A Computer-Aided Content Analysis of Charismatic Leadership

Case Analysis - Content Analysis

A computer-aided content analysis of charismatic leadership

Bligh, Kohles, and Meindl (2004) examined the rhetorical content of President George W. Bush’s public speeches before and after the terrorist attacks of September 11^th to determine whether the crisis affected his communications with the American people or the media portrayal of the President. Specifically they wanted to know whether the crisis had generated a shift towards charismatic leadership.  Their sample comprised 74 major speeches and radio addresses given by the President, obtained from the official White House website. A computerised content analysis software programme (DICTION 5.0) was used.  This software is specifically designed for analysis of political discourse and so comprises dictionaries containing words commonly found in American political discourse.  The software automatically divided each speech into 500-word passages for comparison, resulting in a final sample of 117 passages ranging from 100-500 words (58 precrisis and 59 postcrisis). Eight composite variables relating to charismatic leadership were generated by the researchers, including:

-          Similarity to followers: this comprised the variables ‘levelling, familiarity and human interest’ and included words that reflect a speakers desire to speak on a common level’ (Bligh, Kohles, and Meindl, 2004: 218);

-          Values and moral justifications: using dictionaries to identify religious terms, the focus was on speech that draws on values and morality to incite followers to support the leader’s future vision;

-          Adversity: this focused on references to discontent and hardship, and the use of speech to reflect inappropriateness and evil human behaviour.

 Analysis of the data indicated that the President’s speech did become more charismatic after the events of 9/11.  The first study was then followed by a second that sought to determine whether this affected media responses to the President’s rhetoric. Bligh et al. therefore analysed news articles and television news transcripts relating to the President’s speech from January 2001 to March 2002.  The data were obtained using a broad keyword search for ‘Bush and speech’ on the Lexis-Nexus database. The final sample comprised 442 articles and transcripts. They then followed a similar analytical procedure involving the computer software program as in the first study.  Overall, findings confirmed those of the first study.  A third and final study was then conducted based on analysis of Presidential approval ratings to ascertain the precise nature of charismatic leadership in the wake of the crisis.

 While this case is associated with the political realm, it draws on concepts and approaches that are relevant to and have been developed in connection with business organizations. It is worth noting that the availability of speeches over a long period of time allowed a longitudinal ‘before and after’ design to be employed.

 Questions for consideration

Each question should be answered in paragraph form and addressed in your case analysis:

1. To what extent (if at all) does the use of computerized content analysis make the data less prone to subjective interpretation than manual coding of media?
2. Does computerized content analysis eliminate the need for checking inter-rater reliability?
3. Can we infer that 9/11 caused President Bush’s speeches to become more charismatic in tone?
4. To what extent do you agree with the search terms used in the first of the follow-up studies (‘Bush and speech’)?