex indicators. It is shown how measurements from the components of multidisciplinary systems can be combined to form complex indicators and a specific example is given using Z-scores and dot charts. Finally the accumulated effect of bias in the individual component measurements upon the combined indicator is shown.