Question

In the U.S., there are two sources of national statistics on crime rates: (i) the FBI’s Uniform Crime Reporting Program, which publishes summaries on all crimes reported to police agencies in jurisdictions covering virtually 100% of the U.S.; (ii) the National Crime Survey (NCS), which is based on in- terviews with a nationwide probability sample of households. In 1992, 4.9% of the households in the NCS sample told the interviewers they had experi- enced a burglary within the past 12 months. The same year, the FBI reported a burglary rate of 32 per 1,000 households, or 3.2%, for the U.S. as a whole. If the FBI’s 3.2% figure is the true burglary rate for the population of U.S. households in 1992, what is the probability that the National Crime Survey would obtain a result of 4.9% or more? You may assume that the NCS is based on a simple random sample of 50,000 households out of a population of 100 million households. What do you conclude?

Answer 1

X ho thano. X 1s IS na 50000 a binomial urth D- 0032 E(x) op1600 X ~ N(山フべ)

hous o holds 4-9 이○一一0に一the in tho sanoplos had expeusncad 50000told that ural e observued value ol 4.9% at Soooo X is 0.o 5000o)245 2%-ou mou ho uso hold experience loulglade P(X2 2450) hle ara approximatine a binomial disto wrtit use a Comrection - PIx 7 249.5)

P(x 2449.5 1548-8- PLz 21-56) Thepmobabili ex pexian buiglais Zaro, we can da that the the n the samplo biase The responzas obbaned for the not一一honest- ones and can notre ba relied a)