The Impact of Color of the Instructions and Questions on the Respondents’ ‘Survey Recognition, Response Time, and Response Accuracy’ in Designing a Survey

Abstract

Background In recent times, there have been frequent cases of insincere respondents as the survey environment changed from the previous printing basis. Insincere responses, which create errors and missing values, tarnish the meaning of the survey as a data collection tool. This study aims to discover the cause of such insincere responses through a pilot research. In order to solve the problem, the study applied color, — a cognitive information design element —, in designing the survey and analyzed its effectiveness in terms of survey recognition, response time, and response accuracy. Methods A pilot study was conducted through structural in-depth interviews in respect to qualitative research to discover the cause of insincere responses. A new survey tool was designed with cognitive information design elements based on the earlier theories. Lastly, 90 men and women over the age of 19 who were divided into control and experimental groups between October and December 2022. The survey was composed of 13 questions. Afterwards, We used SPSS 21.0 to compare the groups’ answers to verify the effectiveness of applying color in designing the survey. Results According to the results of difference checks and crossover analysis, the experimental group’s t value at –2.180** was higher in the cognitive section’s question ① (definition of data visualization). Also, in the response time section’s question ⑤ (length of time spent on checking the question), the experimental group’s value of 1.950 was shorter than that of the control group. For the response accuracy section’s question ⑦ (confirming the visual checklist required by the questionnaire), the experimental group showed 20 percent higher accuracy than the control group. In the case of question ⑧ (checking how to properly write down one’s answer in accordance with the questionnaire), the experimental group demonstrated 15.5 percent higher answer rate than the control group.