1
Assignment 2 – Sampling
JS 300
Fall 2005
Dr. Kent R. Kerley
This assignment is worth 25 points (5% of course grade) and is due by class time on
Nov. 1, 2005. The paper must be typed in 12-point Times New Roman. Please include a
title page with the title of assignment, your name, course name, instructor’s name, and
date. The assignment must be stapled in the upper left corner. Please do not place the
assignment in any type of folder or binder.
Suppose that you are given the task of conducting a crime victimization survey in
Birmingham using telephone interviewing. You would need a representative sample of
people to call, and you might use the city telephone directory to pick the sample. Each
student should obtain one page out of Bell South’s Real White Pages for Greater
Birmingham. For the purposes of this exercise, your “study population” is the set of
residential listings on the page you have chosen. Your job is to select one sample of n =
20 listings (name and telephone number), using the method outlined below.
1. Simple Random Sampling
Using the methods described in class and in the text, select a simple random sample of 20
residential telephone listings from the single page of the telephone listings you have
chosen, using the random number table (RNT) handed out during class. Since some
telephone listings will be ineligible for selection, you will need to select more than 20
random numbers to have enough (25 likely will be sufficient). Assign each listing on
your page (including those which are ineligible: see below) a unique number from 1 to
X, where X is the total number of listings on your page. Count by telephone numbers,
not names (i.e., each telephone number is an element). Before looking at the RNT decide
what row and column of the table you want to start on, to pick numbers. Starting there,
select the first 25 or so 3-digit numbers between 001 and X (e.g., between 001 and 305).
Then type out the name and telephone number of each eligible residential telephone
listing which corresponds with each of the random numbers, until you have selected 20
eligible listings. If you run out of random numbers before getting 20 eligible listings, just
go back to the RNT and, picking up from where you left off, continue picking numbers
until you have enough to get 20 eligible listings.
Reject as ineligible any clearly non-residential listings, such as business listings
(including those designated ofc [office], ins [insurance], or att [attorney]), church listings,
or governmental listings, FAX numbers, those designated “Children's Phone,” or any
others where the telephone number could not connect you with an adult in their
residence. Also, to insure that people with multiple telephone numbers do not have a
greater chance of being picked, skip telephone numbers which are the second or latter
number for a given residence (their first listed number is eligible).
Type up a detailed description of what you did, stating how you picked a starting place
on the page of random numbers, the row and column numbers where you started, what
direction you moved within the table, which 3 digits you used (e.g., first 3 digits of each
5-digit number, last 3 digits, etc.), what random numbers you selected, and what names
and telephone numbers the random numbers corresponded with. Also, explain any other
