A Statistically Robust Way to ‘Measure’ Military Morale!
There are relatively few reliable primary or secondary sources that assess levels of morale in armies. As I discussed in a previous post, this makes it extremely difficult for historians to make connections between battle outcomes and that most nebulous of military factors – morale. Considering the great and the good, from Machiavelli to Clausewitz to Foch all argue that morale is, to quote Montgomery, ‘the big thing in war’, this does appear to create a significant problem for historians of war and international relations (IR) specialists alike.
Those scholars who do try to scientifically measure morale, such as Morris Janowitz and Edward Shils, S.L.A. Marshall and Samuel A. Stouffer et. al., in the context of the Second World War, and Leonard Wong et. al., in relation to the 2003 Iraq War, make extensive use of contemporaneously recorded attitudinal surveys. These sources provide historians with valuable information for military organisations at particular places and periods of time. However, their findings are often highly contextualised; their relevance, therefore, to other combatant nations, wars and environments must be understood in that light.
In the majority of cases where attitudinal surveys have not taken place, scholars attempting to assess morale are typically dependent on sources such as personal recollections and memoirs. These too, however, suffer from serious methodological shortcomings, not least the fallibility of individuals’ memories, especially where interviews take place decades after the event, and the impact of prevailing cultural and social interpretations of the meaning of events on the recollections of historical actors.
Contemporaneously recorded diaries or letters are more reliable as historical sources, but it is often difficult to amass a representative sample of such sources for an army. Recent studies, such as Alexander Watson’s Enduring the Great War, this author’s Combat and Morale in the North African Campaign and Jonathan Boff’s Winning and Losing on the Western Front have tried to circumvent this problem by making use of newly discovered or underused official sources, such as censorship summaries of soldiers’ mail and morale reports. Such sources cover morale widely and deeply and tend to express views that represent a considerable body of opinion among troops. However, again, these sources are available in only a minority of cases, and, thus, may be limited as a tool to assess morale in armies across history.
There may, however, be an answer to this methodological problem! In a paper in Anthony King’s 2015 edited volume, Frontline: Combat and Cohesion in the Twenty-First Century I set out to test whether there is a statistically robust way to ‘measure’ morale (an earlier version of this paper can be accessed here).
In addition to the use of attitudinal surveys, combatant letters and diaries, censorship summaries and morale reports, military professionals (particularly commanders and medical officers), theorists and scholars have also used rates of sickness, battle exhaustion, desertion, absence without leave (AWOL) and self-inflicted wounds (SIW) in units as a useful proxy for morale. Militaries tend to record and keep this kind of data, BUT, apart from the testimony of medical officers, and other military professionals, we have little hard evidence that these metrics provide a reliable indication of levels of morale in an army.
To test whether there is a statistically robust relationship between these metrics and morale, I applied this methodology to the case of British Second Army in the Northwest Europe Campaign of the Second World War. Thirty-six bi-weekly censorship summaries, covering the period between 4 April 1944 and 15 October 1945, compiled from 1,494,479 letters sent by officers, NCOs and other ranks in the British Army, were considered (as far as this author is aware, these sources have never before been used in a historical study). These summaries proved a useful source to describe and ‘quantify’ levels of morale (through the use of a numerical morale scale). Where morale was described as ‘excellent’, it was awarded a score of 3. ‘High’ morale was given a score of 2 and ‘good’ morale was scored 1. ‘Satisfactory’ morale was given a score of 0 (neither positive or negative). Morale described as ‘severely tried’ was scored -1, while ‘low’ and ‘very low’ morale were scored -2 and -3 respectively.
Statistics relating to weekly levels of sickness, battle exhaustion and SIW were drawn from the War Diaries of the Deputy Director of Medical Services, Second Army. Convictions by Courts Martial for desertion and AWOL were recorded in the ‘Administrative History of 21st Army Group’ and in the War Diaries of the Deputy Judge Advocate General, 21st Army Group.
In theory, if the tabulated rates of sickness, battle exhaustion, SIW, desertion and AWOL were indeed indicators of morale, they should, broadly speaking, fluctuate along the lines of the shifts in morale levels as encapsulated in the morale scale, produced by use of the censorship summaries.
The results of the correlation analysis showed that the tabulated rates (the combined rate of sickness, battle exhaustion, desertion, AWOL and SIW) had an extremely strong negative correlation with morale (-0.949, P<0.001), i.e. when morale was high, sickness rates, etc. were low, and when morale was low, sickness rates, etc. were high. This is a remarkably strong relationship and shows that these factors when taken together can be used as a quantitative method to assess levels of morale, at the very least for the Army and campaign under discussion. Coincidently, it also offers a quantitative justification of the approach I used in my book Combat and Morale in the North African Campaign.
These findings, it is hoped, will encourage historians and IR specialists to revisit morale and seek to place measurable fluctuations in morale more seriously into explanations of victory and defeat on the battlefield.
This post first appeared at Defence-in-Depth, the blog of the Defence Studies Department, King’s College London.