Our survey examined the various influences that go into a student’s choosing of an area of study. We looked to see if students placed different amounts of emphasis on certain factors depending on their area of study.
Our first null hypothesis states there is no relationship between arts and humanities students placing a different emphasis on availability of jobs and getting a job with a higher salary than would social sciences and science students.
Our second null hypothesis states there is no relationship between students in any area of study placing a greater influence on having a social impact.
We distributed our survey through facebook. Each member of the group posted the survey on their facebook page, which allowed their friends to have an opportunity to complete the survey.
The graph below shows the distribution of students in a specific area of study. There are four areas of studies: SS= Social Science, Hum = Humanities, Sci = Natural Sciences or Mathematics, and Art = Fine arts. The students answering the survey were primarily in the natural and social sciences:
barchart(tally(~Study, data = d, margins = FALSE, format = "count"), auto.key = TRUE)
In our survey, we asked students how confident they are in completing their major. Many students answering the survey were confident or very confident that they will finish their major. The graph below shows the distribution.
barchart(tally(~newConfidence, data = d, margins = FALSE, format = "count"),
auto.key = TRUE)
We also asked students how satisfied they are with their choice of major(s). The majority of students are satisifed with their choice of major(s).
barchart(tally(~newSatisfaction, data = d, margins = FALSE, format = "count"),
auto.key = TRUE)
In our survey we asked students whether certain factors influenced their decision to choose their major. In this section we graph and explain which factors influenced students to choose their majors in the areas of social sciences, humanities, natural sciences or mathematics, and fine arts.
There was a slightly larger number of students in humanities that said that the strength of their program influenced their decision to choose their major. Students in natural sciences, fine arts, and social sciences had more or less similar answers about the influence of program strength on their decision to choose their majors.
mosaicplot(Study ~ ProgramStrength, data = d, las = 2, col = rainbow(5))
Most of the students answered that the easiness of a major had little influence in choosing their major.
mosaicplot(Study ~ MajorEasiness, data = d, las = 2, col = rainbow(5))
Same for easiness of professors; most students did not find that the easiness of professors had a substantial influence in choosing their major.
mosaicplot(Study ~ ProfessorEasiness, data = d, las = 2, col = rainbow(5))
When choosing a major, fine arts students placed the greatest emphasis on interest in their classes. Social science, humanities, and natural science or mathematics students also answered that interest in their classes was an influencing factor in choosing their majors.
mosaicplot(Study ~ Interest, data = d, las = 2, col = rainbow(5))
Many students in natural sciences or mathematics and social sciences said that availability of jobs connected to their major was a significant influence in choosing their majors. On the other hand, job availability connected to their major was less of an influence for fine arts and humanities students.
mosaicplot(Study ~ Jobs, data = d, las = 2, col = rainbow(5))
Getting a job with a high salary had a similar influence on choosing a major for students as availability of jobs connected to their major.
mosaicplot(Study ~ Salary, data = d, las = 2, col = rainbow(5))
Fine arts students found that parental influence had little or no influence in choosing their majors opposed to the students in the other fields of study.
mosaicplot(Study ~ Parents, data = d, las = 2, col = rainbow(5))
Making a social impact influenced humanities students to choose their major slightly more than students in other fields of study.
mosaicplot(Study ~ SocialImpact, data = d, las = 2, col = rainbow(5))
Preparation for graduate school was a greater influence for students in natural sciences or mathematics and social sciences than students in fine arts and humanities.
mosaicplot(Study ~ GradSchool, data = d, las = 2, col = rainbow(5))
Does the availability of jobs and getting a job with a high salary matter to students choosing a major in the areas of fine arts and humanities?
Here is a logistic regression model:
mod = glm(Study%in%c("Art","Hum")~as.numeric(Jobs)+as.numeric(Salary),data=d,family="binomial")
The regression table:
summary(mod)$coef
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.5534 0.6235 2.491 0.01272
## as.numeric(Jobs) -1.1351 0.3629 -3.128 0.00176
## as.numeric(Salary) -0.7037 0.4198 -1.676 0.09373
The coefficients on Jobs and Salary are negative, so it seems to be that the job availability and getting a job with a high salary is not as strong of an influence in choosing a major in fine arts and humanities. However, when we look at the p-values of Jobs and Salary, we observe that for Jobs, the value is less than 0.05 so the coefficient is significant and thus we can assert that job availability is not as strong of an influence in fine arts and humanities, but since the p-value of Salary is not less than 0.05, there is not enough evidence that getting a job with a high salary influences choosing a major in fine arts and humanities.
To have a look at whether there is a relation between the availability of jobs and getting a job with a high salary, we modeled the variables Jobs and Salary below.
mod1 = lm(as.numeric(Salary) ~ as.numeric(Jobs), data = d)
summary(mod1)$coef
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.4633 0.1241 3.733 2.377e-04
## as.numeric(Jobs) 0.6181 0.0528 11.706 3.378e-25
The data shows that there is a relationship between Salary and Jobs, and that people who care about the availability of Jobs within their area of study also care about getting a job with a high salary.
We also observed the influence of social impact on choosing one's major. We found out that making a social impact plays a bigger role in choosing an area of study for students in humanities.
mod2 = lm(as.numeric(SocialImpact) ~ Study - 1, data = d)
The regression table:
summary(mod2)$coef
## Estimate Std. Error t value Pr(>|t|)
## StudyArt 2.125 0.28190 7.538 2.664e-12
## StudySci 1.861 0.09397 19.806 3.949e-46
## StudyHum 2.320 0.15947 14.548 1.005e-31
## StudySS 2.214 0.09530 23.235 8.653e-55
When looking at the data we see that the coefficients for humanities is the greatest. This shows that making a social impact influenced humanities students more than other students when choosing their area of study.
In our modeling analysis section, we observed that there is not enough evidence to reject our null hypothesis that getting a job with a high salary has no influence in choosing a major in fine arts or humanities. To be able to reject our null hypothesis we found that we would need to increase our sample size to 800 respondents to yield a p-value of 0.01 or less.
largerSample = resample(d, size = 800)
mod = glm(Study %in% c("Art", "Hum") ~ as.numeric(Jobs) + as.numeric(Salary),
data = largerSample, family = "binomial")
Regardless of area of study, students who answered the survey placed similar levels of influence on strength of program, easiness of professor, easiness of major, and interest in classes in choosing their majors.
In addition, we found the parental influence for choosing a major played less of a role for fine arts students than other areas of study while preparation for grad school was a bigger influence for natural sciences or mathematics and social sciences students.
Through our analysis we found that availability of jobs influenced fine arts and humanities students, although less than social science and natural science students. The humanities students placed a higher emphasis on making a social impact than students in other fields when choosing a major.
The members who distributed the survey are either in the field of social science or natural science, which may explain the high amount of participation from those fields and lower amounts from fine arts and humanities.
Because the survey was distributed through facebook, the modelers knew the students who participated, which results in a non-random sample.