Exercise 23 Selecting Appropriate
Analysis Techniques for Studies
1. A researcher surveyed two groups of professionals, nurse practitioners
and physicians, and asked them whether or not they supported expanding the role of
nurse practitioners' (NPs) prescribing privileges, answered as either “yes” or “no.” Her
research question is: Is there a difference between NPs and physicians on proportions of
support for expanded prescription privileges? What is the appropriate statistic to
address the research question?
Inferential statistics are conducted to examine relationships, make predictions, and
determine differences among groups in studies. One of the most common statistical tests
chosen to investigate differences between two independent samples is the independent
samples t -test, which is a parametric inferential statistic. The independent sample t-test only
compares two groups at a time, and the independent variable must be continuous and
normally distributed.
2. What statistic would be appropriate for an associational research
question involving the correlation between two non-normally distributed, skewed
continuous variables?
Spearman Rank –Order correlation coefficient is preferred when one or both of the
variables of data significantly deviate from a normal distribution, or when the variables are
ordinal and cannot be treated as an interval/ratio skilled variable.
, 3. A researcher is interested in the extent to which years of practice among
NPs predicts level of support for expanded prescription privileges, measured on a 10-
point Likert scale. She finds that both variables, years of practice and level of support,
are normally distributed. Her research question is: Does years of practice among NPs
predict level of support for expanded prescription privileges? What is the appropriate
statistic to address the research question?
A paired samples t-test. It is a statistical procedure that compares two sets of data from
one group of people. This test is a parametric inferential statistical test. The dependent
variable in a paired samples t-test must be continuous in normally distributed.
4. What statistic would be appropriate for a difference research question
involving the comparison of two independent groups on a normally distributed
continuous dependent variable?
Independent samples t-test. This test compares two groups at a time, and the
independent variable must be continuous and normally distributed.
5. A statistics professor tests her students' level of knowledge at the
beginning of the semester and administers the same test at the end of the semester. She
compares the two sets of continuous scores. Her research question is: Is there a
difference in statistics knowledge from the beginning to the end of the semester? What
is the appropriate statistic to address the research question, if the scores are normally
distributed?
Analysis Techniques for Studies
1. A researcher surveyed two groups of professionals, nurse practitioners
and physicians, and asked them whether or not they supported expanding the role of
nurse practitioners' (NPs) prescribing privileges, answered as either “yes” or “no.” Her
research question is: Is there a difference between NPs and physicians on proportions of
support for expanded prescription privileges? What is the appropriate statistic to
address the research question?
Inferential statistics are conducted to examine relationships, make predictions, and
determine differences among groups in studies. One of the most common statistical tests
chosen to investigate differences between two independent samples is the independent
samples t -test, which is a parametric inferential statistic. The independent sample t-test only
compares two groups at a time, and the independent variable must be continuous and
normally distributed.
2. What statistic would be appropriate for an associational research
question involving the correlation between two non-normally distributed, skewed
continuous variables?
Spearman Rank –Order correlation coefficient is preferred when one or both of the
variables of data significantly deviate from a normal distribution, or when the variables are
ordinal and cannot be treated as an interval/ratio skilled variable.
, 3. A researcher is interested in the extent to which years of practice among
NPs predicts level of support for expanded prescription privileges, measured on a 10-
point Likert scale. She finds that both variables, years of practice and level of support,
are normally distributed. Her research question is: Does years of practice among NPs
predict level of support for expanded prescription privileges? What is the appropriate
statistic to address the research question?
A paired samples t-test. It is a statistical procedure that compares two sets of data from
one group of people. This test is a parametric inferential statistical test. The dependent
variable in a paired samples t-test must be continuous in normally distributed.
4. What statistic would be appropriate for a difference research question
involving the comparison of two independent groups on a normally distributed
continuous dependent variable?
Independent samples t-test. This test compares two groups at a time, and the
independent variable must be continuous and normally distributed.
5. A statistics professor tests her students' level of knowledge at the
beginning of the semester and administers the same test at the end of the semester. She
compares the two sets of continuous scores. Her research question is: Is there a
difference in statistics knowledge from the beginning to the end of the semester? What
is the appropriate statistic to address the research question, if the scores are normally
distributed?
