Questionnaire segmentation design method

Abstract

The invention discloses a questionnaire segmentation design method, which comprises the steps of S1, calculating correlation coefficients of variables in a long questionnaire, and clustering the variables into different data items by clustering the correlation coefficients; S2, segmenting the questionnaire based on the questionnaire segmentation mode, the data items obtained by clustering and thequestionnaire duration to obtain each questionnaire mode; S3, calculating the optimal sample size of each questionnaire mode based on the constraint that the fixed cost precision is the highest or thefixed precision cost is the lowest; S4, randomly issuing questionnaires based on the optimal sample size, and collecting survey data; S5, processing the survey data to obtain processed data; and S6,performing data research based on the processed data, and writing a report. According to the questionnaire segmentation design method provided by the invention, the project answering rate and the dataquality can be improved.

Description

technical field [0001] The present invention relates to the technical field of questionnaire design and questionnaire data processing, and more specifically relates to a questionnaire segmentation design method. Background technique [0002] With the rapid development of big data and artificial intelligence technology, market research data still has its own unique advantages and irreplaceable value in the field of consumer insight. Only research data representing people's "attitude" and big data representing people's "behavior" Only the perfect integration can gain a deeper and more comprehensive insight into consumers. Market research design is the invisible "tentacle" behind consumer insights, and its impact on research results cannot be underestimated. However, compared with the current survey method, which is gradually shifting to multiple and rapid changes in the Internet and mobile Internet, it is relatively lagging behind, and even still follows the traditional quest...

AI Technical Summary

Problems solved by technology

The boredom of the respondents will increase the possibility that the data is not within the estimated range, either exaggerating or reducing the true score of the answer, or inflating or weakening the correlation between variables, which will directly affect the inference of the survey target;( 4) Increased friction between respondents and survey items

Especially for the fixed sample, the interviewees have a negative cooperative attitude, which is not conducive to the normal development of more follow-up survey projects

It can be seen that the lengthy and complicated questionnaire has become an unbearable burden for market research design

Images