A Socioeconomic-Stratified Dataset on Chikui (吃亏) Decision Behaviors among Chinese Adults Assessed via Cognitive–Affective Dual Dimensions (2018–2020)

This dataset relies on the "Decision Hierarchy Theory: Construction and Validation of Cognitive Emotional Integration Framework" project to systematically collect and organize behavioral, eye movement, emotional, and socio-economic background information of the Chinese adult population in the "loss taking" intertemporal decision-making task, aiming to reveal the synergistic mechanism of cognitive and emotional maturity on decision quality. The data collection work was completed in three stages from 2018 to 2020, covering 23 provinces in China. A total of 1433 participants aged 18-78 were recruited, with over 55 occupational categories and a wide range of social representativeness. The first phase of the behavioral experiment was conducted at the Behavioral Laboratory of Shandong University from October 2018 to March 2019, with a total of 187 valid participants recruited (79 males aged 18-25). Using E-Prime 2.0, 13 "lose first, gain later" situations of financial loss were presented, and the computer automatically recorded whether the subjects accepted the loss, decision reaction time (in milliseconds), and PANAS emotion scale pre-test and post test scores; The decision reasons were classified by two independent coders according to the "latent/apparent dimensions", with a consistency of κ=0.87. The second stage of eye tracking experiments was conducted in the same laboratory from June to September 2019. From the first stage, 20 and 19 individuals with high/low self differentiation were screened, respectively (a total of 39 individuals, 10 males and 29 females, with an average age of 21.35 ± 1.28 years). Use the Eyelink 1000 Plus eye tracker (sampling rate 1000 Hz, 17 inch display, resolution 1024 × 768, line of sight 60 cm) to complete 144 attempts of the "probability × latent dimension" task, outputting the percentage of fixation time, fixation frequency, and pupil diameter (in pixels) for each region of interest. The third stage of large-scale questionnaire survey was conducted from December 2019 to January 2020 in the waiting room and online channels of Jinan West Station. A total of 1246 valid questionnaires were collected (806 females, accounting for 64.7%, with an average age of 38.92 ± 15.89 years). The Likelihood of Loss Scale (CLS, α=0.80), Subjective Well Being Scale (SWLS, α=0.88), and the Chinese Social and Economic Status Index (SEI) calculated based on Li Chunling's revised formula were collected, including 8 indicators such as education years, monthly income, management level, and unit nature. After cleaning the data with Excel 2016, SPSS 26.0, and R 4.1.2, three types of tables were formed: ① Behavior and Emotion Data (187 rows x 26 columns), with row labels indicating subject ID and columns containing demographic, CLS score, DSI score, loss making ratio, average reaction time (ms), PANAS difference, expected benefit amount (in hundred yuan), etc; ② Eye tracking data (39 rows x 48 columns), with row labels indicating subject ID and columns containing the percentage of fixation time, fixation frequency, and average pupil diameter (pixel) for high probability, low probability, long vision, and short vision regions of interest; ③ Socioeconomic status data (1246 rows x 15 columns), with row labels as subject IDs and columns containing CLS, SWLS, SEI, and eight sub indicators. The currency unit has been standardized to "hundred yuan". All three types of data are stored in both CSV and Excel formats, with a missing value ratio of less than 0.7%, mainly due to a small amount of income or education years not filled in, and have been processed using multiple imputation method. The reaction time and eye movement indicators have a system error of ± 1 ms or ± 0.1 pixel due to device accuracy, and random effects have been included in the mixed model. The file does not require specialized software and can be read directly using Office, R, or Python.