Rates of computational errors for scoring the SIRS primary scales

Area of Science:

• Psychological assessment
• Clinical psychology
• Psychometrics

Background:

• The Structured Interview of Reported Symptoms (SIRS) is a widely used tool for assessing symptom validity.
• Accurate scoring is crucial for reliable interpretation of psychological assessments.
• Previous research suggests that scoring errors on psychological tests are prevalent.

Purpose of the Study:

• To evaluate the frequency and impact of scoring errors in the SIRS.
• To compare computer-generated scores with hand-tallied scores.
• To identify strategies for minimizing scoring inaccuracies in the SIRS.

Main Methods:

• Item scores from the SIRS were entered into a spreadsheet.
• Computed scores were compared against examiner hand-tallied scores.
• The impact of corrected errors on SIRS classifications was analyzed.

Main Results:

• Approximately 35% of SIRS tests contained at least one scoring error.
• About 8% of examiner-tallied SIRS scale scores were incorrectly summed.
• Correcting identified errors led to a change in classification for only one case.

Conclusions:

• Scoring errors in the SIRS are frequent but generally do not significantly alter diagnostic classifications.
• The findings highlight the need for improved methods to reduce scoring errors in the SIRS.
• Implementing strategies to enhance scoring accuracy is recommended for clinical practice.