While these studies provide valuable insight, they aggregated
CSR activities as one measure, and this practice could be
problematic because various components, different in nature,
may exist in the aggregated CSR measure. According to the
positivity and negativity effects, positive and negative activities
may have different magnitudes of impacts on outcomes. Studies on
impression formation and information integration demonstrate
that an impression does not reflect the entirety of different
attributes to the same extent, but tend to be disproportionately
influenced by negative characteristics of the target (Martijn et al.,
1992). This tendency of over-emphasizing negative information is
often referred to as the negativity effect (for example, Wojciszke
et al., 1993). On the other hand, Taylor and Koivumaki (1976)
examined how people perceive themselves versus others, and
found that people were seen as causing positive behaviors, and
situational factors were regarded as causing negative behaviors.
This tendency is called the positivity effect, and it was found to
operate most strongly for perceptions of intimate ‘‘others,’’ and less
strongly for strangers and liked and disliked acquaintances.