LIMITATIONS
Like all research, some aspects of the present study limit its findings. Confidential interviews preclude
checks of the veracity of the experiences reported and using mainly self-selecting respondents means
the sample is not representative of the general population. The use of self-selection is likely to have
affected the nature of the responses, as bullying tends to only be visible to those who have experienced
bullying; however, this approach may be appropriate when studying groups that have the most
severe experience of a phenomenon (Nielsen & Einarsen, 2008). Furthermore, unsupported targets
may use studies such as this one as an opportunity to air their concerns in the absence of other outlets
(Lutgen-Sandvik, 2006), which may explain the rich vein of experience that the findings reveal.
Equally, former targets who were happy with the outcome of their complaint may have felt it
unnecessary to participate, which may have caused a bias in the responses. Some interviewees referred
to their experiences with union representatives who were also colleagues, but no-one from the union
representative group volunteered to participate, potentially limiting the depth of the findings in this
area. The narrow industry focus and relatively homogeneous sample may limit generalisations and thus
a broader cross-section of workers might well have produced different outcomes.
CONCLUSION