Zogby Methodology 

SAMPLING 

The majority of telephone lists for polls and surveys are produced in the IT department at Zogby International.  Vendor-supplied lists are used for regions with complicated specifications, e.g., some Congressional Districts.  Customer-supplied lists are used for special projects like customer satisfaction surveys and organization membership surveys.

Telephone lists generated in our IT department are called from the 2002 version of a nationally published set of phone CDs of listed households, ordered by telephone number.  Residential (or business) addresses are selected and then coded by region, where applicable.   An appropriate replicate1 is generated from this parent list, applying the replicate algorithm repeatedly with a very large parent list, e.g., all of the US.

Acquired lists are tested for duplicates, coded for region, tested for regional coverage, and ordered by telephone, as needed.

The resulting list is loaded into the CATI application and the randomize function within the CATI software is run to further assure a good mix for the telephone list. 

INTERVIEWING 

Interviews are conducted at Zogby International by professional interviewers trained on our computer-assisted telephone interviewing (CATI) computer system.  A policy requiring one supervisor to no more than twelve interviewers is used.  The sample management module of the CATI system gives all prospective respondent households in the source telephone list the same chance of joining the sample.  Regional quotas are employed to ensure adequate coverage nationwide. 

WEIGHTING 

Reported frequencies and crosstabs are weighted using the appropriate demographic profile to provide a sample that best represents the targeted population from which the sample is drawn from.   The proportions comprising the demographic profile are compiled from historical exit poll data, census data, and from Zogby International survey data.

SAMPLING ERROR

Sampling Error, often referred to as the Margin of Error, is the percentage that survey results are likely to differ from the actual due to the size of the sample drawn.  If a survey were conducted of all the members of a population, the sampling error would be zero.  There are other sources of possible error in survey research such as sample design error and measurement error.

DEFINITIONS

1CATI Computer-Aided Telephone Interview.  This is software application that displays survey questions to interviewer at LAN workstation, stores survey responses keyed in by interviewer on server, and manages list disposition.

2Replicate A sub-list with the same cover characteristics as its parent list.  Replicates are generated from the parent list by selecting every nth  record from the parent list, where n is the size of the replicate / total records in parent list.

http://www.zogby.com/methodology/index.cfm