Overview of the Job Market Process

The process of applying for an academic job follows predictable rhythms, and we advise applicants to review the work of other scholars that thoroughly details these patterns in political science (Carter and Scott Reference Carter and Scott1998; Drezner Reference Drezner1998; Simien Reference Simien2002). We provide a synopsis here, focusing on the creation of the application packet.

Departments begin posting jobs through APSA's eJobs listing service toward the end of the summer. Applications are generally due between early September and December, with most deadlines occurring in October and November. The eJobs listing is a primary source for U.S. academic jobs, although other sources may be consulted.Footnote ³ After selecting a list of potential jobs, the applicant prepares a packet of materials that search committees will consult in making their decisions about which candidates to bring in for interviews and ultimately hire.Footnote ⁴ The typical job market packet consists of a cover letter, a curriculum vitae, three letters of recommendation, and one or two writing samples. As Carter and Scott have noted, “Without being overly ostentatious, the package should be attractive and effectively organized, and the most relevant and important information that you want to convey should jump out at the reader who just ‘skims’ the materials” (Reference Carter and Scott1998, 617). The applicant will assemble the packets and mail them to the institutions to which he or she is applying. The administrative task of mailing 10, 20, 50, or even hundreds of these documents can be daunting; muRL tames the process by automating several of the most time-intensive tasks.

Simplifying the Job Market with Research Tools

Political science training increasingly involves learning software for data processing, statistical estimation, and scientific presentation. Having invested time learning these tools for research purposes, we encourage applicants to exploit them in order to simplify the creation of job packet materials. We employ R to both accelerate the gathering of job data and manage the LaTeX typesetting system to produce a file of ready-to-mail letters.

Why prepare these documents with LaTeX? This free, open-source typesetting system creates beautiful documents so that the user can focus on content and structure. However, until now, LaTeX has been difficult to use for mail-merges, especially when data are being drawn from a spreadsheet. We have simplified and automated this process to allow even novice users to quickly gain LaTeX's typesetting advantages without investing too much time in learning a new way to process documents.

Why do we scrape and process text with R? R is a free language that is well-known for statistical estimation, but it can also collect, organize, and output text as data. For readers unfamiliar with R, a wide array of introductory materials is available (Crawley Reference Crawley2007; Venables, Smith, and the R Development Core Team Reference Venables, Smith and Team2010). R enables users to scrape the web and parse text without being required to learn a new general-purpose computing language. In fact, R users can often harness the strengths of other languages such as Python, C++, or Perl through add-on packages. Jackman (Reference Jackman2006) offers a succinct demonstration of these capabilities as applied to vote data on the web.Footnote ⁵ Scholars have created general tools that can put R to work scraping web pages and manipulating text (Feinerer Reference Feinerer2008; Feinerer, Hornik, and Meyer Reference Feinerer, Hornik and Meyer2008; Temple Lang Reference Temple Lang2009a; Temple Lang Reference Temple Lang2009b). Specific implementations include the automatic collection and processing of U.K. House of Commons voting data (Firth and Spirling Reference Firth and Spirling2003; Firth and Spirling Reference Firth and Spirling2006).

We provide a suite of functions in R that directly imports APSA eJobs data into a spreadsheet, reads any spreadsheet of job information, and outputs a single file of customized letters and mailing labels. The next sections offer examples of how to use these tools to automate data collection and letter creation, so that less time can be spent preparing documents. Users of R and LaTeX should find the implementation of our approach straightforward; for readers who are new to these tools, the process of obtaining muRL, R, and LaTeX is relatively simple and free.Footnote ⁶ There are many reasons why these free and open-source products are well-suited for conducting analyses and writing papers.Footnote ⁷ Here, we provide an efficient method of conducting mail-merges for any purpose, although we focus on political science job application letters.

Figure 1 outlines our method. First, you prepare a spreadsheet of job data, such as that shown in figure 2. This preparation can be done either by hand or automatically by using our apsahtml2csv function (which we detail in the following section). Second, muRL creates a TeX file using write.murl, a function that processes the spreadsheet and incorporates your letter contents. Third, you customize the letter, which may involve altering the TeX file to adjust margins, font sizes, and so on. Last, you compile the TeX file to create a single PDF or PS file containing all the letters.

Figure 1

The muRL Protocol

Figure 2

Typical User-Created Spreadsheet of Job Entries

Note. Columns represent fields of job information; rows represent available positions. Columns include the institution name, job type, deadline, contact individual, department and position titles, and mailing information.

Automated Scraping of APSA Ejobs Data

The first step outlined in figure 1 is preparation of the job spreadsheet. Without muRL, this step involves a substantial amount of data entry. For each job, the user must manually enter the address and other job-specific information into a spreadsheet that will later be used for the mail-merge. If you are applying to positions listed in eJobs, our tool automates the data collection and entry processes for you. Even if you choose to forgo the rest of muRL's functionality, its ability to instantly process many eJobs listings into a spreadsheet can save you considerable time.

Automatically populating a spreadsheet of APSA eJobs information requires two steps: saving the position information presented in the eJobs listing and processing this information using muRL's apsahtml2csv function. To save the job information, first log on to the APSA eJobs webpage. Access to this page requires current APSA membership, a worthwhile investment if you are looking for political science positions. After logging on, locate the jobs that interest you by entering search criteria such as subfield specialty, geographical region, and start date. Figure 3 shows this interface.Footnote ⁸

Figure 3

Typical APSA eJobs Search

When you find a position of interest, navigate to the eJobs summary page for that position. Here, you will find a full description of the position, contact information for the hiring committee, and other related details. These job data will be formatted in the same manner as the example in figure 4.

Figure 4

Typical APSA eJobs Listing

From your web browser's menu bar, select “Save Page As” or “Save As” and save only the page's source html code into a directory (we recommend saving with the extension “.html”). Then repeat this process for every job that interests you. You will now have a directory of .html pages and need only run a single muRL command to convert this raw information into a spreadsheet. If (a) the directory of .html pages is on the desktopFootnote ⁹ and is labeled “eJobs,” (b) the desired output file name is “myjobs.csv,” and (c) you saved all the files with extension “.html,” then the muRL command will be

Of course, you could produce the spreadsheet input to muRL in any way you desire. If you are applying to many political science jobs, however, the apsahtml2csv function will save a great deal of time. Other tools can be combined with our package to create even greater efficiencies in the job application process. For example, you could use apsahtml2csv to create the spreadsheet and then store this information online as a collaborative document using an interface such as Google Documents or Dropbox. Several applicants could split the work of making amendments to the spreadsheet, because a particular position's address, contact person, and so on would be the same for every applicant.Footnote ¹⁰

From the Spreadsheet to the Letters

Once the spreadsheet of job application information has been created either manually or by using the apsahtml2csv function, you can process it by calling read.murl, a single function from our muRL package.Footnote ¹¹ If the spreadsheet is named myjobs.csv on the desktop, read.murl processes the data into an R object (here called “murljobs”) so that write.murl can process it laterFootnote ¹²:

Calling zip.plot(murljobs) creates a map of the zip codes in your spreadsheet, allowing you to see the geographic distribution of jobs. Figure 5 displays the locations of the positions from the data in figure 2's spreadsheet.

Figure 5

Typical muRL Map of Job Locations

Note. Data used were drawn from figure 2.

Next, you can use a single command (write.murl) to create a TeX file. Many options are available to customize the preamble, body, and formatting information to be included in the relevant letters. Full documentation is available for the complete array of options available through arguments in each of muRL's functions. Most important, you should customize your letters to emphasize how your qualifications will benefit the particular department or departments to which you are applying. You can specify the body of the letters directly by using the letter.text argument:

Alternatively, you can create the body of your letters in a separate file, using a text editor such as Emacs, Notepad, or TextEdit. If you save this file as content.txt, you can use the letter.file argument to automatically insert its content into your TeX document:

All that remains is for you to compile the TeX file and send the letters. muRL produces a single file of the letters and associated mailing labels (including postal bar codes). Figure 6 presents an example of a letter created by muRL; the accompanying bar-coded mailing labels are shown in figure 7. If you compile the TeX file using pdflatex, a single PDF file will be created. If the search committee asks that letters be submitted electronically, the PDF file can easily be separated into several letters before uploading or e-mailing. This task can be accomplished by using LaTeX's pdfpages package or other PDF utilities.Footnote ¹³

Figure 6

Typical muRL Application Letter

Figure 7

Typical muRL Mailing Labels

Conscientious job seekers are likely to peruse eJobs frequently to stay apprised of the new openings on the market that continually emerge. Prior to mailing applications, we recommend that you regularly add .html pages to the chosen directory and then process the full set of pages using apsahtml2csv. As positions emerge after you have begun sending out packets, you can simply create a new directory with the .html pages of the most recent positions and use the function whenever you are ready to apply. Since apsahtml2csv scrapes and stores the application closing date, you may also subset jobs by due date, subspecialty, or any other parameter.