Mediamark Research 
Spring 2008 Technical Guide

Data Adjustment Procedures

Ski and Skiing Adjustment

Because Ski is published 8 times annually and Skiing is published 7 times annually, a unique ascription procedure is applied to them in every Spring interviewing wave (i.e., odd numbered waves). Since issues are not published between July and August for Ski, and between May and August for Skiing, claimed readership during this interviewing period cannot reflect levels obtained when the magazines are published regularly. The ascription procedure involves adjusting the audience numbers by using the responses to the frequency question (On the average, out of 4 issues that are published, how many issues of MAGAZINE, do you read or look into? Is it 0, 1,2,3, 4?) to account for the interviewing period when issues are not published. The following levels adjust each response accordingly: .125 (for a 0 out of 4 issues response), .25 (for a 1 out of 4 issues response), .50 (for a 2 out of 4 issues response), .75 (for a 3 out of 4 issues response), and 1 (for a 4 out of 4 issues response).

Back to Table of Contents

Interviewer Estimating Procedures 

MRI requires that answers be recorded for household income, individual employment income and value of owned home for every respondent. In cases where a respondent is unwilling or refuses to provide one or more of these items, MRI instructs the interviewer to estimate a response and to indicate that the information is estimated. Interviewers are trained to use all information about the respondent and his/her neighborhood in making this estimate. The incidence of estimated responses is shown for each of these variables, respectively.

Back to Table of Contents

Effective Sample Size

(Based on Coefficient of Variation) 

Spring 2008 Report

Back to Table of Contents

Statistical Efficiency of Key Demographics

Spring 2008 Report

Back to Table of Contents

Demo/Media Data Ascription

MRI ascribes "no answers" to Personal Interview questions to provide the most complete database of our respondents. This is normally a nominal amount but can, on occasion, reach a percentage or greater than 5% of the total responses. Listed below are he instances (per Wave) when these responses exceeded this 5% threshold.

Back to Table of Contents

For more information regarding Mediamark Research, contact MRI at info@mediamark.com
Copyright 2008,  Mediamark Research