Déjà-vu and Data Viz: The Job Application Process
This week, Dr. Katie Schenk offers advice on how to document and analyze the job application process by getting vulnerable about her own job search experience.
Over recent installments in this column, I have shared my insights into applying for jobs in public health, focusing on the importance of sharing information, clarifying expectations, and making decisions. In today’s installment, I will get vulnerable and share some personal experiences of how that job search process has been working out for me lately, and what I have learned from it.
Inspired by the powerful work of Dr Kinga Stryszowska-Hill, I have documented my recent job search process and the associated outcomes in a data visualization called a Sankey diagram.
Sankey diagram: Background on methods and data
A Sankey diagram is typically used to illustrate relationships flowing between categorical variables. In this case, I am illustrating the relationship between the stages of the job application process that I reached and the eventual outcome of that job application. So, I am cross-tabulating between stage of process (submitted written application / interviewed once / multiple interviews), and outcome (ghosted / rejected / accepted / withdrew or declined).
You can see from my data that over the last 4 years (2019-2022), I have made 143 job applications, and I have had 35 interviews. During this 4-year period, I also engaged in several intermittent periods of contract employment, each of up to 9 months’ duration, when I was less active on the job search.
Following my own advice, I have been highly selective about the jobs to which I have applied. I have applied only to positions that I consider to represent an improvement upon my current situation. I have high-level specialist qualifications and 20+ years of experience, so I’ve focused on applying exclusively to mid-to-senior level specialist positions where my skills and training will be an asset. Most of the positions to which I have applied have titles such as Senior Epidemiologist or Public Health Analyst.
For those of a more data-hungry or methodological bent, contact me directly and I would be happy to share more details of my analytic process. Or tell me where to publish: what’s the professional journal for jobseekers?
The most striking observation highlighted by this visualization is that a whopping 57% of my job applications ended up with me being ghosted — ie, the employer did not respond to let me know whether or not I had been rejected. (Based on the patterns of responses I received, I defined “no response” to mean no response within 3 months — see below.)
You can see that ghosting most commonly happens after the submission of written materials only — but let’s remember that a written job application still requires a substantial investment of my increasingly limited reserves of effort, energy, enthusiasm, optimism, and TIME in order to create and tailor individualized cover letters and application materials. (Not to mention the utterly redundant re-entering of all resume information into the employer’s applicant tracking system: just one of many ways in which the recruitment process shows a lack of courtesy and consideration to candidates — don’t get me started!)
You can also see that 10% of the ghosting is happening after an interview or even multiple interviews. That’s not just a waste of time — it’s also increasingly rude and disrespectful for employers to leave candidates waiting after having met with them personally and started to establish a personal relationship.
What do we have to do as a community in order to communicate to employers that ghosting job seekers is unacceptable? In the next installment of this series, I will focus on the practice of ghosting and explore these patterns more deeply.
Time taken to respond
Another key finding from my dataset is that the mean time between submitting an application and eventually receiving a rejection is 95 days, with a range of 1-565 days. That’s an average of more than 3 months, all the way up to a maximum of 1.5 years. (These data exclude all the ghosting — so really this is an underestimate of waiting time. My fellow biostatisticians will recognize that these data are right-censored.) And yes, you read that right: one employer waited 1.5 years to send me a rejection notice, and another waited an entire year. Up to 13% of the rejections were after more than 6 months (data not shown, but they still hurt my heart).
You can see that there were 3 applications during this entire period where I accepted an offer for employment (albeit temporary) — so what can we learn from this? What are the common features of a successful application?
For each of these 3 successful applications I went through at least 2 interviews following the submission of written materials. In one case, I was required to make 5 separate appointments for virtual interviews in a process that extended over 2 months before finally landing a job offer. (In a different era, would these have been scheduled as in-person panel interviews on a single day?)
I recall another interview process that extended through multiple interviews including a panel, making it through to the final round after 5 weeks. At the end of this period, I received an email saying that a different candidate had been selected for the position, but I was welcome to re-submit my application to be considered for contracting work. Despite my tortured feelings about this process, I even considered doing so — until I saw the original job being actively posted online again by the same individual within minutes! Let’s be clear: this process represents a dishonest bait-and-switch by the employer, who is trying to obtain access to labour under false pretences. And furthermore, this employer does not have a very high estimation of their candidates’ powers of observation.
If that’s how these employers consider it ok to treat candidates, then I think I have had some very lucky escapes.
Careers advisors suggest examining job applications data from the perspective of a funnel that narrows down the pool of applicants through an increasingly restrictive process, from submitting written applications through being interviewed to receiving job offers. They propose that examining the ratios between each stage can offer insights into where the job seeker is experiencing obstacles. From the perspective of an employer recruiting to fill a vacancy, an average of 8.4% of applications for jobs in a national database are converted into interviews (Source: Jobvite). From those interviewees, an average of 36.2% will receive a job offer. According to my data from the last 4 years, I have converted 16% of my applications into interviews, and 2% of my interviews into offers. Does this mean that my written applications are performing well and my presentation within an interview setting is poor, compared to these national benchmarks? Or does it indicate that the perspective of an employer-based analysis cannot readily be translated to the perspective of an individual candidate? I welcome opinions.
Make your own version
I have learned a lot from analyzing my job applications in this way — for example, now I know that the best time to give up hope and assume that I have been ghosted is 3 months after submitting a written application (which then became the definition used in this analysis). Would you be interested in creating your own version of this diagram to reflect your own unique path through the job search journey and seek your own individualized lessons?
To make this diagram, I kept track of my applications data mostly in real-time using Excel, then uploaded the dataset to the Flourish website for data visualization features. Other approaches might include Sankeymatic or programming with R. If there is interest, perhaps I’ll think about offering some instructions and templates — let me know.
FYI @Kinga’s R code is here — her definitions and categories differ from mine, as we are interested in different research questions. Think about the categories that best suit you and reflect your questions. An alternative approach is by @Eric James Stephens PhD. Thanks to @Kinga, @Lewis M., and @Imran Mohammed for the technical recommendations — more information and discussion here.
Questions (join me for discussion in the comments or on LinkedIn)
- Do you find the Sankey chart or the funnel analysis to be helpful tools for analyzing the job search? (Interested in learning more? Want more technical resources to create your own Sankey chart or calculate your own funnel ratios? Get in touch through the chat group.)
- How long is an appropriate response time for an employer to acknowledge a candidate’s application?
- How can we bring the HUMANITY back into HUMAN RESOURCES?
Read previous columns in this series:
- Job Applications: How Much Will You Get Out of Bed For?
- What Does Job Security Mean in Public Health?
- It’s Not You, It’s Public Health
- Rebuilding the Public Health Workforce: A Summary that Wants to Be a Manifesto
- Unity, Community, Immunity, Opportunity: Lessons Learned from Writing About Public Health
- What Actually Works? Careers Advice in Public Health, Part 2
- Using LinkedIn to Your Advantage: Careers Advice in Public Health, Part 1
- It’s a Jungle Out There: Power Balance and Job Applications in Public Health
- Atrophy and Adjunctification: Changes in Public Health Employment Opportunities
- Advice for Building a Career in Public Health — Does Any of It Really Work Anymore?
- Mental Health and Wellbeing Among Public Health Professionals
- Public Health Workspaces
- The Public Health Workforce Is Not Okay: Lessons from the Public Health Frontline
- Dr. Katie Schenk is an infectious disease epidemiologist and public health informatics specialist. She has been working on the public health frontline for governmental Health Departments throughout the COVID-19 pandemic. Currently, Dr. Schenk is serving as a member of the US Medical Reserve Corps at COVID-19 vaccination and testing sites. She teaches Public Health and Global Health at American University in Washington DC and George Mason University, VA. Previously, Dr. Schenk led a portfolio of social and behavioral research studies on children and families impacted by HIV and AIDS in sub-Saharan Africa at the Population Council. Visit her website: https://kdspublichealth.com/about-dr-katie-schenk/ Follow her on Twitter: @skibird613 and LinkedIn: dr-katie-schenk-4a884b84
- Students of Public Health2023.01.23Students Who Rocked Public Health 2022
- Students of Public Health2022.12.01Deadline Extended to Nominate a Student Who Rocked Public Health in 2022
- JPHMP Direct Voices2022.10.19Preview Issue for Public Health Workforce Interests and Needs Survey
- Uncategorized2022.10.12Partnering for Success in One Ohio County