Automatic detection of the autocorrelation-type measurement error component
1986

Automatic Detection of Measurement Errors in Analytical Samples

Sample size: 50 publication Evidence: moderate

Author Information

Author(s): K. M. Hangos, J. L. Nagy, L. Leisztner

Primary Institution: Hungarian Academy of Sciences

Hypothesis

The presence of autocorrelation-type measurement error components can be detected using a proposed numerical method.

Conclusion

The proposed numerical method effectively detects autocorrelation-type measurement errors, showing that the evaluation method significantly influences measurement error detection.

Supporting Evidence

  • The numerical method proposed can be used for automatic validation and quality control of analytical measurements.
  • Results showed that the peak height ratio evaluation method is more sensitive to detecting measurement errors than the peak area ratio method.
  • Empirical variances of the peak height ratios were found to be smaller than those of the peak area ratios.

Takeaway

This study helps scientists find mistakes in their measurements by using a special math method, making sure their results are correct.

Methodology

The study used control samples with varying concentrations of ethyl alcohol and analyzed them using a gas chromatograph, applying both numerical and graphical methods to detect measurement errors.

Limitations

The study's findings may be influenced by the specific conditions of the measurements and the choice of control samples.

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication