Needs assessment

From CEOpedia
Needs assessment
See also


Needs assessment - is a systematic process of identifying the gaps between current results and desired results, ranking those gaps by importance and using the ranking to decide where effort and money should go. In the sense popularised by Roger Kaufman, a need is not a wish and not a solution but a gap in results. Kaufman and Stakenas put it bluntly: "By restricting the use of "need" to a gap in results, confusion can be avoided"[1].

The idea sounds obvious. It isn't, in practice. Managers routinely say "we need a new CRM", "we need a leadership course" or "we need two more nurses", and each of those statements is a solution dressed up as a need. A needs assessment pushes the question back one step: what result is missing, how big is the shortfall, and who suffers from it? Only after that does it make sense to talk about courses, software or headcount. The method is used in corporate training and development, in education, in health and social services, in community planning and in public administration, wherever a limited budget meets a long list of demands.

Background and definitions

Needs assessment took shape in educational planning and programme evaluation in the United States during the 1970s. By 1981 Kaufman and Stakenas could note that needs assessments were becoming a regular part of educational planning, with (citing Witkin) "kits, admonitions, plans, concepts, and junk" on offer. Most of the procedures developed in that period used some form of discrepancy model, asking the planner to document the gaps between "what is" and "what should be"[1].

Two strands of thinking shaped the field.

Kaufman's results-based view. Kaufman distinguished gaps in resources (not enough money or staff), gaps in processes (methods that don't fit) and gaps in results. Only the last type deserves the label "need"[1]. He also separated three kinds of results:

  • Products - en route results an organisation produces, such as courses completed or skills acquired,
  • Outputs - what the organisation delivers to society, for example graduates who are able to support themselves,
  • Outcomes - the effect in society itself, such as citizens who are economically and socially self-sufficient.

The point of the hierarchy is practical. If a needs assessment starts at the level of processes ("we need more e-learning"), it may end up funding a solution that never closes the real gap.

Witkin and Altschuld's planning view. Belle Ruth Witkin and James W. Altschuld wrote what became the standard practitioner's guide, Planning and Conducting Needs Assessments (1995)[2]. They also define need as a discrepancy between the current state and a desired state, and they organise the work into three phases: pre-assessment (exploring what is already known), assessment (collecting and analysing data) and post-assessment (setting priorities and moving towards action). They further distinguish levels of need. The first level covers the people a programme is supposed to serve: students, clients, patients, customers. Further levels deal with the people who deliver the service and with the resources behind it.

Types of needs assessment

Training needs assessment

The best-known business application. William McGehee and Paul W. Thayer, in Training in Business and Industry (1961), proposed that training decisions rest on three linked analyses[3]:

  • Organisation analysis - where in the company training is needed, given its goals, resources and climate,
  • Operations (task) analysis - what a job requires and what a person must know and be able to do to perform it,
  • Man (person) analysis - which employees lack those skills and knowledge and therefore need training.

Later authors, notably Irwin L. Goldstein, kept the three levels and renamed them organisational, task and person analysis[4]. The model is still the backbone of training needs analysis. Its outputs feed directly into a training plan and into a precise training objective for each course; later, training evaluation compares the results against the same gaps. Task analysis draws on job analysis and on job descriptions, while person analysis often uses performance appraisal records, a skill test or a competency model.

One warning from the training literature deserves repeating. Not every performance gap is a training gap. If people know how to do the job but lack tools, time or incentives, a course changes nothing. A proper assessment separates "can't do" from "won't do" or "aren't allowed to do".

Community and public programme needs assessment

In health, social work and local government, needs assessment answers a wider question: which problems in a population should a programme tackle first? Jonathan Bradshaw's taxonomy of social need is often used here[5]. He distinguished:

  • Normative need - defined by experts against a standard (e.g. a recommended number of hospital beds per 1,000 people),
  • Felt need - what people themselves say they want,
  • Expressed need - felt need turned into action, such as a waiting list,
  • Comparative need - the gap between one group or area and a similar one that receives a service.

Each type tells a different story. A neighbourhood might show high normative need for dental care and almost no expressed need, because nobody expects a clinic to open. Sound community assessments combine all four and involve stakeholders early, since residents often see problems that statistics miss.

Other applications

  • Organisational needs assessment, linked to strategic planning and organizational development,
  • Customer needs assessment in marketing, which overlaps with research on customer needs,
  • Post-disaster needs assessment, which follows a damage assessment and estimates what affected people require for recovery,
  • Research needs assessment, which maps a gap in knowledge before a research programme is funded.

Steps of a needs assessment

Models differ in detail, but most follow a similar sequence.

1. Define scope and purpose. Who is the target group? Which results are in question? Who will use the findings, and what decision hangs on them? This is the pre-assessment phase in Witkin and Altschuld's terms. It includes gathering existing reports so that nobody pays twice for the same data[2].

2. Describe the desired state. Desired results come from strategy, regulations, professional standards, benchmarking against other organisations or a competency framework. Vague targets kill the exercise. "Better customer service" can't be measured; "complaints resolved within 48 hours in 90% of cases" can.

3. Collect data on the current state. Common sources, grouped as in a University of Florida extension guide[6]:

  • Archival data - demographic and census figures, organisational records, previous evaluation reports, minutes of advisory committees,
  • Communication methods - surveys, interviews, focus groups and the nominal group technique,
  • Analytic methods - fishbone (cause-and-effect) diagrams, task analysis, success mapping.

In organisations, add performance data: sales figures, error rates, audit findings, employee performance reviews. A depth interview with a few experienced staff often explains numbers that a questionnaire only describes. Self-reports carry their own risks (people overrate their skills; interviewers steer answers), so interviewer bias and social desirability are worth checking for.

4. Analyse the gaps. For each result, compare current and desired levels. Then ask why the gap exists. Lack of knowledge? Poor processes? Missing equipment? Tools like a SWOT analysis or a Pareto analysis help sort causes.

5. Set priorities. There are always more gaps than money. Priorities can be set by size of gap, importance, number of people affected, cost of closing the gap compared with the cost of ignoring it, or urgency. Formal tools include weighted scoring in a decision matrix, the analytic hierarchy process and other multi-criteria methods. In training research the Borich model is the most common choice.

6. Report and move to action. The post-assessment phase translates priorities into solutions, communicates results to decision-makers and evaluates how well the assessment itself worked[2].

The Borich model

Gary Borich proposed his model in 1980 for follow-up studies of teacher education graduates[7]. Respondents rate each competency twice on the same scale (usually 1-5): how important it is to their job, and how well they can perform it (their current ability or attainment). A raw difference would treat a gap in a trivial skill the same as a gap in a critical one. Borich's fix is to weight the gap by the group's average importance rating.

The calculation runs in three steps[8]:

  1. For each respondent i, the discrepancy score is importance minus ability: \( D_i = I_i - A_i \)
  2. The weighted discrepancy score multiplies it by the mean importance of the competency across all respondents: \( WDS_i = D_i \times \bar{I} \)
  3. The mean weighted discrepancy score (MWDS) is the average of the weighted scores.

\[ MWDS = \frac{\sum_{i=1}^{N} (I_i - A_i) \times \bar{I}}{N} \]

where N is the number of respondents. Because the mean importance is a constant for a given competency, the MWDS equals the mean importance multiplied by the mean discrepancy. Competencies are then ranked from highest to lowest MWDS; the top of the list marks the strongest training need. On a five-point scale the score can range from -4 to 20[8]. A negative value means people rate their ability above the importance of the skill. No training need there.

Example of needs assessment

Hypothetical example. A regional insurance office wants to decide which of three topics to put first in next year's training for its customer service team. Six advisers rate each competency on a five-point scale for importance (I) and current ability (A).

Ratings for "Handling complaints"
Adviser Importance (I) Ability (A) Discrepancy (I - A) Weighted discrepancy (× 4.67)
1 5 3 2 9.33
2 5 2 3 14.00
3 4 3 1 4.67
4 5 3 2 9.33
5 4 2 2 9.33
6 5 4 1 4.67
Mean 4.67 2.83 1.83 8.56

Mean importance is 28 / 6 = 4.67. The weighted scores add up to 51.33; divided by six they give an MWDS of 8.56 (the same as 4.67 × 1.83).

The same calculation for the other two topics:

Competency Mean importance Mean ability Simple gap MWDS Rank
Handling complaints 4.67 2.83 1.83 8.56 1
Product knowledge 4.50 3.83 0.67 3.00 2
Using the CRM software 3.67 3.00 0.67 2.44 3

Look at the last two rows. Both show the same simple gap of 0.67 points. Yet product knowledge ranks higher because advisers consider it more important (4.50 against 3.67), and the weighting picks that up. Complaint handling wins by a wide margin. So the office schedules a complaint-handling workshop first and folds CRM tips into on-the-job coaching.

Even so, the numbers are only the start. A short follow-up with two or three advisers might reveal that complaints drag on because the office has no escalation procedure, not because people lack skill. In that case the "training need" is really a process gap, and a workshop alone would miss the target.

Advantages of needs assessment

  • Puts results before solutions, which reduces spending on popular but useless programmes,
  • Gives a documented, defensible basis for budget decisions, important in public and grant-funded programmes,
  • Involves the people affected, which tends to raise acceptance of the eventual programme,
  • Produces a baseline against which later results can be measured, linking planning with performance management and evaluation,
  • Helps separate training problems from problems of motivation, tools or organisation,
  • Scales well: a five-person team and a national health service can use the same logic,
  • Formal prioritisation (Borich, weighted scoring) makes trade-offs visible rather than leaving them to whoever argues loudest.

Limitations of needs assessment

  • Time and cost. A thorough assessment with surveys, interviews and data analysis can take months, and decision-makers often want answers sooner,
  • Self-reported ratings are unreliable. People overestimate their own ability, and importance ratings drift upward because everything looks important when you're asked,
  • Methodological criticism of the Borich model: it calculates means for items measured on ordinal scales, an approach Narine and Harder describe as "subject to debate in modern times"[9]. They proposed a Ranked Discrepancy Model as an alternative, though the two produced strongly correlated results in their study,
  • The desired state is often contested. Experts, managers and clients may disagree about "what should be", and the assessment can turn political,
  • Surveys measure wants and opinions; translating them into needs in Kaufman's sense (gaps in results) requires judgement,
  • Findings date quickly when technology, regulation or markets change,
  • There's a risk of "assessment fatigue": staff surveyed repeatedly with no visible follow-up stop answering honestly.

Related approaches

Needs assessment overlaps with several other tools. Gap analysis in strategy compares current and target performance at company level. Competency mapping and skills inventories support the person-level part of training needs analysis and skills development planning. Programme evaluation looks backward at whether a programme closed the gaps it was meant to close, while needs assessment looks forward. And brainstorming or the nominal group technique are often used inside a needs assessment to generate and rank candidate needs in a workshop setting.

References

Footnotes

  1. ↑ Kaufman R., Stakenas R.G., Needs assessment and holistic planning, Educational Leadership, May 1981
  2. ↑ Witkin B.R., Altschuld J.W., Planning and Conducting Needs Assessments: A Practical Guide, Sage Publications, 1995
  3. ↑ McGehee W., Thayer P.W., Training in Business and Industry, John Wiley & Sons, 1961
  4. ↑ Goldstein I.L., Training in Organizations: Needs Assessment, Development, and Evaluation, Brooks/Cole, 1993
  5. ↑ Bradshaw J., The concept of social need, New Society, 1972
  6. ↑ Benge M., Turner S., Conducting the Needs Assessment #6: Phase 2 - Assessment, University of Florida IFAS Extension, 2023
  7. ↑ Borich G.D., A needs assessment model for conducting follow-up studies, Journal of Teacher Education, 1980, pp. 39-42
  8. ↑ Caillouet O., Harder A., Conducting the Needs Assessment #8: The Borich Model, University of Florida IFAS Extension, 2022
  9. ↑ Narine L., Harder A., Comparing the Borich model with the Ranked Discrepancy Model for competency assessment, Advancements in Agricultural Development, 2021
Needs assessment — recommended articles
Training objective — Training plan — Training evaluation — Job analysis — Competency modeling — Gap in knowledge — Decision matrix
Retrieved from CEOpedia