A colleague received a letter from a health board recently, inviting her to a job interview. To her surprise, the letter forewarned that she would be asked questions on such matters as her approach to work, her reliability, her motivation, her project management skills and her teamwork. She was also advised that these areas would be explored in depth and related back to specific work-related situations. And so they were.

This is a new approach to selection interviewing. Alerting candidates to the questions in advance - or showing them the "goalposts" before the game begins - is already a popular practice. Many of our blue-chip companies are using it, and it's standard in the public sector. Its rising popularity is recognition of the fact that serious flaws have been identified in the traditional approach to interviewing.

The effectiveness of the interview process is influenced by a host of factors. These include the number of interviewers, their professionalism and even the time of day or day of the week that the interview is held.

For example, it is well known that no two interviewers interpret and assess information in the same way. But it is worse than that. Even the same interviewers will reveal fluctuations in their assessment practices over time.

Research tells us that interviews are normally contrived, interrogative conversations involving a meeting - usually between strangers - which rarely lasts for more than an hour. It is, therefore, an artificially distorted and fairly stressful situation, despite the best efforts of some interviewers to reduce the anxiety levels.

Having undergone intensive preparation for the all-important job interview, an interviewee may come across as the "all-singing, all-dancing" dream ticket, who is presentable, fluent and quick-thinking. But to suppose that this pattern of behaviour will persist in the different circumstances of the workplace, over a long period of time, would be a most unwarranted assumption.

Surveys have also shown that:

Interviewers usually look for themselves; that is, they respond disproportionately favourably to the attitudes of a candidate that are ethically, sexually, racially and culturally similar to their own.

Candidates are often accepted or rejected in the first few minutes, with interviewers spending the rest of the time looking for evidence to support their initial impressions.

Interviewers are not good at recalling accurately what they have heard in interviews.

The evaluation of one interviewee is often affected by the contrast with earlier interviewees.

Interviewers' capacity to come to a sensible decision can be hampered by the proportion of time they spend talking, as opposed to listening and learning.

In recent years, however, there may have been an improvement in the ability of the selection interview to deliver the right results. In fact, some estimate that the predictive accuracy of the interview has just about doubled.

This is attributable to what is called the "structured" or "competency-based" interview. This approach entails a study of the vacant post so that the appropriate attitudes or competencies of a model employee are identified in advance. When these have been agreed, they are then weighted accordingly in interviews. For example, for an analyst programmer, technical competence may be twice as important as organisational skills, and so would command twice as many marks in the scoring system.

Candidates are then asked questions about what they did in other work situations in respect of the relevant competencies, and scored accordingly. The premise is that past behaviour is the best guide to predicting future form. It's hard to argue with this logic, given that bookmakers have been making millions out of it for years in relation to horses and sports teams.

The Department of Justice, Equality and Law Reform, in collaboration with the Civil Service Commission and consultants Pearn Kandola, recently developed such an assessment system for selecting prison officers. Having familiarised themselves with the prison environment and the jobs therein, a competency framework was developed. This identified the key skills and characteristics required for effective performance in the prison officer role. This framework now forms the basis of the selection process.

In competency-based interviews, all candidates are generally asked the same questions. This helps increase the consistency of treatment amongst candidates, enhances the relevance of the interview, and reduces bias and subjectivity and the likelihood of legal challenge from disgruntled interviewees.

It is probably the fear of legal action under the Employment Equality Act 1998, and the Freedom of Information Act 1997, that has accelerated the take-up of this new style of interviewing in the Irish public sector.

In this regard, it is interesting to note that the Civil Service Commissioners advise their interviewers to adopt the competency-based interview tool "to determine a material and substantial reason for choosing one candidate against another in order to defend claims of unfair discrimination".

Under the Employment Equality Act, 1998, discrimination is now prohibited on the grounds of sex, marital status, family status, sexual orientation, religion, age, disability, race or membership of the travelling community. Prior to the passage of the Act, such prohibition only applied to sex and marital status.

The threat of legal action also arises under the Freedom of Information Act, which gives job applicants "the right to be given reasons for decisions taken by public bodies that affect them". Arising from cases already brought under the Act, the Information Commissioner has determined that shortlisting boards and interviewers must make notes available to unsuccessful interviewees, provide access to the criteria and marking schemes used by them and advise interviewees of their marks.

However, this new approach to interviewing is not without its flaws. For example, some interviewers complain that it doesn't allow enough flexibility to pursue points of interest as they arise during an interview. Furthermore, when candidates are familiar with this style of interviewing, it's easier for them to "fake" answers. Given that a high proportion of candidates openly admit to lying at interviews, this is a real problem.

It's also hard to identify appropriate competencies for a job. For example, it has been established that male and female styles of management often involve different techniques or competencies. Both may be effective, but that's little help when trying to draw up a competency framework. Furthermore, it presents interviewers with a dilemma when interviewees offer important information which can't be accommodated in a rigid competency framework or candidate assessment form.

Also, the structured interview focuses on work experience, which means that younger candidates, or candidates whose current jobs don't necessitate such competencies, are disadvantaged. It's also true that most jobs require a number of different competencies, instead of simply the four or five explored at interview. There is a danger that where a competency-based interview is used in isolation, key aspects of the interviewee's past may be overlooked (frequent job changes, hobbies, educational record, and so on).

Such is the jobs climate at present that the new approach to interviewing overlooks the fact that interviewers now have to sell jobs to interviewees, almost more than interviewees have to sell themselves to employers.

Despite these limitations, the new style of interviewing is coming to a venue near you. That is, if it hasn't arrived already.

Dr Gerard McMahon is a lecturer at the Dublin Institute of Technology faculty of business and is the Institute of Personnel and Development's specialist tutor in selection interviewing. E-mail: ppl1@indigo.ie