In chriswu1996/williamsgraduates: Find facts about the graduating class of Williams College

Abstract

The Williams Graduates package uses the data published online by the Williams College Registrar to compute statistics about the gender distribution of the graduated classes. This is done for sixteen years from 2000 to 2015 by scanning data directly from the text versions of the published documents. Data of 8354 undergraduate students were collected and analyzed to produce a series of statistical summaries in terms of timeplots to understand the change in gender distribution of the students as time progresses.

Introduction

The gender distribution of the students is an important factor of the college. The package williamsgraduates implements string manipulation on the data about each student in the graduated class from yearly catalogs and organizes the information of each student to an accessible dataframe.

The dataframe is further manipulated, and the data is presented with different timeplots of the gender distribution. These summary functions offer some preliminary analysis on the data, however the data can be used to perform more complex analyses, potentially with information combined from external data sources.

Data

The information used to construct the data in this package was taken from the website of the Office of The Registrar of Williams College in the form of pdf documents. The documents for each year from 2000 to 2015 were collected and manually converted into text files and pared down to only the pages that contained information on graduated students.

The information on students of the graduated class is first read line by line into the function.

> readLines(system.file("extdata", str_c(2000,".txt"), package = "williamsgraduates"), warn = FALSE)

library(stringr)
readLines(system.file("extdata", str_c(2000,".txt"), package = "williamsgraduates"), warn = FALSE)[1:10]

A series of operations is then conducted to produce the output dataframe. Lines that separated by line breaks are first joined back together into a single line. The "*" and "+" signs in front of the students' names are then removed. Suffixes of the name like "Jr." or Roman numerals are also removed. The first name, last name, department honor and latin honor are then extracted and put into a dataframe. The gender of the student is estimated from the student's first name using gender and genderdata packages.

> readgraduates(2000)

options(width=100)
library(williamsgraduates)
data("graduates2000")
graduates2000[1:10,]

However, since there are 8354 names to be estimated in total and the gender package takes a relatively long time to estimate the gender of each name, the data used for showstats is a pre-loaded dataframe using data().

Use readgraduates

The information within the locally stored text files can be accessed by using the readgraduates function which uses the method described above. This function outputs a dataframe containing the first name, last name, major(if available), department honors, latin honors and estimated gender for all graduated students in any given year from 2000 to 2015. However, as the gender package that estimates the gender of students takes a approximately 3-5 minutes to process over 500 names for each year, using, for example, data(graduates2000) to obtain the pre-loaded dataframe that contains the information on graduated students of 2000 would be a faster way. The function is used in the following way:

> readgraduates(2000)

The only argument is:

year: Selects the year of graduated students to be scanned.

Use showstats

The function showstats uses the data from the readgraduates function to perform appropriate analysis and generates timeplots of the gender distribution of the student population. This function has only one argument and is used in the following way:

> showstats(type)

The argument type has six values (graduates, latin honors, department honors, latin honors detail, department honors detail), their results are as follows:

> showstats("graduates")

williamsgraduates::showstats("graduates")

> showstats("latin honors")

williamsgraduates::showstats("latin honors")

> showstats("department honors")

williamsgraduates::showstats("department honors")

> showstats("latin honors detail")

williamsgraduates::showstats("latin honors detail")

> showstats("department honors detail")

williamsgraduates::showstats("department honors detail")

Conclusion

The summary statistics indicate that the estimated gender distribution of graduated students in 2015 was around 50.7% female, and the mean gender distribution of graduated students from 2000-2015 was around 50.2% female, a very even distribution. The graduated students had the most percentage of female students in 2006 at around 53.1%. The graduated students had the least percentage of female students in 2002 and 2011, both at around 47.4%. The graphics produced above yield some interesting results that might be worth noting. While the graduated students had on average an equal amount of male and female students, the mean gender distribution of students that graduated with latin honors was 54.56% female and 54.65% female for students that graduated with department honors. This means that on average, there were proportionately more female students who has attained latin and department honors than male students from 2000-2015. However, it is also worth noting that the gender package could not produce a gender estimation to less common names or non-English names, which might have caused the data to be biased.

Bibliography

Hadley Wickham (2016). stringr: Simple, Consistent Wrappers for Common String Operations. R package version 1.1.0. https://CRAN.R-project.org/package=stringr

Lincoln Mullen (2015). gender: Predict Gender from Names Using Historical Data. R package version 0.5.1.

Lincoln Mullen (2017). genderdata: Historical Datasets for Predicting Gender from Names. R package version 0.5.0. https://github.com/ropensci/genderdata

H. Wickham. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York, 2009.

Office of The Registrar of Williams College http://web.williams.edu/admin/registrar//catalog/archive.html

chriswu1996/williamsgraduates documentation built on May 13, 2019, 7:28 p.m.