19
Moreover, the decrease in the percentages of items containing dramatic pictures and the
increase in the average shot length even significantly contradicted the expectation
concerning the NOS newscasts. For these reasons, we might provisionally conclude, first,
that in some aspects television newscasts have become more sensational during ‘The
Battle of the Newscasts’, and second, that the latest player (SBS) has played the biggest
role in this process.
The ambiguity of our results may be caused by the many uncontrolled theoretical
factors possibly influencing the form and content of the news. For instance, NOS is
somewhat less dependent on advertising revenues because it is partly financed by
compulsory user fees. This might have contributed to a lesser inclination at the NOS
news desk to compete for viewers and to accommodate its newscasts, which could
explain the relative small support for Hypothesis 2, and consequently for Hypothesis 1.
The ambiguity of our results also may be related to the measurement of
‘sensationalism’. We constructed 16 indicators, which on face validity were supposed to
represent one single concept: sensationalism. Reflecting on the results, one can observe
that for some indicators the results were much in line with the expectations. Notably, the
results for ‘personalization’ and ‘close-ups’ were significantly in line with all five
hypotheses. Besides, the results for ‘eyewitness camera’, ‘camera perspective’ and
‘zoom-in’ were also in line with all hypothesis, in three instances significantly. These
techniques have in common that they intend to make information more proximate in a
spatial way, or in the case of subjective ‘camera perspective’, in a sensory way. In this
respect, they seem to relate more directly to the ‘vividness’ concept than most other
indicators, for instance ‘inserted music’ or ‘shot length’.
At present, the measurement of the sensationalism concept, even if defined in terms
of ‘vividness’ and ‘salience’, is largely dependent on face validity. Future studies on
differences in sensationalism (e.g. Grabe et al., 2001) trends in sensationalized news (e.g.
this paper), as well as studies on the effects of sensationalized news (e.g. Grabe et al.,
2000, Zillmann & Brosius, 2000) might provide an adequate measurement of this
concept.
Convention