two presentations are exactly alike, and what works well in
one instance may be unsuitable in another. Nevertheless, some
general points can be made with respect to what constitutes
a good seminar. You should consider the following in preparing
- The most
common mistake made by a speaker is to assume that the audience
has as detailed an understanding of the subject as does the
speaker. This is only very rarely the case. It is best to
devote at least 5 minutes to fairly general background, explaining
in particular what general factors motivate the more specific
research to be discussed thereafter. What is the big picture?
Why should anyone care? What specific questions are to be
addressed by the speaker's research?
- The second most common mistake is
to assume that it is better to give a cursory overview of
many results rather than a cogent presentation of a subset
of all that the speaker has accomplished. The motivation
tends to be that speakers want to look as though they've
accomplished a great deal, but the net result tends to be
that no one in the audience has a clue what actually got
done because insufficient time is devoted to any particular
topic so as to make it comprehensible. CLARITY IS ALWAYS
TO BE PREFERRED OVER QUANTITY.
- To assist in fostering clarity,
the density of information on a single slide should be such
that it takes about 2 minutes to discuss it. This is obviously
a ROUGH guideline, since there are occasionally good reasons
to violate this rule in either direction. However, if you
need MORE than 2 minutes for a slide, chances are it is too
dense in information, and the audience will be confused trying
to follow your discussion. Consider breaking it into two
or more slides or into a slide that "grows" (e.g.,
in a PowerPoint presentation, new portions could be flown
in after a stripped-down version is initially discussed;
in an overhead presentation, additional transparencies can
be laid over the first accomplishing the same thing). Note
that for a 20 minute talk, this suggests 12 to 13 slides/foils/images
is about the right number.
- In choosing topics for slides, consider
doing your conclusions FIRST. Now, evaluate each possible
slide based on whether it provides evidence that permits
you to arrive at your conclusions. If no, discard it (even
if it took you 6 months of work...). Short talks need maximum
- Text in a font size smaller than
16 is illegible to most of the audience. Look at every one
of your slides from the back of a lecture hall and THEN decide
whether you think they are acceptable.
- Garish graphics detract from a talk
rather than enhancing it. Avoid the impulse to animate, colorize,
provide wild backgrounds, UNLESS IT CLEARLY ENHANCES THE
POINT YOU ARE TRYING TO MAKE. Color can be very effective
in highlighting small portions of an otherwise large graphic
where the discussion will be focused. Animation can be useful
in making 3-dimensional structures more comprehensible. Just
remember that the science is the message.
- Practice your talk in front of an
audience at least once or twice, and do that only after having
practiced alone to the point where you know what you WANT
to say as each individual slide comes up. Practice speaking
slowly if you have a tendency to rush.
- Point at things as you discuss them.
Speakers often assume that the audience must be looking at
exactly the same part of the slide as they are, but the audience
has not had the benefit of several prior runs.
- Project confidence! Chances are,
you know more about the subject you are presenting than anyone
else in the audience. Open the audience's eyes. It's OK to
be nervous, but focus on a measured pace and a clear voice
to overcome any deleterious effects of nerves.