Abstract:
This study investigates gender bias in students' handwriting evaluation in the
Tamale Metropolis in the Northern Region of Ghana. The study employed
purposive, convenience, and simple random sampling for selecting students, as
well as examiners. One-way ANOVA was used to examine the impact of
different script levels on the identification of students’ gender based on their
handwriting. The t-test was used to assess the impact of examiners’ level of
experience in the identification of students’ gender based on their handwriting.
Students’ gender was modelled using a binary logistic regression model. The
study employed a test of proportion to estimate a cut-off for significant
identification of gender by examiners. The study found that gender
identification through handwriting becomes easier as students progress from
basic through to tertiary education levels, with upper primary and tertiary level
scripts being the least and most identified respectively. It was also revealed that
only a few examiners demonstrate the ability to correctly identify students'
gender based on their handwriting with a significant cut-off point of 65%. The
study also found that the identification of students’ gender based on handwriting
is influenced by the examiner's experience and script level. Finally, the binary
logistic regression model showed that the estimated probability of accurate
gender identification by examiners increases with the examiner's experience
level. Further studies must use larger and more diverse handwriting samples,
explore additional factors such as specific writing tasks and handwriting
characteristics, and replicate the study in different contexts to enhance the
generalizability of the study finding findings.
