大米消费决策研究 - Projects

Household purchases and restaurant procurement

Households consider taste, size, packaging, and meals; catering buyers also need standards, cost control, and reliable supply. I separated these interview groups so research questions could reflect distinct purchasing situations and support recommendations relevant to each.

One staple, two purchasing situations. Research framework or retained data structure; record counts do not imply valid samples or causal findings. — Household and catering situations determine different purchasing questions. Interview-framework reconstruction.

Literature, consumer text, interviews, and questionnaires

The research combines 11 references, online consumer text, semi-structured interviews, and questionnaires. Each contributes a different view: variable ideas, natural expression, purchasing context, and comparable items. Together they organize the inquiry without independently establishing causal relationships.

Connect consumer expression with buying reasons. Research framework or retained data structure; record counts do not imply valid samples or causal findings. — Consumer text, interviews, and questionnaires contribute different evidence types. Research-framework illustration.

Segmentation, co-occurrence, and sentiment

Retained outputs include segmentation, keywords, frequencies, co-occurrence matrices, and sentiment-related tables. The full-text table has 2514 rows and the sentence table 4029. These are different analytical units and cannot be added as independent respondents; preserving the unit gives scale a clear meaning.

Turn scattered expression into analysis material. Research framework or retained data structure; record counts do not imply valid samples or causal findings. — Full-text and sentence rows are different analytical units. Retained analysis-table structure.

667 questionnaire records

Raw and processed questionnaires each retain 667 records. The processed table adds six dimensions covering product quality, price promotions, perceived cost, brand awareness, brand quality, and loyalty, calculated from associated item means. Row-level checks support this transformation; retained records do not equal valid samples or establish scale validity.

From 667 records to six evaluation dimensions. Research framework or retained data structure; record counts do not imply valid samples or causal findings. — 667 retained records support six derived dimensions; record count is not the valid sample size.

Recommendations for different buying situations

The research organizes discussions around positioning, pricing, product improvement, and channels. Household questions concern taste and quality recognition; catering questions concern cost, requirements, and supply relationships. Interview and chapter materials support these directions, while full quantitative conclusions require correspondence with the final report.

Which value belongs to which audience? Research framework or retained data structure; record counts do not imply valid samples or causal findings. — Recommendations connect purchasing situations with product and procurement value. Draft research implications.

My independent work

I independently completed questionnaires, interviews, text and statistical analysis, and report writing, submitted the Zhengda Cup project, and received a Sichuan provincial second prize. The work turned everyday preferences into specific questions supported by different materials. Responsibilities and the prize are author-confirmed; the reconstructed framework explains methods without implying causal or commercial impact.

My contribution: an independent research project. Research framework or retained data structure; record counts do not imply valid samples or causal findings. — Independent research and a provincial second prize are author-confirmed. Reconstructed materials explain the method.