Must know (final 1/27)

Question 1

T test think of —-

Answer 1

accuracy. statistical test used to determine if the means of two different gaussian data sets vary significantly.

Question 2

F test think of _____

Answer 2

Precision. statistical test used to determine if the SDs of two different gaussian data sets vary significantly.

Question 3

Data set

Answer 3

a group of measurements of like kind

Question 4

Population

Answer 4

a collection of people or things from which we may collect sample data to draw a conclusion.

Question 5

Sub-population or Sample

Answer 5

a data set randomly collected from a defined population.

Question 6

Frequency Distribution

Answer 6

a graphic display of data comparing the frequency vs measurement.

Question 7

Mean

Answer 7

average of all the results

Question 8

Median

Answer 8

the middle value of the range of values

Question 9

Mode

Answer 9

the most frequent measurement

Question 10

Skewness

Answer 10

a measure of asymmetry

Question 11

Standard Deviation (SD)

Answer 11

A quantitative expression to describe the dispersion of a data set. Not applicable if data is non-gaussian. Standard deviation is the most commonly used description of spread. “Average” distance from the center of the data and every value in the data set. Represented by SD, s, or σ. Sidenote: SD2 is also known as the variance.

Question 12

Coefficient of Variation (CV)

Answer 12

SD as a percent of the mean. Normalized measure of dispersion. Helpful for comparing SDs with different units and different magnitudes (i.e. high and low controls). CV (%) = (SD/mean) x 100

Question 13

Accuracy

Answer 13

the agreement between the best estimate of a quantity and the true value or the closeness to the true value. Inaccuracy is often expressed as the plus or minus percent of the estimate from the true value.

Question 14

Precision

Answer 14

the agreement between random measurements. Imprecision is most often measured by the standard deviation, the variance, or standard error.

Question 15

Random error

Answer 15

= good accuracy + poor precision. An error that can be either positive or negative resulting in imprecision.

Question 16

Systematic error

Answer 16

= poor accuracy + good precision. An error that is always in one direction resulting in inaccuracy. The difference between the mean and the true value.

Question 17

SE =

Answer 17

= Constant Error (CE) + Proportional Error (PE)

Question 18

Total Error =

Answer 18

= RE + SE

Question 19

Total Error =

Answer 19

= RE + (CE + PE)

Question 20

Sensitivity

Answer 20

the % of sick patients who test positive. Sens = (TP/(TP+FN))*100

Question 21

Specificity

Answer 21

the % of well people who test negative. Spec = (TN/(TN+FP))*100

Question 22

Predictive value (+)

Answer 22

the probability that a positive test result predicts the patient has the disease. P(+) = (TP/(TP+FP))*100

Question 23

Predictive value (-)

Answer 23

is the probability that a negative test result predicts the patient does not have the disease. P(-) = (TN/(TN+FN))*100

Question 24

Efficiency (of a test)

Answer 24

the percent of test results that reflect the patients actual state of wellness or sickness. E = ((TP+TN)/TOTAL)*100

Question 25

Prevalence (of disease)

Answer 25

number of people in a population having the disease. Prevalence = ((TP+FN)/TOTAL)*100

Question 26

The highest sensitivity is desired when

Answer 26

The disease is serious and should not be missed. The disease is treatable.

Question 27

The highest specificity is desired when

Answer 27

The disease is serious but not treatable or curable. Knowing the disease is absent has public health or psychological value.