Introduction to Environmental Geosciences - Lecture 3

Introduction to Environmental Geosciences - Lecture 3: Data Quality

• Data quality is a measure of how reliable a data set is in relation to expected information.
• It assesses how well the data fits the specific needs or purpose.
• Data quality is determined by several factors:
  • Validity: Accuracy of the data
  • Completeness: Fullness of the data set
  • Timelessness: Up-to-date nature of the data
  • Consistency: Data uniformity across different sources
  • Integrity: Data reliability and trustworthiness
  • Accuracy: How close the measurements are to the true value.
  • Precision: How close repeated measurements are to each other.

Evaluating Data Quality

• To ensure data quality, scientists
  • Use quality control standards
  • Replicate their measurements.
  • Evaluate replicate measurements for accuracy.
  • Record and sign off on measurements immediately after.
  • Repeat measurements after a period of time (e.g., monthly, quarterly)
  • Ensure complete information suitable for the intended purpose

Replicate Measurement

• Scientific measurements are often done in replicates to identify misleading data and ensure data reliability.
• All laboratory measurements are typically performed in triplicate.

Accuracy and Precision

• One way to evaluate data quality is by evaluating and using precision and accuracy.
• Replicate measurements improve precision determination.
• Accuracy involves using quality control standards (substances with known values) of the parameter.
• Quality standards allow evaluation of a method's performance after certain measurement intervals.

Precision Evaluation

• Standard deviation (%RSD) is frequently used to evaluate measurement precision in scientific experiments.
  • Formula: [(SD/Mean) x 100].
• Small %RSD indicates tightly clustered data around the mean, suggesting high precision.
• Larger %RSD indicates more spread-out data.

Accuracy Calculation (Example: pH)

• For accuracy assessments, specifically with certified reference soil, measure the percentage recovery from certified values.
  • Formula: (Measured Value / Certified Value) * 100
• Another method is to calculate the error percentage first and then subtract it. from 100% to get the accuracy.
  • Formula:(Actual value - Measured value) / Actual value * 100

Class Activity

• Evaluate precision and accuracy for given pH measurements of water samples and quality control samples.
• Evaluate the data quality of previous lab measurements.