Personal Section

“It does appear that on many, many different human attributes - height, weight, propensity for criminality, overall IQ, mathematical ability, scientific ability - there is relatively clear evidence that whatever the difference in means - which can be debated- there is a difference in standard deviation, and variability of a male and female population”.

As soon as then-Harvard president Lawrence H. Summers made this statement on January 14th, 2005 as part of a speech during the NBER Conference on Diversifying the Science and Engineering Workforce, a storm of controversy arose over his “sexist” pronouncement. But was he really wrong? Conventional wisdom and Educational Testing Service standardized test statistics do in fact indicate that males may well be “better” at math than are females. However, there is no single explanation for gender differences in math related behaviors. When I began this research endeavor, I myself was an above average mathematics student. Was there really a limit to how far I could improve in mathematics because of my gender or was the theory of gender influencing mathematical ability just an outdated sexist rumor?

To answer this burning question, I looked to the students at my high school to help me determine if these gendered differences really exist. My research was supervised by my social studies projects teacher, Mr. Stephen Sullivan. With the help of various articles, dissertations and math teachers, I was able to compose an academic achievement questionnaire, survey and a mathematics test which were then successfully administered to 164 high school-aged subjects.

Although mathematics is the basic subject of this research, most of the mathematics that I had to learn was used to interpret the statistics garnered through my surveys in order to yield results. Mr. Sullivan taught me the basics of statistical analysis so that I could understand the pages of numbers placed in front of me after I had inputed the results of my surveys. Subjects’ responses were interpreted using t-tests, linear and multiple regressions, ANOVA and correlations.

The completion of this project gave me a new outlook on science. Although I have always loved the logical, problem-solving quality of mathematics, I have regarded the sciences as a field that was only necessary if I wanted to go into medicine or work in a laboratory. Throughout high school, I never enjoyed the more quantitative sciences such as Chemistry, preferring to take courses such as Environmental Science and Forensics. However, this study showed me that science helps solve problems everyday in the lives of ordinary people. It is not just for the person who wants to launch a rocket into space or who seeks to cure a rare type of infectious disease. Science can help dispel sexist rumors or show that along with aptitude, attitude toward mathematics and social stereotyping can affect the math scores of females.

I believe that the main reason that I was able to complete three years of strenuous research was because I knew that in the end, the results that I found would relate to my own life. I chose a problem that I was truly interested in finding solving, one which I knew that the results of which might change how society regards a widely held theory. To all who might like to undertake a project combining science and mathematics, I suggest picking a research topic about which you are truly passionate. If I were not honestly interested in learning the results of my research, I would have given up when I realized that I had to hand-grade 164 surveys!

Research Section

onventional wisdom suggests that males are simply “better” at math than are females. Lawrence Summer’s suggestion that females were inherently inferior to males in mathematic domains was controversial and politically incorrect, but was it inaccurate? The purpose of this study is to determine the degree to which gender influences mathematical ability and the relative influences of study habits, class attendance, and internalization of societal stereotypes. Previous studies have shown that males consistently outperform females on the math subsection of the Scholastic Aptitude Test (SAT), PSAT, ACT as well as math and science Advanced Placement (AP) exams. It has been suggested that influence from significant adults (parents and teachers) who hold gender stereotypical beliefs regarding mathematics can negatively impact females’ math scores. It has also been shown that parents tend to promote math and science activities in boys rather than girls during middle childhood. Length of study time and class attendance was considered in this research as it has been proven that these factors are directly proportional to academic performance. The current study sought to determine any inherent gender inferiority in regards to mathematic ability.

From the outset, it was predicted that there would be no significant difference in the mathematical aptitudes of males and females. Yet it was predicted that math scores would be significantly influenced for girls but not for boys by number of absences/ missed math classes, average math study time per evening, attitude response scores (ability, attitude, stereotypes, perceived task value) and aptitude.

After informed consent was obtained, the surveys, academic achievement questionnaire and mathematics quiz were distributed to 164 high school-aged subjects, aged 14 to 18. The surveys were administered to one of the most racially diverse schools on Long Island so the sample may be considered reasonably representative of the typical American suburban high school population. The academic achievement questionnaire consisted of four constructs: ability/expectancy, perceived task value, attitude, and gender stereotyping. The ability/expectancy construct included: a) a current expectancy for success in mathematics performance b) future expectancy for math performance c) self-concept of math ability and d) actual effort expended on math. The perceived task value construct included a) intrinsic interest value b) attainment value/ importance and c) extrinsic utility value. Students’ ID numbers, instead of names, were affixed to the surveys to ensure confidentiality. All surveys and tests were administered in math classes in a “classroom setting”. Students were instructed to “do your best” on the math problems, and were told that some material might be difficult. All questions were drawn from recent New York State Math A Regents exams, Part I (multiple choice section). At the conclusion of the survey/test session, all papers were immediately collected by the student researcher. Surveys were scored and then all information was entered into the database and analyzed via Statview 5.0 software. The school’s Guidance Department Chair then provided each student’s PSAT math subscore to the faculty advisor, listed by student ID number. At no point did the student researcher ever see a student’s name and survey/test results together.

As predicted, it was found that gender has no direct effect on mathematical aptitude in high school students. When a series of unpaired t-tests for gender effect was conducted it was discovered that boys were much more likely to endorse the stereotype that “boys are better than girls at math”. Aptitude as determined by PSAT math subscores was the only significant predictor of math test scores for males and the most significant predictor for females. However, it was discovered that missing class and attitude towards mathematics were also very significant predictors of math scores in females. The following table shows the regressions of all of the variables tested for both males and females:

	r2	p-value
AptitudeÞ math score (males)	.850	<.0001
AptitudeÞ math score (females)	.508	<.0001
Days missedÞ math score (m)	.024	.1848
Days missedÞ math score (f)	.192	<.0001
AttitudeÞ math score (m)	.029	.1414
AttitudeÞ math score (f)	.279	<.0001
AbilityÞ math score (m)	.023	.1882
AbilityÞ math score (f)	.024	.1541
Stereotyping Þ math score (m)	.017	.2563
StereotypingÞ math score (f)	.017	.2338
Perceived task valueÞ math score (m)	.003	.6513
Perceived task valueÞ math score (f)	.029	.1191
Study minutesÞ math score (m)	4.127E-5	.9561
Study minutesÞ math score (f)	.004	.5688

It is unclear why only females’ math scores are affected by attitude and class attendance in addition to aptitude while males’ scores are only affected by aptitude. Perhaps stereotypes have infiltrated the classroom, causing female students to feel that if they miss a day of class, they will fall much farther behind than their male counterparts due to their inherent inferiority in math. According to Ross (1992) “knowing something about a person’s attitude helps others explain and predict that individual’s actions [;]the study of people’s attitude thus provides a key to the human behavior”. The more positive a female student’s attitude towards math in general, the better she seemed to perform on the Math A Test. This may be due to societal stereotyping. The idea that females are inherently inferior to males in math may create in many females a more negative attitude. Perhaps, this negative attitude only influences females because most males expect that they will do well based on their gender. Those girls who reject such negative thoughts can outperform peers who feel defeated from the very moment they enter the classroom.

This research suggests that there is no significant difference in the mathematical aptitudes of male and female high school students. However, just thinking that females are inferior can cause negative changes in many girls math scores. Harvard’s Lawrence H. Summers probably never considered how harmful his speech was when he made it; Simply pointing out the “fact” that females are inferior to males in mathematics may have convinced a thousand more little girls that it was so.

Works Cited

• Asoyne, Eugene Ifeanyi Ben (2002). “Psychological Differences in Attitude Toward and Academic Achievement in Mathematics between Male and Female Students In Imo State Secondary Schools, Nigeria.” Nigeria: Walden University.
• a.Ben-Zeev, Talia. et. al.(2005). “Math is hard (Barbieä , 1994): Responses of Threat vs. Challenge-Mediated Arousal to Stereotypes Alleging Intellectual Inferiority.” Gender Difference in Mathematics. Cambridge: Cambridge University Press.
• b.Ben-Zeev, Talia. et. al.(2005). “Stereotypes and Math Performance.” Handbook of Mathematical Cognition. Ed. Jamie Campbell. New York, Psychology Press, 235-49.
• Catsambis, Sophia (2005). “The Gender Gap in Mathematics: Merely a Step Function?”
• Gender Difference in Mathematics. Cambridge: Cambridge University Press.
• Crombie, Gail et. al.(2005). “Predictors of Young Adolescents’ Math Grades and Course Enrollment Intentions: Gender Similarities and Differences.” Sex Roles. Vol. 52(5-6), Pages 351-357.
• Schuman, Howard, Howard Walsh, Camille Olson, and Barbara Etheridge (1985). “Effort and Reward”. Social Forces. Vol. 63(4). Pages 945-966.
• Stipek, Deborah (2005). “The Emergence of Gender Differences in Children’s Perceptions of Their Academic Competence.” Journal of Applied Psychology. Vol. 26(3), Pages 276-295.
• Summers, Lawrence H (2005). Address. NBER conference on Diversifying the Science and Engineering Workforce. Cambridge, 14, January.

• Social Science