‘He blogged vs She blogged’: A Corpus-Based Language Analysis of 11,000 Blog Posts

Abstract

This article aims at analysing differences in language use by men and women bloggers of Pakistani English e-newspapers. The corpus-based analysis compares two gendered corpora comprising 11258 blog posts in all (6706 posts from 1674 men and 4552 from 1212 women bloggers) collected from blog archives of leading English e-newspapers of Pakistan. The data comprised blog posts uploaded during November 01, 2008 to August 31, 2015. Using automated text analysis tools—Linguistic Inquiry & Word Count (LIWC) and AntConc—the study employed a quantitative top-down approach for analysing both the corpora along the 93 in-built language variables of LIWC2015. With XLSTAT, the LIWC-computed results were subsequently subjected to statistical analyses through data normality measures, two-tailed hypothesis tests, Bonferroni’s Correction, and effect size calculations. As a result, several gender differences in language use were found, some consistent with the previous research conducted in the contexts outside Pakistan.