The Psychological Assessment of Political Leadersof the three motive scoring systems, together with a wealth of other
information, are available in Motivation and Personality: Handbook of
Thematic Content Analysis (Smith 1992).
Before scoring for research purposes, scorers should have attained
a high degree of agreement with expert scoring on the practice mate-
rials. The running text manual provides step-by-step instructions for
doing this (see also Smith 1992, 526-632). The usual standard of
interscorer reliability for research purposes is category agreement
(see Smith 1992, 529) and r both greater than .85.
Psychometric and Interpretation IssuesMotives of Leaders or Speech Writers?
When the speeches and other verbal material of political leaders are
scored for motive imagery, do the resulting scores reflect the motives
of the leaders themselves or the motives of their speech writers? And
if speech writers are trying to emphasize salient cultural values in
order to create the broadest popular appeal, do the scores really
reflect anybody's personality? On the other hand, candidates do select
their speech writers. Speech writers, in turn, know how to craft
words, phrases, and images to fit the style and personalities of their
clients. (See Crown 1968, 34-38, on Kennedy and Sorenson; Safire
1975 on writing for Nixon versus writing for Agnew; Noonan 1990
on Reagan; Scott 1993 on Clinton; and Gelderman 1997 on presi-
dential speech writers generally.) Finally, for any important speech
such as an announcement of candidacy or an inaugural, the candidate
and other close associates review and rework successive drafts until
they feel appropriate and comfortable. In 1960, for example,
Kennedy personally wrote out a late draft of his inaugural and
inserted some scorable motive images into Theodore Sorenson's
penultimate typed draft.
Using more spontaneous interview material may reduce (though
it does not fully eliminate) this problem, since many "spontaneous"
interviews are carefully prepared, even scripted. Yet in a larger sense,
concerns about the effect of the speech writer may not matter. What-
ever their status, prepared texts do exist as the leader's words; they
are taken as the leader's words; and they have effects as the leader's
words. We can assume that scores based on these words are a reason-