Content deleted Content added
m →Professional status: Corrected dangling participles. Amended wording. Deleted redundant word. Removed or corrected loaded/emotive phraseology. |
m →Gender and handwriting: Corrected number. (Leading zero before decimal point was missing.) |
||
Line 197:
=== Gender and handwriting ===
A 1991 review of the then-current literature concluded that respondents were able to predict the gender of handwriting between 57 and 78% of the time.<ref name="ReferenceB">{{Cite journal|doi=10.1080/0141192910170204|title=Sex Differences in Handwriting: A comment on Spear|year=1991|last1=Hartley|first1=James|journal=British Educational Research Journal|volume=17|issue=2|pages=141–145}}</ref> However, most of these samples, as well as subsequent studies, are based on small sample sizes that are collected nonrandomly. A much larger and more recent survey of over 3,000 participants only found a classification accuracy of 54%.<ref>{{Cite web |url=http://blog.survata.com/identifying-gender-by-handwriting-youre-probably-not-as-good-at-it-as-you-think#:~:text=Gender%20equality&text=Men%20identified%20male%20handwriting%20successfully,of%2049%25%20to%2045%25 |title=Identifying gender by handwriting — you're probably not as good at it as you think |access-date=2020-09-17 |archive-date=2020-08-10 |archive-url=https://web.archive.org/web/20200810042715/https://blog.survata.com/identifying-gender-by-handwriting-youre-probably-not-as-good-at-it-as-you-think#:~:text=Gender%20equality&text=Men%20identified%20male%20handwriting%20successfully,of%2049%25%20to%2045%25 |url-status=dead }}</ref> As statistical discrimination below 0.7 is generally considered unacceptable,<ref>{{cite journal |title=Receiver Operating Characteristic Curve in Diagnostic Test Assessment |journal=Journal of Thoracic Oncology |date=1 September 2010 |volume=5 |issue=9 |pages=1315–1316 |doi=10.1097/JTO.0b013e3181ec173d |last1=Mandrekar |first1=Jayawant N. |pmid=20736804 |doi-access=free }}</ref> this indicates that most results are rather inaccurate,<ref>{{Cite book|doi = 10.1007/978-3-030-01424-7_60|chapter = Handwriting-Based Gender Classification Using End-to-End Deep Neural Networks|title = Artificial Neural Networks and Machine Learning – ICANN 2018|series = Lecture Notes in Computer Science|year = 2018|last1 = Illouz|first1 = Evyatar|last2 = (Omid) David|first2 = Eli|last3 = Netanyahu|first3 = Nathan S.|volume = 11141|pages = 613–621|arxiv = 1912.01816|isbn = 978-3-030-01423-0|s2cid = 52909281}}</ref> and that variation in results observed is likely due to sampling technique and bias.<ref>{{cite journal |last1=Bradley |first1=Sean |title=Handwriting and Gender: A multi-use data set |journal=Journal of Statistics Education |date=March 2015 |volume=23 |issue=1 |pages=1 |doi=10.1080/10691898.2015.11889721 |s2cid=123033133 |doi-access=free }}</ref>
The reason for this bias varies; hypotheses are that biology contributes due to average differences in fine motor skills among males and females,<ref name="ReferenceB"/> and that differences arise from culture and gender bias.<ref>{{Cite journal|doi = 10.1080/00224545.1996.9712254|title = Inferring Gender from Handwriting in Urdu and English|year = 1996|last1 = Hamid|first1 = Sarah|last2 = Loewenthal|first2 = Kate Miriam|journal = The Journal of Social Psychology|volume = 136|issue = 6|pages = 778–782|pmid = 9043207}}</ref><ref>{{Cite journal|doi=10.1080/0141192890150304|title=Differences between the Written Work of Boys and Girls|year=1989|last1=Spear|first1=Margaret Goddard|journal=British Educational Research Journal|volume=15|issue=3|pages=271–277}}</ref><ref>{{Cite journal|doi=10.1080/00224540209603929|title=Judging Gender from Samples of Adult Handwriting: Accuracy and Use of Cues|year=2002|last1=Burr|first1=Vivien|journal=The Journal of Social Psychology|volume=142|issue=6|pages=691–700|pmid=12450344|s2cid=39650656}}</ref>
| |||