The word “interview” is derived from the Latin language which means “see each other.” It is a process to get to know each other and also to select and recruit a candidate for a job. Thrill and Bovey (1987) refer to it as a planned conversation. Selection processes however take different styles and methods of assessing. It is an area organisations need to continuously improve in order to yield desired results.

Organisations mostly employ the use of quantitative and qualitative interviews approaches which may be complimented by use of either the consensus approach or other approaches. Quantitative interviews make use of numerical values to evaluate participants’ response. Methods like psychometric testing fall into this category and usually they contain close ended questions that are delivered in the same format and order to every respondent. Qualitative interviews are mostly useful for explaining individual differences between participant’s experiences and outcomes. Participants through qualitative interviews are allowed to describe their responses using their own words as opposed to being restricted to predetermined categories. Though each method has its own pros and cons, businesses should be cognisant of the nitty - gritties of each if they are to increase chances of effective selection.

Unlike conversations in daily life which are usually reciprocal exchanges, professional interviews involve interviewers who are in charge of structuring and directing the questions. However, it is the selection skills of panellists that determine the calibre of the candidate picked.

Qualitative interviews method allows the candidates to reveal their level of emotions, the way in which they have organised the world, their thoughts about what is happening, their experiences and basic perceptions. Mostly the tasks of the interviewer is to provide a framework within which the candidate can respond in a way that represents accurately and thoroughly to their point of view about the job. Interviewers have the flexibility to use their knowledge, expertise and other skills to explore interesting or unexpected ideas raised by the participants. With this approach a participant may say more than intended. Conversely an interviewer maybe reactive to such responses as well as personalities, moods and other interpersonal dynamics between parties.

The challenge with this approach comes when the interviewers attach a numerical value to these responses. Compared to interviews based on standardised measures, qualitative assessments are prone to a number of intervening variables. Organisations need to make sure that all participants who have this role of conducting interviews have the prerequisite skills and experience to do so. Analysing and interpreting qualitative interviews is more subjective. Training interviewers who conduct interviews is therefore key.

Of importance in any selection process is what is known as inter ratter reliability. Any organisation needs to ensure adequate levels that strive for accuracy and consistency in assessments. The prevalent practice has been to iron out the inconsistencies by making use of averages. On a panel of say eight ratters, we need to critically understand what makes one ratter give the same candidate a 5/5 for a response while another ratter gives a 1/5 for the same response. Training may be the answer and experience in interviewing is also critical. There is need to make use of the internal ratter reliability method to measure the degree of agreement among ratters. Instead of just using the averaging method, consider the need to start calculating the number of ratings that are in agreement. It is quite an easy method which if used will yield better results. Consider the example below:

                                                                                                                                                2/4

From the interviewed four candidates ratters would have consistently agreed on their assessment of two candidates. To make use of this tool, there is also a need to set rules of thumb for what percentage agreement shows a high agreement and a low agreement. If for instance panellists agree on 90% agreement level at least there is all reason to go ahead with the recruitment. The establishment of an organisational intra ratter reliability score helps organisations to see the consistency in ratings given by ratters across multiple instances. The intra rate reliability shows the degree of agreement among ratters. If various ratters do not agree, it is an indication that ratters are not able to pick the candidate’s attributes consistently. Effective selection takes place when ratters are more like automatons, behaving more like rating machines. By so doing they behave like independent witnesses who demonstrate their independence by disagreeing slightly. Candidates   1 and 3 from the table demonstrate that aptitude in panellists.

The need to focus on ratter reliability represents the extent to which the data collected is a correct representation of the variables being measured. Measurement then shows the extent to which ratters assigned the same score to the same variable.

In view of these glaring highlights organisations should seriously think about the inherent cons of averaging candidates out if the right candidate is to be offered the job ultimately.

Emmanuel Jinda is the Managing Consultant of PROSERVE Consulting Group, a leading supplier of Professional Human Resources and Management services locally, regionally and internationally. He can be contacted at Tel: 263 773004143 or 263 242 772778 or visit our website at www.proservehr.com