Introductory Biostatistics

are preventable (5166 American women died from cervical cancer in

1977). In an e¤ort to find out who is being or not being screened for

cervical cancer (Pap testing), data were collected from a certain com-

munity (Table E3.1). Is there a statistical relationship here? (Try a few

di¤erent methods: calculation of odds ratio, comparison of conditional and

unconditional probabilities, and comparison of conditional probabilities.)

TABLE E3.1

Race

Pap Test White Black Total

No 5,244 785 6,029

Yes 25,117 2,348 27,465

Total 30,361 3,133 33,494

3.2 In a study of intraobserver variability in assessing cervical smears, 3325

slides were screened for the presence or absence of abnormal squamous

cells. Each slide was screened by a particular observer and then re-

screened six months later by the same observer. The results are shown in

Table E3.2. Is there a statistical relationship between first screening and

second screening? (Try a few di¤erent methods as in Exercise 3.1.)

TABLE E3.2

Second Screening

First Screening Present Absent Total

Present 1763 489 2252

Absent 403 670 1073

Total 2166 1159 3325

3.3 From the intraobserver variability study above, find:

(a)The probability that abnormal squamous cells were found to be

absent in both screenings.

(b)The probability of an absence in the second screening given that

abnormal cells were found in the first screening.

(c)The probability of an abnormal presence in the second screening

given that no abnormal cells were found in the first screening.

(d)The probability that the screenings disagree.

3.4 Given the screening test of Example 1.4, where

sensitivity¼ 0 : 406

specificity¼ 0 : 985

142 PROBABILITY AND PROBABILITY MODELS