Data management errors

Non-sampling errors can be categorized into laboratory error and data management error, with laboratory error further subdivided into measurement, data interpretation, sample management, laboratory procedure and methodology errors. 

Representativeness Sampling design error Field procedure error Data interpretation error Sample management error Data management error... 

Completeness Data management error Field and laboratory procedure error... 

In the course of sample tracking, data evaluation, and interpretation, field sample IDs may be entered into several different field forms, spreadsheets, and data bases, and appear on maps and figures as identifiers for the sampling points. Because the field records and computer data entry during sample receiving at the laboratory are done for the most part manually, errors in sample ID recording are common. To reduce data management errors, sample numbers must be simple, short, and consecutive. 

Data management errors that may take place during data reduction and reporting have a potential to ruin an otherwise perfect analysis. These errors may have significant effects on the accuracy, precision, and completeness of a data set as shown in Example 4.6 on page 200. 

Laboratories prevent data management errors by implementing QA policies and procedures for data reduction, verification, and software management and by establishing a high standard of professional ethics. 

The following data management errors are known to occur at the laboratories ... 

The assessment phase itself, however, is not invincible from error either. Data management errors may hinder the success of DQA, particularly for projects with insufficient planning, followed by disorderly implementation. Effective assessment is possible only when built on a foundation of solid project planning and supported by well-organized implementation of the planned data collection. 

For sample results with complete and thoughtfully compiled data packages, these questions may be answered immediately upon evaluation. However, if QC check data and support documentation are missing or are inaccurate due to data management errors at the laboratory, the chemist s decision on data quality may be delayed. The chemist will request that the laboratory provide additional data in order to evaluate them at a later date. The loss of continuity in the data evaluation process due to poor quality of data packages is counterproductive and may cause delays in the scheduled project report delivery to the client. 

A complete knowledge of the data quality that arises only from Level 4 validation enables the data user to make project decisions with the highest level of confidence in the data quality. That is why Level 4 validation is usually conducted for the data collected to support decisions related to human health. Level 4 validation allows the reconstruction of the entire laboratory data acquisition process. It exposes errors that cannot be detected during Level 3 validation, the most critical of which are data interpretation errors and data management errors, such as incorrect computer algorithms. 

Validation of the data management system is typically done in two rounds. First, correctly completed data forms are entered to ensure that the system is not flagging any good data. In the second round, completed data forms with intentional data errors are entered. All errors must be identified by the system. 

The management of this screening workload distributed across multiple workstations can be rather labor-intensive and error prone without appropriate sample and data management tools. This places a large resource burden on screeners who could otherwise spend their time on higher-value activities such as more rigorous data evaluation. 

As we already know, an environmental sample is a fragile living matter that can be severely damaged at every step of its existence. Due to the inherent nature of environmental media and a host of potential errors associated with sampling, analysis, and data management, the collection of environmental chemical data is not an exact science. In fact, all environmental chemical data are only the estimates of the true condition that these data represent. In order to make these estimates more accurate, we must examine the sources of errors and take measures to control them. 

The chemist interprets the results of trip and equipment blank analyses to identify sample management errors during sampling, sample handling, and decontamination procedures and to determine whether these errors may have affected the collected sample representativeness. The chemist qualifies the data according to the severity of the identified variances from the SAP specifications and may even reject some data points as unusable. Example 5.8 shows a logical approach to the interpretation of the trip and equipment blank data. 

Having collected optimal quality data, first-rate data management is also critical. Many data that are collected can now be fed directly from the measuring instrument to computer databases, thereby avoiding the potential of human data entry error. However, this is not universally true. Therefore, careful strategies have been developed to scrutinize data as they are entered and once they are in the database. The double-entry method requires that each data set be entered twice (usually by two operators) and these entries compared by a computer for any discrepancies. This method operates on the model that two identical errors are probabilistically very unlikely, and that every time the two entries match the data are correct. In contrast, dissimilar entries are identified, the source (original) data located, and the correct data point entry confirmed. 

Not all factors that influence the reliability and representativeness of data are measurable. Those that are measurable will usually be found if proper data-handling processes are followed. However, there are many immeasurable factors that can severely bias data and that are not readily identified, even by good data-handling and data management procedures. Some of these nonmeasurable errors are ... 

Discrepancy reports are prepared for investigator review and correction. The sponsor translates the computer output into user-friendly reports. There is direct communication between the investigational site coordinator and the data manager for any error messages that may need clarification. 

Clear procedures (SOPs) for conducting such audits must be established, detailing the sampling procedures for CRFs and acceptable error rates. Information is available in literature on error levels and data verification procedures (DGGF, 2003 Zhang, 2004 Society of Clinical Data Management, 2005). 

Statistics finds important applications by engineers, scientists, and indnstrial managers when planning and execnting research, plant design and operation, marketing and sales programs. Improved operations, better and more nniform products, increased safety, and additional profitability often result. Experimental data contain errors or uncertainties for a variety of factors. 

Horstmann, Gay. Bigjana. 4th ed. Hoboken, N.J. John Wiley Sons, 2010. This book describes the most recent version of the Java language, along with its history and special features of the Java interpreter. Snow, Colin. Embrace the Role and Value of Master Data Management, Manufacturing Business Technology, 26, no. 2 (February, 2008) 92-95. This article, written by the VP and research director of Ventana Research, covers how to implement data management and avoid the errors that cost businesses millions of dollars each year. 

---