In this case we are want to determine
if there are differences in how the leaders and nurse professionals
view EPB changes in their unit workflow and practice setting. The
three questions below will guide the selection of appropriate statistical
methods.
-
What is the response
(Y) of interest and how is it measured? Two of the survey’s
rating questions are being analyzed in this case. There are five
response options for each of these questions expressing the degree
of agreement with each proposition. This is referred to as a 5-point
Likert scale and is an ordinal scale of measure. Since this is an
ordinal scale, we are interested in the frequency (or proportion)
of response in each of the five agreement categories.
-
Are predictor variables
mentioned in the problem statement? If so, how many and what are
their measurement levels? The problem statement concerns differences
in the perception of EBP by organization level. In this case we are
interested in comparing the opinions of leaders vs. nurse professionals.
This means there is one predictor variable (X), the organization
level.
-
What are you being asked
to deliver? A data description, an interval estimate, an answer to
a question, or a predictive model? We are being asked to answer a
question. A test of hypothesis is an appropriate statistical method
to answer a question based on a sample of data. For this case we
will make two bivariate comparisons, where the response variable (Y)
is the rating and the predictor variable (X) is the organization level.
The appropriate statistical method is a chi-square test for independence.