Jorge Quitério Search
Business

Situational leadership in security programmes

Situational leadership holds that the appropriate style depends on the person and the task. Applied to security, it explains why a signed policy changes little and why awareness training is effective only as coaching. Two small models illustrate the diagnosis and the forgetting curve.

Consider a senior engineer who joins a team with fifteen years of experience, writes excellent code and needs no supervision. Three weeks later, the engineer pushes a production change without a ticket, stores a credential in a repository, and is surprised when asked about it.

This is probably not carelessness. Nobody explained the local procedures, because senior people are usually left to work independently. On engineering the engineer was an expert; on the team's security procedures, a beginner. The person was managed, but the task was overlooked.

An experienced driver who moves from Luxembourg to the United Kingdom is in the same position. The driving skills transfer completely; the habits of keeping to the right and of looking left first at junctions are now wrong. Nobody would hand such a driver the keys without pointing out the difference, yet new senior colleagues are often given full autonomy without being told the local rules.

The leadership module of my MBA walks through the classic approaches in order: trait, skills, behavioural and situational. The last, situational leadership, addresses this problem directly. This article explains the model, how much evidence stands behind it, and what it says about policies and awareness training.

The relevant literature is as follows.

Hersey and Blanchard (1969) proposed the life cycle theory of leadership: the right leadership style depends on the "maturity" of the follower. Blanchard and colleagues (1985) revised it into SLII, which describes followers by competence and commitment on a specific task.

Northouse (2016) summarises the trait, skills, behavioural and situational approaches, and notes the situational model's strengths (practical, prescriptive, intuitive) and weaknesses (limited empirical research, ambiguous definitions of commitment).

Thompson and Vecchio (2009) tested three versions of the theory empirically and found only limited support for its prescriptions. That matters: the model is a useful diagnostic language rather than an established law, and I use it in that sense.

Blake and Mouton's Leadership Grid, later presented by Blake and McCanse (1991), plots concern for results against concern for people. Wiseman (2017) describes the accidental diminisher, the well-meaning leader who shrinks the team's capability by helping too much.

On the learning side, Ebbinghaus (1885) measured the forgetting curve on himself, and Cepeda and colleagues (2006) synthesised hundreds of experiments showing that spaced practice beats massed practice for long-term retention.

The model

SLII places a person, on a given task, at one of four development levels: D1 (enthusiastic beginner: low competence, high commitment), D2 (disillusioned learner), D3 (capable but cautious) and D4 (self-reliant achiever). Each calls for a style: S1 directing, S2 coaching, S3 supporting, S4 delegating. The leader's skill is diagnosing correctly and changing style as the person develops.

Development level is task-specific

The most important detail is easily missed: the level is task-specific. The senior engineer is D4 on engineering and D1 on the security procedures, at the same time. They need S4 on one and S1 on the other. Treating them as "senior, therefore autonomous" on everything is what produced the surprise in the introduction.

Policy acknowledgement and competence

This also explains why publishing a policy changes so little. Asking people to read and sign a document is equivalent to giving a beginner a manual and expecting competent performance. It informs; it does not build competence. Competence comes from practice, shadowing and feedback: direction first, then coaching, then gradually less.

Awareness training as coaching

A yearly slide deck with a quiz is, at best, S1 delivered once. Memory research is clear about what follows: without reinforcement, retention decays quickly. Spaced, short follow-ups that explain why and show how behave like S2 coaching, and each one makes the next decay slower. Language-learning applications schedule their reviews on the same principle, returning to each word at increasing intervals. The model below illustrates the difference.

Structure and relationship

The Leadership Grid adds a second lens. When introducing standards it is tempting to sit in the "authority-compliance" corner: high on results, low on people. That gets compliance, but rarely ownership. Clear structure gives a leader credibility, and the relationship is what turns compliance into ownership. In practice: explain the reason for each control, and listen when a team says it does not fit their work.

Two common errors

The first is Wiseman's accidental diminisher: the security lead who fixes every finding personally because it is faster. It is faster in the short term, but the team learns to wait.

The second comes from my own field. A control is in place or it is not. Most leadership decisions are trade-offs. Bringing the binary mindset of compliance into people management is a category error.

The figures below illustrate two results: the same person can need opposite styles on two tasks, and a single yearly training session is forgotten, while spaced coaching keeps knowledge alive.

Suggested settings:

  1. Diagnosis. Switch between the two tasks. For the same person: D4 and S4 on engineering work, D1 and S1 on the security procedures.
  2. Development. On the security procedures, raise competence past the middle. With commitment still high, the diagnosis jumps to D4 and the style relaxes to delegating. Now lower commitment to about 0.6: the same competence reads as D3, capable but cautious, and the style becomes supporting.
  3. Declining commitment. Keep competence low and drop commitment below the middle. That is D2, the disillusioned learner, and the right style is coaching, not more directing.

The forgetting model is deliberately simple: retention halves every "half-life", and each spaced follow-up lengthens the half-life. Its parameters are illustrative, not measured.

  1. Annual session. With the defaults, the single session leaves an average retention over the year of about 4%. By the next annual session, almost nothing is left.
  2. Spaced coaching. The same first session with a short follow-up every four weeks keeps average retention near 90%.
  3. Less frequent follow-ups. Stretch the follow-ups to every 13 or 26 weeks. Retention falls, but even two or four reinforcements beat one big session. That is the spacing effect.
  4. Pessimistic parameters. Set the strengthening per follow-up to ×1.2. Spacing still helps, just less: the conclusion does not depend on optimistic parameters.

The forgetting model. In the model, retention after tt days is R(t)=2−t/hR(t) = 2^{-t/h} with half-life hh. With one session, the average over a year is 1365∫03652−t/h dt=h365ln⁡2(1−2−365/h)\frac{1}{365}\int_0^{365} 2^{-t/h}\,dt = \frac{h}{365\ln 2}(1 - 2^{-365/h}), which for h=10h = 10 days is about 4%. Each follow-up resets RR to 1 and multiplies hh by the strengthening factor, so with kk follow-ups the half-life grows geometrically, and the area under the curve grows with it. Any factor above 1 makes spacing better than one session; the size of the gain depends on the factor, which is what the empirical literature measures.

Situational leadership is not settled science; its prescriptions have only partial empirical support. As a diagnostic language it remains useful in security, because it raises the right question: competent and committed at what?

  • Diagnose development level per task. Seniority in engineering says nothing about familiarity with local procedures.
  • Treat a signed policy as information. It does not create competence.
  • Run awareness as spaced coaching instead of an annual event.
  • Explain the reason for each control, and consider the response.

References

  1. P. Hersey and K. H. Blanchard, "Life cycle theory of leadership", Training and Development Journal, 23(5), 26-34, 1969.
  2. K. Blanchard, P. Zigarmi and D. Zigarmi, Leadership and the One Minute Manager, William Morrow, 1985.
  3. P. G. Northouse, Leadership: Theory and Practice, 7th ed., SAGE Publications, 2016.
  4. G. Thompson and R. P. Vecchio, "Situational leadership theory: A test of three versions", The Leadership Quarterly, 20(5), 837-848, 2009.
  5. R. R. Blake and A. A. McCanse, Leadership Dilemmas, Grid Solutions, Gulf Publishing, 1991.
  6. L. Wiseman, Multipliers: How the Best Leaders Make Everyone Smarter, revised and updated edition, HarperBusiness, 2017.
  7. H. Ebbinghaus, Über das Gedächtnis, Duncker & Humblot, 1885.
  8. N. J. Cepeda, H. Pashler, E. Vul, J. T. Wixted and D. Rohrer, "Distributed practice in verbal recall tasks: A review and quantitative synthesis", Psychological Bulletin, 132(3), 354-380, 2006.