Objective To develop the Rating Scale for Schizophrenic Features in Writings（ RSFiW） and analyze its reliability and validity. Methods The preliminary scale was developed by combining literature review and comparing the written language characteristics of 30 patients with schizophrenia hospitalized in Chongqing Mental Health Center from April 2015 to August 2018 with 32 healthy subjects recruited externally between September 2018 and April 2021. The preliminary scale was used to assess 157 patients with schizophrenia （from inpatients of several mental hospitals in Chongqing between September 2018 and April 2021） to carry out item analysis and exploratory factor analysis. Confirmatory factor analysis and internal consistency testing were conducted with an additional 154 patients（ Inpatient and outpatient patients with schizophrenia in Chongqing Mental Health Center during the same period of time） rated with the adjusted scale. Criterion validity was tested using the Positive and Negative Syndrome Scale and the Scale for Thought， Language， and Communication as validity scales. Results The scale contained 15 items that were decomposed into 5 factors of casual-deliberate， planned， fluent， offset， and redundant-abbreviated， explaining 53.912% of the total variance. Confirmatory factor analysis showed a good model fit（ χ2/df=1.248， CFI=0.930， TLI=0.901， SRMR=0.071， RMSEA=0.040）. Cronbach's α for the total scale was 0.565. Content validity was good. The factor scores correlated with some of the factor scores of the validity scale. Retest reliability Spearman r was 0.546. The inter-rater reliability for most items were moderate to strong with statistical differences（ weighted Kappa=0.325-0.787， P＜0.05）. The area under the receiver operating characteristic（ ROC） curve was 0.999（ P＜0.001）. When the maximum cutoff point of the Youden index was 5.5， the sensitivity was 99.4% and the specificity was 96.9%. Conclusions The development process of this scale is reasonable， and the reliability and validity are acceptable.