In a study published in 2011, two researchers from the School of Nursing at Auburn University evaluated levels of "empathy" in a sample of N = 598 senior students in baccalaureate-level nursing programs in the southeastern United States.(McMillan LR, Shannon DM.Psychometric analysis of the JSPE Nursing Student Version R: Comparison of senior BSN students and medical students attitudes toward empathy in patient care.ISRN Nursing,2011: 726063.)

Empathy was defined as "a predominantly cognitive (rather than emotional) attribute that involves an understanding (rather than a feeling) of experiences, concerns, and perspectives of the patient, combined with a capacity to communicate this understanding".(Definition from Hojat M.Empathy in Patient Care.New York: Springer, 2007.)

Previously, Hojat had developed a questionnaire that could be used to measure empathy levels among physicians and medical students.McMillan and Shannon modified that questionnaire to apply to nursing students.The questionnaire posed 20 statements (for example, "Patients feel better when their nurses understand their feelings") and students responded to each statement by choosing one of 7 options (Strongly disagree, Disagree, Somewhat disagree, Neither agree or disagree, Somewhat agree, Agree, Strongly agree).Based on a nursing student's responses to the 20 statements, he or she would receive a "score" between 20 and 140 points;in theory, higher point valuespotentiallyreflect higher levels of empathy.

In his research, Hojat had found that the mean level of empathy among medical students was 115.In their sample of nursing students, McMillan and Shannon found a mean level of empathy of= 114.57 and a standard deviation of= 10.94.

McMillan and Shannon used these statistics to test the following statistical hypotheses:

Null Hypothesis: = 115versusAlternative Hypothesis: 115

To obtain a p-value for these hypotheses, they applied a One-Sample t-Test.The p-value they reported from the test was p = 0.22 = 22%.

In this discussion,please address the following questions:

(a)By testing these "statistical hypotheses", what was McMillan and Shannon's"research hypothesis"?

(b)Do the results from the One-Sample t-Test support McMillan and Shannon's researchhypothesis?Why or why not?

(BTW - If you apply the One-Sample Z-Test using the statistics from McMillan and Shannon's sample, you will find that the p-value they reported is incorrect.The p-value from the Z-Testshould be 0.337 = 33.7%; this implies that the p-value from the t-test should be slightly larger.It appears that McMillan and Shannon miscalculated their Z test statistic.Unfortunately, such mistakes are not uncommon, even in articles in "peer-reviewed" journals.)