IDENTIFYING PEOPLE LIKELY TO RESPOND ACCURATELY TO SURVEY QUESTIONS

• US 20130110584A1
• Filed: 10/26/2012
• Published: 05/02/2013
• Est. Priority Date: 10/28/2011
• Status: Active Grant

• Current Assignee: Lightspeed
• Original Assignee: Global Market Insite, Inc.
• Inventors: Eggers, Mitchell, Drake, Eli
• Granted Patent: US 9,639,816 B2
• Time in Patent Office: Days
• 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

Abstract

A survey data structure is described that specifies how to administer a survey to each of a group of people. The data structure contains an entry that specifies text that poses a fabrication gauge question for which a percentage of the population who can accurately answer the fabrication gauge question affirmatively is known. The data structure further contains a plurality of entries that each specify text posing an unlikely question for which the percentage of the population who can accurately answer the unlikely question affirmatively is small. The data structure specifies how to conduct a survey that serves as a basis for determining which people to whom the survey is administered are responding truthfully to the survey.

Claims

1. A method in a computing system, comprising:

causing to be posed to each of a plurality of people;

a plurality of unlikely questions, each of whose affirmative answer is accurate for only a relatively small percentage of the population, and

a fabrication gauge question whose affirmation answer is accurate for a known percentage of the population regardless of whether it is answered accurately by people to whom it is posed;

receiving responses to the posed questions provided by the people to whom they were posed;

determining a percentage of people who answered the fabrication gauge question in the affirmative; and

discarding people from the plurality of people in decreasing order of the number of unlikely questions to which they answered affirmatively until the percentage of affirmative answers to the fabrication gauge question among undiscarded people of the plurality is within range of the known percentage.

2. The method of claim 1 wherein the unlikely questions and fabrication gauge questions are caused to be posed in the same survey with substantive questions to which the respondents'"'"' answers are sought.

3. The method of claim 1 wherein the unlikely questions and fabrication gauge questions are caused to be posed in same survey with routing questions for routing respondents to further surveys for which they are expected to qualify.

4. The method of claim 1 wherein the unlikely questions and fabrication gauge questions are caused to be posed in a separate survey from substantive questions to which the respondents'"'"' answers are sought.

5. The method of claim 1, further comprising causing to be posed to each of a plurality of people, together with the unlikely questions, at least one likely question, each of whose affirmative answer has a relatively high likelihood of being accurate,

6. A computer-readable medium having contents adapted to cause a computing system to perform a method for analyzing survey responses, the method comprising:

receiving a plurality of completed survey responses, each survey response being generated by a person and specifying an answer to each of a plurality of questions, the plurality of questions comprising a question of a first type and a plurality of questions of a second type, the first question having a selected answer that is accurate for a known percentage of people, each of the questions of the second type having a typical answer and an atypical answer,

of the responses completed, selecting responses to discount such that (a) the number of undiscounted responses containing the selected answer to the first question divided by the total number of undiscounted responses is at least approximately equal to the known percentage, and (b) no unselected response specifies more atypical answers to questions of the second type than any selected response.

7. The computer-readable medium of claim 6 wherein the plurality of questions further comprise a question of a third type having at least two answers,

the method further comprising, for each of the answers to the question of the third type, counting the number of unselected responses that specified the answer to the question, to the exclusion of selected responses that specified the answer to the question.

8. The computer-readable medium of claim 6 wherein the plurality of questions further comprise a question of a third type having a free-form answer,

the method further comprising, for each of the answers to the question of the third type, compiling the free-form answers specified by the unselected responses in response to the question, to the exclusion of the free-form answers specified by the selected responses.

9. A computer-readable storage medium storing a survey data structure specifying how to administer a survey to each of a plurality of people, the data structure comprising:

an entry specifying text posing a fabrication gauge question for which a percentage of the population who can accurately answer the fabrication gauge question affirmatively is known; and

a plurality of entries each specifying text posing an unlikely question for which a percentage of the population who can accurately answer the unlikely question affirmatively is small

such that the data structure specifies how to conduct a survey that serves as a basis for determining which people to whom the survey is administered are responding truthfully to the survey.

10. The computer-readable storage medium of claim 9 wherein the entries further comprise at least one entry specifying text posing a substantive question whose answer is sought from the people.

11. A computer-readable storage medium storing a survey response data structure specifying how each of a plurality of people have responded to a survey, the data structure comprising a plurality of entries each corresponding to a person who has responded to the survey, and each comprising:

a first component containing data indicating the person'"'"'s response to a fabrication gauge question for which a percentage of the population who can accurately answer the fabrication gauge question affirmatively is known; and

a plurality of second components each indicating the person'"'"'s response to an unlikely question for which a percentage of the population who can accurately answer the unlikely question affirmatively is small

such that the data structure serves as a basis for determining which of the people responded truthfully to the survey.

12. A computer-readable storage medium storing an adjusted survey response data structure indicating results for a survey in which each of a plurality of people has responded to a plurality of questions, the data structure comprising a plurality of entries each corresponding to a person who has responded to the survey, and each comprising data that, for each of one or more substantive survey questions, indicates an answer given by the person to whom the entry corresponds in response to the substantive survey question, the data structure containing entries corresponding only to people determined to be likely to have responded to the survey accurately, to the exclusion of people determined to be unlikely to have responded to the survey accurately,

such that the contents of the data structure can be used to determine answers given to substantive questions only by people determined to be likely to have responded to the survey accurately, to the exclusion of people determined to be unlikely to have responded to the survey accurately.