Rajiv Gandhi University of Knowledge Technologies: Resume Screening Bias Exercise for HR Interns

A blind resume screening exercise works by scoring the same set of resumes twice, once with identifiers visible and once with them removed, and then comparing the two sets of scores. If the ranking shifts, the difference points to what the identifiers were doing to your judgement. This guide gives you six invented sample resumes, a rubric and a comparison method you can run alone or with a partner in a couple of hours.

What a blind-screening exercise can and cannot show an HR intern

Resume screening is where many hiring decisions quietly start. A recruiter may look at a few hundred resumes and spend seconds on each. In that time, details unrelated to the job, such as a name, a photograph, an address, a college label or an age, can colour the reading without anyone intending it.

A blind-screening exercise is a small experiment on yourself. It does three useful things.

• It shows whether your scoring changes when identifiers disappear.
• It forces you to define what “good” means before reading, which is the main protection against drift.
• It produces a table you can show a mentor as evidence of careful thinking.

It cannot prove that you or anyone else is biased, because six resumes and one reader are a tiny sample. Treat the output as a prompt for reflection and for a better process, not as a verdict on anyone’s character.

Step one: define the role and the scoring rubric before you read any resume

The order matters. Write the rubric first, so the criteria cannot be bent to suit a favoured candidate. For the exercise, use an invented role: a junior data analyst trainee at a mid-sized company. The scenario is illustrative.

Notice what the rubric leaves out: name, gender, age, home town, family background and the name of the college. Those are exactly the details that the second round will remove.

Step two: six invented sample resumes with identifiers marked

All six resumes below are fictional and illustrative. The names, towns and details are invented, any resemblance to a real person is coincidental, and they are deliberately short. In your own copy, keep the identifiers in round one and strip them in round two.

The right-hand column is the substance. The designers of this exercise made R6 look impressive on identifiers but thin on evidence, and R1, R3 and R5 look modest on identifiers but strong on evidence. That is intentional so the comparison has something to reveal. In your own exercise, you can invent different mixes, but keep the substance and the identifiers independent of each other.

Step three: score round one with identifiers visible

Give yourself the full resume text as written, including identifiers, and score each resume against the rubric. Work fast, as a real screener would, and write each score in a table before moving on. Do not look back at earlier scores while scoring a later resume.

Step four: remove identifiers and score round two blind

Wait at least a day, or ask a partner to prepare the stripped versions so you cannot recognise them. Then redact the following and relabel resumes with random codes such as C-41 or C-77 so the order is not guessable.

• Name, and any initials that reveal community or region.
• Photograph, date of birth, age, marital status and parents’ details.
• Home address and town. Keep only the fact that the candidate is able to work in the required location, if relevant.
• College name, replaced with “degree, field, year”.
• Gaps in dates, replaced with total months of experience and learning, so that the rubric looks at evidence rather than at a pause.
• Pronouns and gendered words in summaries.

Score again using the same rubric and a fresh table. Resist the urge to remember what you gave before.

Step five: compare the two rounds with a before-and-after table

The comparison is the point of the exercise. Use one row per resume and compute the shift.

Worked illustration of reading the comparison

The numbers below are invented to show how to interpret a result. They are not findings about real people.

In this invented result, R6 fell from first to fifth once the institute name was hidden, while R5 climbed from sixth to second once the village address and career gap disappeared. A reader looking at those shifts would ask: which details was I rewarding in round one that the rubric never mentioned? The largest absolute changes (R6 at -8 and R5 at +6) are the places to examine first.

A few cautions apply when you read your own table. A change of one or two points may be ordinary noise from scoring fast. A pattern across several resumes is more informative than a single jump. And with only six resumes, you can only say “this is what happened in my exercise”.

Step six: turn the findings into process changes

The exercise is only useful if it changes how you screen. Here are practical changes an intern can propose to a mentor.

• Write the rubric before opening applications, and share it with everyone who screens.
• Remove identifiers in a first pass, where the tool or a coordinator allows it.
• Score criterion by criterion across all resumes (all skills scores first, then all projects scores) rather than resume by resume, which reduces the halo effect of an impressive first line.
• Ask two people to score independently and discuss only the gaps.
• Keep a short note of why each resume was advanced or rejected, in terms of the rubric.
• Review the shortlist as a whole to see whether any group of qualified candidates is disappearing for reasons unrelated to the role.

RGUKT Basar students: connecting the exercise to training and placement resources

Students at Rajiv Gandhi University of Knowledge Technologies, Basar, in Telangana, have a close view of the other side of screening, because the university’s Training and Placement Division works with employers on campus. The division’s official page describes its role as helping students develop professionally and acting as the link between students and employers, and lists activities such as mock interviews, group discussions and pre-placement talks, along with personality-development sessions on communication and career planning.

That context gives the exercise a practical use. Mock interviews and group discussions are often easier to prepare for when you understand how a resume gets shortlisted in the first place. Running the exercise on your own resume, with a classmate redacting it, can show which parts of it carry evidence and which parts only carry labels. The division’s notices page, the T&P Cell notices, is where recruitment drives and training announcements are posted, so check there for current details, and look at the placement policy page for the rules that apply to students. These pages are described here as they appeared in search excerpts or at fetch time, and they change, so read the current versions.

The university also lists a Career Guidance Cell that offers free counselling on job hunting and career skills, which is a natural place to take the results of this exercise for feedback on your own resume. Where an internship is part of your programme, a circular such as the internship guidelines and evaluation sheet for an earlier academic year shows the kind of document your department issues, though the details for your batch may differ.

Verification steps for RGUKT Basar students planning an HR screening exercise

1. Ask the Training and Placement office whether practice resume reviews are offered and who conducts them.
2. Check the current T&P notices for resume formats the division expects, so your redacted versions match them.
3. Ask your academic section for the internship circular that applies to your batch and whether a non-technical HR project is acceptable.
4. Ask the Career Guidance Cell whether they can review your exercise results and how to book a session.
5. If you use real classmates’ resumes, get their written consent and remove all personal data before sharing.

Running the exercise as a short project, and an optional online way to practise

If you want a structured way to try HR tasks over a short period, the free two-week online internship at CareerSwami includes an offer letter and a completion certificate after successful completion, and this exercise can serve as a self-directed HR task. A realistic 14-day plan looks like this.

1. Days 1 to 2: define the role and write the rubric.
2. Days 3 to 4: invent or adapt the six sample resumes.
3. Days 5 to 6: score round one.
4. Day 7: have a partner redact and recode the resumes.
5. Days 8 to 9: score round two.
6. Days 10 to 11: complete the comparison table and write observations.
7. Days 12 to 14: draft process recommendations and a one-page summary.

Limits of blind screening that interns should state honestly

• Removing names does not remove every signal. Hobbies, project titles and school names inside bullet points can still reveal identity.
• A rubric can carry its own bias. If it rewards a particular kind of certificate, it may favour candidates who could afford it.
• Screening is only one stage. Interviews and referrals bring bias back in unless they are also structured.
• Employers may have legal and policy obligations on hiring and data handling. Ask a qualified person at the organisation rather than assuming.

Frequently Asked Questions

How many resumes do I need for a blind screening exercise?

Six is enough to learn the method, but too few to draw conclusions. If you want patterns, repeat the exercise with more resumes and more than one scorer.

Can I use real resumes for the exercise?

Only with the owners’ written consent, and with personal data removed before anyone else sees them. Invented resumes are safer for learning.

What should be redacted in a blind resume review?

Name, photograph, age, marital status, address, college name and career gap dates are typical, replaced by evidence-focused summaries such as total months of experience.

What does a large score change between rounds mean?

It means something outside the rubric influenced the first round. Look at which identifiers were present on that resume and which rubric criterion you scored differently.

Does blind screening guarantee fair hiring?

No. It reduces one source of distortion at one stage. Structured interviews, clear criteria and reviews of the full pipeline are also needed.

Should the rubric be weighted by role?

Yes. A data role weights projects and skills heavily, while a customer-facing role may weight communication more. Decide weights before reading and keep them fixed.

How can the exercise support my RGUKT placement preparation?

Running it on your own resume shows which lines carry evidence. You can then take the revised resume to the placement office or career guidance cell for feedback.

Related Human Resources guides

• TERI School of Advanced Studies: Human Resources Internships in Sustainability and Green Jobs
• Sarla Birla University: Compensation Benchmarking Basics for an HR Internship Project
• Dr. C.V. Raman University: Final-Year Students' Guide to Converting an HR Internship Into a Job Offer
• All Human Resources internship guides