58 SITUATING HUMAN RESOURCE MANAGEMENT
(see, e.g. Boselie, Paauwe, and Jansen 2001) is that the institutional context is
largely ignored. A related concern is that the bulk of the research is American
and the pattern of results may be different in Europe and elsewhere. One rea-
son for this might be the legislative framework which requires organizations to
apply many of the HR practices about which firms in the USA and to a lesser
extent the UK and possibly also Australia have choice.
A final concern is that much of the evidence is cross sectional. There may
be an association but it is not possible on the basis of such evidence to assert
causality. The usual critique is that successful organizations may be more likely
to introduce HR practices. In one recently reported study, where longitudinal
data were available, there was even some evidence to support this direction
of causality (Guest et al. 2003). The complexity of the causal links has been
explored by Schneider et al. (2003) with data on aspects of job satisfaction
and performance in twenty-five large US firms over an eight-year period. They
found more evidence to support the view that successful organizational per-
formance leads to job satisfaction than vice versa. At the same time, they did
not find a consistent unidirectional causal link. Therefore, while firm success
seems to lead to more satisfied workers, it is also possible, at least to some
extent, that happy workers lead to firm success. Since there are continuing
doubts in the research on HRM and performance about conceptualizing and
operationalizing the independent variable, the dependent variable and the
relationship between them, there is considerable scope for error. Given all
this built-in error, it is perhaps surprising that the great majority of published
findings are so robustly positive.
While there are a number of academics who would argue that the growing
body of evidence does support a link, possibly even a causal link, between
HRM and various measures of performance, others are more cautious and
would argue that the case for a link between HRM and performance remains
unproven, due to the lack of the reliability and validity of the accumulated
studies. Behind the body of evidence, there are also questions about the temp-
tation to publish positive rather than negative findings. Certainly, articles tend
to emphasize significant positive results that explain a very small proportion of
the variance and to ignore negative or insignificant associations. This empha-
sis on positive results may give a misleading impression about the scale and
significance of such results. The concern is that some groups of policymakers
may be less cautious than many academics in interpreting and generalizing
from positive results.
There is a large body of ‘grey research’, undertaken by consultants, who typ-
ically highlight and market any positive links between HRM and performance.
Organizations such as the Chartered Institute of Personnel and Development
(CIPD) in the UK and their parallel organizations in other countries have an
understandable interest in identifying and promoting a link between HRM
and performance. The CIPD in particular has been a sponsor of research