Taken from Strategic HR Review (May/June 2006)

Geoff Trickey, managing director at Psychological Consultancy Ltd explains how using psychometric testing at Her Majesty’s Customs and Excise helped discover connections between competency and performance.

Bridging the gap between traits (the language of psychologists) and competencies (the language of the workplace) at Her Majesty’s Customs and Excise (HMCE), was made possible through the development of a new algorithmic mapping technology that matches personality profiles to competency requirements. This approach puts the power of psychometrics behind the competency rating process, assisting in a company’s recruitment strategies.

The roots of personality testing
Personality assessment is relevant to staff selection and development because it highlights any aspects of temperament that are a particular asset or a potential hindrance to achieving success in any role. The term “temperament” is used in this context because, although it’s often used as a synonym for “personality,” it emphasizes the underlying aspects of personality, the deeper roots as distinct from more superficial or controlled interpersonal strategies, shaped by cultural influences such as manners and social etiquette. Temperament undoubtedly influences job performance and, at the extremes, may determine whether an individual will take to a particular role “like a duck to water” or will be unable to function effectively in it at all. The basic elements of temperament are whether an individual will be prickly or approachable, compliant or unrestrained, assertive or easygoing, passionate or calm, conservative or adventurous. We all fall somewhere between these extremes.

Personality measurement in business
Although personality reports for organizational use may strive to present their findings in more business-focused and user-friendly ways, there’s still a false assumption that those who use such reports will be versed in the structure of personality, trained in test interpretation and draw inferences about the probable behavior of any candidate and their likely competence for the role, but often this isn’t the case. In many cases there will be no one on the selection panel with these skills. The language of personality assessment will be familiar to psychologists and trained HR professionals, but employers are likely to take a more immediate and pragmatic perspective to the competencies of prospective employees. To survive in business, they must have employees who will add value and make a net contribution rather than a net decrement to company finances.

Matching personality to competency
Over the past decade at Psychological Consultancy Ltd (PCL), professional business psychologists based in the UK, we’ve repeatedly been asked to report our assessments in language that reflects the competency framework of the organization. Since personality assessments are used to predict workplace competence, it makes sense to address those competencies directly and, by so doing, bring the psychometric contribution into line with other approaches to staff recruitment (competency-based interviews, assessment centers and the evaluation of resumés, career histories and so on).

Planning an automated system
It’s clear that ensuring the consistency of competency ratings from personality data can only be achieved by a systematic approach and that this needs to be automated. At PCL, we envisaged a database driven system in which:
• the relationships between personality traits and competency definitions could be defined; and
• test scores could be combined and weighted to generate competency ratings.

Applying the principles at HMCE
A pilot project commissioned by the HR team at Her Majesty’s Customs and Excise (HMCE) provided PCL with an ideal opportunity to develop an online system able to automate the generation of competency ratings from personality data (the techniques used formed the framework for PCL’s PROFILE:MATCH™ product). The team at HMCE identified seven competencies (confidential) that captured their definition of “talent” within the organization, the characteristics that differentiate “the best” from “the rest” at HMCE. PCL was invited to contribute to the assessment aspects of its talent management project that needed to be widely available and inclusive. In mid-2005, we worked with the HMCE HR team to develop online assessments, that would generate competency ratings within the HMCE framework, and the key to this task was to develop new ways of aligning personality measures to competencies.

Implementing the measuring system
The new system reflected the complex relationships between personality and HMCE’s required competencies. We surveyed 3,000 volunteer participants, and captured their attributes, temperament and competencies using a series of mathematical algorithms (see box, above right for an example of the competencies, temperament-algorithms relationship pattern). The reliability of these ratings at HMCE was high, due to the involvement of large numbers of contributing personality items (for example “I get annoyed with people when they make mistakes”) from multiple “personality scales” (such as self-esteem, composure and sensitivity).

The “virtuous cycle”
Recognizing the speculative nature of both the algorithms and the competency definitions, the spirit at the start of field trials was very much “Let’s start somewhere and work from that first approximation.” We anticipated the need to progressively fine-tune the system in the light of experience, this is typical in the development of selection systems. Hard results from the surveys were then used to establish a “virtuous cycle,” in which successes, as well as mistakes, were fed back into the system to drive further developments and to shape and refine the process.

Analyzing results, making revisions
Because of the large volume of data involved we were able to analyze the measurement quality of our competency ratings. We discovered that the ratings were finely incremented and normally distributed for all seven competencies. Clearly, the algorithms protected the psychometric qualities of the personality data from which ratings were drawn. Initial relationships between competency ratings and “talent” ratings for existing staff were positive, but not as strong as we would have liked. In the spirit of the “virtuous cycle” described above, PCL and HMCE immediately revised the competency definitions and algorithms and recoded the data accordingly. Re-analysis showed that the correlation of competency ratings with performance had increased to show a highly significant and potentially predictive relationship between the two.

Moving forward
The rating of participants correlated very well with the profiles of a selected pool of existing staff considered to exemplify the talent characteristics valued by HMCE. HMCE (now part of Her Majesty’s Revenue and Customs,HMRC) plan to repeat the talent survey later this year, based on their refined competency framework.

home