Using Judgmental Forecasts

Judgmental forecasts are based on subjective views – often the options of experts in the field. Suppose a company is about to market an entirely a new product, or the board is looking at plans for 25 years in the future. They won’t have any relevant historical data for a quantative forecast. Sometimes there is a complete absence of data, and at other times the data is unreliable or irrelevant to the future.

 Quantative forecasts are always more reliable, but when you don’t have the necessary data, you have to use a judgmental method. There are five widely used methods:

  • Personal insight. This uses a single person who is familiar with the situation to produce a forecast based on his or her own judgment. This is the most widely used forecasting method – but is unreliable and often gives very bad results.
  • Panel consensus. This collects together a group of experts to make a forecast. If there is no secrecy and the panel talk freely and openly, you can find a genuine consensus. On the other hand, there may be difficulties in combining the views of different people.
  • Market surveys. Sometimes even groups of experts don’t have enough knowledge to give a reasonable forecast about, for example, the launch of a new product. Then market surveys collect data from a sample of potential customers, analyze their views and make inferences about the population at large.
  • Historical analogy. If you are introducing a new product, you might have a similar product that you launched recently, and assume that demand for the new product will follow the same pattern. If a publisher is selling a new book, it can forecast the likely demand from the actual demand for a similar book it published earlier.
  • Delphi method. For this you contact a number of experts by post and give each a questionnaire to complete. Then you analyze the replies from the questionnaires and send summaries back to the experts. You ask them if they would like to reconsider their original replay in the light of summarized replies from others. This is repeated several times – usually between three and six – until the range of options is narrow enough to help with decisions.

 My Consultancy–Asif J. Mir – Management Consultant–transforms organizations where people have the freedom to be creative, a place that brings out the best in everybody–an open, fair place where people have a sense that what they do matters. For details please visit, Line of Sight


Conducting an Interview

You may not have the time or inclination to create structured situational interviews. However, there are several things you can do to increase the standardization of the interview or otherwise assist the interviewer to ask more consistent and job relevant questions. They include:

  1. Base questions on actual job duties. This will minimize irrelevant questions based on beliefs about the job’s requirements. It may also reduce the likelihood of bias, because there’s less opportunity to ‘read’ things into the answer.
  2. Use job knowledge, situational, or behaviorally oriented questions and objective criteria to evaluate the interviewee’s responses. Questions that simply ask for opinions and attitudes, goals and aspirations, and self-descriptions and self-evaluations allow candidates to present themselves in an overly favorable manner or avoid revealing weaknesses. Structured interview questions can reduce subjectivity and therefore the chance for inacurate conclusions, and bias. Examples of structured questions include: (a) situational questions like, “Suppose you were giving a sales presentation and a difficult technical question arose that you could not answer. What would you do?”; (b) past behavior questions like, “Can you provide an example of a specific instance where you developed a sales presentation that was highly effective?”; (c) background questions like, “What work experiences, training, or other qualifications do you have for working in a teamwork environment?”; (d) job knowledge questions like, “What factors should you consider when developing a TV advertising campaign?”
  3. Train interviewers. For example, review laws with prospective interviewers and train them to avoid irrelevant or potentially discriminatory questions and to avoid stereotyping minority candidates. Also train them to base their questions on job-related information.
  4. Use the same questions with all candidates. When it comes to asking questions, the prescription seems to be “the more standardized, the better.” Using the same questions with all candidates can also reduce bias “because of the obvious fairness of giving all the candidates the exact same opportunity.”
  5. Use rating scales to rate answers. For each question, provide a range of possible ideal answers and quantative score for each. Then you can rate each candidate’s answers against this scale. This ensures that all interviewers are using the same standards.
  6. Use multiple interviewers or panel interviews. Doing so can reduce bias, by diminishing the importance of one interviewer’s idiosyncratic opinions, and by bringing in more points of view.
  7. If possible, use structured interview form. Interviews based on structured guides usually result in the best interviews. At the very least, list your questions before the interview.
  8. Control the interview. Limiting the interviewers’ follow-up questions (to ensure all interviewees get the same questions), using a larger number of querstions, and prohibiting questions from candidates until after the interview are other “structuring” techniques.

My Consultancy–Asif J. Mir – Management Consultant–transforms organizations where people have the freedom to be creative, a place that brings out the best in everybody–an open, fair place where people have a sense that what they do matters. For details please visit, Line of Sight