MT 520: SAMPLE EXAM
QUESTIONS

Sample Exam Questions for UW-Madison Med Tech Students

3/3/97 Please note: These sample questions are nearly ready. Test yourself. Answers will be posted in a few weeks.

PT criteria for acceptable performance are given in three of the following formats:

• A. Target value fixed concentration limits
• B. Target value fixed 3SD range
• C. Target value fixed percent
• D. Target value 3SD of selected test group
• E. Target value 3SD of reference value range

For a BUN method such as studied in Real World episode #1, which of the following would be considered application characteristics:

• A. Performance time of 15 minutes or less
• B. Sample size of 50 ul or less
• C. Precision - SD or 2 mg/dL or less
• D. Chemical reaction - urease for specificity
• E. Calibration - primary aqueous standards
• F. Accuracy - bias<3 mg/dL

Which characteristics are the focus of method evaluation

• A. Application characteristics
• B. Methodology characteristics
• C. Performance characteristics

Which experiments provide information about random error (of only the test method)?

• A. Replication experiment
• B. Interference experiment
• C. Recovery experiment
• D. Comparison of methods experiment

What is the per cent recovery when a 100 mg/dL addition is made to a sample containing 100 mg/dL and a value of 180 mg/dL is measured (assume all dilution effects have been accounted for)?

• A. 100%
• B. 90%
• C. 80%

In a replication experiment, an F-test was calculated to compare the performance of the test and comparative methods. The calculated F-value was 4 and the critical F-value was 2.6. This means:

• A. There is a difference in the precision of the two methods
• B. There is no difference in the precision of the two methods
• C. The test method performs better that the comparative method
• D. The Comparative method performs better than the test method

Useful estimates of errors are provided by t-test statistics when (2 answers):

• A. Proportional error is absent
• B. The correlation coefficient is high
• C. The t-value is small
• D. The mean of the data is close to X_c
• E. There is no random error

What is the systematic error at a critical concentration of 100 mg/dL when the following statistics are obtained from a comparison of methods experiment: slope=1.00, y-intercept=6.0 mg/dL, s_y/x=4.0 mg/dL

• A. 4.0 mg/dL
• B. 6.0 mg/dL
• C. -6.0 mg/dL
• D. 1.0 mg/dL
• E. 10.0 mg/dL

For a comparison of methods experiment, the correlation coefficient was 0.990. This indicates that:

• A. Random error between methods is only 1%
• B. Systematic error between methods is only 1 mg/dL
• C. Random error between methods is small compared to the concentration range studied
• D. Random error is acceptable
• E. Systematic error is acceptable

The following statistical summary was obtained from a urea nitrogen comparison of methods experiment: N-316, a=-0.31 mg/dL, s_a=0.23 mg/dL, b=1.032, s_b=0.009 , s_y/x=0.97 mg/dL, s_x=13.2 mg/dL, r 0.997, x=21.1 mg/dL, bias=0.40 mg/dL, s_d=1.08 mg/dL, and t=6.58

What is the proportional error?

• A. -0.31 mg/dL
• B. 1.032%
• C. 3.2%
• D. 0.3%

What is the constant error?

• A. -0.31 mg/dL
• B. 3.2%
• C. 13.2 mg/dL
• D. 6.58

What is the random error between methods

• A. 0.997
• B. 2(0.97) mg/dL
• C. 13.2 mg/dL
• D. 0.40 mg/dL

What is the 95% confidence interval for the slope?

• A. 1.032+2(0.009)
• B. 0.31 +2(0.23)
• C. 21.1+2(13.2)
• D. 1.032+2(0.97)

What is the systematic error at a concentration of 21 mg/dL?

• A. 0.40 mg/dL
• B. -0.31 mg/dL
• C. 21.36 mg/dL
• D. 21.1 mg/dL

Is this systematic error statistically significant?

• A. Yes, at p=0.05
• B. No, not at p=0.05

Is this systematic error clinically significant?

• A. Yes, using the performance standard in Question 38
• B. No, not when using the performance standard in Question 38

What is the systematic error at a concentration of 27 mg/dL?

This website is best viewed by Internet Explorer or Netscape2.0 or better.