Prepare for your exams

Study with the several resources on Docsity

Earn points to download

Earn points by helping other students or get them with a premium plan

Guidelines and tips

Prepare for your exams

Study with the several resources on Docsity

Earn points to download

Earn points by helping other students or get them with a premium plan

Community

Ask the community

Ask the community for help and clear up your study doubts

University Rankings

Discover the best universities in your country according to Docsity users

Free resources

Our save-the-student-ebooks!

Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors

Instructions on how to use sas software to analyze response rates in a survey using chi-square tests and confidence intervals. A real-life example of a survey study where participants were contacted via letter or phone before receiving the survey. How to read in the data, compute frequencies, create a two-way table, and perform chi-square tests and confidence interval calculations. It also includes questions for further exploration.

Typology: Lab Reports

Pre 2010

1 / 3

Download Analyzing Response Rates in a Survey: Chi-square Test and Confidence Intervals using SAS and more Lab Reports Statistics in PDF only on Docsity! Lab 13: Two-way tables STT 421: Summer, 2004 Vince Melfi We’ve already used proc freq to help in analyzing the output of simulations. Today we’ll see that proc freq also performs tests for two-way tables as described in Chapter 9 of the text. 13.1 Chi-square tests Exercise 9.31 in the text presents data collected to determine whether contacting people by phone or letter before sending them a survey will increase the response rate. Specifically, one group of people received a letter before getting the survey; one group received a phone call before receiving the survey; and one group didn’t receive any information before the survey arrived. For this study, a response was defined as returning the survey within 2 weeks. The data are contained in the file U:\msu\course\stt\421\summer04\survey.dat. For reference, the contents of the file are given next. 01 yes letter 171 02 no letter 220 03 yes phone 146 04 no phone 68 05 yes none 118 06 no none 455 First comes the record number (1 through 6), then response status (yes or no) then contact method (letter, phone, none), then the number in the group. For example, there were 171 people who responded and received a preliminary letter; there were 220 people who didn’t respond and received a preliminary letter. The following SAS code reads in the data, drops the useless variable record and prints the raw data. data survey; infile ’U:\msu\course\stt\421\summer04\survey.dat’; input record response $ contact $ number; drop record; proc print data = survey; run; Since the data contains the numbers in each group rather than a separate observation for each subject, we’ll need to use the weight statement in proc freq. For example, the following SAS code will compute the frequencies for each variable response and contact. proc freq data = survey; weight number; 1 run; 1. What proportion of subjects responded? 2. What proportion of subjects received a phone call? We want to create a two-way table and see whether contacting the subject before the survey arrives is related to whether the subject responds. To create the table we use the tables statement. To peform the chi-square test of association we use the chisq option. proc freq data = survey; weight number; tables contact * response / chisq; run; 13.1.1 Explanation and questions 1. Each cell of the two-way table contains the count in the cell, the percentage of the whole dataset in the cell, the percentage of that row’s data in the cell, and the percentage of that column’s data in the cell. For example, the upper left cell represents those who received a letter but didn’t respond. The number 220 tells us that there were 220 people in this category. The percentage 18.68 tells us that these 220 people are 18.68% of all people in the study. The percentage 56.27 tells us that 56.27% of those who received a letter didn’t respond. The percentage 29.61 tells us that 29.61% of those who didn’t respond received a letter. 2. What percentage of those who were not contacted responded? What proportion of those who were contacted by phone responded? Does it seem that contacting the sub- jects increases response? 3. After the table comes the output from the statistical tests. We’ll concentrate on the first line labeled “Chi-Square.” The other lines present other statistical methods for analyzing these data. What is the value of the Chi-Square statistic? Would you reject the null hypothesis of independence at the level α = 0.01? 2