January 10th, 2018
Critique of Inferential Statistics
Inferential statistics provide insights on whether to accept or reject the suggested relationships among variables as presented in a study’s hypotheses (Nieswiadomy, 2008, p. 259; Katz, et al. 1995). Halbesleben, Wakefield, Wakefield and Cooper (2008), present such statistics to identify whether the relationships suggested in their hypotheses are significantly indicated in their study results.
Firstly, the authors hypothesized that burnout is “associated with perceptions of a less safe environment for patients” (p. 564). Secondly, they suggested that burnout would be “associated with lower likelihood to report errors and near misses” (p. 565). Controlling for other variables (hours and tenure) that could affect the outcome, hence lead to erroneous findings, the authors presented appropriate analyses to assess the hypothesized relationships. For instance, from the regression results it is evident that there was an inverse relationship between burnout and patient safety grade (coefficient for exhaustion was minus 0.40 and that for depersonalization was minus 0.16), with these relationship being significant at 99% and 95% confidence interval respectively, thus supporting hypothesis one. Hypothesis two received mixed results. Even after controlling for demographics, neither of the measures was shown to be significantly associated with number of events reported at 95% (β coefficients of minus 0.02 for exhaustion and minus 0.01 for depersonalization, were less than (minus or plus) 0.05, the α, value denoting acceptable error at 95%), thus contrary to hypothesis two. Such β coefficients with respect to near-miss reporting frequency were however in significant negative relationships with measures of burnout (at α = 0.05 for exhaustion, and α = 0.01 for depersonalization) thus supporting hypothesis two.
Halbesleben, J. R. B., Wakefield, B. J., Wakefield, D. S. & Cooper, L. B. (2008). Nurse burnout and patient safety outcomes: nurse safety perception versus reporting behavior. Western Journal of Nursing Research, 30(5), 560-577. doi: 10.1177/0193945907311322.
Katz, R.T., Campagnolo, D.I., Goldberg, G., Parker, J.C., Pine, Z.M., & Whyte, J. (1995). Critical evaluation of clinical research. Archives of Physical Medicine and Rehabilitation. 76, 82-93.
Nieswiadomy, R. M. (2008). Foundations of Nursing Research (5th Ed). Upper Saddle River, New Jersey: Pearson Education, Inc.