Panel Interview

What a panel interview is

A panel interview (المقابلة أمام لجنة) is a hiring interview in which one candidate sits before several interviewers in the same session. Each interviewer, or some of them, asks questions, and the interviewers then combine their views in one decision. The number of interviewers in one session defines a panel interview. The number of sessions and the number of candidates do not.

The name is confused with two other arrangements, and the difference lies in who sits in the session. Sequential interviews are several sessions, and in each session a single interviewer faces the candidate. A group interview is a single session with several candidates. A panel is a single session in which several interviewers face one candidate. Changing any one of these three features gives a different arrangement.

What a panel adds and what it does not

A panel adds two things. First, several interviewers hear the same answer at the same moment, and each judges it from a different role. The manager judges it against the work, the human resources representative against the conditions for hiring and the candidate’s readiness to start, and a colleague against what working alongside the holder will involve. Second, one session replaces three, so the candidate describes the work history once and does not repeat it to three people.

A panel does not add two other things, because neither depends on the number of interviewers:

• The quality of the questions. A larger number of interviewers does not improve the questions. Three interviewers who ask questions that nobody wrote in advance produce three separate impressions, and those impressions are not scores that can be compared with the scores of another candidate. Four features make an interview comparable: uniform questions, one order of questions, a rating scale and independent rating. They are covered under structured interview. A panel neither supplies these features nor excludes them.
• The type of question. The number of interviewers does not change what a question measures. A behavioural question remains a behavioural question when three interviewers ask it. Question types are covered under behavioral interview questions.

A structured interview describes how an interview is run, and a panel interview describes who attends it. One session can combine the two features, and a session can also have one without the other.

How a panel differs from neighbouring stages

• A prescreening interview. It is a short conversation that decides whether an application continues or stops, as covered under prescreening interview. A panel comes after it, at the point where competences are compared and not where conditions are checked.
• An assessment centre. It is a set of exercises taken from the job and observed by trained assessors, as covered under assessment centre. A panel is a session of questions and answers and does not include exercises, even if one room hosts both.
• An automated interview. The candidate answers at a time of the candidate’s choosing, with no interviewer present, as covered under automated interview. In a panel the candidate meets the interviewers face to face or remotely at one moment, so the interviewers can react to each answer.

When a panel suits and when it does not

A panel suits a job whose holder will work with more than one party in the establishment: a manager who wants a result, a team that works with the holder, and a human resources function that knows the terms of hiring. In that case one session brings together what would otherwise be three sessions, and each party hears the candidate’s answer directly and not through someone who relays it.

A panel does not suit the first screening stage. That stage needs a short answer that decides pass or stop, and such an answer does not justify three interviewers. A panel also does not suit a job whose holder will work with one manager only, because the additional members then add interview time and no additional role to judge the answer. This is an inference from the effect of the number of interviewers, not a ruling.

Where the value of a panel can be lost

• Who speaks first. If one interviewer states an opinion aloud before the others have written their judgements, that opinion may become the starting point for the interviewers who follow. A difference in judgements may then turn into an apparent agreement that does not reflect what each person saw. A group agreeing on what none of its members wants is described under Abilene paradox.
• One impression colouring the rest. A candidate who speaks fluently at the start of a session may be rated high on qualities that nobody asked about, and the same may happen with several interviewers at the same moment. The mechanism is covered under halo and horn effect.
• Overlap and gaps in the questions. If the establishment does not divide the competences to be measured among the interviewers in advance, several interviewers may ask about the same quality while another quality receives no question. The overlap and the gap become visible only when the sheets are compared.
• The burden on the candidate. Sitting before several people may make the situation harder for the candidate, but it does not make the session a stress interview. In a stress interview the discomfort is intended and produced on purpose. In a panel it is a side effect of how the session is arranged, and the difference is covered under stress interview. The effect on a candidate of what the candidate meets during the process is covered under candidate experience.

How a disciplined panel is built

The practices below are practical recommendations drawn from the four problems above. They are not conditions and not settled rules:

• Roles fixed before the session. The establishment names a chair to manage the time and to move between the questions, and it assigns to each interviewer the competences that the interviewer will ask about.
• A sheet for each interviewer, and a written score before anyone speaks. Each interviewer writes a score on a described scale before any discussion begins. The scale can follow the model of the graphic rating scale.
• The order of speaking in the discussion. It begins with the written scores and then turns to the examples on which they rest. The most junior interviewer speaks before the most senior, so that the junior interviewers do not simply follow the view of the person who holds the highest post.
• A rule for disagreement. Before the session, the panel agrees what happens if the gap between two scores reaches a stated size. The interviewers review the examples, or the establishment holds an additional interview, and the gap is not buried in the average.
• Panel size. Each additional member increases the number of people the candidate addresses and the time the panel needs for discussion. A member contributes only the competences assigned to that member. So the size is set by the number of competences assigned and not by the number of people who wish to attend.
• Introducing the panel to the candidate. The establishment gives the candidate the names and roles of the panel members before the session, so that the candidate knows who is being addressed.

What is recorded after a session

A session that ends with no written record leaves only memory, and memory may lean toward the last thing said. It is advisable to keep three things from each session: the sheet of each interviewer with scores and examples, the joint decision and its reason in two lines, and what the panel agreed to verify if a disagreement remains. This is a practical recommendation.

The establishment keeps the sheets with the rest of the hiring data. If the candidate is hired, the sheets that the establishment needs move to the employee file, whose content is covered under employee file.

A worked example: three interviewers and assumed scores

Suppose that a panel at a company in Riyadh interviews candidates for a senior accountant job. The panel consists of the finance manager, a human resources representative and a senior accountant from the team. Each gives an overall score from 1 to 5. The figures are assumed, to show the mechanism, and they were not measured in any case.

• The first candidate, with scores written before the discussion. The manager gave 4, the human resources representative gave 3 and the colleague gave 2. The average is 3.00, from (4 + 3 + 2) ÷ 3, and the range between the highest score and the lowest is two points.
• The second candidate. All three gave 3. The average is 3.00, from (3 + 3 + 3) ÷ 3, and the range is zero.

The two candidates have the same average, but the scores carry different information. The range of two points for the first candidate shows that the three scores differ, and the difference is a question to put to the interviewers in the discussion. The average alone hides it.

Now suppose that the panel discussed before anyone wrote a score. The manager spoke first and said that the first candidate was strong. The human resources representative then wrote 4, the colleague wrote 3 and the manager wrote 4. The new average is about 3.67, from (4 + 4 + 3) ÷ 3, and the first candidate now leads the second by about 0.67 points. The cause of the lead is the order of speaking and not what the candidate said. The example shows that a panel which discusses before it writes does not gather three independent judgements. The scores that it gathers have been shaped by the first speaker.

The panel interview and the equal opportunity duty in the Labor Law

Item 4 of Article 61 of the Labor Law requires the employer to refrain from anything that nullifies or weakens equality of opportunity or of treatment in employment and occupation. The Arabic text of that item is:

أن يمتنع عن القيام بكل ما من شأنه إبطال أو إضعاف تطبيق تكافؤ الفرص أو المعاملة في الاستخدام والمهنة، سواء من خلال الاستبعاد أو التفريق أو التفضيل بين المتقدمين للعمل أو العاملين لديه على أساس العرق أو اللون أو الجنس أو السن أو الإعاقة أو الحالة الاجتماعية أو أي شكل من أشكال التمييز الأخرى.

The English rendering is ours, not an official translation. The employer must refrain from anything that nullifies or weakens the application of equality of opportunity or of treatment in employment and occupation, whether through exclusion, distinction or preference among applicants for work or among the employer’s workers, on the basis of race, colour, sex, age, disability, social status or any other form of discrimination.

Item 4 of Article 61 of the Labor Law does not mention an interview or a panel, and it does not require scores to be recorded. Linking written scores to this duty is our reading and not the text. The reading is that a score which each interviewer wrote against a described criterion before the discussion lets an establishment show on what basis it preferred one candidate to another. A judgement that leaves no trace except in the discussion of the panel gives the establishment nothing to show. Implicit bias in interviews is covered under implicit bias.

In the sources we reviewed, we found no text in the Labor Law or its Implementing Regulation that requires a panel interview in private establishments or fixes the number of its members. We also found no source that sets an ideal number of interviewers, or a measured value for how well this kind of interview predicts performance. The figures in the example above are assumed and were not measured.

This is an explanation of the concept and of the statutory provisions cited, not legal advice.