Jump to content


Photo
- - - - -

Analyzing The Relationship Between Ap Computer Science Score, Gender And Race Using Two-Way Anova

computer statistics applied science

  • Please log in to reply
3 replies to this topic

#1 JunyanZhang

JunyanZhang

    Curious

  • Members
  • Pip
  • 2 posts

Posted 02 May 2017 - 02:17 AM

In this paper, I used two-way ANOVA (analysis of variance) to explore whether gender and race would influence student’s academic performance in some special majors. To be more specific, I would like to know how students perform when they are studying computer science. I used two-way ANOVA to test the null hypothesis and use the F value to find the answer to the question. Data was collected based on the survey sent out by me.   

 

 

When I was reading articles related to Statistics, I found a really interesting one.  Randy Olson is an author who wrote several controversial gender-related articles. In his article, “Average IQ of Students by College Major and Gender ratio”, he gave a really interesting plot.  What the plot demonstrates is that if one major has lower female student ratio, the average IQ of students will be higher. Those so-called male-dominated majors (for instance, physics, astronomy, computer science) tend to have higher average IQ. However, the estimate of IQ was based on SAT scores, which, in my point of view, was not representative and even misleading. I was really curious about the relationship between the academic performance and gender. In addition to gender, I was also interested in knowing how the races influenced the academic performance. It is widely accepted that in the field of computer science, many minority groups are underrepresented. How does the combination of gender and race influence students’ performance when studying computer science? In this paper, I want to explore how the gender and race factors influence students’ academic performance in the field of Computer Science.

 

Attached File  Research Essay.pdf   124.9KB   4 downloads



#2 Buffy

Buffy

    Resident Slayer

  • Administrators
  • 8946 posts

Posted 03 May 2017 - 12:15 AM

Cool. What did you find?

 

Can you summarize that for folks that are too lazy to download and read your entire paper? 

 

Are there aspects you haven't resolved that some of our members might be able to give you feedback on?

 

We're all about discussion here, so we prefer it if you'd expound a little more in your posts and give people something to respond to.

 

Welcome to Hypography, Henry!

 

 

No history can be a faithful mirror. If it were, it would be as long and as dull as life itself. It must be a selection, and, being a selection, must inevitably be biased, :phones:
Buffy


#3 spartan45

spartan45

    Advanced Member

  • Members
  • PipPipPip
  • 84 posts

Posted 03 May 2017 - 03:22 PM

In this paper, I used two-way ANOVA (analysis of variance) to explore whether gender and race would influence student’s academic performance in some special majors. To be more specific, I would like to know how students perform when they are studying computer science. I used two-way ANOVA to test the null hypothesis and use the F value to find the answer to the question. Data was collected based on the survey sent out by me.   

 

 

When I was reading articles related to Statistics, I found a really interesting one.  Randy Olson is an author who wrote several controversial gender-related articles. In his article, “Average IQ of Students by College Major and Gender ratio”, he gave a really interesting plot.  What the plot demonstrates is that if one major has lower female student ratio, the average IQ of students will be higher. Those so-called male-dominated majors (for instance, physics, astronomy, computer science) tend to have higher average IQ. However, the estimate of IQ was based on SAT scores, which, in my point of view, was not representative and even misleading. I was really curious about the relationship between the academic performance and gender. In addition to gender, I was also interested in knowing how the races influenced the academic performance. It is widely accepted that in the field of computer science, many minority groups are underrepresented. How does the combination of gender and race influence students’ performance when studying computer science? In this paper, I want to explore how the gender and race factors influence students’ academic performance in the field of Computer Science.

 

attachicon.gifResearch Essay.pdf

Have you considered the sort of things I have noticed, mentioned below:  

I remember being in a mixed class with a young woman German language teacher who seemed to exploit the fact the boys appeared to be weak at learning a foreign language compared with the girls. She would make fun of the boy’s poor language ability and bully those especially weak at the subject.  Only years later did I find out male and female brains are wired differently, giving girls a distinct advantage at language. Much is made of Einstein failing a college entrance exam. Interestingly, the subject that caused the problem was his poor French language exam results.

http://www.nytimes.c...?pagewanted=all

 I have listed the 3 main differences that may affect academic abilities between men and women, using information from the reference at the foot of this post.

  1. Mathematical abilities. An area of the brain called the inferior-parietal lobule (IPL) is typically significantly larger in men, especially on the left side, than in women. This section of the brain is thought to control mental mathematical ability, and probably explains why men frequently perform higher in mathematical tasks than do women. Interestingly, this is the same area of Einstein’s brain that was discovered to be abnormally large. The IPL also processes sensory information, and the larger right side in women allows them to focus on, "specific stimuli, such as a baby crying in the night."
  2. Language. Two sections of the brain responsible for language were found to be larger in women than in men, indicating one reason that women typically excel in language-based subjects and in language-associated thinking.
  3. Spatial ability. Men typically have stronger spatial abilities, or being able to mentally represent a shape and its dynamics, whereas women typically struggle in this area. Medical experts have discovered that women have a thicker parietal region of the brain, which hinders the ability to mentally rotate objects–an aspect of spatial ability. Research has shown this ability in babies as young as 5 months old, negating any ideas that these abilities were strengthened by environmental influences.

http://www.mastersof...-womens-brains/


  • JunyanZhang likes this

#4 JunyanZhang

JunyanZhang

    Curious

  • Members
  • Pip
  • 2 posts

Posted 04 May 2017 - 11:41 PM

Cool. What did you find?
 
Can you summarize that for folks that are too lazy to download and read your entire paper? 
 
Are there aspects you haven't resolved that some of our members might be able to give you feedback on?
 
We're all about discussion here, so we prefer it if you'd expound a little more in your posts and give people something to respond to.
 
Welcome to Hypography, Henry!
 
 
No history can be a faithful mirror. If it were, it would be as long and as dull as life itself. It must be a selection, and, being a selection, must inevitably be biased, :phones:
Buffy


Sure I can expound more. However, the paper includes tables and special symbols that can not be displayed in the post. So I will just add my conclusion for people who are interested . Sorry for the inconvenience!

According to the P value, we have the following conclusion:

P(gender) is greater than a(alpha), so we fail to reject the null hypothesis that the population means of each gender group are equal. There is no significant difference between the mean female score and the mean male score.

P(race) is less than a , so we reject the null hypothesis that the population means of each race group are equal. There is significant difference between the mean score of students with different races.

P(interaction) is greater than a , so we fail to reject the null hypothesis there is no interaction between the gender factor and the race factor.

So based on the ANOVA analysis, the average score obtained by female students is not significantly different from the average score obtained by male students. However, the races do have an effect on the average score.



Also tagged with one or more of these keywords: computer, statistics, applied science