Voterlistsonline.com offers the following options:
1.No polling sample - All records fitting the selected criteria are exported.
2.Random Sample - The user enters the number of records they wish to export, for example 5,000. All the records fitting the select criteria are assigned a random number and ordered by this random number. The first 5,000 records (per this example) would then be selected for export.
3.Targeted Random Sample - The user selects a target field, for example congressional district. By clicking on the “allocate” button, the application will display the list of values in the selected field and count of available records per each. Using the congressional district example, all of the congressional districts and the count of records per each one meeting the selected criteria would be displayed. The user then goes through the list and enters the number of records per each congressional district they wish to pull. For example, the user might want 1,000 records from each congressional district. The final file will consist of the total number of records selected by the user, allocated as the user specifies based on the target field selected.
4.Cluster Sample with target group of - Records fitting the selected criteria are ordered using a geographic sort. The user then selects the number of clusters they wish to pull, and how many records should be in each cluster (cluster size). For example, the user may specify 100 clusters of 25 records each (for a total of 2500 records). The records fitting the selected criteria would then be segmented into 100 clusters, and the top 25 records from each cluster will be selected for export. (Note that if you choose to export voters instead of households, you will very likely get multiple members of each household as they will be ordered together geographically).
5. Cluster Sample with target group of --fieldname- - Performs the same as a cluster sample with no target group defined, except that when a target group is selected the application will pull the requested number of clusters per each value in the target field. For example, if the user specifies 100 clusters with a size of 25 with a target group of “congressional district”, the application will pull 100 clusters of size 25 for each available congressional district in the sample universe. Thus if the user has a query defined which contains three different congressional districts, the final pull would be 7500 records instead of 2500 if no target group were selected.
6. Normalized Random Sample - The Normalized Random Sample option allows the user to make a random pull proportioned according to the counts of available records in a group of selected target fields. For example a user might wish to generate a 25,000 record sample normalized by gender and party. In this case the user would enter the total number of records desired (in this case 25,000) and then select which clauses in their query should define the universe from which to base the normalized sample. If the user does not select any clauses, the universe will consist of the entire population of non-moved/non-dead records in the state(s) being queried. To normalize on gender and party the user would then select gender and party from the list of fields. The pull is made by first counting the numbers of each gender and party combination in the defined universe, and then making a proportionate pull based on those percentages to arrive at the total number of records requested.
Pricing for Samples is by record.