Entry Errors Research Articles

A simplified procedure for entry of raw genotypic data.

Genotypic data entry is a time-consuming and tedious task in genetic linkage studies. Success of the study is dependent upon the accuracy of the data. To simplify the process and eliminate data entry errors, we developed procedures that enable scientists to rapidly enter large quantities of data and compare the data entered with those entered by another individual. These procedures significantly reduce the time required to enter data while improving its integrity. The procedure relies on quick association of a single-digit pattern number with the visual image of an animal's genotype. This association is facilitated by a standard set of rules applicable to all possible outcomes (pseudocode provided). Fewer data entry errors are made because the procedure reduces required keystrokes by 75% and short-term memory load. These procedures have been used as independent programs operating in PC environments (programmed in FORTRAN) as well as linked with a relational database on an IBM 9377 [programmed in CSP(AE/AD) and SQL/DS].

Accuracy of bar codes versus handwriting for recording trauma resuscitation events

To compare the accuracy of computerized bar code data entry with conventional handwritten data entry during videotaped trauma resuscitations. Laboratory; video simulation. Twenty-four emergency nurses. The nurses viewed videotapes of four different major trauma resuscitations during a single session. Each nurse recorded resuscitation events by handwritten entry in two cases and by bar code entry in another two. A computerized bar code system was designed specifically for contemporaneous charting of rapidly occurring events during trauma resuscitations. The handwritten and bar-coded records then were compared with a master list of events, and the number of entry errors were counted. Errors were defined as "omissions" (failing to record an event), "commissions" (recording an event that did not occur), or "inaccuracies" (errors in recording details of an event). Differences in the number of entry errors between the two recording methods were compared using unpaired t-tests. Differences in the number of errors after adjusting for the different nurses, different case being viewed, and order of viewing were analyzed using balanced analysis of variance techniques. P < .05 was considered significant. The mean +/- SEM number of total errors per record for bar codes was 2.63 +/- 0.24 compared with 4.48 +/- 0.30 for handwriting (P < .0001). The mean number of omissions per record for bar codes was 2.25 +/- 0.21 compared with 3.65 +/- 0.27 for handwriting (P = .0001). The mean number of inaccuracies per record for bar codes was 0.38 +/- 0.10 compared with 0.83 +/- 0.12 for handwriting (P = .0038). There were no commission-type errors. Computerized bar code data entry of trauma resuscitation events had significantly fewer entry errors than handwritten data entry in a laboratory setting. Potential advantages of bar code data entry include keyless data entry, automatic time-stamping, standardization of documentation, legibility of the medical record, and "point-of-care" data capture.

Influences on inferences. Effect of errors in data on statistical evaluation

Inadvertent random and systemic errors introduced into data sets and manipulation of data are well-defined sources of discrepancies in statistical evaluation of clinical trials. In this study, the authors show the influence of errors on the widely used statistical result, P values.Using data from a retrospective study of patients with Hodgkin disease treated at the University of Minnesota between 1970 and 1984 and observed to 1988, we introduced various errors into the data to study the impact on results.Inadvertent random and systemic errors affect statistical results. Data entry and transcription errors, vague definitions of endpoints and prognostic factors, and the omission and selection of patients are examples of frequent errors that affect statistical evaluation.The results and inferences of many studies are sensitive to systemic errors and data manipulation. Great care must be given to the clear definitions of terms, exclusion and inclusion criteria, group assignments, treatment protocols, and the subgroups on which statistical analysis is performed. Clinicians and statisticians must work together to improve the performance and interpretation of clinical trials.

Influences on inferences. Effect of errors in data on statistical evaluation.

Inadvertent random and systemic errors introduced into data sets and manipulation of data are well-defined sources of discrepancies in statistical evaluation of clinical trials. In this study, the authors show the influence of errors on the widely used statistical result, P values. Using data from a retrospective study of patients with Hodgkin disease treated at the University of Minnesota between 1970 and 1984 and observed to 1988, we introduced various errors into the data to study the impact on results. Inadvertent random and systemic errors affect statistical results. Data entry and transcription errors, vague definitions of endpoints and prognostic factors, and the omission and selection of patients are examples of frequent errors that affect statistical evaluation. The results and inferences of many studies are sensitive to systemic errors and data manipulation. Great care must be given to the clear definitions of terms, exclusion and inclusion criteria, group assignments, treatment protocols, and the subgroups on which statistical analysis is performed. Clinicians and statisticians must work together to improve the performance and interpretation of clinical trials.

The laboratory workstation: a data management model for a small laboratory section.

Many small sections of clinical chemistry laboratories fail to benefit fully from laboratory information system (LIS) installation because the volume does not justify cost of interfacing the instruments in these sections to the LIS. The opportunity for manual data entry errors remains a problem in these sections. This paper describes the design of a laboratory workstation that serves as a data hub that makes it feasible to use one instrument interface line from the LIS to acquire data from four instruments. This approach reduces the likelihood of data entry error and improves the efficiency of personnel in the laboratory section. Further, the tools necessary to create such a workstation are commercially available and do not depend on microcomputer programming and support personnel within the chemistry laboratory.

An electronic medical record system with direct data-entry and research capabilities

The transfer of medical records from a paper system to a computer-based system is inevitable. However, the widespread acceptance of electronic medical records has been delayed by problems such as high cost, inefficiency, data entry errors and poor physician acceptance. We have developed a database system that has overcome these difficulties and now serves as an electronic medical record. Our system has been in use for a year and a half, and currently contains information on over two thousand patients. The database provides an electronic radiation oncology chart containing patients' demographic information, technical treatment data and dictated reports. All dictated notes are captured, including consultation notes, treatment summaries, on-treatment visits, letters and follow up reports. The system provides data validation upon entry, required few additional software or hardware purchases, and allows for efficient retrieval of data. Unlike other database systems which require the hiring of data entry clerks to input the data, ours combines transcription and data entry. The database runs on a local area network of computers and uses a commercially available relational database package. It makes extensive use of mouse interface features such as pull-down menus, pop-up lists, buttons, multi-page forms, and scrolling fields, making the system easy to use with minimal training. Many custom features are built in, such as help screens, control functions, audit trails, and a system that keeps track of each patient's referring and other relevant physicians. For research purposes, the system has the capability to perform survival analyses on arbitrary user-defined subsets of patients. Data may also be exported transparently to statistical packages for other types of analyses.

A computer-aided procedure for measuring lashley III maze performance

As an animal swims through the Lashley III maze, an observer types into a Macintosh computer the path taken. The computer program, Observe Software, then breaks the string of choices into two-step sequences and counts the number of such sequences. These data are then sent to a spreadsheet, where the sequences are sorted into forward and backward responses. Forward choices are Correct Path, T Choice Errors and Cul Entry Errors. All backward choices are errors, by definition. They are classified as T Choices, Cul Entries, Cul Exits, and Return to Start. The animal's behavior is then described by the various error classes plus a measure called Learning Index. Examples of learning by rats and mice are presented.

Research in Physical Medicine and Rehabilitation

This paper describes important aspects of preliminary data analysis to be taken after data are checked for clerical entry errors and before the primary statistical analysis is performed. These include description and graphic display of each variable, recoding categorical data, transforming continuous data into another continuous variable and recording continuous to categorical data. Missing values and outlying data points are identified and several techniques are recommended to minimize mistakes in variable recording. Related variables measured with different units may be combined by using the z transformation and converted back to one of the original units for ease of interpretation. Finally, both categorical and continuous variables are checked for reliability by using kappa or the intraclass R.

The effects of data entry error: An analysis of partial verification

Modern data entry technology has greatly reduced entry errors by building quality control mechanisms such as cross-field and range edits directly into data entry programs. In many application areas the increased accuracy of modern data entry techniques has made 100% verification unnecessary to assure high standards of data quality. Verification of randomly selected forms combined with 100% verification of critical fields provides a cost efficient alternative to 100% verification of all fields. This paper analyzes the effects of data entry error on study findings and provides a basis for designing a partial verification scheme for achieving quality control goals in a cost efficient manner.

Research in Physical Medicine and Rehabilitation

This paper describes important aspects of preliminary data analysis to be taken after data are checked for clerical entry errors and before the primary statistical analysis is performed. These include description and graphic display of each variable, recoding categorical data, transforming continuous data into another continuous variable and recoding continuous to categorical data. Missing values and outlying data points are identified and several techniques are recommended to minimize mistakes in variable recoding. Related variables measured with different units may be combined by using the z transformation and converted back to one of the original units for ease of interpretation. Finally, both categorical and continuous variables are checked for reliability by using kappa or the intraclass R.

An examination of the efficiency of some quality assurance methods commonly employed in clinical trials

The cost and efficiency of training clinical centre staff and of duplicate data entry in clinical trials is reviewed. Training is an essential component of quality assurance programmes and it is usually carried out at regular intervals in long-term clinical trials. Initial training of staff and regular retraining is important to assure standardization and it can lead to reduced trial costs. Interim training for new staff and for remedial purposes is less efficient than regularly scheduled training sessions. Regional centres for training and the use of computer aided instruction are two ways such interim training can be made more efficient and standardized. Duplicate data entry or verification can result in a substantial reduction in data entry errors depending on the nature of the data being keyed. Selective verification should especially be considered for important fields for which consistency checks cannot be performed, that are alphabetic or that are several characters in length. Quality assurance procedures should be implemented to monitor data entry accuracy in clinical trials.

The outcome index. A method of quality assurance in the special care area.

A method of quality assurance for a surgical intensive care unit is described. A system outcome score is devised, incorporating only easily obtained objective components that reflect the likelihood of death. Through the use of a derived outcome index, the actual mortality rate is compared with the predicted mortality rate as a method of monitoring the quality of care provided. Subroutines exist to identify errors in data entry, to detect malicious interference in patient care, to add nonscoring components for the purposes of clinical studies, and to facilitate retrieval of a concise summary of the major events during the stay of every patient admitted to the intensive care unit.

A Microcomputer-Based Data Acquisition and Reporting System for Clinical Pathology Data from Animal Drug Toxicology Studies

A microcomputer-based system has been developed for the acquisition, reporting, and evaluation of clinical biochemistry and hematologic data. The system consists of an IBM microcomputer interfaced with several peripheral devices. Associated computer programs were developed inhouse and customized for use in animal drug toxicity studies. The system has the following advantages: (1) Data are directly transmitted to the microcomputer via interface devices, thereby minimizing the possibility of data entry errors and eliminating the need for handwritten records; (2) individual animal reports and group summary reports are readily generated; (3) individual animal data and group summary values are routinely compared during the data collection phase against a historical reference range database, and “outliers” are flagged, allowing quick identification of aberrant values or parameters that may have been affected by experimental treatments; and (4) treatment groups are tested in the data evaluation phase against con...

Human interface with logging operation optimization programs

This paper presents a case study involving the comparison of two programs which were designed for use in the logging industry. Of particular interest in this comparison are the human factors problems likely to arise during field use of such programs. This paper discusses the need for application of consistent sets of human factors design principles. A specific major conclusion related to the field use of these programs is that it is advisable to provide effective redundancy techniques against catastrophic loss of data in an environment where the user has multiple taks to perform. This paper also discusses the consequences of "good" as well as "poor" examples of menu design, display design, data entry error checking, and handling of erroneous keystroke and keystroke sequences.

Automated data collection using bar code

Using bar code technology to automate data collection provides a rapid and reliable alternative to paper-and-pencil tracking or keyboard entry into pocket or laptop computer. An array of bar code symbols is printed on a data menu, with a unique symbol corresponding to each possible observation. Bar code symbols are scanned using hand-held readers, which record the event, log the observation time, and store data for transfer to a personal computer. Advantages of the bar code monitoring system include: (1) ease of use by staff with minimal technical training, (2) reduced data entry errors and increased entry speed, (3) reliable portable operation, and (4) low-cost hardware. While the bar code system described here is used for behavioral monitoring in a residential treatment setting for the developmentally disabled, with minimal modification this system can be adapted for use in a wide variety of research and clinical applications.

Assessing the quality of data entry in a computerized medical records system.

A primary care university clinic discovered major errors in its medical records after instituting a computerized medical records system. A quality assessment tool was devised to measure the error rate and provide clues to the sources of these errors. During the implementation of this tool a marked reduction in data entry errors occurred suggesting its use had a positive impact on our medical record quality.

Long workdays versus restdays: assessing fatigue and alertness with a portable performance battery.

A test battery designed to assess psychological/behavioral fatigue was used to compare restdays to a workweek of five 12-hr days at a data entry job simulation. Across both workdays and restdays, the battery was presented at regular intervals to test for fatigue effects and diurnal variations. Increased data entry errors across the workday and the workweek as well as subjective reports of increased tiredness on workdays indicated that the work regimen was fatiguing. Test battery performance paralleled those results. On workdays, as compared with restdays, grammatical reasoning was faster but less accurate; digit addition was slower; simple, dual, and choice reaction times were slower; and hand steadiness decreased. The results demonstrated the sensitivity of the battery to long hours of work. The results are discussed in terms of work-rest differences, changes across the workweek, diurnal variations, and cognitive demand.

OTV aeroassist with low [formula omitted

OTV aeroassist is viewed in the light of maximizing vehicle performance by minimizing aerobrake weight. An aero-entry strategy is developed which shows the optimality of a low L D lifting brake concept entering at the mid-point of its control capability. Entry error analysis is used to justify a 0.12 L D . A guidance and control scheme for efficient aerobraking is described which utilizes a predictor corrector algorithm and continuous roll for lift vector nulling. Extensive results of a 4 degree of freedom trajectory simulation confirms the effectiveness of the 0.12 L D . Good results for an even lower L D of 0.08 are also indicated.

Airborne Message Entry by Voice Recognition

This report presents the results of an experiment that measured performance in a simulated ASW message entry task with two modes of data input—vocal and manual. The subjects (Ss) were 12 Naval enlisted men. The independent variable was message data entry mode—vocal or manual. The dependent variables were: time to enter 20 lines of text, data entry errors that were corrected by the Ss, and errors that remained undetected. All Ss were trained to use the voice recognition system with a 100 word vocabulary set. The task was for the S to read one line of message text from a display and then re-enter the text below the displayed text via either voice recognizer or keyboard until 20 lines of text had been entered. Keyboard entry was found to be slightly faster (11%) than voice recognition input. While the number of initial errors (corrected) in the vocal input mode was over three times greater than the number for manual input, the remaining input errors (uncorrected) were about the same.

Non-significant, self-validated part identification numbers

The numbering and identification of parts within a manufacturing organization is of immense importance, for it serves as a major means of communicating information throughout the organization. The oldest, and still most prominent method of part identification is the use of meaningful, alphanumeric part numbers. With the transition to computer-assisted manufacturing systems, however, many of the weaknesses of this type of system have come to light. A newer approach to part numbering, the use of all-numeric, non-significant part numbers as a vehicle for identifying parts and accessing the required information about each, may be a major contributor to improved communications in the computer age. This paper explores both approaches to part identification, addressing the advantages and drawbacks of each. It is concluded that the use of non-significant numbering is indeed the best approach to a flexible, simple, and standardized identification system, allowing access to all pertinent part information without incorporating that information into the part number. Two case studies, one of a leading compressor manufacturer, and the other a producer of greeting cards, are presented, demonstrating the feasibility of a conversion from significant to non-significant systems and the benefits obtainable. The problem of reducing data entry errors and maintaining data integrity with non-significant part numbers is analyzed and a proposed algorithm presented. Finally, routines are demonstrated for both generating and then vetting and validating non-significant numbers.