25.11.2014 Views

Biostatistics

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

62 CHAPTER 2 DESCRIPTIVE STATISTICS<br />

72.0 97.5 130.0 68.1 86.4 70.0 73.0<br />

59.7 89.6 76.9 74.6 67.7 91.9 55.0<br />

90.9 70.5 88.2 70.5 74.0 55.5 80.0<br />

76.9 78.1 63.4 58.8 92.3 100.0 84.0<br />

71.4 84.6 123.7 93.7 76.9 79.6<br />

45.6 92.5 65.6 61.3 64.5 72.7<br />

77.5 76.9 80.2 76.9 88.7 78.1<br />

60.6 59.0 84.7 78.2 72.4 68.3<br />

67.5 76.9 82.6 85.4 65.7 65.9<br />

Source: Data provided courtesy of Dr. N. Thilothammal.<br />

(a) For these data compute the following descriptive measures: mean, median, mode, variance,<br />

standard deviation, range, first quartile, third quartile, and IQR.<br />

(b) Construct the following graphs for the data: histogram, frequency polygon, stem-and-leaf plot,<br />

and boxplot.<br />

(c) Discuss the data in terms of variability. Compare the IQR with the range. What does the<br />

comparison tell you about the variability of the observations?<br />

(d) What proportion of the measurements are within one standard deviation of the mean? Two<br />

standard deviations of the mean? Three standard deviations of the mean?<br />

(e) What proportion of the measurements are less than 100?<br />

(f) What proportion of the measurements are less than 50?<br />

Exer cises for Use wit h Large Data Set s Availabl e on th e Foll owing Websit e: www .wiley.com /<br />

c ollege/da niel<br />

1. Refer to the dataset NCBIRTH800. The North Carolina State Center for Health Statistics and<br />

Howard W. Odum Institute for Research in Social Science at the University of North Carolina at<br />

Chapel Hill (A-20) make publicly available birth and infant death data for all children born in the<br />

state of North Carolina. These data can be accessed at www.irss.unc.edu/ncvital/bfd1down.html.<br />

Records on birth data go back to 1968. This comprehensive data set for the births in 2001 contains<br />

120,300 records. The data represents a random sample of 800 of those births and selected variables.<br />

The variables are as follows:<br />

Variable Label<br />

PLURALITY<br />

SEX<br />

MAGE<br />

WEEKS<br />

MARITAL<br />

RACEMOM<br />

HISPMOM<br />

GAINED<br />

SMOKE<br />

Description<br />

Number of children born of the pregnancy<br />

Sex of child ð1 ¼ male; 2 ¼ femaleÞ<br />

Age of mother (years)<br />

Completed weeks of gestation (weeks)<br />

Marital status ð1 ¼ married; 2 ¼ not marriedÞ<br />

Race of mother (0 ¼ other non-White, 1 ¼ White; 2 ¼ Black; 3 ¼ American<br />

Indian, 4 ¼ Chinese; 5 ¼ Japanese; 6 ¼ Hawaiian; 7 ¼ Filipino; 8 ¼ Other<br />

Asian or Pacific Islander)<br />

Mother of Hispanic origin (C ¼ Cuban; M ¼ Mexican; N ¼ Non-Hispanic,<br />

O ¼ other and unknown Hispanic, P ¼ Puerto Rican, S ¼ Central=South<br />

American, U ¼ not classifiable)<br />

Weight gained during pregnancy (pounds)<br />

0 ¼ mother did not smoke during pregnancy<br />

1 ¼ mother did smoke during pregnancy

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!