

99
臺大管理論叢
第
28
卷第
1
期
(2)
Thus, the ARs and CARs are estimated as follows:
(3)
(4)
where
AR
i,t
are the abnormal returns of the
i
th
firm on day
t
; and
CAR
i,[Tn, Tm]
are the
cumulative abnormal returns of the
i
th
firm around the CSR announcement date period [
T
n
,
T
m
].
3.3 News and Media Measures
Media coverage (
MEDIA
) is defined in this study as the number of days that a
particular firm is mentioned within the media by referring to Vega (2006).
MEDIA
, which
is used in this study as a proxy for ‘media reputation’, represents the extent to which the
firm is included in media reports.
(5)
where
NEW
i,t,n
is a dummy variable which is equal to 1 if firm
i
is mentioned in the
headline or leading paragraph of a news article on day
t
; N is the total number of articles
in which firm
i
is mentioned in the headline or leading paragraph;
MEDIA
i,t
refers to the
media exposure of firm
i
on day
t
(for example,
MEDIA
is equal to 3 if there are 3 news
on firm
i
that the name of the firm is mentioned in the headline or leading paragraph; if
there is no news on firm
i
, then
MEDIA
is equal to 0). We also provide the measurement of
average media around the CSR announcement date.
(6)
(7)