Survey assignment method

• US 7,269,570 B2
• Filed: 12/18/2000
• Issued: 09/11/2007
• Est. Priority Date: 12/18/2000
• Status: Expired due to Fees

• Current Assignee: GFK Custom Research, LLC (Nielsen Holdings plc)
• Original Assignee: Knowledge Networks, Inc. (Nielsen Holdings plc)
• Inventors: Krotki, Karol P.
• Primary Examiner(s): BOSWELL, BETH V
• Application Number:
  US09/740,515
• Publication Number:
  US 20020077881A1
• Time in Patent Office: 2,458 Days
• Field of Search: 705/1, 705/10
• US Class Current: 705/7.32
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 30/0203 Market surveys; Market polls
• CPC Class Codes: G06Q 30/0204 Market segmentation

Abstract

A survey assignment method uses a weighting factor to compensate for members which are temporarily removed from a panel. This weighting factor allows for non-proportionate samples to be made from the panel. Additionally, a non-proportionate selection weighting factor can also be used. By using these weighting factors, a relatively random selection of the members for survey can be done.

Claims

1. A computer-implemented method comprising:

identifying a group within an available survey panel including a plurality of members, the group having predetermined characteristics;

determining a weight for each of the members, said weight being derived to match the members to demographics of the general public;

selecting a first set of members from the group for a first survey;

temporarily removing the selected first set of members from the available survey panel;

adjusting weights of remaining members of the available survey panel to compensate for the removal of the first set of members from the available survey panel and thereby to make the remaining members of the available survey panel match the demographics; and

selecting, with a processor, additional members from the available survey panel for a second survey, selection probabilities of the additional members being respectively proportional to the adjusted weights to compensate for the removal of the first set of members from the available survey panel;

wherein the step of selecting, with a processor, additional members includes;

determining an interval value that is equal to a summation of the adjusted weights of the remaining members of the available survey panel divided by the number of the additional members to be selected;

selecting a random number that is between zero and the interval value;

assigning an identification (ID) number to each of the remaining members of the available survey panel;

determining a cumulative weight for each of the remaining members of the available survey panel, the cumulative weight of each particular remaining member being determined by a summation of the weight of the particular remaining member and the weights of other remaining members whose ID numbers are smaller than the ID number of the particular remaining member; and

choosing the additional members from the remaining members, each of the additional members having a cumulative weight that matches the sum of the random number and an integer multiple of the interval value.

2. The method of claim 1, wherein the step of adjusting weights includes increasing the selection probabilities of remaining members in the group.

3. The method of claim 1, wherein weights of the remaining members of the panel are modified to compensate for the group members removed from the panel by using a weight factor proportionate to the number of original members in the group over the number of remaining members in the group in the available panel.

4. The method of claim 1, wherein multiple groups are identified within the available survey panel, and members of these multiple groups are selected for the survey and removed from the available survey panel, and wherein selection probabilities of the multiple groups are modified to compensate for the removal of the members of the group from the available survey panel.

5. The method of claim 1, wherein the step of selecting a first set of members includes:

determining an additional interval value that is equal to a summation of the weights of the members of the available survey panel divided by the number of the first set of members to be selected;

selecting a an additional random number that is between zero and the additional interval value;

assigning an additional identification (ID) number to each of the members of the available survey panel;

determining an additional cumulative weight for each of the members of the available survey panel, the additional cumulative weight of each particular member being determined by a summation of the weight of the particular member and the weights of other members whose additional ID numbers are smaller than the additional ID number of the particular member; and

choosing the first set of members from the available survey panel, each of the first set of members having an additional cumulative weight that matches the sum of the additional random number and an integer multiple of the additional interval value.

6. The method of claim 1, wherein one or more weighting factors are multiplied to the weights of the remaining members of the available survey panel to compensate for the removal of people who have been previously given a survey within a certain time period.

7. The method of claim 6, wherein the weighting factors include an original weighting factor to match the members of the available survey panel to the demographics of the general population, a removal weighting factor to compensate for all of the members of the panel removed from the panel and a selection weighting factor to deal with a non-proportionate selection.