Portrait

What you see

Each comparison asks “Which is more like you at work?” and shows two cards side by side (stacked on a phone), with an Item pill such as Item 4 / 18 above them. Each card has a work-style name and a one-line, first-person statement. There are six styles in all:

You choose by selecting a card or pressing the left or right arrow key; on a desktop each card shows its arrow key. The card you pick is highlighted briefly, and the next pair appears. You cannot go back. There is no time limit and you can pause. When every pair is answered, the line under the cards reads “All 18 answered.” and you submit the set. It is recorded, never marked right or wrong. The instruction is simply: choose which of two work styles sounds more like you; there are no right or wrong answers.

Repeated pairs

Here is the design detail that matters most. Every pair of styles in use appears at least once, and then some pairs come back later, each time with the sides decided afresh, so about half the time the cards are swapped. Those repeats are the consistency checks. You are not told which comparisons are repeats.

What changes as levels rise

More styles and more comparisons give a finer-grained profile, and more chances to contradict yourself. Sessions start at level 3. If every repeated pair was answered consistently you move up a level; if none was, you move down.

What it measures

Not an ability. Portrait is a self-report preference instrument: it records which of two ways of working you say is more like you. Our taxonomy treats it that way. It is marked as unrated, contributes nothing to any dimension score, and appears on your report only as a response-style summary.

The format is a pairwise forced choice. Instead of rating each statement on its own ("how much does this describe you, 1 to 5?"), you must choose between two statements. The idea goes back to Thurstone's work on comparative judgement, and modern versions are scored with Thurstonian item response models. Forced choice exists mainly to reduce socially desirable responding: when both options sound reasonable, you cannot simply agree with everything flattering. In a meta-analysis of high-stakes testing, forced-choice personality measures showed an overall score inflation of only 0.06 standard deviations, far less than single-statement scales, and pick-one formats like this one were among the faking-resistant designs.

Forced choice has a cost that is worth understanding. Because every pick is a trade-off, a simple count of wins gives a relative profile: it tells you which styles you prefer over which, within your own answers, not how much of each you have compared with other people. That is why commercial tools invest in the statistical models above.

Our six styles are a practice typology, not a validated personality model. They loosely echo familiar trait territory (Steward sounds conscientious, Explorer sounds open to experience, Connector sounds sociable and agreeable), but no claim is made that your Portrait profile predicts anything.

How it is scored here

• Consistency across repeated questions. For every pair that appeared more than once, did you choose the same style each time, whichever side it was on? The figure is the share of repeated pairs you answered consistently.
• Preference for one side. How lopsided your left and right picks were: the difference between left and right choices as a share of all comparisons. A high value suggests answers were driven by position rather than content.
• Work-style profile. For each style, the percentage of its appearances in which you chose it. This is descriptive only, and never presented as good or bad.

How employers use questionnaires like this

Personality and work-style questionnaires are rarely pass-fail on their own. Common uses:

• Fit to a role profile. SHL describes its OPQ as assessing role fit and future performance. An employer defines which characteristics matter for a role, and your profile is compared with that. A strong Steward profile might suit a quality or operations role; a strong Explorer profile, a research or product role.
• Interview prompts. Many employers use the report to choose follow-up questions: "Your profile suggests you prefer to decide quickly. Tell me about a time that went wrong."
• Development. Tools like the Hogan Personality Inventory are used from entry-level hiring to leadership development.

What they are looking for in your answers is a plausible, consistent person. Saville's own candidate guidance puts it plainly:

Please be as honest as you can. There is no one right or wrong answer; job roles vary and there are many ways of being effective in any one job.

Can you fake it, and should you?

People can raise their scores when told to fake a rating-scale questionnaire, and all Big Five traits are about equally fakeable in that setting. Real applicants do inflate a little: compared with non-applicants they score higher on conscientiousness (d = .45) and emotional stability (d = .44), especially on traits they see as job-relevant. But forced choice cuts that inflation sharply, and the inflation that remains shows up mostly on the traits candidates think the job wants, which is exactly what consistency checks and profile comparisons are built to notice. Social desirability itself does not predict job performance.

The practical case against engineering your answers is simple. You do not know which profile the employer wants; you are likely to contradict yourself; and if it works, you may be hired into a role that fits someone else.

Where you will meet it

Portrait is a two-statement version of the same family of task. Vendors' blocks are often larger (three to six statements) and their statements more numerous and subtle, but the skill of answering quickly, honestly and consistently is identical.

The method

1. Pick a real context before you start

   Fix one concrete setting: your final-year project, an internship, a college society. Answering "at work" in the abstract invites answering as an ideal worker. Answering about a real team keeps you honest and consistent.

2. Ask 'which do I do more often?', not 'which is better?'

   Every statement in a good forced-choice block is a strength. The question is frequency and preference, not virtue.

3. Set a tie-break rule and keep it

   When both fit, decide by what you do under time pressure, or what others would say about you. Use the same rule every time; that is what keeps repeated pairs consistent.

4. Read the words, ignore the side

   Repeated pairs are often shown with the sides swapped. If you notice yourself reaching for the same arrow key, slow down and read both cards.

5. Keep a steady pace

   Give each pair enough time to read both cards and apply your rule, then move on. Long deliberation invites second-guessing about what an employer might want; rushing invites side habits.

Worked example 1: the swapped pair

Early in a session you see Driver on the left and Analyst on the right, and choose Driver. Twenty comparisons later the same pair returns with Analyst on the left, and you choose Analyst.

What happened, and how to fix it

This pair counts as inconsistent: you chose Driver once and Analyst once. Notice that both choices were the left card, which suggests position, not preference, decided at least one of them. If this happened on one of three repeated pairs, your consistency figure would be 2 of 3, or 67%.

The fix is the tie-break rule. Suppose your rule is "what I do under deadline pressure". In your final-year project, when the deadline loomed, did you push the group to decide, or insist on checking the data? Whichever it was, give that answer both times.

Worked example 2: reading your own report

At level 1 (12 comparisons, styles Driver, Analyst and Connector), a candidate chose the left card 10 times and the right card 2 times. Connector appeared 8 times and was chosen 6 times. All three repeated pairs were answered consistently.

Work out the three figures

• Consistency: 3 of 3 repeated pairs consistent, so 100%. Under the session rules that also moves the candidate up a level.
• Preference for one side: the difference between left and right picks is 10 − 2 = 8, and 8 ÷ 12 is about 67%. That is high. Even with consistent repeats, such a lopsided pattern is worth checking: were the left cards really more like them, or was it habit?
• Connector in the profile: 6 ÷ 8 = 75% of its appearances.

None of these numbers is good or bad as a description of the person. The first two only describe how usable the answers were.

Tips and traps

Levelling up

Higher levels add styles and comparisons. With six styles there are fifteen different pairs, and more of them repeat, so consistency depends entirely on a stable internal rule rather than on remembering what you said. Treat each session as a chance to find out whether your self-description is stable.

1. Week 1
   • Complete one Portrait session using a single real context
   • Write your tie-break rule in one sentence before starting
   • Read your profile and consistency figure
2. Week 2
   • Repeat with the same context and rule; compare the two profiles
   • If a style moved a lot, think about why: context, mood or unclear rule
   • Prepare one STAR story for your top style and one for your lowest
3. Week 3
   • Complete a longer session without pausing
   • Check your side-preference figure; if it is high, slow down on close calls
   • Try PortraitXT and compare what each format says about you

Check yourself

5 practice questions

1. What does Portrait's consistency figure measure?
2. Why do employers use forced-choice formats for personality questionnaires?
3. A candidate chose the left card 14 times and the right card 6 times in 20 comparisons. What is their side-preference figure?

and 2 more

Create a free account to answerSign in

Sources

1. Brown, A., & Maydeu-Olivares, A. (2011). Item response modeling of forced-choice questionnaires. Educational and Psychological Measurement, 71(3), 460–502. doi.org/10.1177/0013164410375112
2. Brown, A. Thurstonian models for preference judgements. annabrown.name
3. Cao, M., & Drasgow, F. (2019). Does forcing reduce faking? A meta-analytic review of forced-choice personality measures in high-stakes situations. Journal of Applied Psychology, 104(11), 1347–1368. doi.org/10.1037/apl0000414
4. Viswesvaran, C., & Ones, D. S. (1999). Meta-analyses of fakability estimates: Implications for personality measurement. Educational and Psychological Measurement, 59(2), 197–210. doi.org/10.1177/00131649921969802
5. Birkeland, S. A., Manson, T. M., Kisamore, J. L., Brannick, M. T., & Smith, M. A. (2006). A meta-analytic investigation of job applicant faking on personality measures. International Journal of Selection and Assessment, 14(4), 317–335. doi.org/10.1111/j.1468-2389.2006.00354.x
6. Ones, D. S., Viswesvaran, C., & Reiss, A. D. (1996). Role of social desirability in personality testing for personnel selection: The red herring. Journal of Applied Psychology, 81(6), 660–679. doi.org/10.1037/0021-9010.81.6.660
7. SHL. Occupational Personality Questionnaire (OPQ). shl.com
8. WikiJob. SHL OPQ32 test guide (candidate-reported format). wikijob.co.uk
9. Saville Assessment. Introducing Wave Professional Styles (candidate preparation). savilleconsulting.dk
10. Hogan Assessments. Hogan Personality Inventory. hoganassessments.com
11. PrepInsta. Accenture and Capgemini recruitment patterns (candidate-reported). prepinsta.com/accenture, prepinsta.com/capgemini

Put it into practice: PortraitSign in to try it