Gender-linked differences in the use of references to quantity and references to location in written language: A gender schema approach

A study was conducted to examine the influence of gender based schematic processing on usage of the language features References to Quantity and References to Location in written language. Using narratives obtained by Parkhill (1991), these language features were evaluated for differences in usage by individuals who varied on the dimensions of gender and gender-role orientation. Parkhill's participants included 153 men and 121 women attending East Texas State University who were first categorized using the Bem Sex Role Inventory (BSRI; Bem, 1974). From these cases, randomly selected students from each BSRI category resulted in a sample of 60 women (ages 18--51) and 60 men (ages 18--50) who authored stories in response to a problem-solving prompt. The narratives were rated by four trained judges for use of the two language features. Each sample received two independent ratings and features. Each sample received two independent ratings and the number of occurrences for each variable was aggregated across judges with the mean divided by the total word count of the sample to obtain a score representing proportional usage for each writer. A square-root transformation of the proportional scores was performed so that the data distributions more closely approximated the assumptions for appropriate use of analysis of variance (ANOVA) procedures. A 2 x 4 factorial ANOVA (completely randomized design, fixed effect model, equal sample sizes) was conducted to test the effects of gender and BSRI category placement on each of the dependent variables. No significant differences in usage were observed between genders; however a significant difference between BSRI categories was observed, supporting one hypothesis of the study. Those individuals placed in the Masculine BSRI category were found to use significantly more References to Location than those in the Undifferentiated category.